- Inkscape Devel - Inkscape Mailman

zoom and precision
by bulia byak 12 Dec '03

12 Dec '03

I've augmented my libnr precision and 256x zoom patch with a fix to store SVG paths with 8 digits instead of 6 (changing %g to %.8g in printfs in svg/svg-path.c). Since I've been using this patch for some time and have not seen any problems, so I decided to commit it now. This is code that is in use literally every second, so by putting it in early in the release cycle we'll be able to catch more bugs with it. Please work and play at higher resolutions than you usually do and submit any bugs to the tracker. Note that you'll need to rerun configure after updating because I patched libnr/gen_nr_config.c which is recompiled and rerun on configure (I think). _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Open Source myths
by Jon A.Cruz 12 Dec '03

12 Dec '03

Here's a nice little article on software development http://www.onlamp.com/pub/a/onlamp/2003/12/11/myths.html Of all the issues raised, the ones I've thought about I'm in agreement with. The few I hadn't consciously considered before seem to be right on track also. In all, it seems that these are things that those in the Inkscape community are doing, however it's always good to explicitly state what's good and why. Among other things it might help us improve and also just keep us on track. (so at the very least, read the article and keep things in mind)

1 0

C++-ification
by Bryce Harrington 12 Dec '03

12 Dec '03

Just a quick head's up, we're about to proceed with the renaming of all .c files to .cpp. Nathan is working on updating Makefiles &tc. presently and if it works out, he'll be committing the changes soon. The expected near-term effects to be aware of: * Change histories on code files will be reset * Potential build issues on some platforms * Everyone will need to cvs update -Pd, make distclean, and rebuild * Compilation with gcc will cease to work as support for gcc is ended * If you are working on a branch or out of sync with CVS HEAD, your next merge may be more complicated than usual due to the file rename These issues should be transitory, but please expect some turbulence for a bit. Thanks, Bryce

2 2

Re: [Sodipodi-list] Re: 0.8 multiplier
by MenTaLguY 12 Dec '03

12 Dec '03

On Thu, 2003-12-11 at 13:06, Lauris Kaplinski wrote: > SVG spec is resolution agnostic. Yes, but I finally read this part of the SVG spec... > All its units are just > defined as certain ratios of undefined base unit (pixel), that > may or may not be related to some real-world unit. For the initial coordinate system, the SVG specification refers to CSS2's definition of length units, which broadly specifies two classes of units: 1. relative units which do NOT have fixed relationships with real-world units * em and ex: determined by the current font size * px: determined by the user agent, generally some multiple of device pixels 2. absolute units which _are real-world units_ (subject to the user-agent's ability and desire to match them) * mm: real-world mm * cm: 10 mm * in: 2.54 cm * pt: 1/72 in * pc: 12 pt > Sodipodi document has real-world dimensions. > To map resolution agnostic SVG onto real-world paper, it uses > mapping: > 1 SVG point == 1 typographic point (To be totally pedantic, typographic points are 0.0138 in, which is not _quite_ the same as Postscript/SVG points [1/72 = 0.01388888...]; I think everyone uses the Postscript definition nowadays though.) > As 1SVG point == 1.25 SVG pixels, and pixels are SVG base units, > SVG base unit == 0.8 typographic points on all sodipodi documents. Unless a viewBox is set, the relationship between user-space base units (pixels) and the absolute units is totally user-agent dependent. (this is why all examples in the SVG spec specify a viewBox) Thinking about this, if Sodipodi intends to be completely interoperable, it needs to always write a default viewBox according to the base units/points mapping it imposes. e.g. for a document 100ptx100pt, the viewBox should be "0 0 125 125". -mental

1 0

menu reorg
by bulia byak 12 Dec '03

12 Dec '03

I propose to break Edit menu into two, Edit and Object. Otherwise it's pretty long already, and will be still longer when we add boolean ops. I think everyrhing up to and including Clear All/Select All should remain in Edit, and the remainder of items should go to Object. If there are no objections, I will commit (it's a cosmetic patch so it's unlikely to break anything). _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

0.37 plan
by Bryce Harrington 12 Dec '03

12 Dec '03

It's very cool to have 0.36 out the door after so much work! It's being well received by the community, and the feedback is that we've definitely made notable improvement over the baseline. So what's next? First, as we talked about, we're going to keep things low key the rest of the month, and enjoy the holidays. This next release is going to be a bit different than this past one. For 0.36 our goal was to strive to get some new features in and revamp the look and feel, to prove that this project brings new value to the community and to gain experience working as a project. For the 0.37 release our focus is going to change. We want to make the codebase itself better, so that it is easier and more enjoyable for us to work on in the future. Doing this will require undertaking several sizeable endeavors: * Convert the codebase to C++ * Change from function prefixing to true C++ namespaces * Adopt a new code style * Remove all obsolete code * Organize all of our documentation The above work will touch pretty much all the files in the codebase, and since that could be a significant amount of work, I think we should use a divide-and-conquer approach to it. A way to achieve this is to divide the codebase into several distinct "subsystems"; this way the size of the commitment each of us must make will be tractible, and so that we can work together as a team to achieve them. I'll establish a list of these subsystems shortly; there is a diagram on the Documentation page of the website to give some hints as to what this list will be. In addition, there are several features we'd like to pursue, such as reorganizing the dialogs, Fred's new booleans work, and Nathan's grid code. There are also a number of bugs we decided to leave to this release, that should receive attention. We can look forward to seeing these features and bug fixes folded in, as we proceed with the principle objective of the code conversion and cleanup. Bryce

1 0

Plugins - first cut at requirements list
by Bryce Harrington 12 Dec '03

12 Dec '03

I've posted my first stab at a list of requirements for a plugin/extension/module system here: http://www.inkscape.org/cgi-bin/wiki.pl?PlugIns Ted's done a good bit of work on implementing support for a module system, so hopefully this list can be made to mesh with that work. The intent of making a requirements list is to firmly identify what we want out of this system, so that as folks work on implementing it, it is clear what it needs to do. Please look it over and add your thoughts of what should be in there. Bryce

1 0

new booleans patch
by bulia byak 12 Dec '03

12 Dec '03

I've been sent a new patch that implements boolean operations, offset/inset path, and path outline. The license is public domain. It's been uploaded to the tracker. I've just compiled it and it seems to work as advertised, although I did not test it much yet. I will now incorporate it properly (creating commands for all functions) and test as much as I can. If there are no problems, I think we must get this into CVS right now, before the release. Rationale: 1) it's a killer feature, 2) it's orthogonal to everything else, so even if something is broken in the new functions, this will not affect the rest of Inkscape. Everyone is welcome to download and test the patch too (or at least review the code). Big thanks to the author! _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

3 3

status bar discussion
by bulia byak 11 Dec '03

11 Dec '03

Please review and contribute: http://inkscape.org/cgi-bin/wiki.pl?StatusBar _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] New guide prototype
by bulia byak 11 Dec '03

11 Dec '03

>G'Day! >Peter and I have been working on a new grids and guides idea which extends >the control handle idea of Dia and Visio 'correctly' (We can justify this >claim quite strongly). There is a test bed version of it at: Thanks, I really enjoyed it! The ability to move a guide so that all objects aligned to it would move too was something I always dreamed of. But I'm not sure I fully understand the difference between the two tools. Is it correct that the first tool behaves exactly as the current selector, i.e. scales and moves regardless of snaps (but keeps the snaps where breaking them is not required for moving), while the second one attempts to scale or move preserving snaps if possible? Note that currently in selector, the edge of the object is not the same as the resize handle, as the handle is always at some distance from the edge. When you implement the "snap marks" you will need to attach them to the object's bounding box, not the handles. And the center of an object can now be moved to any point, so snapping the center onto something must take into account (this is particularly important with text objets, where I often need to align e.g. the baseline of a letter, which is neither the center nor the edge). _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

Re: [Inkscape-devel] color selector changes
by bulia byak 11 Dec '03

11 Dec '03

>>- The GTK+ selector is shakey: when you move the small circle inside the >>triangle, the triangle shakes and jitters. I think this is because the >>color itself produces HSV, but its position is determined by the RGB, and >>the conversion/rounding errors when the color goes to RGB and back make it >>unsteady. Can you ensure that the color wheen is connected to the HSV >>value? Also, when you move the circle to the black or white corners, the >>triangle snaps to the red, i.e. loses its current hue value. I think this >>will also be fixed if you make it reflect the HSV value, not RGB. >> > >Guess what? Don't bug me about this one. ;-) > >Actually... it is just the standard GTK+ color selector. My code only >creates one and then listens for changes. All those details are handled by >the GTK+ version you're running. > >Although it could be a bit of feed-back loop. I'll check to make sure >that's minimized. Yes, that seems to be a feedback problem. I commented out the last gtk_signal_emit (GTK_OBJECT (csel), cs_signals[CHANGED]); in sp_color_gtkselector_gtk_changed(), and jittering is gone, although the object color is not updated. So you just need to make sure this signal you emit is caught by the right listeners (i.e. the canvas objects) and not the color widget itself. Maybe encosing the emit into a pair of block/unblock functions might help. >Actually "color palettes" were something I was going to add. For the GTK+ >wheel, I just need to feed an array of colors for the palettes. Of course we need a separate color palette widget, ideally dockable to the document window, but for now a small palette inside the picker would be nice too. >So we might need "color palettes", "color schemes" and "swatches". I think >details of this would be best hashed out on jabber and on the Wiki. Yes, and a current color widget too (see wiki). >True... but it's built into the GTK+ widget. It does not have alpha (the >one difference). Of course, a simple work-around is to just hide the bottom >entry area when the color picker has one. Would that work better or be >distracting? I don't like that option too. Would be ideal to find a way to tweak the standard GTK widget. >I've split the second notion off into the notebook with tabs. Thus the >"Mode" switch should change to only determine the colorspace to use. I was >planing on "RGB", "CMYK" and "Any", with "Any" being the default. Later we >might also add "Spot". Then it must be named more descriptively, e.g. "Store color as:", but then "Any" sounds kind of funny. >But... I was just thinking the other day that another approach might help. >If we did named "color picker sets" then the user could change the set >named "full" and the set named "brief". They could also create new sets and >use them per location, or just stay with the default sets. I was thinking >that this has the benefit of being simple for beginning/normal users, yet >allow full customization for power users. And the coding would not be too >different than for a two-state enum'ed or boolean'ed set. Maybe, although I don't feel a need for this. Maybe just me. I think the GTK color wheel can be the default picker most people use, and the others are just "for completeness" and for special purposes. The wheel makes it possible to enter RGB values anyway, and it's not that you'll want to move RGB sliders very often, because HSV is much more natural to an artist. >In either case, hooking the color selections up to the new overall >preferences setting dialog would be handy. Yes, that's where it belongs IMHO, without cluttering the actual picker interface. Please add that to the PreferencesDialog wiki page. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Re: inkscape helping
by Jonathan Phillips 11 Dec '03

11 Dec '03

Although I'm new to the project and everything (congrats to ourselves on the new release), it is important for me to tell everyone I'm involved with about some personal things. My sister, Charis Gohil, passed away today at 3 AM CST. It was an amazing peaceful moment here as it was the first snow, and we (family and my sister) had our last words before the end of her time on earth. I started a journal about my sister, but it is not really that relevant anymore, but has pictures of her and myself. http://protofunk.org/?page=bio/charis/index I loved my sister very much and am happy that she is in a better place :) jon On Wed, 2003-12-10 at 21:46, Bryce Harrington wrote: > On Wed, 10 Dec 2003, Jonathan Phillips wrote: > > > stuff... > > > > Bryce, I'd be on top of all these changes but my sister's state has > > worsened in the last two days. She is probably going to pass away in the > > next day or two. I will get back as soon as possible. Sorry I can't help > > get this release out the gate. I'll be on board and running at 200% > > soon. I'm going to be offline here for a bit. > > > > This is not all doom and gloom here, it is actually been one of the best > > times with my sister ever...fyi... > > Thanks for the note; don't worry about Inkscape, family comes first. > We've got things well in hand. > > Btw, what's your sister's name? > > Bryce -- Jonathan Phillips <jon@...15...>

2 1

RE: [Inkscape-devel] color selector changes
by bulia byak 11 Dec '03

11 Dec '03

>Well... I just checked in some changes for the color selector. It needs a >little more polish (mode, CMYK types, and the popup to select subsets of >possible pickers) but should be ready for some initial testing and use. Thanks! My initial impressions: - Do the slider strips have to be so wide (i.e. tall)? I'd like to see them take less vertical space. - Before, the slider strips expanded horizontally when you resized the fill&stroke dialog. This was nice as it allowed to get a better resolution at the expense of screen space. Now they stay the same length no matter how you resize the window. - RGB, CMYK, and HSV are all familiar abbreviations to a designer. But "GTK+" is confusing. Is this a new color model? Better call it by what it shows, e.g. "Wheel". - The sliders are always black and therefore are invisible on dark areas of the strip (this was the problem with the old picker too). They need to be made either e.g. black interior with white outline, or some pseudo 3D, or (as in Gimp) pairs of black triangle at top and white at bottom, or automatically inverting when going from light to dark and back (as is the pointer on the GTK+ wheel). - The GTK+ selector is shakey: when you move the small circle inside the triangle, the triangle shakes and jitters. I think this is because the color itself produces HSV, but its position is determined by the RGB, and the conversion/rounding errors when the color goes to RGB and back make it unsteady. Can you ensure that the color wheen is connected to the HSV value? Also, when you move the circle to the black or white corners, the triangle snaps to the red, i.e. loses its current hue value. I think this will also be fixed if you make it reflect the HSV value, not RGB. - BTW when you resize the window, it's the color wheen in the GTK+ tab that must be made larger so you can pick colors with better precision. - When the fill&stroke dialog has focus, the GTK+ triangle has some ugly non-antialiased one-pixel black border. Without focus, it looks OK, no border. - In Gimp, a similar GTK+ color wheel also has a small palette where you can remember colors (in the bottom right area). Can we have it, too? - "Color name" should really be "Hex" or something like that. Besides, it's redundant as we have the RGBA field below. - The "previously selected color" in the GTK selector is always black. It must reflect the color we got from the program or from preferences; i.e. for "apply to selected objects", when I select a new object, both halves of the color swatch must show the same color, and then the left side should stay the same while the right one is changing. - The "Mode" switch at the top seems redundant (and incomplete), now that we have the multi-tab interface for different pickers. - The bottom right button that allows you to select which pickers are shown is also redundant, IMHO. Instead, the color selector must remember (both within and across sessions) the last picker used and show it when activated. With this, removing pickers is not necessary and frees no extra space. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

New guide prototype
by Nathan Hurst 11 Dec '03

11 Dec '03

G'Day! Peter and I have been working on a new grids and guides idea which extends the control handle idea of Dia and Visio 'correctly' (We can justify this claim quite strongly). There is a test bed version of it at: http://bowman.csse.monash.edu.au/~pmoulder/guides-LATEST.tar.gz To run it you will need python-gtk2 and python-numeric (also called numpy). Please run it and tell us if the behaviour feels right - it is easier to try variations whilst the code is python, rather than C++ (although it is easier to know whether the behaviour feels right once we implement it in inkscape proper) Feel free to ask questions about our intentions, and suggest extensions that would be helpful. Clearly the prototype is but a taste of what can be done with this new approach (We have several whiteboards worth of ideas to implement :), but it is important to work out the basic interface now. In particular, what do people think about the two tools behaviour? Should this be folded into a modifier key on the selection tool? Can it be done well with only one tool? Is the breaking of constraint behaviour (which control point disconnects when you move something) reasonable, or can you suggest a better heuristic? (Currently the rule that is broken is chosen arbitrarily, perhaps it would be better to break the oldest created relationship, or perhaps the least recently moved?) What do people think of the indicators for how control points are connected to guide lines and guide points? Suggestions? In case you're wondering, our intent is to provide Dia level diagraming support, as well as strong CAD type design elements. Some things which we know how to do, but haven't added include: adjustable spacers, enforced alignment and distribution, clipart with programable control handles (like visio), and many other goodies :) Thanks, njh

2 3

Inkscape 0.36 - "See, even a 5 year old can use it."
by Ted Gould 11 Dec '03

11 Dec '03

Inkscape Release 0.36 "See, even a 5 year old can use it." This is the second release of Inkscape and there has been a tremendous amount of work that has been done on the project. Many patches and ideas that had been 'waiting in the wings' have now been applied to the codebase. Now Inkscape is easier to use for users as the interface is less complex and more intuitive. In Brief: * Main GUI has been significantly redesigned * Usability of dialogs, keyboard shortcuts, and mouse is improved * Preliminary support for layer grouping, path markers, and extension modules * Much more user and developer documentation - doxygen, docbook, man page, wiki, diagrams, & more * Many bugfixes, minor enhancements, and usability tweaks What is Inkscape? The Inkscape application is an open source SVG editor with capabilities similar to Illustrator, CorelDraw, Visio, etc. Supported SVG features include basic shapes, paths, text, alpha blending, transforms, gradients, node editing, svg-to-png export, grouping, and more. Enhancements * CSDI interface has been traded for an SDI one: Now each document window has its own menubar at the top and its own detachable toolbar at left. This increases Inkscape's compliance with the GNOME HIG and decreases confusion for the majority of users. * A secondary (horizontal) detachable toolbar is introduced for operations and controls specific to each tool. At the moment this toolbar is mostly empty but we have big plans on how to fill it with all kinds of useful things for convenient access. * Dialogs now always stay on top of the document window (even if you have several document windows open). They are easy to open or activate (using shortcuts displayed in their title bars), close (Ctrl+w), or defocus (Esc). Moreover, dialogs remember their size and position across sessions. * New commands: Paste style (apply the style of the copied object to the current selection), Fit page width to window. * Introductory layers support. There is now code in the code base to start handling layers. GUI elements to follow - watch this slot! * Arrowhead and marker support. A much requested feature which increases the usability of Inkscape for everyday tasks. * Using mouse is now less painful thanks to the increased mouse tolerance (you don't have to be exactly over an object to grab it) and introduced drag tolerance (very small drags, which are usually accidental, are now considered clicks and do not move anything). * Undo is now unlimited (that is, limited only by available memory). * Lots of new keyboard shortcuts. Notably, Esc now consistently deselects, cancels selection, and cancels drag or transformation of any kind (so far only in selector and node edit). Arrows, Ctrl+a, and Tab/Shift?+Tab act on nodes in node editor exactly as they do on objects in Selector. There are at least two ways to activate each tool, one with an F<number> key (Corel Draw style) and another with a single letter (Adobe Illustrator style). Space toggles between Selector and your current tool. Dialogs are consistently activated by Ctrl-Shift-<letter> combinations. * There's no lower limit on the size of exported PNGs anymore. You can use Inkscape to create even 1x1 PNG images. * The standard xml:space attribute on text objects is now honored. For newly created objects it has the value of "preserve" which means all whitespace characters are shown exactly as typed. * Enhanced modules system. Now modules can be defined using external files so no C coding is required for an extension. * Mac OS X and Windows build fixes. These builds are getting more mature and working better. * Increased code documentation. Come check it out, the Inkscape codebase is getting more readable every day. * Numerous bugfixes and usability improvements. Infrastructure * The web presence of Inkscape is dramatically different than with the last release. If you haven't already, you should check it out at: http://www.inkscape.org. * Many people have been contributing to the Wiki, which has a sizable amount of content today: http://www.inkscape.org/cgi-bin/wiki.pl. * Doxygen documentation builds are now happening on HEAD nightly: http://www.inkscape.org/doc/doxygen/html. * CVS snapshots are taken on a nightly basis: http://www.inkscape.org/cvs-snap/. What lies ahead * More useful stuff in the secondary toolbar - all kinds of controls that are relevant to a specific tool will be put into that toolbar for convenient access. * A new, extensive Preferences dialog. Lots of settings that are now hard-coded will be settable. * Dialogs reorganization: less dialogs overall, with their functions more logically grouped. * More precision, maximum zoom of 256x or more. * Complete keyboard control for node editing, including editing control handles of a node. * GUI selection of arrowheads for path start/end/mid markers. Known problems * At least on KDE, floating dialogs may not stay on top when you switch between maximized document windows. You may need to click on the window's title bar to bring the dialogs to top. (See src/dialogs/dialog.events for a possible workaround and comments, feel free to suggest a patch.) * Also on KDE, detached toolbars do not stay on top. (This is a KDE problem.) * On Windows, dialogs do not stay on top of document in this version. * Some keyboard shortcuts (e.g. arrow keys for nudge) do not work if your mouse pointer is not over the document window. * When using Copy/Paste? (or Paste style) commands, you cannot copy gradients or arrowheads from one document to another. Workaround: use the XML editor to copy the content of the <defs> element from the source document to the target document, then copying gradients and arrowheads will work. * Pasting text directly into a text object on the canvas does not work. Workaround: open the Text dialog (Ctrl+Shift+t) and paste into the text editing area there, then click Apply. * 'warning: failed to load external entity "glade/icons.svg" - This is not a fatal problem, but merely an indication that Inkscape could not locate its icon images. This problem can also be detected when some of the button icons show up as random noise. The usual fix is to go through the 'make install' process so that the icons.svg file gets copied to the expected location. * 'colinear!' - This is a libart warning. The original libart intersector has all kinds of numerical problems, and it is believed that this warning is emitted if it tries to find intersection of two almost parallel lines. Robert Crosbie recently noticed this still occurs when using the Calligraphic tool. More information at: http://www.inkscape.org/

1 0

New Jabber server
by Bryce Harrington 11 Dec '03

11 Dec '03

Hi all, dmandell has graciously set up a permanent Jabber room for Inkscape on his server, and I'd like to propose we transition to that instead of the one on jabber.org. The reasons for this are: * Fewer users on it so probably not so laggy * We've got the ear of the admin (useful in case of problems) * Forewarning in case of server software changes * Fewer random visitors We're going to test it a bit this evening, and assuming things go well, and no one identifies any issues with transitioning, we will update the website docs to point there instead of jabber.org. Bryce

1 0

color selector changes
by Jon A. Cruz 11 Dec '03

11 Dec '03

Well... I just checked in some changes for the color selector. It needs a little more polish (mode, CMYK types, and the popup to select subsets of possible pickers) but should be ready for some initial testing and use. As part of it, I was simplifying the pickers to juse use SPColor objects instead of all the overloaded different component primitive types. This might be a good thing to continue... as it could facilitate moving to better color, custom spot color, etc. Oh, and the 'API' (as it were) for creating new pickers will probably undergo a little revision. At the moment the gtkcolorselector might be an interesting place to start.

2 1

0.36 release candidate tagged in CVS
by Bryce Harrington 11 Dec '03

11 Dec '03

The one priority-9 bug that was identified in testing was fixed by Bulia, thus closing the last task required for the Inkscape 0.36 release. It builds with both gcc and g++. Folks are itching to get into dev mode for 0.37. We're ready to cut the release. RELEASE_0_36_0 is now tagged in CVS. Unless we discover any major issues during packaging, this should be considered the official release source. If there are issues that require fixing for this release, we can create a branch from this tag point to work on them, and allow the main branch to return to development. This tagged version can be checked out via the command: cvs co -r RELEASE_0_36_0 -d inkscape_0_36_0 inkscape Ted, can you finalize and post the official tarball at your earliest convenience? Nathan and Mael, once this tarball is available, please proceed with creating the Debian and Windows binary packages, and I'll do the same for RPM. Once all of these packages have been created, then we will put out an announce to Freshmeat, et al. Hopefully we can achieve this within a day or two. Meanwhile, everyone should consider CVS HEAD to be unfrozen and available for open development once again. Thanks, Bryce

3 4

last changes pleease
by bulia byak 11 Dec '03

11 Dec '03

Hoping it's not too late, could you guys please take the latest CVS for the release? Just a couple last-minute cosmetic changes: h and v shortcuts for flipping, and the Edit menu is split into Edit and Object. This is important for making a good first impression, as the old Edit menu was way too big. _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

RE: [Inkscape-devel] 0.36 release candidate tagged in CVS
by bulia byak 10 Dec '03

10 Dec '03

> > What about doc/keybindings.txt? Whoever is preparing the release, pray > > remove this old crap and replace it with a copy of KeyboardShortcuts >from > > wiki! > > (ideally this needs to be automated) > >Ah, true... > >Ted, I've marked this as a task for you. Bulia, can you cut a copy of >your doc from wiki and put it into CVS HEAD? Ted can pick it up from >there when he's ready. Done, doc/keybindings.txt removed, doc/keyboard-shortcuts.txt added (manually created for now). _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

Website problems
by Jon A.Cruz 10 Dec '03

10 Dec '03

I've noticed some problems on the website in regards to it's markup. The most visible problems showed up under Mozilla 1.0.2. I then went to validate the pages (at http://validator.w3.org) and found that there are actual problems. Whoever can make changes should probably go through and use validators and do some cleanup. In hand with those pointing out problems, the 3.2 and 4.0 HTML references at http://www.htmlhelp.com are *very* useful, since among other things they are machine generated from the DTDs and include listing what tags/elements can contain others, and which can be contained. A *must* use for anyone doing web work. :-) Anyway, on to some specific problems: * The validator doesn't like the keywords meta tag for some reason. * Anchor tags can't contain H1 tags. Just swap and put the A's inside the H1's and all will be happy-happy * ALT attributes missing on images. It's polite to have them. :-D * List nestling broken. This is probably the main cause of the problems I was seeing. * Literal "<defs>" in the text, instead of "<defs>" On the list nestling, the problem is with UL's contained in other UL's. The validator complains because a UL can't contain another UL. However, it's legal to include a UL inside of a LI. So at the moment the last list entry is closed before opening a new UL. That closing </LI> tag should be moved to *after* the enclosed UL. Then all should be happy-happy. :-) (I've gone ahead and sent this out to the list since these are subtle things that are easy to trip over and not be aware of, however once understood they're easy problems to avoid. Plus the two links up there are very useful)

2 1

RE: [Inkscape-devel] 0.36 release candidate tagged in CVS
by Bryce Harrington 10 Dec '03

10 Dec '03

On Wed, 10 Dec 2003, bulia byak wrote: > What about doc/keybindings.txt? Whoever is preparing the release, pray > remove this old crap and replace it with a copy of KeyboardShortcuts from > wiki! > (ideally this needs to be automated) Ah, true... Ted, I've marked this as a task for you. Bulia, can you cut a copy of your doc from wiki and put it into CVS HEAD? Ted can pick it up from there when he's ready. Bryce

1 0

Re: [Inkscape-devel] new booleans patch
by bulia byak 10 Dec '03

10 Dec '03

The updated patch from the author and my hookup patch (which puts the new commands into the menu and as shortcuts - ugly, just for testing) are here: http://sourceforge.net/tracker/index.php?func=detail&aid=857377&group_id=93… Bryce, can you give Fred developer access to the CVS so he could create a branch for testing the new functions? (Or maybe I can now do this too as an admin, but I don't know how :) _________________________________________________________________ Get holiday tips for festive fun. http://special.msn.com/network/happyholidays.armx

2 2

Re: libgsf + Inkscape
by Jody Goldberg 10 Dec '03

10 Dec '03

On Tue, Dec 09, 2003 at 09:49:19PM -0800, Bryce Harrington wrote: > On Tue, 9 Dec 2003, Dom Lachowicz wrote: > > > > I just committed a bunch of documentation. We're a bunch closer to 100% API > > documentation now. Nice work. > Very cool. May want to think about a webpage too... at the least, a > list of features. Yah. I know. I tend to be better at writing code than web pages. Probably when gsf 2.x comes out or someone steps forward. > Possibly libgsf and neon would obviate any need for gnome-vfs but I > haven't really looked in enough depth to say for sure. I haven't > confirmed that gnome-vfs actually supports WebDAV, and if it does, > how good the support is. > > WebDAV is composed of a number of different methods, some basic (GET, > PUT, HEAD), others more advanced (MOVE, MKCOL, PROPFIND, etc.) So when > a lib or app says it supports WebDAV you have to check 'how much'. Neon > supports a wide range of them, according to its feature list. > > http://www.webdav.org/neon/ Interesting I'll have to have a look. As mentioned our plans had been to use libcurl for network related activity. However, I have not checked on its webdav status any more than gnome-vfs. > Have a look at Neon and let me know what you think. Would it be > feasible to use it in conjunction with libgsf? I'll have a look. > The direction we're heading is to make wrapperization of the app easier, > whether that be with Python bindings (via SWIG?), Bonobo, or even a kind > of SVG Canvas. Being able to plug inkscape into other things opens lots > of very interesting uses of it. That would be very interesting. Abi and Gnumric had previously discussed supporting embedding by just rendering an svg image. Something like this would be nice.

2 1

Re: [Inkscape-devel] new booleans patch
by bulia byak 10 Dec '03

10 Dec '03

Are you subscribed to the inkscape-devel list? Let's move our discussion there, because many people will be interested in your code. >the code is "borrowed" from the >sp_path_selected_combine(), and there's a check on the >number of elements in the selection (looks like it, at >least). It only checks that there are at least two objects selected: if (g_slist_length (il) < 2) return; So we need to fix it for the case when there are more than two. If you can make intersect and union functions accept any number of objects in one go, that would be the best. And for difference functions, we need to check that there are exactly two objects selected, and complain if not (for now you can just print a g_warning and do nothing). > > Inset and offset, although very very useful, are > > unfortunately more buggy. > > When you do them repeatedly, the number of nodes > > grows very fast, and soon > >oops... yes the number will grow fast (exponentially). >isn't there already a function to cleanup a path in >sodipodi? None that I know of. i recall seing somewhere discussion on how >to simplify a path... besides it's just a lame >approximation of the offset, there's probably wiser >algorithms (that i do not konw about) You can discuss this on the list with people who have much better knowledge of algorithms than I, such as Nathan Hurst. >well, the soutline/offset/inset algorithm is quite >stupid: build an offset path, and treat it like a >polygon. the "difficulty" lies in building the offset >path. the code is still rough, and doesn't like null >segments in the path, path with multiple subpath, >closed paths, etc... i have to admit it hasn't been >tested thoroughly; just outlining a path twice as you >suggested brought up a number of serious problems i'll >have to investigate. on top of that, the result of the >offset is transformed in a svg path description (so if >d="", it means the algorithm completely failed), and >the precision used to store the coordinates in the d >attribute seems insufficient. We plan to increase it from 6 digits to at least 8. Do you think 8 will be enough for you? >technically, coordinates >are rounded to a 1/32 grid by the code in livarot; >dumping the path with less precision in the xml tree >can cause ugly things. the cusps you mentioned are >usually small segments making sharp turns because >coordinates were rounded. using a round or bevel join >"removes" them (btw, how do you set the miter limit in >inkscape?) I don't think setting meter limit is implemented. Anyone? I think now the best course of action is this. I'll ask Bryce to give you access to the CVS so you could create a branch for testing and development of the new commands. You will store the entire Inkscape tree there, so that people would be able to checkout and test it without patching. To hook up your commands to the menu and as shortcuts, I did a simplistic patch, which I will now upload to the Inkscape tracker along with your updated patch; you can use my patch as a starting point. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

Re: [Inkscape-devel] new booleans patch
by bulia byak 10 Dec '03

10 Dec '03

After some testing I can say that the four basic operations (union, intersect, diff, symdiff (which is similar to combine, except that it adds nodes where sides intersect)) are very stable. I could not force them to do something wrong or misbehave. The only problem is that when more than one object is selected, these commands only operate on the first two objects while others simply disappear. But I think it will be easy to fix. The quality of these operations seems to be good: no extra points, no curve distortion. Inset and offset, although very very useful, are unfortunately more buggy. When you do them repeatedly, the number of nodes grows very fast, and soon some very strange artifacts appear (such as dense "clusters" of nodes that really distort the shape). Also, when you do inset or offset to a path with many nodes, it becomes extremely slow, and Inkscape may hang for minutes. Sometimes, a path simply disappears (the object is there, but its d="" attribute is empty). The outline stroke function works acceptable. Sometimes the strokes are distorted (usually near ends or sharp cusps) but the number of nodes in the resulting shape is not too big, and for simple strokes it gives good results. >Actually, having looked at the patch, I think we really need to leave >this 'til the next release. Three reasons: First, it's in C++ and we >had decided to hold off on changes that commit us to C++ until 0.37. What is the value of preserving plain C compilability? Will people complain if we require C++ for 0.36? I, for one, do not care what is the compiler, so long as the program works :) >Second, it's a good bit of code, and deserves to be tested. The best way to test is to release! We're not doing 1.0, after all. It's just a 0. release. >Third, we >will need to do some additional development (buttons, etc.) to integrate >it. I could certainly live without buttons - menu commands and shortcuts are quite enough for the time being. Besides, we could borrow buttons from the other patch, and I think I remember seeing the icons somewhere in the sodipodi innards already. So my opinion is this: let's include the four boolean operations and outline path for this release, and then help the author debug the inset/offset operations for the next release. _________________________________________________________________ Get holiday tips for festive fun. http://special.msn.com/network/happyholidays.armx

2 1

Re: [Inkscape-devel] non-ASCII chars
by bulia byak 10 Dec '03

10 Dec '03

>So... it's often good to do 'sanitizing' of input. In this case, where it's >direct user input, speed should not even be a problem. OK, what about: u = g_utf8_get_char (utf8); if (u == (gunichar) -1) { g_warning ("Bad UTF-8 character"); return pos; } if (!g_unichar_isprint (u)) { g_warning ("Non-printable character"); return pos; } is this enough sanitizing? :) _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

3 2

Status of priority-6 bugs for 0.36 release
by Bryce Harrington 10 Dec '03

10 Dec '03

All of the priority 9 bugs and all of the bugs listed on the roadmap for required the 0.36 release have been resolved and closed. (It's been pretty amazing how much intensity was placed into bug fixing this last week; 20 bugs closed!) The remaining priority-6 bugs are listed and commented on below. My thought was that the priority-6 bugs "would be nice to close" but were not required. We've closed a number of them, and I think the remainder can be left for a future release. If anyone feels differently on any of these, speak up now. Bryce [ 851467 ] Export selection bounds too small There is a patch for this that John Stracke is developing for Sodipodi. If we wait a tad longer it might become more matured. See Sodipodi Bug#758775 for the patch. [ 841633 ] Icons problem for win32 I'd like to get confirmation that this still occurs, as I suspect the GtkButton rework may have resolved it. It doesn't cause a crash and sounds like its platform dependent, though. If the problem still exists, then it should be made high priority for the next release. [ 850936 ] .32 crashes when setting printer destination It no longer crashes but it does emit an ERROR message. This should receive high priority in the next release but may be acceptable for 0.36. [ 853804 ] SVG COMPLIANCE: Paths shown will fill when fill:none The issue is that "file: none;" and "fill:none;" behave differently for the given case. This is important for importing SVG but perhaps one we can leave 'til next release as it doesn't impact general use. [ 852619 ] XML editor crash There are a couple XML editor related bugs, but they appear unrelated. This particular bug is uncovered after a sequence of steps that may not be encountered by most users in general usage. [ 50549 ] spiral crash This is an easy bug to reproduce, however the spiral shape does not seem like one that would normally get heavy use. [ 847886 ] transformation inaccuracies The change from float to double addressed part of the bug, the remainder appears to be a legitimate issue but may not be noticeable to regular users. [ 851937 ] Postscript file output does not respect document size More of a feature request than a bug, I think. [ 851131 ] Print preview does nothing I'm not sure if print preview is implemented, as it's never worked for me. I've commented out the menu item. If this *does* work for anyone, let me know so we can leave it in but mark it a 'known issue', and get it fixed. Otherwise we should list high on the feature request list. [ 853858 ] errors when copying objects between docs Bulia estimates this will require too much effort to fix for 0.36. It's listed as a known issue on the release notes page. [ 850971 ] Crash while deleting too much using the XML editor This bug is in regards to deleting the sodipodi namedview element via the XML editor. I looked into the code to see if there was an easy way to trap it but to do it right it looks like it would take some research. Since it's an unusual error I don't think a quick fix would be appropriate, and some time should be taken during the next release cycle to find a good solution. [ 851779 ] Style property "currentColor" not used properly This is another compliance issue for importing SVG files. Specifically, from gnuplot.

4 4

RE: libgsf + Inkscape
by Bryce Harrington 10 Dec '03

10 Dec '03

On Tue, 9 Dec 2003, Dom Lachowicz wrote: > > > I'll work on getting at least the API documentation up-to-snuff during > >the > > > coming week. On the whole, the parts I envision a project like Inkscape > > > using strongly resemble STDIO's FILE operations enough for a strong > > > similarity to be found. > > I just committed a bunch of documentation. We're a bunch closer to 100% API > documentation now. Very cool. May want to think about a webpage too... at the least, a list of features. > >The Inkscape file management is one area in need of major reworking. > >Since Inkscape is at root a file editor, having powerful file management > >capabilities is important. As well, some currently only blue-sky ideas > >are brought within feasible reach by having support for things like > >WebDAV, archive access/extraction, etc. Several libs including libgfs, > >gnome-vfs, and neon have been suggested. We will be using one of these > >rather than creating our own, so we can focus our energies into SVG > >editing. ;-) > > I'm not familiar with neon, but I can definitively say that LibGSF better > fits your needs than GnomeVFS. At a bare minimum, we're able to piggyback > directly on top of GnomeVFS if you so choose, and with much less hassle. > With LibGSF, you get superb portability to other platforms and a wide > variety of inputs and outputs. Your application is inherently pluggable, at > least in terms of "file sinks." Possibly libgsf and neon would obviate any need for gnome-vfs but I haven't really looked in enough depth to say for sure. I haven't confirmed that gnome-vfs actually supports WebDAV, and if it does, how good the support is. WebDAV is composed of a number of different methods, some basic (GET, PUT, HEAD), others more advanced (MOVE, MKCOL, PROPFIND, etc.) So when a lib or app says it supports WebDAV you have to check 'how much'. Neon supports a wide range of them, according to its feature list. http://www.webdav.org/neon/ > As a side note, I've grown a bit bored with working on LibRSVG. I'll have to > check out Inkscape. We'd love to have you, we've been quite impressed with what you've done with librsvg. The project's quite fun - progress is swift, there's a good teamwork attitude, and plenty of room for new ideas. > >Alan had mentioned it also supported WebDAV? > > If GnomeVFS supports WebDAV, so would we. If not, we could add a plugin to > LibGSF probably pretty easily. Have a look at Neon and let me know what you think. Would it be feasible to use it in conjunction with libgsf? > >Cool, it sounds like this would take care of several user feature > >requests we've had such as cut and paste operability between Inkscape > >and AbiWord. There's also been interest voiced into bonobo. Most of us > >don't have much depth in the tech behind these so the more help/advice > >available, the easier it'd make it. > > Yeah, if it'd be possible to get at least image/png and image/svg on the > clipboard, that'd be awesome. And if you could make Inkscape into a > BonoboControl, that'd be sweet too. With LibGSF, you'd be able to load and > save things from BonoboStreams insanely easily. The direction we're heading is to make wrapperization of the app easier, whether that be with Python bindings (via SWIG?), Bonobo, or even a kind of SVG Canvas. Being able to plug inkscape into other things opens lots of very interesting uses of it. That, plus being able to interact with files (or svg streams generally) through network connections... I'm not sure I can imagine all the uses, but it would certainly make things possible that haven't been before. > >Yes I noticed these in the API docs, but I was wondering what the > >motivation for inclusion of them in this was? Are there special XML > >file management capabilities it provides over what'd be available from a > >generic XML library? > > Well, there is a really easy routine to create a XmlParseCtx from a > GsfInput. There are also a bunch of nice SAX utils. There is also a nice set > of functions for writing XML documents. I'm not sure if either is > appropriate for your needs, but they're there if you need them. Cool, thanks. Bryce

1 0

real world usability test
by bulia byak 10 Dec '03

10 Dec '03

Today my 5 years old daughter spent about 20 minutes in Inkscape, drawing stars. She enjoyed it a lot. She had no problem understanding the role of the selector button ("click it and then you can move your stars around"), node edit button ("click it and you can do funny things to a star by moving those small dots in it") and the star itself. She also used ctrl-z often. The main reason for joy is that the program did not crash :) There were some problems; the biggest bummer was when she accidentally grabbed the handle of the main toolbar and detached it - then she clicked on the canvas and the toolbar was gone! Bad, bad Inkscape. Also, the lack of a palette was a bit of an inconvenience, although I switched the fill&stroke dialog to HSV and she had no problems dragging the hue slider to get different colors for her stars (with two other sliders set to maximum). Overall, I think the program passed this test :) I don't think it would be that enjoyable with Sodipodi, with its myriad of confusing buttons and sinking windows... _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Re: [Inkscape-devel] max undo
by bulia byak 10 Dec '03

10 Dec '03

> > One thing: right now subsequent arrow-key movements are stored as >separate > > undo steps. I'd like to combine all adjacent moves in one direction >(only > > horizontal or only vertical) in one step. Is there already a mechanism >for > > that, or do I have to dig into the undo system to implement it? > >Yes, there's already a facility present for that: >sp_document_maybe_done(). Thanks for the pointer. I tried this but it did not work until I replaced: doc->actionkey = NULL; by doc->actionkey = key; in sp_document_maybe_done(). This fix seems logical to me, but it's not my code so please confirm that it's correct. I've just committed a version that puts subsequent same-direction arrow-key movements into one undo step (in both selector and node edit) so you can test how it works. Previously, maybe_done was only used in the item properties dialog (which is planned for elimination btw). _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Re: [Inkscape-devel] max undo
by bulia byak 10 Dec '03

10 Dec '03

>Realistically, are users ever going to backtrack through more than ~300 >levels of undo anyway? Easy :) Real drawing sessions may span hours and hours. Right now, however, a crash is a more likely outcome of being in Inkscape for too long, but in the future... My opinion is that a vector app must have unlimited undo, or at least very large. It's the promise of the vector: everything is preserved, no loss of information. Unlimited undo is part of that appeal. One thing: right now subsequent arrow-key movements are stored as separate undo steps. I'd like to combine all adjacent moves in one direction (only horizontal or only vertical) in one step. Is there already a mechanism for that, or do I have to dig into the undo system to implement it? _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

6 11

Counting Lines
by Ted Gould 10 Dec '03

10 Dec '03

Hey, Bryce asked for this, but I thought it might be useful for other people too. There is a little utility on this page called 'lc' which counts lines. There are others too, but I've only used lc. Just so everyone doesn't have to do it, I ran it on the Inkscape codebase and threw it into the attached Gnumeric file ;) http://www.chris-lott.org/resources/cmetrics/ --Ted

1 0

Inkscape 0.36 pre-release 1
by Bryce Harrington 09 Dec '03

09 Dec '03

A pre-release version of Inkscape is available for review and testing in source and RPM forms here: http://sourceforge.net/project/showfiles.php?group_id=93438 Please download, test out, and report any significant problems you encounter to the Inkscape bug tracker. We anticipate releasing 0.36 soon, unless any show-stopper bugs (such as frequent crashes) are discovered. Thanks, Bryce

1 0

Re: Map Symbols [was Re: [Sodipodi-list] Getting lines straight]
by Jonathan Phillips 09 Dec '03

09 Dec '03

Maybe a novel idea would be to make separate project on sourceforge that is svg clipart. Like svg-clipart.sf.net would be great. The benefits would be open source designing of symbols, maps, etc for the SVG-compliant community. It might also be nice to push the artwork developed via different SVG apps to a standard, well defined place that might promote interoperability and compliance of all,or many :) ofthe different SVG drawing programs. It would be totally cool to use CVS to design with and allow ppl. from all over the world to work on designs. Thoughts? (I hope I haven't just gotten a little too utopian) Jon On Fri, 2003-12-05 at 07:03, Pat Suwalski wrote: > Áki G. Karlsson wrote: > > Hi > > > > I would be more than happy to participate in a svg map symbols project, > > contributing images like below that are most relevant for my present > > work... > > > > I am just worrying that maybe the sodipodi list doesn't want to be > > subjected to it and that it would perhaps be better served as, say, a > > part of wikitravel... > > Should I be starting a "map symbols" clipart section on the sodipodi web > site? > > Right now, I'm still a little fazed by the whole sodipodi.com thing. > > --Pat > > > > ------------------------------------------------------- > This SF.net email is sponsored by: IBM Linux Tutorials. > Become an expert in LINUX or just sharpen your skills. Sign up for IBM's > Free Linux Tutorials. Learn everything from the bash shell to sys admin. > Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click > _______________________________________________ > Sodipodi-list mailing list > Sodipodi-list(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/sodipodi-list -- Jon Phillips UCSD Graduate Researcher Computational Media PO BOX 948667 La Jolla, Ca 92037 858.361.2811 www.protofunk.org jon@...15...

8 14

max undo
by bulia byak 09 Dec '03

09 Dec '03

I have increased max undo from 128 to 512, as 128 was way too low imho. In the next version it will be in preferences, but 512 may be the default. Any objections? And by the way, has anyone tried my 256x max zoom patch? _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

7 6

text tool news
by bulia byak 09 Dec '03

09 Dec '03

* src/sp-text.c: fixed typing non-ascii chars; copied (with fixes) treatment of xml:space from sodipodi 0.33 * src/sp-text.h: small fix from sodipodi 0.33 * src/text-context.c: copied sp_text_context_preedit_reset and related code from sodipodi 0.33; xml:space='preserve' is added to newly created text objects; esc deselects; uncatched keys are passed on; misc fixes Please test the text tool thoroughly! What I did NOT copy from sodipodi is the new code within #ifdef SP_TEXT_NEW_CONTENT (no idea what it is, it's off by default) and the pervasive change by which SPObject now has "children" but SPText has not, so all text->children were replaced by much uglier ((SPObject *) text)->children. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

xml:space default?
by bulia byak 09 Dec '03

09 Dec '03

As a result of my and mental's copying code from sodipodi, now it's possible to set xml:space=preserve via xml editor on a tspan, and the multiple adjacent spaces will not be combined into one as they are now. Does anyone else think it might be a good idea to add this attribute by default to all newly created text objects? BTW this would fix this bug: https://sourceforge.net/tracker/index.php?func=detail&aid=851005&group_id=9… _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

RE: [Inkscape-devel] non-ASCII chars
by bulia byak 09 Dec '03

09 Dec '03

On the other hand, why do we have to guard against any control chars at all? If we get a string in UTF8, suffices it to check that it's valid UTF8, and complain if it's not. If it is valid, we can insert it without any further checks. Any objections? >From: "bulia byak" <archiver_1@...19...> >To: inkscape-devel(a)lists.sourceforge.net >Subject: RE: [Inkscape-devel] non-ASCII chars >Date: Tue, 09 Dec 2003 06:09:38 +0000 > >OK, I solved this one. Indeed the problem was gchar (i.e. signed) - which >was treated as less-than-zero when it run into this stupid bit of code: > >if ((spnew[i] < 32) && ((spnew[i] != 9) && (spnew[i] != 10) && (spnew[i] != >13))) spnew[i] = 32; > >which was meant to replace control chars with spaces, but ended up >replacing all non-ascii chars with spaces when used with gchar instead of >guchar. > >I'll now rewrite this brain damage properly, using g_unichar_iscntrl(). > > >>From: "bulia byak" <archiver_1@...19...> >>To: inkscape-devel(a)lists.sourceforge.net >>Subject: [Inkscape-devel] non-ASCII chars >>Date: Tue, 09 Dec 2003 05:36:50 +0000 >> >>I have merged the interesting parts of 0.33's sp-text.c and >>text-context.c, but this did not help with the cyrillic chars being >>untypeable. I started to dig into this specifically and now I have a >>strong suspicion that the culprit is the inkscape's change from guchar to >>gchar in functions that insert new chars into the string as they are >>typed. I cannot test it quickly because it requires changing in lots of >>places (and they cannot be simply cast, at least not with g++), but this >>seems likely. For example sp_text_insert used to receive the char to >>insert as a guchar, but now it gets gchar. As it is supposed to be in >>UTF8, perhaps the encoding goes wrong and the result is the space >>characters in the string (or at least something that looks like spaces). >> >>So: >> >>- Does anyone else think that guchar/gchar replacement could break this? >> >>- Why was this replacement necessary in the first place? >> >>- Is it OK if I reverse it in the text code? >> >>_________________________________________________________________ >>Add photos to your e-mail with MSN 8. Get 2 months FREE*. >>http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo… >> >> >> >>------------------------------------------------------- >>This SF.net email is sponsored by: SF.net Giveback Program. >>Does SourceForge.net help you be more productive? Does it >>help you create better code? SHARE THE LOVE, and help us help >>YOU! Click Here: http://sourceforge.net/donate/ >>_______________________________________________ >>Inkscape-devel mailing list >>Inkscape-devel(a)lists.sourceforge.net >>https://lists.sourceforge.net/lists/listinfo/inkscape-devel > >_________________________________________________________________ >The new MSN 8: smart spam protection and 2 months FREE* >http://join.msn.com/?page=features/junkmail >http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms… > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

RE: [Inkscape-devel] non-ASCII chars
by bulia byak 09 Dec '03

09 Dec '03

>On Tue, 2003-12-09 at 01:26, bulia byak wrote: > > On the other hand, why do we have to guard against any control chars at >all? > > If we get a string in UTF8, suffices it to check that it's valid UTF8, >and > > complain if it's not. If it is valid, we can insert it without any >further > > checks. Any objections? > >Well, define "complains"? Issue a g_warning and insert nothing. Actually this cannot happen unless there's a bug somewhere else. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

sodipodi diff
by bulia byak 09 Dec '03

09 Dec '03

(sorry my gaim keeps segfaulting, no idea why, never used it before... email is more reliable anyway :) I did a diff between the hydra branch and Sodipodi 0.33 (only /src). It's 450Kb so I cannot even put it on the tracker - maybe put into CVS? Where's the best place? Anyway, I would propose that everyone would review the diff and delete those parts that are already in Inkscape, or those that make to sense, or are already resimplemented differently, etc. Anything worth transferring should be transferred (and then deleted from the diff). Hopefully we can thus eventually reduce it to zero. I have already removed some small things that I copied to Inkscape. What I'd like to discuss is this: - arikkei_* functions are now used extensively all over the codebase, for things like number to string conversions. Are they worth borrowing? What are their advantages compared to atoi and snprintf used before? - Sodipodi 0.33 can type Cyrillic characters in the text tool. Inkscape cannot. Sodipodi also can display fancy chars such as em-dash (ctrl-u 2014), Inkscape cannot. I think this is due to all the Unicode new stuff in 0.33. We must have this too. Unfortunately it's quite big and complex and I'm not sure I can transfer this correctly. Any takers? - The "duplicate" checkbox in transformation dialog is nice. It looks simple so I will transfer it (after 0.36 release). _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

5 9

RE: [Inkscape-devel] non-ASCII chars
by bulia byak 09 Dec '03

09 Dec '03

OK, I solved this one. Indeed the problem was gchar (i.e. signed) - which was treated as less-than-zero when it run into this stupid bit of code: if ((spnew[i] < 32) && ((spnew[i] != 9) && (spnew[i] != 10) && (spnew[i] != 13))) spnew[i] = 32; which was meant to replace control chars with spaces, but ended up replacing all non-ascii chars with spaces when used with gchar instead of guchar. I'll now rewrite this brain damage properly, using g_unichar_iscntrl(). >From: "bulia byak" <archiver_1@...19...> >To: inkscape-devel(a)lists.sourceforge.net >Subject: [Inkscape-devel] non-ASCII chars >Date: Tue, 09 Dec 2003 05:36:50 +0000 > >I have merged the interesting parts of 0.33's sp-text.c and text-context.c, >but this did not help with the cyrillic chars being untypeable. I started >to dig into this specifically and now I have a strong suspicion that the >culprit is the inkscape's change from guchar to gchar in functions that >insert new chars into the string as they are typed. I cannot test it >quickly because it requires changing in lots of places (and they cannot be >simply cast, at least not with g++), but this seems likely. For example >sp_text_insert used to receive the char to insert as a guchar, but now it >gets gchar. As it is supposed to be in UTF8, perhaps the encoding goes >wrong and the result is the space characters in the string (or at least >something that looks like spaces). > >So: > >- Does anyone else think that guchar/gchar replacement could break this? > >- Why was this replacement necessary in the first place? > >- Is it OK if I reverse it in the text code? > >_________________________________________________________________ >Add photos to your e-mail with MSN 8. Get 2 months FREE*. >http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo… > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] unlimited undo
by bulia byak 09 Dec '03

09 Dec '03

>Okay, unlimited undo is now in CVS. I've added that to the release notes - please check if it's correct. _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

non-ASCII chars
by bulia byak 09 Dec '03

09 Dec '03

I have merged the interesting parts of 0.33's sp-text.c and text-context.c, but this did not help with the cyrillic chars being untypeable. I started to dig into this specifically and now I have a strong suspicion that the culprit is the inkscape's change from guchar to gchar in functions that insert new chars into the string as they are typed. I cannot test it quickly because it requires changing in lots of places (and they cannot be simply cast, at least not with g++), but this seems likely. For example sp_text_insert used to receive the char to insert as a guchar, but now it gets gchar. As it is supposed to be in UTF8, perhaps the encoding goes wrong and the result is the space characters in the string (or at least something that looks like spaces). So: - Does anyone else think that guchar/gchar replacement could break this? - Why was this replacement necessary in the first place? - Is it OK if I reverse it in the text code? _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

unlimited undo
by MenTaLguY 09 Dec '03

09 Dec '03

Okay, unlimited undo is now in CVS. It's always kind of cool when you add a new feature by _deleting_ code. ^_^ -mental

1 0

button.c again
by bulia byak 09 Dec '03

09 Dec '03

So what's the resolution with button.c? I still cannot compile it, now with one more error: button.c:62: error: invalid conversion from `void (*)(SPAction*, int, void*)' to `void (*)(SPAction*, unsigned int, void*)' button.c: In function `void sp_button_toggle_set_down(SPButton*, int)': button.c:189: error: invalid conversion from `void (*)()' to `void*' button.c:191: error: invalid conversion from `void (*)()' to `void*' _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Re: [Inkscape-devel] sodipodi diff
by bulia byak 09 Dec '03

09 Dec '03

>>I can at least have a look at it... I think as part of the merge it >>might be important to understand why it doesn't currently work, too. >> >> > >I can lend a hand looking into this a little. Though it's been ages since >I've done MSVC stuff here, I have worked in i18n on and off for the past 10 >years. That would be fantastic! This is an important issue, as it makes the program almost useless unless you only do English-language art. We must get it right. Fortunately, it may be that the Unicode fonts stuff is not relevant to the issue. This is because I can actually type cyrillic (or paste fancy chars from other programs) into the text dialog, and when applied, it is properly displayed on the canvas. This may mean that only the text tool is somehow broken, but font display works right. I will look into it but I would welcome anyone's help too. I've filed a detailed bug so we don't forget about this: https://sourceforge.net/tracker/index.php?func=detail&aid=856528&group_id=9… It would be very nice to have this fixed before the release. _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

reduce, reuse, recycle
by Jonathan Phillips 09 Dec '03

09 Dec '03

In terms of getting ready for the 0.36 release of Inkscape, what do you all think about commenting out a few unimplemented features. IMO I think we should comment out: print preview (in the file menu) about modules (in the help menu) No use in confusing the new users. Jon -- Jonathan Phillips <jon@...15...>

3 3

Re: [Inkscape-devel] Ruler issues
by Michael D. Seymour 08 Dec '03

08 Dec '03

On Mon, 2003-12-08 at 13:28, Alan Horkan wrote: > > 2) The markings used in the ruler gradations aren't anti-aliased. It > > looks like Inkscape uses some sort of customized GTK ruler. > > > > One should note that the GTKRuler class is being moved out of GTK+ > > because it's too specialized/little-used. One would think that just > > Huh? > Dia, Abiword, Gimp, Sodipodi and I'm sure plenty more use rulers. That's my argument as well. > Maybe not enough projects are using the GTKRuler specifically, and when > you say moved, where to? what happens to programs that used it? Have no idea. The comment was in the GTKRuler entry in the GTK documentation: http://developer.gnome.org/doc/API/2.0/gtk/GtkRuler.html "NOTE this widget is considered too specialized/little-used for GTK+, and will in the future be moved to some other package. If your application needs this widget, feel free to use it, as the widget does work and is useful in some applications; it's just not of general interest. However, we are not accepting new features for the widget, and it will eventually move out of the GTK+ distribution." > - Alan -- Michael D. Seymour "Data is not the plural of anecdote" e-mail: michael_d_seymour@...93... -- Robert L. Park web: http://www.astro.queensu.ca/~seymour/

1 0

Re: [Inkscape-devel] button size
by bulia byak 08 Dec '03

08 Dec '03

> See what y'all think... I think the size contrast may be a bit more pronounced. However, making secondary buttons smaller is bad, they're hardly visible already. Especially the raise/lower buttons are hard to decipher, they're too complex and dark, perhaps someone will have to redraw them eventually. The main toolbar buttons may be made a bit larger, IMHO. Also, can we rearrange the toolbars so that the corner is occupied by the secondary toolbar, not the primary? Rationale: secondary toolbar may ultimately need lots of space for all its controls, so every pixel counts. The primary, on the other hand, is unlikely to grow significantly. And if you do that, can you provide a bit of separation between the toolbars by moving the primary one to the bottom a little? Ultimately I would like to see in that interval a "handle" that will let me drag the toolbar e.g. to the right edge or make it floatable, but until then, let it be just a few pixels of empty space. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

6 7

RE: [Inkscape-devel] sodipodi diff
by bulia byak 08 Dec '03

08 Dec '03

The diff is uploaded to inkscape_web cvs module as /misc/diff33.diff _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

5 6

Inkscape 0.36-pre1
by Ted Gould 08 Dec '03

08 Dec '03

There are files on Sourceforge: http://sourceforge.net/project/showfiles.php?group_id=93438 This is aimed at being an incremental build up to the final release. All of the bugs that are in the roadmap are currently fixed, with some more in the tracker. If anyone has any bugs that restrict Inkscape from being usable - this is a good time to bring them up. Also, there is work on the release notes for 0.36. The last ones weren't very good, so this time we're planning ahead so they are reasonable. If you have contributed code or bug fixes or anything to Inkscape between 0.35 and 0.36 make sure your contribution is in the Release notes. We don't want anyone's contributions to be left out. Wiki page: http://www.inkscape.org/cgi-bin/wiki.pl?ReleaseNotes I'm very excited about the 0.36 release - I hope everyone gets a chance to try this build out so that we can make it even better. --Ted

1 0

[Fwd: Re: button.c]
by MenTaLguY 08 Dec '03

08 Dec '03

Has anyone else had difficulty building button.c since I made this change? -----Forwarded Message----- From: MenTaLguY <mental@...3...> To: bulia byak <archiver_1@...19...> Subject: Re: button.c Date: Sun, 07 Dec 2003 19:07:50 -0500 On Sun, 2003-12-07 at 02:32, bulia byak wrote: > Mental, your latest button.c only compiles if I change > > g_signal_handlers_block_by_func (G_OBJECT (button), G_CALLBACK > (sp_button_perform_action), NULL); > gtk_toggle_button_set_active (GTK_TOGGLE_BUTTON (button), down); > g_signal_handlers_unblock_by_func (G_OBJECT (button), G_CALLBACK > (sp_button_perform_action), NULL); > > by > > g_signal_handlers_block_by_func (G_OBJECT (button), (void*) G_CALLBACK > (sp_button_perform_action), NULL); > gtk_toggle_button_set_active (GTK_TOGGLE_BUTTON (button), down); > g_signal_handlers_unblock_by_func (G_OBJECT (button), (void*) G_CALLBACK > (sp_button_perform_action), NULL); > > Is that the right fix? No. Casting a function pointer to a normal pointer isn't safe or portable -- you can get away with it on x86, but on many architectures (like e.g. Itanium), a function pointer is really a much larger "function descriptor" datatype. Actually, this should compile as-is (it did compile for me), so it's something funky with one of our build configurations (not sure whose yet). I'm using gcc 3.3.2 with CFLAGS="-g -Wall", and building with glib 2.2.3. What compiler, compiler options, and version of GLib are you using, and what error(s) do you get when you try to compile? -mental

3 5

Ruler issues
by Michael D. Seymour 08 Dec '03

08 Dec '03

This will get submitted as a bug, but I believe it warrants discussion. 1) It seems the ruler co-ordiante system is flipped in the y-axis w.r.t. the SVG co-ordinate system. The rulers have their origin on the bottom left of the given viewport, and the co-ordinates increase up and to the right. This goes against the SVG spec, which states that the origin of the viewport is on the top left, and co-ordinates increase to the right and down. When creating a rectangle and manually specifying the (x,y) = (0,0), the top left corner is placed at the top left of the viewport. Odd that nobody noticed this earlier. 2) The markings used in the ruler gradations aren't anti-aliased. It looks like Inkscape uses some sort of customized GTK ruler. One should note that the GTKRuler class is being moved out of GTK+ because it's too specialized/little-used. One would think that just about any drawing/display app needing co-ordinates in its display canvas would use them. -- Michael D. Seymour "Data is not the plural of anecdote" tel: (416) 545-1060 -- Robert L. Park e-mail: michael_d_seymour@...93... web: http://www.astro.queensu.ca/~seymour/

1 0

Re: [Inkscape-devel] docked toolbar steals focus
by bulia byak 08 Dec '03

08 Dec '03

>Ick. > >To be honest I still don't totally understand how events are handed off >to event contexts in Sodipodi/Inkscape. :/ Neither do I. But you have much better overall knowledge of the program's innards. I start to feel that this problem is not for me :( I'll keep poking but... _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

Re: [Inkscape-devel] [Fwd: Re: button.c]
by bulia byak 08 Dec '03

08 Dec '03

>On Sun, 7 Dec 2003, MenTaLguY wrote: > > Has anyone else had difficulty building button.c since I made this > > change? Here's what I see: button.c: In function `void sp_button_toggle_set_down(SPButton*, int)': button.c:189: error: invalid conversion from `void (*)()' to `void*' button.c:191: error: invalid conversion from `void (*)()' to `void*' $ gcc -v Reading specs from /usr/lib/gcc-lib/i586-mandrake-linux-gnu/3.3.1/specs Configured with: ../configure --prefix=/usr --libdir=/usr/lib --with-slibdir=/lib --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --enable-long-long --enable-__cxa_atexit --enable-languages=c,c++,ada,f77,objc,java,pascal --host=i586-mandrake-linux-gnu --with-system-zlib Thread model: posix gcc version 3.3.1 (Mandrake Linux 9.2 3.3.1-2mdk) Glib is 2.3.2 as far as I can see. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

node edit news
by bulia byak 08 Dec '03

08 Dec '03

(Sorry if this comes in several copies, I'm having trouble sending to the list) I just did a temporary hack: now sp_document_undo emits the selection_changed signal. The node editor catches it and redraws the path. This fixes bug 848355 for now. As an unwanted side effect, selected nodes are deselected after an undo. A better solution would require figuring out what's up with selection_modified signal and using it instead (see my previous mail - attached below). On the bright side, today I implemented duplicate node in node editor (shift-d). As a regular duplicate object (ctrl-d), it places the new node(s) exactly over the previously selected ones (and selects the new node(s) instead). By using together with Tab and Ctrl-a, this makes possible some nice keyboard-only effects. Here are some ideas for eyecandy screenshots for 0.36: - create a filled star with about 30 rays, press ctrl-shift-c, F2, ctrl-a, shift-d, some arrow keys (maybe with shift).... - create a spiral with about 30 revolutions, ctrl-shift-c, F2, ctrl-a, shift-d, some arrow keys... Also, I fixed shift-b (break nodes) so that it breaks all selected nodes and only selects one end of each break (previously it selected both ends). So you can add more fun to the above examples by pressing shift-b, some arrow keys, then ctrl-a again, shift-d, more arrow keys, etc. More usefully, this allows to easily "draw" with keyboard. Use tab to move to the end node of an unclosed path and go like this: shift-d, arrows, shift-d, more arrows, etc. It's an easy way to create rectangular shapes, very useful for diagrams. So what remains in node editor is implementing the []<> keys to edit control points, and voila - a totally keyboard-driven path editor! Something I've dreamed about for long, but did not see to date in any vector editor. P.S. Looks like my previous mail was lost, here it goes: Does anyone have an idea why sp_selection_idle_handler in selection.c is constantly emitting the "selection modified" signal - even two signals, one on the selection and the other on the inkscape object??? There is indeed a constant flood of these signals even when absolutely nothing is happening, not even a mouse move. Is this by design? I'm sure there must be a better way. I was investigating why undo and XML-editor changes in node edit do not work, and found that nodepath context lacks any callback that would update the path when its repr has changed. I tried to hook it up to the "modified" signal but that rendered the program unusable because that signal is "always on" like white noise. Can someone share any insights? It seems to me that something is seriously broken here. I think that it's the repr subsystem that must issue a signal when the document is changed, but that subsystem apparently emits no signals whatsoever. _________________________________________________________________ Don�t worry if your Inbox will max out while you are enjoying the holidays. Get MSN Extra Storage! http://join.msn.com/?PAGE=features/es

3 3

Re: [Inkscape-devel] docked toolbar steals focus
by bulia byak 08 Dec '03

08 Dec '03

> > >Grabbing focus between widgets is accomplished via > > >gtk_widget_grab_focus(). The widget will need to have the >GTK_CAN_FOCUS > > >flag set, however. (we should probably do this for the canvas) > > > > OK, I'll investigate the grabs mechanism. Maybe if the fix is simple it >will > > go into this version. > >I've taken care of the latter half in desktop.c now (setting the >CAN_FOCUS widget flag). > >I guess what remains is just determining when it would be appropriate to >call gtk_widget_grab_focus() on the canvas widget. I cannot make it work. I added gtk_widget_grab_focus (&(dtw->canvas->widget)); into sp_desktop_widget_new in desktop.c. For debugging, I also added gtk_signal_connect (GTK_OBJECT (&(dtw->canvas->widget)), "event", GTK_SIGNAL_FUNC (sp_canvas_event_handler), &(dtw->canvas->widget)); and put some debug output into sp_canvas_event_handler. With or without the grab, it looks like the canvas widget gets keypresses, but does not pass them on to the event context. That is, it does pass it to the event context but only when the mouse is within the canvas. Maybe we need to grab focus on the event context, not on the canvas? But how to do that? I think the event context is not a GTK widget. _________________________________________________________________ Shop online for kids� toys by age group, price range, and toy category at MSN Shopping. No waiting for a clerk to help you! http://shopping.msn.com

2 2

RE: [Inkscape-devel] Gtk Buttons
by bulia byak 07 Dec '03

07 Dec '03

Also a question: if we want to restore the drop-down button lists for some tools, as were used for e.g. shape tools in SP, can we do that with the standard widget? >From: MenTaLguY <mental@...3...> >To: Inkscape ML <inkscape-devel(a)lists.sourceforge.net> >Subject: [Inkscape-devel] Gtk Buttons >Date: Sat, 06 Dec 2003 01:30:33 -0500 > >It turns out that SPIcons weren't equipped to deal gracefully with >situations where they were allocated less space than they were >requested. I fixed this, as well as adjusting the buttons' requested >size slightly. > >While I was mucking around in there, I also removed the code that drew >random garbage in icons that weren't present (it turns out to have been >a deliberate decision in Sodipodi). > >Anyway, CVS HEAD now has standard Gtk buttons. > >-mental ><< signature.asc >> _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

4 3

docs/keybindings.txt
by bulia byak 07 Dec '03

07 Dec '03

That file is old. Can we set it up so that it's automatically updated from the KeyboardShortcuts wiki page? It's where I keep all shortcuts up-to-date, and it's much better categorized too. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] Marker screenshots
by bulia byak 07 Dec '03

07 Dec '03

>I put a screenshot and example svg of path arrowheads / markers on the >website... Markers are good, but have you seen my message "node edit news" with screenshots suggestions? I'd like to see them implemented instead of the second screenshot with stars. Sodipodi can do stars, too - but we need to show something only we can do :) _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

button.c
by bulia byak 07 Dec '03

07 Dec '03

Mental, your latest button.c only compiles if I change g_signal_handlers_block_by_func (G_OBJECT (button), G_CALLBACK (sp_button_perform_action), NULL); gtk_toggle_button_set_active (GTK_TOGGLE_BUTTON (button), down); g_signal_handlers_unblock_by_func (G_OBJECT (button), G_CALLBACK (sp_button_perform_action), NULL); by g_signal_handlers_block_by_func (G_OBJECT (button), (void*) G_CALLBACK (sp_button_perform_action), NULL); gtk_toggle_button_set_active (GTK_TOGGLE_BUTTON (button), down); g_signal_handlers_unblock_by_func (G_OBJECT (button), (void*) G_CALLBACK (sp_button_perform_action), NULL); Is that the right fix? _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Marker screenshots
by Bryce Harrington 07 Dec '03

07 Dec '03

I put a screenshot and example svg of path arrowheads / markers on the website... http://www.inkscape.org/screenshots/ Bryce

1 0

[Fwd: [Sodipodi-list] [Fwd: Sodipodi bug 758775]]
by MenTaLguY 07 Dec '03

07 Dec '03

-----Forwarded Message----- From: John Stracke <francis+keyword+sodipodi.6389c0@...99...> To: Sodipodi <sodipodi-list(a)lists.sourceforge.net> Subject: [Sodipodi-list] [Fwd: Sodipodi bug 758775] Date: Sat, 06 Dec 2003 12:00:29 -0500 (I sent this to Lauris at first, but maybe he's been too busy with the 0.33 release. Anybody else have commit access?) I've just uploaded a patch for sodipodi bug #758775 (the one about export bounding box not considering stroke width). It's not complete (it doesn't do text, and it's overeager--it takes the view that it's better to err in the direction of including extra blank space than to cut off some of the diagram), but I believe it's solid enough and useful enough to check in. https://sourceforge.net/tracker/?func=detail&atid=104054&aid=758775&group_i…

1 0

RE: [Inkscape-devel] Bug priorities
by bulia byak 07 Dec '03

07 Dec '03

850971: dupe of 840375 _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

2 1

[Fwd: [Sodipodi-list] [Fwd: Sodipodi bug 758775]]
by Jonathan Phillips 06 Dec '03

06 Dec '03

we have this problem too...its in our bug tracker... any takers... jon -----Forwarded Message----- From: John Stracke <francis+keyword+sodipodi.6389c0@...99...> To: Sodipodi <sodipodi-list(a)lists.sourceforge.net> Subject: [Sodipodi-list] [Fwd: Sodipodi bug 758775] Date: Sat, 06 Dec 2003 12:00:29 -0500 (I sent this to Lauris at first, but maybe he's been too busy with the 0.33 release. Anybody else have commit access?) I've just uploaded a patch for sodipodi bug #758775 (the one about export bounding box not considering stroke width). It's not complete (it doesn't do text, and it's overeager--it takes the view that it's better to err in the direction of including extra blank space than to cut off some of the diagram), but I believe it's solid enough and useful enough to check in. https://sourceforge.net/tracker/?func=detail&atid=104054&aid=758775&group_i… -- Jonathan Phillips <jon@...15...>

4 4

RE: [Inkscape-devel] Bug priorities
by bulia byak 06 Dec '03

06 Dec '03

842565 - fixed, please test and close bug 846963 - I think is fixed too 848355 - fixed temporarily, a better fix needs more work, but the bug may be closed because 852973 stays, and the better fix will have to fix them both 842564 - I would mark this "feature not a bug" 843123 - fixed 848694 - again feature not a bug, a discussion is on wiki (see CurrentColorWidget) 842627 - I cannot reproduce, anyone else? 853858 - this is what we discussed recently about copying <defs>, I think it's unlikely to be fixed before the release _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

Re: [Inkscape-devel] docked toolbar steals focus
by bulia byak 06 Dec '03

06 Dec '03

> > So the toolbar steals focus and won't let it go no matter how you try. >Since > > this involves moving focus not between windows but between widgets, I'm >not > > sure how to address this. > >Well, this basically arises as the buttons (by default) are set up to >accept keyboard focus on click, but the canvas widget is not. > >The toolbar buttons should probably not accept keyboard focus, though, >as they (almost) all have keyboard shortcuts anyway. I have fixed that >much in CVS. Yes, this problem is gone today - thanks a lot! >Grabbing focus between widgets is accomplished via >gtk_widget_grab_focus(). The widget will need to have the GTK_CAN_FOCUS >flag set, however. (we should probably do this for the canvas) OK, I'll investigate the grabs mechanism. Maybe if the fix is simple it will go into this version. http://sourceforge.net/tracker/index.php?func=detail&aid=854425&group_id=93… > > > > (see my comment at bottom). Basically we need all buttons to give focus >to > > the canvas at once, and all text fields must do the same on esc or enter > > (and not grab focus until you explicitly click on them). > >Well, in the case of dialogs we at least want basic standard navigation >keys like e.g. tab to advance focus to work, as well as accomodate any >IMs that might be active. With dialogs, keyboard focus now works more or less as it should: you can keyboard-navigate within a dialog, and Enter or Esc leave it. When we'll have spinbuttons and textfields in the aux toolbar, we'll need to make sure they behave consistently (tab for moving between widgets, esc to send focus to canvas, enter to set the value and send focus to canvas) and never grab focus unless clicked (or unless activated by a shortcut). As for shortcuts, I propose this: - all top-level menus must have hot-keys and be activated by alt-hotkey, e.g. alt-f must open the File menu - all text labels next to spinbuttons and textfields in the aux toolbars must also have hotkeys (underscored letters), and: -- either we make sure that these hotkeys never clash with the top menu hotkeys, and both are called by alt-hotkey -- or we can use independent hotkeys for the aux toolbar and use a different modifier (e.g. alt-shift-hotkey) for them Ideas? _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

Inkscape website above soft quota
by Bryce Harrington 06 Dec '03

06 Dec '03

There is a 100M soft-limit for project files on SourceForge. We're at 124M. 3.4M backup 0 cgi-bin 0 htdocs 9.1M inkscape 544K inkscape_project 106M inkscape_web 296K wiki 4.1M wiki_db 19M inkscape_web/cvs-snap 40M inkscape_web/doc 40M inkscape_web/doxygen 3.3M inkscape_web/inkscape_web 1.8M inkscape_web/magnethead 2.3M inkscape_web/screenshots Perhaps the doxygen files could be trimmed down? Or else maybe we should consider hosting them from another location? Bryce

1 0

Re: [Inkscape-devel] a few bugs I need tested...
by bulia byak 06 Dec '03

06 Dec '03

>I think I tend to prefer having the feature requests end up in Wiki, >since that allows them to be elaborated on, discussed, revised, >etc. which is valuable to do in order to make them implementable, and >that's exactly the sort of thing Wiki's are good for. The RFE tracker >works good as an "inbox" since it gives a consistent way for users to >submit to, along with bugs and patches. I think any small and unconnected requests may go to the tracker, but any larger requests that need discussion, or groups or interrelated requests, must be on wiki under "New features discussion" (hierarchized and categorized if possible). _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

1 0

Bug priorities
by Bryce Harrington 06 Dec '03

06 Dec '03

I've gone through the non-prioritized bugs in the SF bug tracker and assigned them priorities of 9, 6, or 3. We should not release until all priority 9 bugs are resolved. We should also seek to close as many of the priority 6 bugs as we can within the time available. Priority 3 bugs can be left until after the release, although many will be quick fixes and would be nice to get resolved. I've assigned a few bugs to people, and would encourage everyone to look through the list for ones they know how to fix, and take them if they're not taken already. There are seven priority 9 bugs, none of which have owners yet. Most have to do with saving/opening, and thus can impact user confidence a great deal. Many of the priority 6 bugs are crashes, and are thus important for that reason, even if they occur infrequently. Bryce

1 0

a few bugs I need tested...
by Jonathan Phillips 06 Dec '03

06 Dec '03

Hello, Well, I've finished transferring as many bugs that still have problem in the most recent build of inkscape-20031203, to the inkscape bug tracker on sf.net. There are still a few I need help with either because I do not know how to test them, or they require win32/mac os X. The ones to check from http://sf.net/projects/sodipodi bug tracker: #642384 Misrendered file -- text not aligned #834520 Crash when using a huge R1/R2 value for star shape #830980 LandScape Printing #768804 Sodi doesn't refresh font list; crashes reading deleted font #751439 UTF8 in font names creates problems Please report back to me which one you looked at and the outcome because I'm trying to keep track of the transfer with a spreadsheet. Also, if the bug still exists, please post the outcome on the bug tracker. Thanks, and after these 5 are looked into, the transfer of bugs is complete. Then we need to hack on them ;) Thanks, Jon -- Jonathan Phillips <jon@...15...>

2 3

Re: [Inkscape-devel] [Fwd: [Sodipodi-list] [Fwd: Sodipodi bug 758775]]
by bulia byak 06 Dec '03

06 Dec '03

> > It's actually not as simple as just taking the stroke width into > > account, either -- when you have mitred corners, they can stick out FAR > > beyond the stroke width. > >Markers too. > > > SVG filters will also make finding the real bounding box increasingly > > interesting, when we implement them. Maybe the renderer subsystem can give us the true bounding box of an object? I think this would make sense. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] rotate/scale transformation crash
by bulia byak 06 Dec '03

06 Dec '03

>Draw a line/curve >Bring up transform dialog >Enter a value to rotate or scale >Apply >Apply again. > >This causes a segfault for me. Anyone else? Works for me, no segfault _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

2 1

style.c crash
by bulia byak 06 Dec '03

06 Dec '03

Latest cvs crashes when doing convert to curves, I localized this crash to: /* markers */ p += g_snprintf (p, c + BMAX - p, "marker:%s", from->marker[SP_MARKER_LOC]); p += g_snprintf (p, c + BMAX - p, "marker-start:%s", from->marker[SP_MARKER_LOC_START]); p += g_snprintf (p, c + BMAX - p, "marker-mid:%s", from->marker[SP_MARKER_LOC_MID]); p += g_snprintf (p, c + BMAX - p, "marker-end:%s", from->marker[SP_MARKER_LOC_END]); in style.c. Bryce, I guess this is your recent code, so perhaps you can guess how to fix this. _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

2 2

Gtk Buttons
by MenTaLguY 06 Dec '03

06 Dec '03

It turns out that SPIcons weren't equipped to deal gracefully with situations where they were allocated less space than they were requested. I fixed this, as well as adjusting the buttons' requested size slightly. While I was mucking around in there, I also removed the code that drew random garbage in icons that weren't present (it turns out to have been a deliberate decision in Sodipodi). Anyway, CVS HEAD now has standard Gtk buttons. -mental

2 1

rotate/scale transformation crash
by Robert Crosbie 06 Dec '03

06 Dec '03

>From the latest build: Draw a line/curve Bring up transform dialog Enter a value to rotate or scale Apply Apply again. This causes a segfault for me. Anyone else? Cheers, Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] dockable toolbars
by bulia byak 06 Dec '03

06 Dec '03

>On Sat, 2003-12-06 at 01:07, bulia byak wrote: > > Why oh why can't they (gtk authors) make them transient by default??? > >They do set some of the same WM hints on them as on transients, though. >In Metacity, for example, they are treated more or less like transients. > >KWM obviously sees things differently. I'm not sure who, if anyone, is >to blame. > > > In which case, I wonder if it's actually better to go without >dockability, > > at least for this release... > >I'll leave that decision to you. If at least some WMs get it right, let's leave detachability in place. I'll add a note to the Release Notes for KDE users. I will also try to find out if I can file a bug with KDE or check if such bug is already filed. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Feature Freeze for 0.36
by Bryce Harrington 06 Dec '03

06 Dec '03

A lot of new features have been prepared for 0.36, and as proposed earlier I think it's time for us to freeze further new features and shift focus to fixing bugs and putting the polish on what we've done so far. I expect we will want to be in feature freeze mode for at least a few days, and perhaps up to a week, before finalizing and packaging the release. So what we need to focus on is: * Finalizing & thorough testing features added for this release * Documentation of new (and existing) features * Fixing all high importance bugs (with criticality scores of 6+) * Testing portability to Win32 and Mac OSX * Preparing for the 0.36 announcement (PR, website, screenshots, etc.) Bryce

1 0

[Fwd: Re: [Inkscape-devel] Gtk Buttons]
by MenTaLguY 06 Dec '03

06 Dec '03

For the list: what do you all think about the primary toolbar handle question? -----Forwarded Message----- From: Jonathan Phillips <jon@...15...> To: MenTaLguY <mental@...3...> Subject: Re: [Inkscape-devel] Gtk Buttons Date: Fri, 05 Dec 2003 23:02:12 -0800 THEY LOOK GR*! The difference between primary and secondary is now quite obvious. I understand that the nudge of the canvas will go away (happens when moving from tools with secondary options and those without) when we put options for all tools into the 2ndary toolbar. QUESTION: How should that weird space above the primary tool be dealt with. I didn't even realize that the primary toolbar was movable, and thought that space was just an accident until I clicked on it. It seems that there is a GTK error that doesn't show the shading on the handle of the primary toolbar. Thoughts on this? It is a problem that it is not obious that the primary is movable. That space where the handle is located, is more cosmetic, but looks unpolished. jon

2 1

docked toolbar steals focus
by bulia byak 06 Dec '03

06 Dec '03

Mental and whoever else can debug this, here's the problem: - start inkscape - press ctrl-arrows: canvas is scrolling (at least when the mouse is over canvas) - use e.g. F2/F1 keys to switch tools; ctrl-arrows still work - now click on any main toolbar button: ctrl-arrows don't work anymore EVEN after you click on the canvas! So the toolbar steals focus and won't let it go no matter how you try. Since this involves moving focus not between windows but between widgets, I'm not sure how to address this. The same applies to all other keys defined in event-context.c. Maybe we need to rewrite the handler in there somehow. Maybe there's a way to also fix the problem with these keys only working when the mouse is over the canvas. I could not fix this at my first try, now we need to focus on this because disabling keys that used to work is a bad usability problem. If you figure out how to fix this, perhaps you'll also be able to fix this bug: http://sourceforge.net/tracker/index.php?func=detail&aid=854425&group_id=93… (see my comment at bottom). Basically we need all buttons to give focus to the canvas at once, and all text fields must do the same on esc or enter (and not grab focus until you explicitly click on them). _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] Gtk Buttons
by bulia byak 06 Dec '03

06 Dec '03

WOW! One fix: in toolbox.c b = sp_button_new (size, type, action, tt); must be b = sp_button_new (size, (SPButtonType) type, action, tt); Also, there's now again some slight jumping in the aux toolbar when switching tools. Perhaps adjust aux_toolbox_size_request? >From: MenTaLguY <mental@...3...> >To: Inkscape ML <inkscape-devel(a)lists.sourceforge.net> >Subject: [Inkscape-devel] Gtk Buttons >Date: Sat, 06 Dec 2003 01:30:33 -0500 > >It turns out that SPIcons weren't equipped to deal gracefully with >situations where they were allocated less space than they were >requested. I fixed this, as well as adjusting the buttons' requested >size slightly. > >While I was mucking around in there, I also removed the code that drew >random garbage in icons that weren't present (it turns out to have been >a deliberate decision in Sodipodi). > >Anyway, CVS HEAD now has standard Gtk buttons. > >-mental ><< signature.asc >> _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] dockable toolbars
by bulia byak 06 Dec '03

06 Dec '03

> > On the other hand, I could not find a way to make the floating toolbars > > transient :( gtk_handlebox provides no clues on this whatsoever, and was > > unable to figure out the GTK_WINDOW of the detached toolbar. So at least >on > > KDE, toolbars sink miserably when floating which makes them almost >unusable. > >You'd need to dip into the internal GtkHandleBox structure, I think. Why oh why can't they (gtk authors) make them transient by default??? >GtkHandleBox sucks, to be honest. It was just the best I could do on >short notice. :/ In which case, I wonder if it's actually better to go without dockability, at least for this release... _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

status: standard GTK buttons
by MenTaLguY 06 Dec '03

06 Dec '03

I'm nearly done a patch that converts SPButton to a subclass of GtkToggleButton. Works great, with one issue -- apparently, an SPIcon widget refuses to paint itself when placed in a GtkButton. I'm going to keep trying to narrow down the problem, but if anyone has any ideas... -mental

1 0

RE: [Inkscape-devel] crash when moving with arrow keys
by bulia byak 05 Dec '03

05 Dec '03

>Yeah, that's what it was. The preferences had a wacked out value. I'd love to know how that could happen. Right now there's no code that sets this value, only reads it. Perhaps a bug somewhere in the repr system. For now I will create a function to read a pref value with min and max specified, which will return the default value if what is read is outside the limits. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Turn your development effort into Instant $$$!!
by Michael D. Seymour 05 Dec '03

05 Dec '03

Hey all, Looking to make some extra money? Mark Shuttleworth, the guy who got to the ISS as a Soyuz Cosmonaut is a big Python fan. Seems he wants to fund the development of Python scripting of "OpenOffice, Abiword, Gnumeric, Agnubis, Sodipodi and other Gnome applications", among other things: http://www.markshuttleworth.com/bounty.html Worth a look, even if other things have to take first priority. Truth, Mike S. -- Michael D. Seymour "Data is not the plural of anecdote" tel: (416) 545-1060 -- Robert L. Park e-mail: michael_d_seymour@...93... web: http://www.astro.queensu.ca/~seymour/

6 9

Re: [Inkscape-devel] DOM SVG Canvas
by Ian Grant 05 Dec '03

05 Dec '03

> The work Mental had been doing assumed C. Also I suspect serious > consideration would be given to keeping it in C if that would make it > easier to reuse in other apps. But we really haven't talked about it. I favour C. Does providing a C interface compromise any C++ class wrapper for that interface? Whilst I have heard of many C++ class interfaces wrapping C libraries I have never heard of the converse. Is it totally impossible because of C++ method names? -- Ian Grant, Computer Lab., William Gates Building, JJ Thomson Ave., Cambridge Phone: +44 1223 334420

6 8

RE: [Inkscape-devel] crash when moving with arrow keys
by Robert Crosbie 05 Dec '03

05 Dec '03

Yeah, that's what it was. The preferences had a wacked out value. id="nudgedistance" value="-1.07374e+09" /> Thanks, Rob. http://members.rogers.com/rcrosbie >From: "bulia byak" <archiver_1@...19...> >To: swingincelt@...19..., inkscape-devel(a)lists.sourceforge.net >Subject: RE: [Inkscape-devel] crash when moving with arrow keys >Date: Fri, 05 Dec 2003 23:09:02 +0000 > >I cannot reproduce this crash with the latest CVS, so I have to guess: > >1. if you have ~/.inkscape/preferences, delete it, restart inkscape > >2. if the problem persists, please send me your ~/.inkscape/preferences.xml >- i suspect there's something wrong with your options.nudgedistance value > >>I was just trying to use the arrow keys to move an object and inkscape >>crashed. Sucks because I lost the changes to the document I was working >>on. I did a few experiments and it seems that moving a selection with the >>arrow keys is broken. >> >>This line >>M 100 100 L 100 200 >>moved here after pressing right arrow or alt+right arrow >>M -1.34217e+09 100 L -1.34217e+09 200 >> >>Inkscape will likely freeze or crash if you try to move objects a few >>times with the arrow keys. >> >>Cheers, >>Rob. >>http://members.rogers.com/rcrosbie >> >>_________________________________________________________________ >>Help STOP SPAM with the new MSN 8 and get 2 months FREE* >>http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms… >> >> >> >>------------------------------------------------------- >>This SF.net email is sponsored by: SF.net Giveback Program. >>Does SourceForge.net help you be more productive? Does it >>help you create better code? SHARE THE LOVE, and help us help >>YOU! Click Here: http://sourceforge.net/donate/ >>_______________________________________________ >>Inkscape-devel mailing list >>Inkscape-devel(a)lists.sourceforge.net >>https://lists.sourceforge.net/lists/listinfo/inkscape-devel > >_________________________________________________________________ >Protect your PC - get McAfee.com VirusScan Online >http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

1 0

RE: [Inkscape-devel] crash when moving with arrow keys
by bulia byak 05 Dec '03

05 Dec '03

I cannot reproduce this crash with the latest CVS, so I have to guess: 1. if you have ~/.inkscape/preferences, delete it, restart inkscape 2. if the problem persists, please send me your ~/.inkscape/preferences.xml - i suspect there's something wrong with your options.nudgedistance value >I was just trying to use the arrow keys to move an object and inkscape >crashed. Sucks because I lost the changes to the document I was working >on. I did a few experiments and it seems that moving a selection with the >arrow keys is broken. > >This line >M 100 100 L 100 200 >moved here after pressing right arrow or alt+right arrow >M -1.34217e+09 100 L -1.34217e+09 200 > >Inkscape will likely freeze or crash if you try to move objects a few times >with the arrow keys. > >Cheers, >Rob. >http://members.rogers.com/rcrosbie > >_________________________________________________________________ >Help STOP SPAM with the new MSN 8 and get 2 months FREE* >http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms… > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

crash when moving with arrow keys
by Robert Crosbie 05 Dec '03

05 Dec '03

I was just trying to use the arrow keys to move an object and inkscape crashed. Sucks because I lost the changes to the document I was working on. I did a few experiments and it seems that moving a selection with the arrow keys is broken. This line M 100 100 L 100 200 moved here after pressing right arrow or alt+right arrow M -1.34217e+09 100 L -1.34217e+09 200 Inkscape will likely freeze or crash if you try to move objects a few times with the arrow keys. Cheers, Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] dockable toolbars
by bulia byak 05 Dec '03

05 Dec '03

>- The secondary toolbar needs more material and more cosmetic work. In >selector, its buttons should be divided into three groups by small spaces >or dividers: group buttons, z-order buttons, rotate and flip buttons. I did button grouping, added some spacing around and between groups. I think it's starting to look acceptable. Now we need to add more stuff there. On the other hand, I could not find a way to make the floating toolbars transient :( gtk_handlebox provides no clues on this whatsoever, and was unable to figure out the GTK_WINDOW of the detached toolbar. So at least on KDE, toolbars sink miserably when floating which makes them almost unusable. Any ideas? _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

RE: [Inkscape-devel] a few bugs I need tested...
by bulia byak 05 Dec '03

05 Dec '03

>#642384 Misrendered file -- text not aligned it's actually 837614. The file uses the font "Trebuchet MS" - if you don't have it installed and there's no alignment, no surprise. Try to find this font (in MS core fonts?) and test the file with it. >#834520 Crash when using a huge R1/R2 value for star shape Seems to be fixed - now the proportion filed only accepts a numver between 0 and 1. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

libnr precision and maximum zoom
by bulia byak 05 Dec '03

05 Dec '03

Following our year-ago sodipodi discussions, I tried to increase the maximum zoom. It did not work because at about 70x, paths were simply disappearing. So I went into libnr and simplistically replaced all floats with doubles and all shorts with ints. And it worked! Now I can draw at 256x and there does not seem to be any serious problems. I have uploaded the (very simple) patch to the patch tracker. Please test and report your experience. What should be done: - The patch must be reviewed by someone more knowledgeable than me. I don't quite understand what I did (even though it worked). - The precision of SVG files must also be increased. Currently it's 6 digits; will 8 digits be enough? - There are some nasty renderer bugs that are made more evident by this change. See this wiki page: http://www.inkscape.org/cgi-bin/wiki.pl?RendererIssues If someone can have a look at the bugs listed there it would be really great. _________________________________________________________________ Browse styles for all ages, from the latest looks to cozy weekend wear at MSN Shopping. And check out the beauty products! http://shopping.msn.com

1 0

jabber room (again)
by Kees Cook 05 Dec '03

05 Dec '03

I'm having trouble with jabber.org again. If other people are too, try "inkscape" in conference.gristle.org. The admin there says he fixed the permission trouble. (And if it still doesn't work, can people send me a private email with their jabber client and version?) -- Kees Cook @outflux.net

2 1

RE: [Inkscape-devel] dockable toolbars
by bulia byak 04 Dec '03

04 Dec '03

>- We need to decide if floating toolbars are to be shared among documents >(as are dialogs), or stay local to each document. FOr the latter, it's a >simple sp_transientize call on the floating toolbar window so it becomes >transient to the current document. For the former, it's a bit more complex, >see any of the dialogs I did recently for the code. I think floating >toolbars must behave as all other dialogs, i.e. be always transient to the >current document. On a second thought, no, since floating toolbar still remembers its "native" window and reappears docked there when you close it, it's better if a toolbar always stays local to its window, no matter docked or not. So we must find a way to make the floating toolbars transient to their native window only. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

dockable toolbars
by bulia byak 04 Dec '03

04 Dec '03

First, WOW! I did not expect the toolbars to be dockable so soon. Mental, you rock :) Second, if we want to make a release with this dockability, there are a couple of things that must be fixed asap. - We need to decide if floating toolbars are to be shared among documents (as are dialogs), or stay local to each document. FOr the latter, it's a simple sp_transientize call on the floating toolbar window so it becomes transient to the current document. For the former, it's a bit more complex, see any of the dialogs I did recently for the code. I think floating toolbars must behave as all other dialogs, i.e. be always transient to the current document. We'll also need to remember in prefs their docked/undocked status, size and position. - Both toolbars must either be not resizable (perhaps easier in the short term), or they must reflow the buttons so they all fit. Now you simply cut them off when resizing. - The secondary toolbar needs more material and more cosmetic work. In selector, its buttons should be divided into three groups by small spaces or dividers: group buttons, z-order buttons, rotate and flip buttons. Do you think I could start putting material into the secondary toolbar now, or should we better wait for Gtkmm? I want it to have at least a few non-button controls for the release so that people can have a better idea of what it's for. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

test
by bulia byak 04 Dec '03

04 Dec '03

test - please ignore _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

test
by bulia byak 04 Dec '03

04 Dec '03

test, please ignore

1 0

wiki updates -> help menu + similar applications
by Jonathan Phillips 04 Dec '03

04 Dec '03

I've added two items to the wiki. 1.) Under "User Interface Discussion," I added: "Help Menu." I laid out how the menu is currently, how it could possibly be changed to look, and also possible additions to it, to make it a "real" help menu. 2.) "Similar Application" after the "External Links" button. I went out on a limb and decided that we need screenshots of other apps functional parts. I guesss under these app pages, people could also list any feature(s) they really like as an abstract attempt to sway developers to add that functionality to Inkscape. Please post any screenshots, or links to screenshots, or links in general to other vector apps on the appropriate page (or make another one). I would like to help out with the help as soon as I can. I'm about done adding all these bugs to the bug tracker from sodi, and then after that, I would like to look into being the major player in helping with the help menu. I'm still green on OSS development, but helping in this way seems most plausible to get my feet wet. thx, Jon -- Jonathan Phillips <jon@...15...>

1 0

Re: [Inkscape-devel] adobe illustrator 10 screenshots (fwd)
by Alan Horkan 04 Dec '03

04 Dec '03

> I've started to put screenshots of illustrator 10 on my own personal > website. > > http://protofunk.org/projects/group/inkscape/reference/adobeillustrator/10/ > I think we should get shots of some other vector editing apps too, for > comparison sake. Dia screenshots http://matrix.netsoc.tcd.ie/~horkana/dia/ visio screenshots http://matrix.netsoc.tcd.ie/~horkana/dia/visio/screenshots/ I have a few of Adobe Illustrator and Jasc Web Draw screenshots lying around and the rest of my screenshots are from non vector graphics applictions but may still occasionally provide useful ideas so I'll try and put them up soon. Later Alan H.

1 0

Feature freeze soon?
by Bryce Harrington 04 Dec '03

04 Dec '03

How would everyone feel about a feature freeze on Friday, with a release as soon as we feel it is stable enough and all the critical bugs are fixed (hopefully within a week or so)? Also, how do we stand on the following bugs? Are some fixed or do they still need attention? * Fix bug: on quit, still says I have unsaved changes, even though I saved [swingincelt] * Fix bug: can't open compressed svg unless it has .svgz extension. [swingincelt] * Fix bug: autodetect fails if no extension is added to filename (.svg) [swingincelt] * Fix bug: save as/save should use "SVG with xmlns:sodipodi namespace" as default * Fix bug: files are sometimes erroneously saved without the extension namespaces (typically the first time) [tgould, mental] Thanks, Bryce

2 1

adobe illustrator 10 screenshots
by Jonathan Phillips 03 Dec '03

03 Dec '03

I've started to put screenshots of illustrator 10 on my own personal website. http://protofunk.org/projects/group/inkscape/reference/adobeillustrator/10/ Send me your requests. The shots might look slightly weird because I'm running illustrator 10 with crossover office 2.1.0, and the fonts look off or something. I'll post more screenshots later. I put one shot up of all the dialogs in illustrator which should be very useful. I'm having a problem getting screenshots of the menu items extended, but I'll figure out away to get shots of those as well. Thoughts? I think we should get shots of some other vector editing apps too, for comparison sake. Jon -- Jonathan Phillips <jon@...15...>

1 0

editing nodes
by Robert Crosbie 03 Dec '03

03 Dec '03

Here's a problem I noticed while editing nodes with the XML editor. Don't have time right now to submit a SF bug right now. Here's what I did: Draw a path Edited some points manually using XML dialog Moved different points using GUI The points I edited using XML dialog reverted back to their original values. Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Minor bug
by Bob Jamison 02 Dec '03

02 Dec '03

Hi, all. I know that I should be using SF's bug machine, but it appears to be heavily overloaded at the moment. I just checked out the CVS trunk, and built it. It is -really- looking good. The current key bindings for selecting a drawing tool are wonderful. However, neglecting to release a key before its corresponding tool is used causes side effects. For example: Hold down the 'p', and start drawing using the freehand tool. You will get an error and a crash: ** ERROR **: file draw-context.c: line 1142 (spdc_add_freehand_point): assertion failed: (dc->npoints > 0) aborting... A suggested fix for this particular problem would be ultra simple. Repeated keystrokes are hitting this function, but any invocations after the first are being processed without giving them a new point. (Which seems proper). Instead of failing hard, fail soft: At draw-context.c line 1142, replace: g_assert (dc->npoints > 0); with if (dc->npoints < 1) return; This 'soft fail' seems to be indempotent, and I don't see any side effects. g_assert() is a bit harsh, anyway, IMHO. HOWEVER, this just fixes a particular symptom. The general case of repeating keys during a draw operation needs attention. Keep up the good work! Bob Jamison Titan Corp.

2 2

doxygen & manpage
by Kees Cook 02 Dec '03

02 Dec '03

tvon: can you correct the groups for the inkscape_web/doc/doxygen directories/files to be "inkscape", please? gouldtj: can you update your update.dox.sh to get the latest copy from inkscape_project? I added automatic manpage generation to it. -- Kees Cook @outflux.net

2 2

button size
by MenTaLguY 01 Dec '03

01 Dec '03

I made the tool selection buttons along the side larger than the buttons for various commands up top. While I'd experimented with this a while back and abandoned it, it's since been suggested a number of times (most recently by Jon), and Xara appears to do so as well, I decided to try it again. See what y'all think... -mental

3 2

I'm here, I swear
by Ted Gould 01 Dec '03

01 Dec '03

Hey guys, I just wanted to say that I'm not MIA, I was just gone over the holiday weekend (in the US it was Thanksgiving). I'm now catching up on the ~120 devel messages and ~180 CVS messages... I've already seen a couple for me to look at - I'll hopefully get through all tonight. --Ted

1 0

Fonts problem with win32 build
by Mael 01 Dec '03

01 Dec '03

Hello, Concerning the fonts problem in the win32 build (fonts are rendered in italic), I think this is because of the function nr_type_calculate_position in nr-type-directory.c which uses the name of the font and string comparisons to test if it contains "italic" "bold", etc .. But for example on my french win32 fonts are named with "italique" instead of "italic" and "gras" instead of "bold" and so on, so this function gives wrong results. Sodipodi_0_33_1_Setup_Win2000_XP.exe seems to have the same problem excepted that it finds normal as default style instead of italic (so it's less visible :), when I try to set the style of the font to "gras" or "italique" it always goes back to "normal". M

1 0

select within group
by bulia byak 01 Dec '03

01 Dec '03

I want to enable ctrl-click to select within group, but I don't see how it can be done. I've delved deep into the arena code and found that if a group is "transparent" then its members can be selected one by one. But how can I do that on ctrl-click? In select-context.c, the handler that checks what button/key was pressed seems to already get a selected item as an argument from the arena code. Will I have to analyze this preselected item on my own, or is there a more elegant way? Please enlighten me! -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: [Inkscape-devel] button size
by bulia byak 01 Dec '03

01 Dec '03

For now I would vote for: #define TOOL_BUTTON_SIZE 28 #define AUX_BUTTON_SIZE 22 but I think we need to try to add more stuff to the secondary toolbar to see how it will look. > Well, as it is the toolbar size forces the window to a fairly large > minimum height. I'm a little uncomfortable forcing a much larger > minimum size. We have a lot of room to grow :) Currently the default window is very small - 400x400 if without the toolbar strut. BTW maybe we should increase this a bit, say to 640x480. But ultimately the size and maximized state of the document window should be remembered from previous session. > Do you think it would be okay to permit the button bar to be cut off if > the window is resized smaller? I think the main toolbar should not be cut (and it's not likely to be cut anyway), but the secondary toolbar must be permitted to be cut off if the window is small. > Well, perhaps I should use the GTK detachable toolbar widget there, > though the plain GTK one is kind of weak. :/ What is wrong with it? Is there an alternative? Maybe in Gtkmm? It would be nice to have this done without coding it from scratch :) -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Warnings for unimplemented SVG styles
by Bryce Harrington 01 Dec '03

01 Dec '03

I've put in a change to the codebase for some warning messages. Previously, if Inkscape encountered style properties that were not implemented, it would issue a generic warning such as: Unimplemented style property id: 1042 I've created specific warnings for each feature, so now it will report: Unimplemented style property id SP_PROP_FONT_SIZE_ADJUST: value: -4 This won't affect the quantity of warnings issued, just their specificity. I figured as we strive to gain better SVG compliance, we'll want to know more precisely what features we're missing. The warnings are all in style.c. If you're interested in trying your hand at adding a new SVG style property, have a look-see there at the sp_style_merge_property function. Bryce

1 0

Re: [Inkscape-devel] button size
by bulia byak 01 Dec '03

01 Dec '03

Also, I think it makes sense to establish a certain difference in style between the primary and secondary buttons. The primaries may be more 3Dish, with shades and highlights; the primaries should be much simpler, with less colors, without any highlights or gradients. This contrast need not be too drastic, but still it may be helpful. Thus, from this viewpoint, the current node editor secondaries are OK, while for the selector, only the three last buttons (rotate and flip) are OK, others must be made simpler, flatter and more laconic. The same goes for zoom secondaries which are currently "too beautifil". -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: [Inkscape-devel] button size
by Robert Crosbie 01 Dec '03

01 Dec '03

IMHO, I think the top bar could use the larger buttons too. Rob. http://members.rogers.com/rcrosbie >From: MenTaLguY <mental@...3...> >To: Inkscape ML <inkscape-devel(a)lists.sourceforge.net> >Subject: [Inkscape-devel] button size >Date: Sun, 30 Nov 2003 18:12:50 -0500 > >I made the tool selection buttons along the side larger than the buttons >for various commands up top. > >While I'd experimented with this a while back and abandoned it, it's >since been suggested a number of times (most recently by Jon), and Xara >appears to do so as well, I decided to try it again. > >See what y'all think... > >-mental ><< signature.asc >> _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

[Fwd: Re: [Sodipodi-list] Win32 0.33 prerelease beta2]
by MenTaLguY 30 Nov '03

30 Nov '03

For those of you working on the Inkscape Win32 stuff... (you might want to hit the sodipodi list archives and follow the whole thread though) -----Forwarded Message----- From: Mitsuru Oka <oka326@...90...> To: Lauris Kaplinski <lauris@...39...> Cc: sodipodi-list(a)lists.sourceforge.net Subject: Re: [Sodipodi-list] Win32 0.33 prerelease beta2 Date: Mon, 01 Dec 2003 04:20:09 +0900 Lauris Kaplinski <1070186315.8512.36.camel@...91...> wrote: >Hello! > >I wonder, if gtk+ is built with multibyte charset, and sodipodi >uses Unicode, is the result functional at all? >We can switch back to multibyte, if it is needed, although this makes >things much uglier and error-prone. Yeah, It's reasonable. I found that dirty hack also works on win98: Index: nr-type-w32.c =================================================================== RCS file: /cvs/gnome/sodipodi/src/libnrtype/nr-type-w32.c,v retrieving revision 1.18 diff -u -r1.18 nr-type-w32.c --- nr-type-w32.c 29 Nov 2003 23:50:52 -0000 1.18 +++ nr-type-w32.c 30 Nov 2003 19:15:43 -0000 @@ -15,6 +15,8 @@ #include <stdio.h> #include <tchar.h> +#include <glib.h> + #include <libarikkei/arikkei-dict.h> #include <libarikkei/arikkei-strlib.h> @@ -565,11 +567,16 @@ nr_type_w32_typefaces_enum_proc (LOGFONT *lfp, TEXTMETRIC *metrics, DWORD fontType, LPARAM lParam) { if (fontType == TRUETYPE_FONTTYPE) { - LOGFONT lf; + GSList *list; + LOGFONT *lf; + + lf = nr_new (LOGFONT, 1); + *lf = *lfp; - lf = *lfp; + list = g_slist_append (*(GSList **)lParam, lf); + *(GSList **)lParam = list; - EnumFontFamiliesEx (hdc, &lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, lParam, 0); +/* EnumFontFamiliesEx (hdc, &lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, lParam, 0); */ } return 1; @@ -578,12 +585,14 @@ static void nr_type_w32_init (void) { + GSList *font_que = NULL; LOGFONT logfont; int pos, i; /* g_print ("Loading W32 type directory...\n"); */ - hdc = CreateDC (L"DISPLAY", NULL, NULL, NULL); + /* hdc = CreateDC (L"DISPLAY", NULL, NULL, NULL); */ + hdc = CreateDCA ("DISPLAY", NULL, NULL, NULL); arikkei_dict_setup_string (&familydict, 131); arikkei_dict_setup_string (&namedict, 537); @@ -591,7 +600,14 @@ /* read system font directory */ memset (&logfont, 0, sizeof (LOGFONT)); logfont.lfCharSet = DEFAULT_CHARSET; - EnumFontFamiliesEx (hdc, &logfont, (FONTENUMPROC) nr_type_w32_typefaces_enum_proc, 0, 0); + EnumFontFamiliesEx (hdc, &logfont, (FONTENUMPROC) nr_type_w32_typefaces_enum_proc, (LPARAM)&font_que, 0); + + while (font_que) { + LOGFONT *lf; + lf = (LOGFONT*)font_que->data; + EnumFontFamiliesEx (hdc, lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, 0, 0); + font_que = g_slist_remove (font_que, (gpointer)lf); + } /* Fill in lists */ NRW32Families.length = families_len; Best regards,

1 0

Re: [Inkscape-devel] paste style
by bulia byak 30 Nov '03

30 Nov '03

> It seems like when copying to the clipboard, there needs to be a > mechanism to scan the selected objects for any referenced objects > and also copy the defs for those. The recipient of the paste > would then need to add those refs to its own set of defs. It's > possible that the receiving document could a) have a different def > by the same name, or b) have the same def by a different name; in > case (a) it would seem prudent to rename the incoming style, and > in case (b) perhaps it should normally merge by default but allow > the user an option to not merge when pasting. That sounds like a lot of code. Maybe we can simplify it a bit as follows: - a new clipboard is made for storing defs - whenever anything is cut or copied, the entire defs from the current document are placed on that defs clipboard - whenever clipboard objects (or style) are pasted, the contents of the defs clipboard are merged with the current document defs: -- same name, same content: disregard -- same name, different content: rename both defs and the pasted objects' references (this is the most tricky part) -- different name, same content: just add, no merge An advantage of this approach is that you don't need to scan your objects and select relevant defs when copying. However, you'll still have to scan and re-id what you are pasting in case of name conflicts (although these are likely to be rare). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

paste style
by bulia byak 30 Nov '03

30 Nov '03

2003-11-29 bulia byak <bulia@...23...> * src/selection-ch emistry.c src/selection-chemistry.h src/verbs.c src/verbs.h src/shortcuts.c src/interface.c: new function, verb, shortcut (shift-ctrl-v), menu command: paste style (apply the style of copied object to selection). If more than one object is copied, the first one in the clipboard list is taken as the source of the style. One problem with both paste and the new paste style: gradients are not copied between documents. This is because the CSS refers to a gradient which is stored separately in <defs>, and it's not copied or pasted when you copy or paste an object using it. The same will apply to bitmap fills and masks when we implement them. Did anyone think about how to solve this in the most generic and elegant way? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

3 2

g_free cleanup
by Peter Moulder 30 Nov '03

30 Nov '03

This is a low-priority cleanup that may be objectionable, so I'll post here first. According to the g_free documentation, g_free accepts a NULL argument (as a no-op), yet much code explicitly checks for NULL before calling g_free. Disadvantage of the transformation: - Involves a function call even in the NULL case (unless g_free is a macro or inline function that tests for NULL before passing to the "real" version). I've left a couple of places untransformed where it looks like NULL has a significant probability. If NULL is unlikely, then we're better off having more readable code [and saving a couple of bytes of object code]. - May make diff3 against sodipodi code harder. Strong candidates for applying the transformation are where the tested pointer is something long like `style->text->font_family.value', and src/module.c where it uses a IF_NOT_NULL_FREE macro. Btw, some (but not all) of the searching for this pattern was using a regexp to reduce the probability of accidentally transforming `if (foo) g_free (bar)' for distinct `foo' and `bar'. Any comments on where this transformation (removing the `if') should be applied? In absence of comments, I'll probably commit the patch in its current form (other than ChangeLog date). pjrm. Index: ChangeLog =================================================================== RCS file: /cvsroot/inkscape/inkscape/ChangeLog,v retrieving revision 1.56 diff -d -p -U5 -r1.56 ChangeLog --- ChangeLog 29 Nov 2003 09:20:28 -0000 1.56 +++ ChangeLog 30 Nov 2003 03:34:38 -0000 @@ -7,10 +7,33 @@ Make utest.c,utest.h formatting conform more to doc/coding-style.txt. Have utest.c built into libutest.a (see src/utest/Makefile.am comments). + * src/document.c (sp_document_new), file.c (file_open_ok, + sp_file_open_dialog, sp_file_save_dialog, sp_file_do_import), + main.c (sp_main_console), src/module.c (sp_module_finalize, + sp_module_input_finalize, sp_module_input_build, + sp_module_output_finalize, sp_module_output_build), + src/sp-anchor.c (sp_anchor_set), src/sp-image.c (sp_image_set), + src/sp-text.c (sp_string_release, sp_string_read_content, + sp_string_set_shape), src/sp-use.c (sp_use_release), src/style.c + (sp_style_unref, sp_style_merge_property, + sp_style_merge_from_parent, sp_text_style_unref, + sp_style_read_istring), src/bonobo/embeddable-document.c + (sp_bonobo_stream_read, sp_embeddable_document_ps_save), + src/dialogs/stroke-style.c (sp_stroke_style_scale_line), + src/dialogs/text-edit.c (sp_text_edit_dialog_text_changed), + src/helper/sodipodi-ctrl.c (sp_ctrl_build_cache), + src/modules/ps.c (sp_ps_print_image), + src/widgets/font-selector.c (sp_font_preview_set_phrase), + src/widgets/gradient-image.c (sp_gradient_image_size_allocate), + src/xml/repr-action.c (coalesce_chgattr, coalesce_chgcontent, + free_action), src/xml/repr.c (repr_finalize, sp_repr_merge): + Minor cleanup: replace `if (POINTER_EXPR) g_free (POINTER_EXPR)' + with the simpler (more readable) `g_free (POINTER_EXPR)'. + * src/xml/repr-action.c (reverse_log): Add doc comment. Rename some vars. * src/xml/repr-action.c (sp_repr_replay_log, sp_repr_undo_log), src/xml/repr-action.h, src/document-undo.c (sp_document_cancel, sp_document_redo, sp_document_undo): Remove the unused doc Index: src/document.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/document.c,v retrieving revision 1.9 diff -d -p -U5 -r1.9 document.c --- src/document.c 29 Nov 2003 21:55:57 -0000 1.9 +++ src/document.c 30 Nov 2003 03:34:38 -0000 @@ -334,12 +334,12 @@ sp_document_new (const gchar *uri, unsig name = g_strdup_printf (_("New document %d"), ++doc_count); } doc = sp_document_create (rdoc, uri, base, name, advertize, keepalive); - if (base) g_free (base); - if (name) g_free (name); + g_free (base); + g_free (name); return doc; } SPDocument * Index: src/file.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/file.c,v retrieving revision 1.10 diff -d -p -U5 -r1.10 file.c --- src/file.c 30 Nov 2003 03:29:01 -0000 1.10 +++ src/file.c 30 Nov 2003 03:34:38 -0000 @@ -99,11 +99,11 @@ file_open_ok (GtkWidget *widget, GtkFile gchar *filename; filename = g_strdup (gtk_file_selection_get_filename (fs)); if (filename && g_file_test (filename, G_FILE_TEST_IS_DIR)) { - if (open_path) g_free (open_path); + g_free (open_path); if (filename[strlen(filename) - 1] != G_DIR_SEPARATOR) { open_path = g_strconcat (filename, G_DIR_SEPARATOR_S, NULL); g_free (filename); } else { open_path = filename; @@ -112,11 +112,11 @@ file_open_ok (GtkWidget *widget, GtkFile return; } if (filename != NULL) { gchar* key; - if (open_path) g_free (open_path); + g_free (open_path); open_path = g_dirname (filename); if (open_path) open_path = g_strconcat (open_path, G_DIR_SEPARATOR_S, NULL); key = (gchar*)g_object_get_data (G_OBJECT (fs), "type-key"); sp_file_open (filename, key); g_free (filename); @@ -142,11 +142,11 @@ sp_file_open_dialog (gpointer object, gp { #ifdef WIN32 char *filename; filename = sp_win32_get_open_filename (open_path, "SVG files\0*.svg;*.svgz\0All files\0*\0", _("Select file to open")); if (filename) { - if (open_path) g_free (open_path); + g_free (open_path); open_path = g_dirname (filename); if (open_path) open_path = g_strconcat (open_path, G_DIR_SEPARATOR_S, NULL); sp_file_open (filename, NULL); g_free (filename); } @@ -199,11 +199,11 @@ sp_file_save_dialog (SPDocument *doc) char *filename; unsigned int spns; filename = sp_win32_get_save_filename (save_path, &spns); if (filename && *filename) { sp_file_do_save (doc, filename, (spns) ? SP_MODULE_KEY_OUTPUT_SVG_INKSCAPE : SP_MODULE_KEY_OUTPUT_SVG); - if (save_path) g_free (save_path); + g_free (save_path); save_path = g_dirname (filename); save_path = g_strdup (save_path); g_free (filename); } #else @@ -320,11 +320,11 @@ sp_file_do_import (SPDocument *doc, cons const gchar *e, *docbase, *relname; SPRepr * repr; SPReprDoc * rnewdoc; if (filename && g_file_test (filename, G_FILE_TEST_IS_DIR)) { - if (import_path) g_free (import_path); + g_free (import_path); if (filename[strlen(filename) - 1] != G_DIR_SEPARATOR) { import_path = g_strconcat (filename, G_DIR_SEPARATOR_S, NULL); } else { import_path = g_strdup (filename); } Index: src/main.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/main.c,v retrieving revision 1.9 diff -d -p -U5 -r1.9 main.c --- src/main.c 27 Nov 2003 06:14:27 -0000 1.9 +++ src/main.c 30 Nov 2003 03:34:38 -0000 @@ -372,11 +372,11 @@ sp_main_console (int argc, const char ** } } fl = g_slist_remove (fl, fl->data); } - if (printer) g_free (printer); + g_free (printer); inkscape_unref (); return 0; } Index: src/module.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/module.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 module.c --- src/module.c 16 Nov 2003 07:09:42 -0000 1.5 +++ src/module.c 30 Nov 2003 03:34:38 -0000 @@ -27,20 +27,10 @@ #include "document.h" #include "module.h" #include "modules/db.h" -/** - \def IF_NOT_NULL_FREE - - This is a quick little macro to check if a value is null, and - if it is not, then free the data. It's used in most of the - finalize functions, and then also when updating values. It - doesn't do that much, but it makes the code easier to read -*/ -#define IF_NOT_NULL_FREE(x) if ((x) != NULL) g_free((x)) - /* SPModule */ static void sp_module_class_init (SPModuleClass *klass); static void sp_module_init (SPModule *module); static void sp_module_finalize (GObject *object); @@ -52,11 +42,11 @@ static SPModule * sp_module_new (GType t /** \var module_parent_class This is the parent class for the modules. It should be - GObject - but no promises */ + GObject - but no promises */ static GObjectClass * module_parent_class; /** \return The type value for a SP Module \brief A quick way to get the type value for a SPModule @@ -157,12 +147,12 @@ sp_module_finalize (GObject *object) sp_module_db_unregister (module); if (module->repr) sp_repr_unref (module->repr); - IF_NOT_NULL_FREE(module->name); - IF_NOT_NULL_FREE(module->id); + g_free (module->name); + g_free (module->id); G_OBJECT_CLASS (module_parent_class)->finalize (object); return; } @@ -285,11 +275,11 @@ sp_module_unload_default (SPModule * mod should be created. \param repr The XML structure of the module definition. This function is used to build an SPModule (really a subclass of that though) from an XML description stored in a SPRepr tree. If the class - has it's own build function, then that function is used. Otherwise + has its own build function, then that function is used. Otherwise the generic 'sp_module_private_build' function is used. A reference to the SPRepr structure is added in this function so it should not need to be added by each individual subclass. */ @@ -424,14 +414,14 @@ sp_module_input_finalize (GObject *objec { SPModuleInput *imod; imod = (SPModuleInput *) object; - IF_NOT_NULL_FREE(imod->mimetype); - IF_NOT_NULL_FREE(imod->extension); - IF_NOT_NULL_FREE(imod->filetypename); - IF_NOT_NULL_FREE(imod->filetypetooltip); + g_free (imod->mimetype); + g_free (imod->extension); + g_free (imod->filetypename); + g_free (imod->filetypetooltip); G_OBJECT_CLASS (input_parent_class)->finalize (object); return; } @@ -470,23 +460,23 @@ sp_module_input_build (SPModule *module, while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "input")) { child_repr = sp_repr_children(child_repr); while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "extension")) { - IF_NOT_NULL_FREE(imod->extension); + g_free (imod->extension); imod->extension = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "mimetype")) { - IF_NOT_NULL_FREE(imod->mimetype); + g_free (imod->mimetype); imod->mimetype = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypename")) { - IF_NOT_NULL_FREE(imod->filetypename); + g_free (imod->filetypename); imod->filetypename = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypetooltip")) { - IF_NOT_NULL_FREE(imod->filetypetooltip); + g_free (imod->filetypetooltip); imod->filetypetooltip = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } child_repr = sp_repr_next(child_repr); } @@ -623,14 +613,14 @@ sp_module_output_finalize (GObject *obje { SPModuleOutput *omod; omod = (SPModuleOutput *) object; - IF_NOT_NULL_FREE(omod->mimetype); - IF_NOT_NULL_FREE(omod->extension); - IF_NOT_NULL_FREE(omod->filetypename); - IF_NOT_NULL_FREE(omod->filetypetooltip); + g_free (omod->mimetype); + g_free (omod->extension); + g_free (omod->filetypename); + g_free (omod->filetypetooltip); G_OBJECT_CLASS (output_parent_class)->finalize (object); } /** @@ -667,23 +657,23 @@ sp_module_output_build (SPModule *module while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "output")) { child_repr = sp_repr_children(child_repr); while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "extension")) { - IF_NOT_NULL_FREE(omod->extension); + g_free (omod->extension); omod->extension = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "mimetype")) { - IF_NOT_NULL_FREE(omod->mimetype); + g_free (omod->mimetype); omod->mimetype = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypename")) { - IF_NOT_NULL_FREE(omod->filetypename); + g_free (omod->filetypename); omod->filetypename = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypetooltip")) { - IF_NOT_NULL_FREE(omod->filetypetooltip); + g_free (omod->filetypetooltip); omod->filetypetooltip = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } child_repr = sp_repr_next(child_repr); } Index: src/sp-anchor.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-anchor.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-anchor.c --- src/sp-anchor.c 2 Nov 2003 00:46:23 -0000 1.2 +++ src/sp-anchor.c 30 Nov 2003 03:34:38 -0000 @@ -130,11 +130,11 @@ sp_anchor_set (SPObject *object, unsigne anchor = SP_ANCHOR (object); switch (key) { case SP_ATTR_XLINK_HREF: - if (anchor->href) g_free (anchor->href); + g_free (anchor->href); anchor->href = g_strdup (value); sp_object_request_modified (object, SP_OBJECT_MODIFIED_FLAG); break; case SP_ATTR_XLINK_TYPE: case SP_ATTR_XLINK_ROLE: Index: src/sp-image.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-image.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-image.c --- src/sp-image.c 31 Oct 2003 07:05:50 -0000 1.2 +++ src/sp-image.c 30 Nov 2003 03:34:38 -0000 @@ -171,11 +171,11 @@ sp_image_set (SPObject *object, unsigned image = SP_IMAGE (object); switch (key) { case SP_ATTR_XLINK_HREF: - if (image->href) g_free (image->href); + g_free (image->href); image->href = (value) ? g_strdup (value) : NULL; sp_object_request_update (object, SP_OBJECT_MODIFIED_FLAG | SP_IMAGE_HREF_MODIFIED_FLAG); break; case SP_ATTR_X: if (sp_svg_length_read_lff (value, &unit, &image->x.value, &image->x.computed) && Index: src/sp-text.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-text.c,v retrieving revision 1.6 diff -d -p -U5 -r1.6 sp-text.c --- src/sp-text.c 16 Nov 2003 23:37:19 -0000 1.6 +++ src/sp-text.c 30 Nov 2003 03:34:39 -0000 @@ -132,12 +132,12 @@ sp_string_release (SPObject *object) { SPString *string; string = SP_STRING (object); - if (string->p) g_free (string->p); - if (string->text) g_free (string->text); + g_free (string->p); + g_free (string->text); if (((SPObjectClass *) string_parent_class)->release) ((SPObjectClass *) string_parent_class)->release (object); } @@ -149,13 +149,13 @@ sp_string_read_content (SPObject *object SPString *string; const gchar *t; string = SP_STRING (object); - if (string->p) g_free (string->p); + g_free (string->p); string->p = NULL; - if (string->text) g_free (string->text); + g_free (string->text); t = sp_repr_content (object->repr); string->text = (t) ? g_strdup (t) : NULL; /* Is this correct? I think so (Lauris) */ /* Virtual method will be invoked BEFORE signal, so we can update there */ @@ -295,11 +295,11 @@ sp_string_set_shape (SPString *string, S sp_chars_clear (chars); if (!string->text || !*string->text) return; len = g_utf8_strlen (string->text, -1); if (!len) return; - if (string->p) g_free (string->p); + g_free (string->p); string->p = g_new (NRPointF, len + 1); /* fixme: Adjusted value (Lauris) */ size = style->font_size.computed; face = nr_type_directory_lookup_fuzzy (style->text->font_family.value, sp_text_font_style_to_lookup (style)); Index: src/sp-use.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-use.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-use.c --- src/sp-use.c 2 Nov 2003 00:30:53 -0000 1.2 +++ src/sp-use.c 30 Nov 2003 03:34:39 -0000 @@ -147,11 +147,11 @@ sp_use_release (SPObject *object) if (use->child) { use->child = sp_object_detach_unref (SP_OBJECT (object), use->child); } - if (use->href) g_free (use->href); + g_free (use->href); if (((SPObjectClass *) parent_class)->release) ((SPObjectClass *) parent_class)->release (object); } Index: src/style.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/style.c,v retrieving revision 1.4 diff -d -p -U5 -r1.4 style.c --- src/style.c 16 Nov 2003 23:37:19 -0000 1.4 +++ src/style.c 30 Nov 2003 03:34:39 -0000 @@ -259,13 +259,11 @@ sp_style_unref (SPStyle *style) /* if (style->object) gtk_signal_disconnect_by_data (GTK_OBJECT (style->object), style); */ if (style->object) g_signal_handlers_disconnect_matched (G_OBJECT(style->object), G_SIGNAL_MATCH_DATA, 0, 0, NULL, NULL, style); if (style->text) sp_text_style_unref (style->text); sp_style_paint_clear (style, &style->fill, TRUE, FALSE); sp_style_paint_clear (style, &style->stroke, TRUE, FALSE); - if (style->stroke_dash.dash) { - g_free (style->stroke_dash.dash); - } + g_free (style->stroke_dash.dash); g_free (style); } return NULL; } @@ -472,11 +470,11 @@ sp_style_merge_property (SPStyle *style, SPS_READ_IENUM_IF_UNSET (&style->font_stretch, val, enum_font_stretch, TRUE); break; case SP_PROP_FONT: if (!style->text_private) sp_style_privatize_text (style); if (!style->text->font.set) { - if (style->text->font.value) g_free (style->text->font.value); + g_free (style->text->font.value); style->text->font.value = g_strdup (val); style->text->font.set = TRUE; style->text->font.inherit = (val && !strcmp (val, "inherit")); } break; @@ -812,11 +810,11 @@ sp_style_merge_from_parent (SPStyle *sty style->writing_mode.computed = parent->writing_mode.computed; } if (style->text && parent->text) { if (!style->text->font_family.set || style->text->font_family.inherit) { - if (style->text->font_family.value) g_free (style->text->font_family.value); + g_free (style->text->font_family.value); style->text->font_family.value = g_strdup (parent->text->font_family.value); } } } @@ -1222,12 +1220,12 @@ static SPTextStyle * sp_text_style_unref (SPTextStyle *st) { st->refcount -= 1; if (st->refcount < 1) { - if (st->font.value) g_free (st->font.value); - if (st->font_family.value) g_free (st->font_family.value); + g_free (st->font.value); + g_free (st->font_family.value); g_free (st); } return NULL; } @@ -1316,11 +1314,11 @@ sp_style_read_ienum (SPIEnum *val, const } static void sp_style_read_istring (SPIString *val, const gchar *str) { - if (val->value) g_free (val->value); + g_free (val->value); if (!strcmp (str, "inherit")) { val->set = TRUE; val->inherit = TRUE; val->value = NULL; Index: src/bonobo/embeddable-document.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/bonobo/embeddable-document.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 embeddable-document.c --- src/bonobo/embeddable-document.c 30 Oct 2003 05:52:06 -0000 1.2 +++ src/bonobo/embeddable-document.c 30 Nov 2003 03:34:39 -0000 @@ -132,11 +132,11 @@ sp_bonobo_stream_read (Bonobo_Stream str do { Bonobo_Stream_read (stream, STREAM_CHUNK_SIZE, &iobuf, &ev); if (ev._major != CORBA_NO_EXCEPTION) { - if (* buffer != NULL) g_free (* buffer); + g_free (* buffer); len = -1; break; } * buffer = g_realloc (* buffer, len + iobuf->_length); @@ -204,11 +204,11 @@ sp_embeddable_document_ps_save (BonoboPe document = SP_EMBEDDABLE_DOCUMENT (closure); /* FIXME: super dirty but functional */ do { - if (filename) g_free (filename); + g_free (filename); filename = g_strdup ("inkscapeXXXXXX"); fd = mkstemp (filename); } while (fd == -1); Index: src/dialogs/stroke-style.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/dialogs/stroke-style.c,v retrieving revision 1.20 diff -d -p -U5 -r1.20 stroke-style.c --- src/dialogs/stroke-style.c 29 Nov 2003 22:03:24 -0000 1.20 +++ src/dialogs/stroke-style.c 30 Nov 2003 03:34:39 -0000 @@ -1101,11 +1101,11 @@ sp_stroke_style_scale_line (SPWidget *sp g_snprintf (c, 32, "%g", dist); sp_repr_css_set_property (css, "stroke-width", c); /* Set dash */ sp_stroke_style_set_scaled_dash (css, ndash, dash, offset, dist); sp_repr_css_change_recursive (SP_OBJECT_REPR (i->data), css, "style"); - if (dash) g_free (dash); + g_free (dash); } } else { for (r = reprs; r != NULL; r = r->next) { double length; double *dash, offset; @@ -1115,11 +1115,11 @@ sp_stroke_style_scale_line (SPWidget *sp sp_convert_distance (&length, sp_unit_selector_get_unit (us), SP_PS_UNIT); g_snprintf (c, 32, "%g", length * 1.25); sp_repr_css_set_property (css, "stroke-width", c); sp_stroke_style_set_scaled_dash (css, ndash, dash, offset, length); sp_repr_css_change_recursive ((SPRepr *) r->data, css, "style"); - if (dash) g_free (dash); + g_free (dash); } } sp_repr_css_attr_unref (css); if (spw->inkscape) sp_document_done (SP_WIDGET_DOCUMENT (spw)); Index: src/dialogs/text-edit.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/dialogs/text-edit.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 text-edit.c --- src/dialogs/text-edit.c 27 Nov 2003 00:55:38 -0000 1.5 +++ src/dialogs/text-edit.c 30 Nov 2003 03:34:39 -0000 @@ -637,11 +637,11 @@ sp_text_edit_dialog_text_changed (GtkTex if (str && *str) { sp_font_preview_set_phrase (SP_FONT_PREVIEW (preview), str); } else { sp_font_preview_set_phrase (SP_FONT_PREVIEW (preview), NULL); } - if (str) g_free (str); + g_free (str); if (text) { gtk_widget_set_sensitive (apply, TRUE); } gtk_widget_set_sensitive (def, TRUE); Index: src/helper/sodipodi-ctrl.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/helper/sodipodi-ctrl.c,v retrieving revision 1.3 diff -d -p -U5 -r1.3 sodipodi-ctrl.c --- src/helper/sodipodi-ctrl.c 30 Oct 2003 11:33:16 -0000 1.3 +++ src/helper/sodipodi-ctrl.c 30 Nov 2003 03:34:39 -0000 @@ -298,11 +298,11 @@ sp_ctrl_build_cache (SPCtrl *ctrl) } else { sr = fr; sg = fg; sb = fb; sa = fa; } side = (ctrl->span * 2 +1); c = ctrl->span ; - if (ctrl->cache) g_free (ctrl->cache); + g_free (ctrl->cache); size = (side) * (side) * 4; ctrl->cache = (guchar*)g_malloc (size); if (side < 2) return; switch (ctrl->shape) { Index: src/modules/ps.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/modules/ps.c,v retrieving revision 1.8 diff -d -p -U5 -r1.8 ps.c --- src/modules/ps.c 16 Nov 2003 23:37:19 -0000 1.8 +++ src/modules/ps.c 30 Nov 2003 03:34:39 -0000 @@ -854,12 +854,12 @@ sp_ps_print_image (FILE *ofp, guchar *px #if 0 fprintf (ofp, "showpage\n"); g_free (data); #endif - if (packb != NULL) g_free (packb); - if (plane != NULL) g_free (plane); + g_free (packb); + g_free (plane); #if 0 if (ferror (ofp)) { g_message (_("write error occured")); Index: src/widgets/font-selector.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/widgets/font-selector.c,v retrieving revision 1.3 diff -d -p -U5 -r1.3 font-selector.c --- src/widgets/font-selector.c 16 Nov 2003 23:37:19 -0000 1.3 +++ src/widgets/font-selector.c 30 Nov 2003 03:34:39 -0000 @@ -617,11 +617,11 @@ sp_font_preview_set_rgba32 (SPFontPrevie } void sp_font_preview_set_phrase (SPFontPreview *fprev, const gchar *phrase) { - if (fprev->phrase) g_free (fprev->phrase); + g_free (fprev->phrase); if (phrase) { fprev->phrase = g_strdup (phrase); } else { fprev->phrase = NULL; } Index: src/widgets/gradient-image.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/widgets/gradient-image.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 gradient-image.c --- src/widgets/gradient-image.c 30 Oct 2003 20:48:05 -0000 1.2 +++ src/widgets/gradient-image.c 30 Nov 2003 03:34:39 -0000 @@ -149,11 +149,11 @@ sp_gradient_image_size_allocate (GtkWidg image = SP_GRADIENT_IMAGE (widget); widget->allocation = *allocation; if (GTK_WIDGET_REALIZED (widget)) { - if (image->px) g_free (image->px); + g_free (image->px); image->px = g_new (guchar, 3 * VBLOCK * allocation->width); } sp_gradient_image_update (image); } Index: src/xml/repr-action.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/xml/repr-action.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 repr-action.c --- src/xml/repr-action.c 29 Nov 2003 02:55:37 -0000 1.5 +++ src/xml/repr-action.c 30 Nov 2003 03:34:39 -0000 @@ -228,12 +228,11 @@ coalesce_chgattr (SPReprAction *action) { /* ensure changes are continuous */ if (strcmp(action->act.chgattr.oldval, iter->act.chgattr.newval)) break; - if (action->act.chgattr.oldval) - g_free (action->act.chgattr.oldval); + g_free (action->act.chgattr.oldval); action->act.chgattr.oldval = iter->act.chgattr.oldval; iter->act.chgattr.oldval = NULL; free_action (iter); prev->next = next; @@ -262,12 +261,11 @@ coalesce_chgcontent (SPReprAction *actio if ( action->repr == iter->repr ) { /* ensure changes are continuous */ if (strcmp(action->act.chgcontent.oldval, iter->act.chgcontent.newval)) break; - if (action->act.chgcontent.oldval) - g_free (action->act.chgcontent.oldval); + g_free (action->act.chgcontent.oldval); action->act.chgcontent.oldval = iter->act.chgcontent.oldval; iter->act.chgcontent.oldval = NULL; @@ -429,20 +427,16 @@ free_action (SPReprAction *action) sp_repr_unref (action->act.del.child); if (action->act.del.ref) sp_repr_unref (action->act.del.ref); break; case SP_REPR_ACTION_CHGATTR: - if (action->act.chgattr.oldval) - g_free (action->act.chgattr.oldval); - if (action->act.chgattr.newval) - g_free (action->act.chgattr.newval); + g_free (action->act.chgattr.oldval); + g_free (action->act.chgattr.newval); break; case SP_REPR_ACTION_CHGCONTENT: - if (action->act.chgcontent.oldval) - g_free (action->act.chgcontent.oldval); - if (action->act.chgcontent.newval) - g_free (action->act.chgcontent.newval); + g_free (action->act.chgcontent.oldval); + g_free (action->act.chgcontent.newval); break; case SP_REPR_ACTION_CHGORDER: sp_repr_unref (action->act.chgorder.child); if (action->act.chgorder.oldref) sp_repr_unref (action->act.chgorder.oldref); Index: src/xml/repr.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/xml/repr.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 repr.c --- src/xml/repr.c 16 Nov 2003 23:37:20 -0000 1.5 +++ src/xml/repr.c 30 Nov 2003 03:34:39 -0000 @@ -173,11 +173,11 @@ repr_finalize (SPRepr *repr) for (rl = repr->listeners; rl; rl = rl->next) { if (rl->vector->destroy) (* rl->vector->destroy) (repr, rl->data); } while (repr->children) sp_repr_remove_child (repr, repr->children); while (repr->attributes) sp_repr_remove_attribute (repr, repr->attributes); - if (repr->content) g_free (repr->content); + g_free (repr->content); while (repr->listeners) sp_repr_remove_listener (repr, repr->listeners); } static void repr_doc_finalize (SPRepr *repr) @@ -859,14 +859,14 @@ sp_repr_merge (SPRepr *repr, const SPRep g_return_val_if_fail (repr != NULL, FALSE); g_return_val_if_fail (src != NULL, FALSE); g_return_val_if_fail (key != NULL, FALSE); if (src->content) { - if (repr->content) g_free (repr->content); + g_free (repr->content); repr->content = g_strdup (src->content); } else { - if (repr->content) g_free (repr->content); + g_free (repr->content); repr->content = NULL; } for (child = src->children; child != NULL; child = child->next) { SPRepr *rch;

3 2

New Inkscape man page
by Bryce Harrington 30 Nov '03

30 Nov '03

Here's my first draft of a new man page for Inkscape. The old one as mostly just a list of options, most of which were invalid. I created this as POD 'inkscape.pod' and generated the man page 'inkscape.1.in' by running pod2man and changing "...Perl Documentation..." to "GNOME Documentation". This is set up to be installed when you run 'make install', so after installation you can view it via 'man inkscape'. Bryce INKSCAPE(1) GNOME Documentation INKSCAPE(1) NAME Inkscape - a 2D vector graphics editing program. SYNOPSIS "inkscape [options] [filename ...]" options: -?, --help -a, --export-area=x0:y0:x1:y1 -b, --export-background=COLOR -d, --export-dpi=DPI -e, --export-png=FILENAME -f, --file=FILENAME -h, --export-height=HEIGHT -i, --bitmap-icons -p, --print=FILENAME -s, --slideshow -w, --export-width=WIDTH -x, --with-gui -z, --without-gui --export-svg=FILENAME --usage DESCRIPTION Inkscape is a GUI editor for Scalable Vector Graphics (SVG) format drawing files, with capabilities similar to Illustrator, CorelDraw, Visio, etc. Supported SVG features include basic shapes, paths, text, alpha blending, transforms, gradients, node editing, svg-to-png export, grouping, and more. . The interface is designed to be comfortable and efficient for skilled users, while remaining conformant to GNOME stan- dards so that users familiar with other GNOME applications can learn it's interface rapidly. SVG is a W3C standard XML format for 2D vector drawing, that allows defining objects in the drawing using points, paths, and primitive shapes. Colors, fonts, line thicknesses, and so forth are specified as 'style' attributes to these objects. The intent is that since SVG is a standard, and since its files are text/xml instead of binary, it is be possible to use SVG files in a sizeable number of programs and for a wide range of uses. Inkscape uses SVG as its native document format, and has the goal of becoming the most fully compliant drawing program for SVG files avail- able in the Open Source community. OPTIONS -?, --help Show help message -a x0:y0:x1:y1, --export-area=x0:y0:x1:y1 Exported area in millimeters (default is full document, 0,0 is lower-left corner) -b COLOR, --export-background=COLOR Background color of exported bitmap (any SVG supported color string) -d DPI, --export-dpi=DPI The resolution used for converting SVG into bitmap (default 72.0) -e FILENAME, --export-png=FILENAME Export document to png file -f FILENAME, --file=FILENAME Open specified document(s) (option string may be excluded) -h HEIGHT, --export-height=HEIGHT The height of generated bitmap in pixels (overwrites dpi) -i, --bitmap-icons Prefer bitmap (xpm) icons to SVG ones -p FILENAME, --print=FILENAME Print document(s) to specified output file (use '| program' for pipe) -s, --slideshow Show given files one-by-one, switch to next on any key/mouse event -w WIDTH, --export-width=WIDTH The width of generated bitmap in pixels (overwrites dpi) -x, --with-gui Try to use X server even if $DISPLAY is not set) -z, --without-gui Do not open GUI and do not use X server (only process files from console). Useful for printing SVG documents. --export-svg=FILENAME Export document to plain SVG file (no "xmlns:sodipodi" names- pace) --usage Display brief usage message KEYBINDINGS See keybindings.txt for a listing of keybindings. DIAGNOSTICS The program returns zero on success or non-zero on failure. A variety of error messages and warnings are printed to STDERR or STD- OUT. If the program behaves erratically with a particular SVG file or crashes, it is sometimes useful to look at this output for clues. For example, it may indicate if a given SVG file contains elements or attributes that are not yet supported by the codebase. EXAMPLES While obviously Inkscape is primarily intended as a GUI application, it can be used for doing SVG processing on the commandline as well. (Note though that other SVG tools and libraries such as librsvg may be better suited for production SVG processing.) Printing an SVG file directly: inkscape filename.svg -p | lpr Converting SVG file to a png: inkscape --export-png=filename.png --export-height=400 --export-width=600 Convert an SVG document using items in the sodipodi namespace to plain SVG: inkscape filename1.svg e --export-svg=filename2.svg ENVIRONMENT DISPLAY to get the default host and display number. FILES $HOME/.inkscape/preferences.xml - The user's preference settings. $HOME/.inkscape/extensions.xml - The filter programs to be used in the application. OTHER INFO The canonical place to find Inkscape info is at http://www.inkscape.org/. The website includes links to other relevant documentation, tutorials, user manual, examples, mailing list archives, the latest released version of the program, and more. SEE ALSO gimp(1) NOTES Ctrl & Shift modify dragging behaviour in some modes Shift click/rubberband usually toggles selection Snapping to guidelines/grid work only for certain modes To break line (or add node) select its endpoints It should import png, bmp, jpg, xpm & gif bitmaps and svg vector images. Open can handle only svg graphics BUGS When importing svg generated with ill2svg, all paths are usually grouped. Ungroup these manually. Many bugs are known; please refer to the website for reviewing the reported ones and to report newly found issues. AUTHORS This codebase owes its existance to a large number of contributors throughout its various incarnations. The following list is certainly incomplete, but serves to recognize the many shoulders on which this application sits: John Bintz, Hans Breuer, Bulia Byak, Chema Celorio, Zbigniew Chyla, Kees Cook, Robert Crosbie, Jon Cruz, Danilo Egan, Frank Felfe, Ted Gould, Bryce Harrington, Nathan Hurst, Thomas Ingham, Bob Jamison, Lau- ris Kaplinski, Lynn Kerby, Petr Kovar, Raph Levien, Vitaly Lipatov, Dmitry G. Mastrukov, Michael Meeks, Frederico Mena, MenTaLguY, Yukihiro Nakai, Christian Neumair, Mitsuru Oka, Christian Schaller, Tom von Schwerdtner, Daniel Yacob, Masatake Yamato, This man page was put together by Bryce Harrington <brycehar@...88... rington.com> HISTORY The codebase that would become Inkscape began life in 1999 as the pro- gram 'GILL', the GNOME Illustrator application, created by Raph Levien. The stated objective for Gill was to eventually support all of SVG. Raph implemented the PostScript bezier imaging model, including stroking and filling, line cap style, line join style, text, etc. A Bonobo Raph's Gill page is at http://www.levien.com/svg/. Work on Gill appears to have slowed or ceased in 2000, perhaps in part due to frus- trations dealing with implementing DOM and SVG. The next incarnation of the codebase was to become the highly popular program Sodipodi, led by Lauris Kaplinkski. The codebase was turned into a very usable illustration program over the course of several year's work, adding several new features, multi-lingual support, port- ing to Windows and other operating systems, and eliminating dependen- cies. Inkscape was formed in 2003 by a number of active Sodipodi developers wanting to take a different direction with the codebase in terms of focus on SVG-compliance, interface look-and-feel, and a desire to open development opportunities to more participants. In a way, this direc- tion marks a return to the codebase's Gill roots. COPYRIGHT AND LICENSE Copyright (C) 1999, 2000, 2001, 2002, 2003 by Authors. Inkscape is free software; you can redistribute it and/or modify it under the terms of the GPL. Inkscape 0.36cvs 2003-11-29 INKSCAPE(1)

2 2

Re: GIMP user interface discussion
by bulia byak 30 Nov '03

30 Nov '03

> > Of course it would be better! So, are you doing it already? When can we expect to see the unsinkability option for utility windows in, say, KDE? > > Well, you're using KDE not me ;) Like I said, Gimp has a much higher profile than us currently. If you push the KDE people they're more likely to listen. We have plans to push for that too, but our main priority is to roll out a program that WORKS NOW. This includes unsinkable dialogs, too. > > > The A4 rectangle in the canvas window is horribly disturbing. > > > > It helps me, but you can switch it off easily. In a next version this preference will be saved for future sessions. > > How about choosing "new" and set A4 to open up an image window, the > edges of which would determine the size of the image, that is A4? The > rulers would then copy the image size. Have you ever worked with any commercial vector editor? They ALL have paper size shown by edges on the canvas, and they all allow to draw both within the paper and outside. That's standard. > > No, WiW is secondary. The biggest problem with Gimp is that dialogs sink. They never do in any Windows or Mac program I've EVER seen. > > :) > > No it's not. The problem with sinking dialogs comes from the > WiW/separate issue. See below. What I am trying to do is to show you that sinking/unsinking dialogs are orthogonal to the WiW. And I'm using Inkscape to demonstrate that. Inkscape has no WiW but its dialogs are unsinkable. > > Besides, what's the purpose of having the main toolbox on the left > while having the other one on the top? It makes no sense. > > > > It makes perfect sence as they are the primary and secondary toolbars, directly linked (change tool on the primary and get all relevant controls on the secondary). > > This is the approach taken by (at least) Photoshop and Xara, and it's very convenient. > > Please compare the following images. > > http://www.suse.de/en/private/products/suse_linux/winerack/images/photoshop… > http://www.pleasewipeyourfeet.com/webtest/pr/gimp.jpg > > Do you really think that having a subtoolbox above the primary toolbox > makes any sense? Have you ever seen trees organized bottom-up? It may be surprising for you, but it's very convenient. Photoshop users were raging when this secondary toolbox appeared (in version 6 I think). Xara is considered one of the most convenient and usable vector app by most people who have ever used it, and it had such a secondary toolbox from the start. > See how many options gimp provides for the brushtool. > How would you fit them in the top toolbar? Plenty of options is not necessarily an advantage. They may be distracting. As for your example: - First, some of the brush options are very unlikely to be needed every time you use brush, and therefore should be removed into an options dialog tab for this tool. - Second, I do indeed think that even all of the Gimp brush tools on your screenshot can be put into a horizontal toolbar without problems. - Third, this actually does not matter - your Tool options is worse NOT because it's not horizontal but because it sinks and grabs focus and clutters taskbar. If it didn't (i.e. if it was transient as are dialogs in Inkscape, or dockable into a document window), the difference between a tool options dialog and our secondary toolbar would not amount to much because both switch their content when you change tools (which is the main point of having such a dialog). > Regarding dialog sinking which you've never seen on any windows/mac app, > to be honest i haven't used PS for a *long* time, just guessing from the > screenshot - the toolbox and docks, how do they behave if you maximise > the image window? Are they always on top? OF COURSE!!! Moreover, MOST people work in Photoshop with their images maximized. This would be impossible if palettes didn't stay on top. And it's pretty much impossible in Gimp for the same reason. > Can you hide them so that all > space on screen belongs to the image window? Yes, easily, just press Tab to toggle all palettes on or off. In Inkscape, we plan to use the ` key for the same because Tab is taken. > It's all the same thing except - if you minimise PS you minimise all, All including dialogs. Believe me, it's a huge difference in usability. > but you end up having a mess on the bottom of the main window when > minimising separate images. > Subtle differences between WiW and non-WiW. I repeat, I don't care about WiW. In fact I like documents in separate windows better, and Inkscape is going to stay that way. What we're speaking about is the behavior of dialogs, not document windows. > BTW Seems like Xara isn't available anymore, just checked the corel > site, or is there any other company providing the sw? See a screenshot: http://www.xara.com/press/graphics/xarax/screenshots/firestorm.jpg This is default interface, pretty cluttered, but in daily work you can easily remove all the toolbars under the menu except the bottom one. That's the secondary toolbar whose content switches as you switch tools. As you can see, quite a lot of options there (shown for the selector tool), all very compact and convenient. > > > As for the transients - as long as 1.) the toolbox and the docks all behave like one group(for the purpose of minimising, > > > bringing forward, always on top etc), > > > and making the image window active switches the images/layers/channels/paths/ dock to that image, it will do a much better job > > > than WiW has ever done. the images should always be kept as separate objects/windows. > > > > Exactly! This is _precisely_ what we're doing in Inkscape and what Gimp refuses to do. > > I don't think so. :) What are you denying? That your description precisely fits the current Inkscape interface? DOwnload the CVS and see for yourself. That Gimp flatly refuses to use transients in the same way? Ask Sven and you'll hear for yourself :) > > I ran the new Gimp and saw severe usability problems at once. You say they can be fixed by changing preferences? Great! But first, not everything can be fixed, and the biggest problem - sinking dialogs - is unfixable. > > Bulia, the main problem is that you're approaching it with destructive > criticism: > > "I ran gimp for the *first* time and *immidiatelly saw* problems A, B, C > which are unsolvable" Hey, I did not call them unsolvable until Sven confirmed that :) > Awwww, c'mon! > > the problem with sinking dialogs or transients or whatever doesn't lie > in the gimp domain, it lies in the freedesktop domain, the freedesktop > proposal should enforce certain rules on Wm for apps like gimp, which > are not WiW. If that it isn't implemented yet, is that a real problem? Of course it is. Users need usable programs NOW. Until there are other widely supported ways to achieve the same goal, we'll use transients. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

release commands
by Kees Cook 30 Nov '03

30 Nov '03

Say, when we cut a release, can we also run this from the inkscape_web directory on SF: pod2html --cachedir=/tmp --infile=../inkscape/inkscape.pod --outfile=doc/inkscape.html And what do people think of adding a "man page" link to the "documentation.php" file to look at this file? -- Kees Cook @outflux.net

2 3

inkscape_web
by Kees Cook 30 Nov '03

30 Nov '03

Hopefully I won't get flamed... 2003-11-30 Kees Cook <inkscape@...62...> * ChangeLog: created for us to track website changes * download.php: added CVS snapshot area link and SF download link * includes/navigation.php: added local link for "download.php" again * screenshots/index.php: added version heading to image list I pushed these changes to the main site (and did a chmod g+w). Now other people can get to the CVS snapshots (there didn't seem to be a link anywhere). -- Kees Cook @outflux.net

2 2

Info for Translators
by Bryce Harrington 30 Nov '03

30 Nov '03

If you're interested in helping with providing translation for Inkscape, here's a good introductory description of how this works: http://www.gtkmm.org/gtkmm2/docs/tutorial/html/ch20s03.html If you're interested in providing translations, download the .po file for your language from here: http://cvs.sourceforge.net/viewcvs.py/inkscape/inkscape/po/ If a .po file for your language does not already exist, then create one by copying the inkscape.pot file that gets generated when you run make in the codebase. Edit the file to add or correct translations of the English strings, and then upload your work via the SourceForge patch tracker: http://sourceforge.net/tracker/?func=add&group_id=93438&atid=604308 (Check the Upload checkbox and add your file or patch.) We will be cutting the 0.36 release within a week or two so if you want to update the translations for your language, best to get them in now! Bryce

2 2

calligraphic lines
by Robert Crosbie 30 Nov '03

30 Nov '03

Does it actually do anything? I just get a spew of console output when I try to use it. x_order_2: colinear! colinear! ... Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

[Fwd: [Sodipodi-list] Sodipodi 0.33 released]
by Jonathan Phillips 30 Nov '03

30 Nov '03

-----Forwarded Message----- From: Lauris Kaplinski <lauris@...39...> To: sodipodi-list(a)lists.sourceforge.net Subject: [Sodipodi-list] Sodipodi 0.33 released Date: Sun, 30 Nov 2003 19:50:35 +0200 Version 0.33 of your favourite vector drawing application is released. Sodipodi 0.33 release notes ================================= This release incorporates lots of bugfixes and enhancements that have accumulated during the 6 month period. - Experimental spsvgview slideshow program from the same codebase (Lauris) - Page border and paper color can be modified (Lauris) - Text linespacing can be modified (Lauris) - Select all and clear all actions (Masatake) - Support mousewheel (Tomas) - Full in-place Gtk+ IM support (Yukihiro) - Support fonts that have assigned mapping at unicode private area (Lauris) - Works again on 64 bit machines (Mitsuru) - New node joining mode, adding new line between nodes (Bob) - XML editor supports moving of branches (Mental) - Text properties can be applied to multiple text objects (Petr) - Basic support for xml:space attribute (Mitsuru) - Tab cycling mode is configurable (Frank, Lauris) - Repeat transformation method (Lauris) - Bugfixes (everyone) Availability ============== http://sourceforge.net/project/showfiles.php?group_id=4054 30/11/2003 Lauris Kaplinski ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Sodipodi-list mailing list Sodipodi-list(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sodipodi-list -- Jonathan Phillips <jon@...15...>

1 0

Re: [Inkscape-devel] calligraphic lines
by bulia byak 30 Nov '03

30 Nov '03

> Does it actually do anything? I just get a spew of console output when I try to use it. > > x_order_2: colinear! > colinear! > ... Oh, and I just suggested to Bryce that we remove the "colinear" from Release Notes :) Indeed this warning is now only triggered by the calligraphic tool. As for the tool, why, it does draw nice lines and it's fun to play with. I'm not sure about its practical usefulness, but I did not play with it too much. Besides, there are some fixes for this tool in Sodipodi which we need to port over (the person who created this tool originally is now working with Lauris on Sodipodi). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: GIMP user interface discussion
by bulia byak 30 Nov '03

30 Nov '03

> Do you really want to drop years of serious UI design work(gimp) just > because > the kde people haven't done their work yet? I never suggested to "drop" anything, only _add_. > Wouldn't it be better to > push hard for the freedesktop standard? Of course it would be better! So, are you doing it already? When can we expect to see the unsinkability option for utility windows in, say, KDE? > no offense intended, but inkscape is pretty much useless for professional work as it is right now. > Take a look at the inkscape screenshot i've made, there are 13! dialogs open. And it's just version 0.36. Please don't dodge the subject. Dialogs reorganization is in our plans (see DialogsReorganization on our Wiki), it's completely irrelevant to what I've been discussing in Gimp. > Where are you going to put them so that they're immidiatelly accesible? If i click on freehand, > i'd like to immidiatelly set the thickness or color of the curves i'd like to draw, How will i do that? See SecondaryToolbar on our Wiki. > Depending upon which tool you choose in inkscape another (sub)toolbox shows up on the top, > so it's either there or it's not, you'll end up looking at a jumping canvas which is just disturbing. Jumping already fixed. > Besides, what's the purpose of having the main toolbox on the left while having the other one on the top? It makes no sense. It makes perfect sence as they are the primary and secondary toolbars, directly linked (change tool on the primary and get all relevant controls on the secondary). This is the approach taken by (at least) Photoshop and Xara, and it's very convenient. > It's not clear which tool is selected. I guess this will be fixed when we return to the standard button widget. > The A4 rectangle in the canvas window is horribly disturbing. It helps me, but you can switch it off easily. In a next version this preference will be saved for future sessions. > > Second, users coming from Windows or Mac will never... > > If you want those users to migrate, the first thing you should do is to look at the UI of all those apps. As i said, > they all look pretty much the same, except that gimp doesn't do WiW and gtk doesn't offer that(luckily!). No, WiW is secondary. The biggest problem with Gimp is that dialogs sink. They never do in any Windows or Mac program I've EVER seen. > There's no space left for innovation here. :) > As for the transients - as long as 1.) the toolbox and the docks all behave like one group(for the purpose of minimising, > bringing forward, always on top etc), > and making the image window active switches the images/layers/channels/paths/ dock to that image, it will do a much better job > than WiW has ever done. the images should always be kept as separate objects/windows. Exactly! This is _precisely_ what we're doing in Inkscape and what Gimp refuses to do. > No ofense intended again, but you guys just installed gimp 1.3.x. toying around with it for a few hours, > it seems like it's not sufficient to realise how well designed the gimp UI really is and how efficient it is. No offence in my comments either. Just first impressions of a new user, which should be at least as valuable to a UI designer as comments from a long-time user. I ran the new Gimp and saw severe usability problems at once. You say they can be fixed by changing preferences? Great! But first, not everything can be fixed, and the biggest problem - sinking dialogs - is unfixable. And second, first-time users don't have to go into preferences; they should get the most usable and _standard_ interface by default. Sinking dialogs are not standard in any UI, sorry. > But keep in mind, that through all those years of gimp development, tons of discussions were going on regarding the UI, > so it's really a great piece of work and it just works. Why would you want to go through all that again? :) > Sorry if i sounded rude but i just fear seeing another promising project going the drain just because the > developers didn't trust something that is actually proven. Don't worry for us. We won't go down the drain, I promise :) -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: GIMP user interface discussion
by bulia byak 30 Nov '03

30 Nov '03

> '+' is a shift-key for a minority of our users. Do you have any statistics to back this up? Whereas I so have some statistics in my favor. On a default install of Mandrake 9.2, grep /etc/X11/xkb/symbols for layouts containing "[ equal, plus" and those containing "[ plus". You will find 61 layouts where plus is in the shifted position and 52 of those where it's non-shifted. So the majority of layouts seems to be on my side, probably that speaks something about the majority of users too. Besides, don't you think that appealing to the majority is irrelevant here? I'm not asking you to break anything for the "majority" in order to make things better for "minority". I'm just asking to _add_ something that your "majority" won't even notice. As far as I can see, "=" is not used anywhere else in Gimp. Why not make it _another_ shortcut for zoom-in? And finally, you did not give any reasons for not making KP_plus and KP_minus shortcuts for zoom as well. For that, you have no excuses whatsoever :) > You are looking at > this with a very US-centric view. :) I've never been to US. I lived most of my life in Russia and now live in Canada. > But it is tedious to repeat these discussions that we all had on bugzilla and on the > gimp-developer list already. I am not going to argue about The GIMP user interface in > private emails. This should be done on the mailing list. No problem. I will bring it up on the Gimp list after we release 0.36 and collect more usability data and feedback on the new dialogs behavior. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Inkscape Project Status 2003-11-29
by Bryce Harrington 30 Nov '03

30 Nov '03

Inkscape Project Status Saturday, November 29, 2003 ------------------------------------------------------------------------ Much progress has been made towards the 0.36 release; in fact only a handful of tasks (mostly bug fixing and my arrowhead work) remains. The user interface of Inkscape has been completely redesigned, a heap of usability improvements added, and documentation and unit testing development is well under way. Our activity percentile on SourceForge has rarely dipped much below 99% and in fact we've been in the top 100 quite often. It is surprisingly common for an issue to be raised by a user, hashed out, and implemented in CVS within a day or two. Needless to say, the open development process has been paying good dividends. The downside to having such open development is that it's easier for new bugs to creep in, however they're getting identified quickly and added to tracking mechanisms, so we're optimistic that we can apply focus on them during feature freeze and make 0.36 at least as stable as 0.35 (and hopefully more!) Once this release is out, we will turn focus to codebase cleanup, including committing to C++ and dropping support for gcc. Next after that will come an aggressive bug hunt. These two efforts will gain us quality, code documentation, wider familiarity with the codebase, and a testing framework that will make the codebase more pleasant for developers to work in. These efforts will set the stage for a series of major architectural refactorings we will undertake thereafter. Statistics Nov 02 Nov 15 Nov 29 ========== ====== ====== ====== Lifetime Rank on SourceForge: 6103 4391 3676 (56.2%) (68.5%) (73.6%) Max Week's Rank on SourceForge: 6023 62 37 (Nov 16) (56.8%) (99.6%) (99.8%) Lines of Code in src/: 115,754 115,901 116,970 Lines of Docs in doc/: 1,135 1,135 2,100 Lines of content in website: 74 1,173 1,624 Size of the Inkscape wiki: 0 3,700 5,737 Hours since last ChangeLog mod: 90 6 2 Bugs open/total: 0/0 9/15 40/51 Features open/total: 0/0 18/18 23/24 Patches open/total: 0/0 1/ 6 5/20 CVS Commits (as per inkscape-cvs): 148 481 963 Inkscape-devel membership: 4 49 61 Inkscape-announce membership: 0 9 14

1 0

printing doubles
by Robert Crosbie 30 Nov '03

30 Nov '03

I like my svg to be really clean and I always hand tweak my drawings to make sure all objects are where I want them to be. One think i dont like is when doubles are printed like this: x="200.000000" y="200.000000" Is there are reason why we store more precision than necessary here? This value comes from sp_repr_set_double which uses sp_xml_dtoa. Doesn't sp_xml_dtoa do the same thing as printf %f formatting? I have found that if I call sp_repr_set_double_attribute and replace the "%f" with a "%g" It will give a much cleaner output: x="200" y="200" Is there a technical reason why we have a custom dtoa function and keep the extra 0's? Cheers, Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Is there a gadget-lover on your gift list? MSN Shopping has lined up some good bets! http://shopping.msn.com

3 2

Snap to grid
by Jonathan C. Sitte 30 Nov '03

30 Nov '03

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I was hoping that you developers were thinking about implementing snap to grid? Thus allowing you to move images around and have them line up in an orderly fashion. Another thing is what do you think of the current user interface? Just wondering what you were doing with that. I know that you are talking about making a layers menu. That will be real helpful. The export and save options are really archaic at the moment as well. Overall the program has allot of potential. I was using Sodipodi allot but now that I found out inkscape I really like the direction you want to take it. If there is anything I can do to help let me know. ;) - -- Jonathan C. Sitte <jcsitte@...67...> Digital Arts & Computer Technology http://www.staticnull.org/jcsitte/ 917.414.2559 M-F 8AM-10PM Public Key: http://www.staticnull.org/jcsitte/0x40E69A47-pubkey.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQE/uIlcbzreREDmmkcRAjR/AJ9hyQnN/cWI/4LszNhP9WAprE5r8ACgkbt+ 5R9ZLhUvqJIvtMjexw6fxm4= =9PI8 -----END PGP SIGNATURE-----

5 10

Fast compiling with both gcc and g++
by Bryce Harrington 30 Nov '03

30 Nov '03

I documented a procedure for building both gcc and g++ copies of the codebase from the same cvs checkout. http://www.inkscape.org/cgi-bin/wiki.pl?CompilingInkscape (Down at the bottom of the page) I would highly encourage developers to adopt an approach such as this, so you can help detect and avoid g++ compilation errors. Bryce

1 0

Updated main roadmap
by Bryce Harrington 30 Nov '03

30 Nov '03

I've gone over our long-term roadmap and revised our list of milestones based on where it appears folks want to go. http://www.inkscape.org/roadmap.php M2 - I'm relabeling our current focus from 'features' to 'Interface Redesign' since between Mental and Bulia's efforts, this has been the lion's share of the effort. M3 - Since there wasn't any issues raised to my proposal to make the next release focus on the C++ and codebase cleanup work, I've formalized that as our next priority. As well, I've tentatively listed the layer dialog as the Gtkmmification objective. M4 - For milestone 4, I'm taking the suggestion of focusing on doing a bug hunt, as it really sounds like we need it. Of course, this doesn't mean we should put off bug fixing until this milestone, but rather at this point we should each set aside our feature implementation efforts in favor of focusing on clearing the bug queue. I think by tag-teaming the bugs all together, it'll be much more enjoyable than just tackling them as they sufficiently annoy us. ;-) Additionally for M4 I've committed us to doing a codebase-wide implementation of Mental's new unit testing framework. Doing it at M4 should give us enough time to get that framework well-tested. I figure doing this in parallel with the bug hunt will be synergistic and will readily leverage each other's work. M5 - Gtkmmification M6 - Plugin System M7 - New features. I figure with the new plugin system fresh out of the shop, focusing on implementing features will give it a good road test. M8 - STLification. M9 - Extract SVG Canvas.

1 0

Re: [Inkscape-devel] New win32 cvs build]
by bulia byak 30 Nov '03

30 Nov '03

> Concise descriptions or hints. Good point, I reviewed all tooltips and edited many of them. > Suggested addition to the view dialog. In the real drawing world the "fit > page in window" view is extremely useful. Equally useful is the view "fit > page width in window". This is especially so since much work is done in > portrait mode. Done, there's now a menu command to fit to page width. Someone has to draw an icon for it, then it can be added as a button. Feel free to suggest a keyboard shortcut too. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: Menu layout feedback [Re: [Inkscape-devel] Snap to grid]
by bulia byak 30 Nov '03

30 Nov '03

> 1.) Inkscape Menu - I agree with Bryce in that it should totally be > removed Agreed > 2.) Secondary Toolbar + Annoying Canvas Moving - I really think that the > secondary toolbar should never go away. Exactly, just leave it empty if there's nothing to show there, but I think ultimately all tools will have some secondary controls to put there. > Also, I think that the tools in the secondary toolbar need to be > differentiated as options extending from the primary toolbar's selected > tool. One way to do this might be to give more space on the left of the > secondary toolbar. Maybe the secondary toolbar should have a darker gray > underneath the buttons, or should be a smaller size. I think space at left plus some spacing between buttons plus, if possible, smaller button size on the secondary toolbar will do. > 3. Tool from Toolbar, Selection Not Obvious - It is not obvious once a > tool is selected that it is selected, visually. Not a problem for me, maybe try a different gtk theme? > 3.) Editing Nodes - When selecting objects and nodes in illustrator and > other programs, you usually can constrain whatever is selected by > pushing the shift button. Then when you move the items selected, they Sure, this is in Wiki already. Please view the KeyboardShortcutsToDo page and leave your comments there. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

GNOME HIG menubar
by MenTaLguY 30 Nov '03

30 Nov '03

Due to popular demand and sensible arguments, I've reverted the menubar to conform to the GNOME HIG. -mental

1 0

Re: [Inkscape-devel] [Fwd: Re: inkscape]
by bulia byak 30 Nov '03

30 Nov '03

> A good idea would be to talk to the gimp developers and work > together on UI usability issues in the future, the results would > be included in next revisions of HiG for similar types of apps. I agree with most of Marek's comments. In particular, the monster multisection toolbox is already removed in the latest CVS. As for working together on usability, I'm all for that! So... Today I finally tested the latest development Gimp (1.3.23). Below are my impressions focusing on my area of interest in Inkscape, with my view on what is good and what is bad in Gimp (and what we should or should not borrow). Just a few first impressions, I did not do in-depth testing. Probably I'll have more to say after I've used it for some time. Feel free to forward this to a Gimp list (drop me a note if you do so, so I could track responses). Summary: Sure the new Gimp is impressive. And MUCH more usable than 1.2. But in the usability department, there are still things I strongly dislike. I do not want to say they are necessarily errors, since this is subjective, but I think there are other people out there who would agree with me on at least some of the points below. The majority of palettes (dialogs) are sill not transient and "sink" readily when you switch to the document window that overlaps them. Working with a maximized document window is therefore next to impossible. Some of the dialogs/palettes are transient, though - I think it's those that can only apply to the current document and are more or less "modal", i.e. you cannot do anything on the canvas while such a window is active (examples are Levels, Color balance, etc.) Making such modal windows transient is indeed a step in the right direction, as in Gimp 1.2 even they could be buried easily (and confusingly for most users). However, having to dig out my primary toolbox from under a pile of windows still drives me crazy. Dockable tabs/palettes are cool, no arguing about that. However, they only dock to each other and not to a document window. The main toolbox is also exempt from docking. This means that although you now have a way to reduce the number of windows on your desktop, you still have at least three windows (document, toolbox, a group of docked palettes) that you'll have to alt-tab your way through. On the plus side, all floating palettes remember not only position but on/off status and all docks across sessions: when you restart Gimp, it reopens all the palettes it had opened last time. I wonder if we have to do the same in Inkscape (currently only size and position of dialogs is remembered). Weirdly, the single-letter keys for switching tools only work in the document window. When you have focus on the toolbox, you cannot press "t" to switch to text tool, even though the tooltip at the text tool button suggests that you can! This is despite the fact that tools seem to be global, i.e. you cannot have different tools active in different document windows. Similarly, I cannot press crtl-z to undo an action that I just did in a palette (e.g. trashing a layer); I must switch to the canvas (and remember, it's my reponsibility to select the correct document window, if I have several open!) and press ctrl-z there! I think this is another example where the idea that palettes are standalone non-transient windows in their own right plays bad for usability. The palettes are so "independent" that they don't even care to track the current document. Switch from one document to another and watch your Layers palette, for example. Does it show you the layers of the new document? Nope! You have to use another palette called Images to select what document is being displayed. Moreover, the Images palette only affects those palettes that are docked together with it - if you have e.g. a Layers palette and a Path palette in two different floating windows, you'll have to add an Images palette (or an Image menu) to _both_ these windows and switch images manually in each of them. I think this is very inconvenient. A much better solution is what is being implemented in Inkscape: all floating windows always reflect the active document (i.e. the one whose window has focus). Also, they are re-transientized to the active document when you switch document windows - this means they always stay on top, but are minimized when the _last_ document window is minimized, are restored when _any_ document window is restored, and do not clutter the taskbar. Another problem with Gimp's palettes being too independent is that you can't leave them using keyboard, because a palette does not know who to give focus to. That is, you can pop up a Layers palette from within a document window using keyboard (ctrl-l), but you can't go back using keyboard except with the help of your window manager (e.g. alt-tab). So if, for example, I have edited some value in a text field in a palette and press Enter, it does affect one of the documents, but the focus remains within the palette and the affected document does not pop up if it's buried. Esc also does nothing within a palette. Moreover, there's even no keyboard way to close a floating palette - Ctrl-w only works in a document window but not in a palette (why?). (By contrast, in Inkscape, Ctrl-w closes a floating dialog, Esc moves focus from it to the document that it's currently reflecting/affecting, and so does Enter in text fields.) Misc small things: Zoom shortcuts are inconvenient. "-" on the keyboard zooms out without shift, but to zoom in I have to press "+" with shift. Keypad - and + do not work at all. Distance measurement tool snaps to horizontal if used with ctrl and to verticals if used with alt. More convenient would be using ctrl only to snap to closest n*step angle where step can be set in preferences (e.g. 45 degrees). (This is what we're going to do in Inkscape.) In fullscreen mode, no floating palettes can be seen. If I switch to them by alt-tab, I don't see my document window anymore! I switch back to document and the palettes are hidden again. P.S. Regarding our recent discussion of Gtkmm/Glade, Gimp seems to use what we use now, i.e. plain C. Does this mean that if we switch to Gtkmm we won't be able to borrow code, such as dockability, from Gimp? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

3 3

Bugs
by Bryce Harrington 29 Nov '03

29 Nov '03

The Sodipodi bugs numbered between 775385 and 830500 that I could verify have been copied into the Inkscape tracker. I was able to verify about half of them. Only one was super critical and in need of being fixed immediately (IMHO). To do this I used a spreadsheet. I listed the numbers of all the bugs, then as I had time I went through each, one-by-one, to check them against Inkscape. Once that was done I created new RT tix for the validated ones. I've put the spreadsheet in inkscape/doc/ in case someone else might like to use it. Bryce

2 1

[Fwd: Re: [Inkscape-devel] New win32 cvs build]
by unknown＠example.com 29 Nov '03

29 Nov '03

Alan Horkan wrote: > On Thu, 27 Nov 2003, Mael wrote: > ... >>You still need to install gtk2 runtime from dropline >>(http://www.dropline.net/gtk/) first, then run the setup and run >>inkscape.exe (and enjoy :) > > > I've installed GTK-Runtime-Environment-2.2.4.1.exe which seems to be the > most up to date version available but I'm not having any joy. Basically I > dont think Inkscape is finding my copy of GTK properly. > Best you can do is add foo/GTK/2.0/lib to your path in the autoexec. A proposed by Mael (using batch file). > I get the following two errors > INTL.DLL is linked to missing export > ICONV.DLL:libiconv_set_location_prefix. > > and > > C:\Program Files\inkscape\inkscape.exe > A device attached to the system is not functioning > > I've tried unsuccessfully both > inkscape-0.36cvs-setup.exe and the earlier > inkscape-0.36-pre0-w32.exe > > I'm running windows 98, but I will probably be able to test against > windows 2000 this evening (and maybe Windows XP later in the week). > > Sincerely > > Alan Horkan I already tried the new one on Win98 and Win2K and it works ok in this point. The problem with the font I already described. There is this one bitmap <print direct> garbage. (BUG) The latest (beta2) sodipodi crashes on the file open dialog, but this is an other point. Adib.

2 1

New win32 cvs build
by Mael 29 Nov '03

29 Nov '03

Hello, For win32 people who want to test the new ui here is an inkscape build from cvs. I've done a *very basic* installer with nsis (http://nsis.sourceforge.net/). Download http://195.221.122.126/tmp/inkscape-0.36cvs-setup.exe [1.65 Mo] You still need to install gtk2 runtime from dropline (http://www.dropline.net/gtk/) first, then run the setup and run inkscape.exe (and enjoy :) Preferences are still saved in directory of install, this could easily be changed to save in user home dir Translations are included in the package I join the basic nsis script, in case someone want to polish/improve with : - create start menu entries and desktop icon - use more eye candy modern_ui from nsis - display of licence - option to launch inkscape at the end - safer uninstaller (be careful currently it removes the dir of install whatever is in..) ..etc.. I've not the time to do extensive tests right now, but some dialogs seem to generate lot of warnings in the console. I'll try to make a list of problems this week-end. M

3 2

Re: Addition: Inkscape
by Bryce Harrington 29 Nov '03

29 Nov '03

On Sat, 29 Nov 2003 webmaster@...86... wrote: > Your entry Inkscape has been added to > > http://www.linuxlinks.com/ > > It will appear after the next update (under 24 hours), and is listed in > both the category "Software/Graphics/Drawing" and in the What's New section, > http://www.linuxlinks.com/New/ Thanks. > To keep our database accurate we would appreciate it if you would let > us know of any changes to your listing, including any change in URL. > > If you would like us to announce new versions of Inkscape please > send information to announce@...86..., or alternatively you can post > details at our Linux news service, http://www.linuxlinks.com/portal/news/submit.php Okay, I've added 'announce@...86...' to our announcement list. > If you haven't already done so, you are more than welcome to add a > reciprocal link to us. I've added LinuxLinks to our Links to Us page. http://www.inkscape.org/cgi-bin/wiki.pl?Links_To_Us Thanks, Bryce

1 0

Re: GIMP user interface discussion
by bulia byak 29 Nov '03

29 Nov '03

> > I have never heard before about "the utility window type hint". > > It is defined in the Free Desktop window manager specification. IIRC > the KDE window manager claims to adhere to that spec: > > http://www.freedesktop.org/Standards/wm-spec It may be aware of it internally, but there are no configuration controls available to the user. Most importantly, I cannot set utility windows to stay on top within their app or be hidden on the taskbar. The freedesktop specification is bright and wonderful, but looks like we (Inkscape) will have to use transients until that specification is better supported in WMs. (Even though transients have their problems, too.) > > For example, if you add a third option, "transient", to the "window > > type hints" list in Preferences and make that the default (both for > > toolbox and for docks), I think the default Gimp will behave in a > > more expected way for the majority of new users (while remaining > > usable for those who prefer another behavior.) > > Transient implies a parent and there's no parent window the toolbox or > docks coule be transient for. Wrong. First of all, when a dock "follows active image", I see no reason why the active image cannot be considered its parent. And when you switch images, re-transientizing all docks to the new active document window actually works pretty well. Test the Inkscape CVS to see what I'm speaking about. And second, even when a dock does not follow active image, it is logical to consider all open document windows a "collective parent" for all docks, so that docks and toolbox stay on top of the topmost document. This is what we're doing in Inkscape, and the feedback has been very positive so far. Which is understandable, given that our approach is a good approximation of what Windows/Mac users come to expect. (And if you feel your document is cluttered with dialogs, we'll soon have a global key that toggles all dialogs and toolbars off and on.) Which brings us to the point, do you want to really win user share from commercialware? After all, in many cases what is required is just change a few defaults, and I don't see any reason not to do that because it won't affect power users who have their preferences set differently anyway. > > I noticed that menu option and it was indeed on, but it did not do > > anything for me. As I now see, there's one more thing I had to do: > > "Activate the focused image" in Preferences must be on. Please make > > that setting on by default, too. > > We discussed this recently on the mailing-list. And the outcome so far > was that the default should be off since it is almost possible to work > with this setting with a window manager that does focus-follows-mouse. First, I may be missing something but I have just tested "Activate the focused image" with "Focus follows mouse" in KDE, and it's very convenient. I see no problems with taht so far. Second, users coming from Windows or Mac will never have "focus follows mouse" set. This is counterintuitive for them. Those who use this setting are more likely to be power users, and power users will be able to configure this in Gimp as they see fit. > Please note that all it changes is the behaviour when you click the > window decorations. There are other ways to activate an image. Not only click. Many use alt-tab as the primary way to switch windows. > > Sure, but zoom keys are the first any user will try to use, why not > > make them more convenient by default? > > It's a convenient setting for most users and it is a logical binding > as well. Surely a lot better than '='. Have you done usability tests with newbies? Do you think they will be relieved to know that zoom out is a no-shift key while zoom in needs a shift? Why argue when we can just test? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: GIMP user interface discussion
by bulia byak 29 Nov '03

29 Nov '03

Hello Sven, Thanks for taking time to answer my comments. I hope you don't mind if I cc my response to the Inkscape list. As I wrote, I only spent one day playing with the new Gimp and did not claim to be able to discover everything it's capable of doing. I'm glad to hear Gimp is so configurable, however the first impressions are important, and maybe you can use the feedback of the fresh-eye, first-time users to change some defaults. > The majority of palettes (dialogs) are still not transient and > "sink" readily when you switch to the document window that overlaps > them. Working with a maximized document window is therefore next to > impossible. Some of the dialogs/palettes are transient, though - I > think it's those that can only apply to the current document and are > more or less "modal", i.e. you cannot do anything on the canvas > while such a window is active (examples are Levels, Color balance, > etc.) Making such modal windows transient is indeed a step in the > right direction, as in Gimp 1.2 even they could be buried easily > (and confusingly for most users). However, having to dig out my > primary toolbox from under a pile of windows still drives me crazy. > > Some people prefer the toolbox and the docks to stay on top, others > hate this. So it's up to you to set the utility window type hint in > the GIMP perferences and to configure your WM to keep these windows on > top. Most window managers seem to handle utility windows like this by > default. I have never heard before about "the utility window type hint". My KDE (3.1.3) does not seem to be aware of such windows at all, I don't see how I can configure that. If you use the hint that most WMs are unaware of, you must either bug the WM developers to implement it (the high profile of Gimp does give you some power to do that, IMHO) or offer some better supported ways to do the same. For example, if you add a third option, "transient", to the "window type hints" list in Preferences and make that the default (both for toolbox and for docks), I think the default Gimp will behave in a more expected way for the majority of new users (while remaining usable for those who prefer another behavior.) > The main toolbox is also exempt from docking. This means that > although you now have a way to reduce the number of windows on your > desktop, you still have at least three windows (document, toolbox, a > group of docked palettes) that you'll have to alt-tab your way > through. > > That is simply wrong. You can dock to the bottom of the toolbox just > like to any other dock. The only difference is that the toolbox > doesn't offer an image menu. We suggest to dock the tool-options to > the toolbox but it's up to you to configure your preferred setup. Indeed! However, unlike tabs, I could not drag-and dock the toolbox to another dock, which is why I did not realize how it can be docked. That is, unlike tabs, you cannot grab the toolbox at an empty menubar space and drag it to another dock. Also, there's no way to add a toolbox to a dock via the "Add tab" menu. I think the interface must be consistent: if the toolbox is dockable, it must behave exactly like all other dockables. > Weirdly, the single-letter keys for switching tools only work in the > document window. > > We hope to get the problem of the global accelerator table sorted out > before 2.0. Things will definitely become a lot better with the UI > manager that will be part of GTK+-2.4. So GIMP-2.2 is supposed to > have a much better global handling of shortcuts. Glad to hear that. It's important. In Inkscape, global keys really work everywhere now, while in Gimp, I cannot even press Ctrl-q to exit if I'm inside a dock! > The palettes are so "independent" that they don't even care to track > the current document. Switch from one document to another and watch > your Layers palette, for example. Does it show you the layers of the > new document? Nope! You have to use another palette called Images to > select what document is being displayed. > > That's simply not true. You can set each dock to "Auto Follow Active > Image" in the dockable menu and it will update with the active > image. If you add an image menu, there's that "Auto" button next to it > which gives the same functionality. Please note that this is the > default configuration that The GIMP will start with when it is > launched for the first time. I noticed that menu option and it was indeed on, but it did not do anything for me. As I now see, there's one more thing I had to do: "Activate the focused image" in Preferences must be on. Please make that setting on by default, too. Still, even with "Auto follow", I see no way to move focus from a dock to the active image using keyboard. > Zoom shortcuts are inconvenient. "-" on the keyboard zooms out > without shift, but to zoom in I have to press "+" with shift. Keypad > - and + do not work at all. > > It may be inconvenient on your keyboard layout, but why don't you > reconfigure it then? Go to the Prefs dialog and enable "Dynamic > Keyboard Shortcuts", hover your mouse over the Zoom menu entry and > press your favorite key. Sure, but zoom keys are the first any user will try to use, why not make them more convenient by default? > Fullscreen mode is actually designed as a way to view your document > w/o any distracting user interface elements. It's not supposed to be > used for editing. However since some people prefer to do the latter, > you can configure the fullscreen appearance of the image window in > Preferences. If you follow my advice to set the utility window type > hint, your toolbox and docks should also float on the image window in > fullscreen mode. Wonderful, but see above on the "utility type hint". -- bb -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: Menu layout feedback [Re: [Inkscape-devel] Snap to grid]
by bulia byak 29 Nov '03

29 Nov '03

> I'd be prepared to write a few simple introductory pages to get things > started, explaining Vector versus Raster graphics for one thing. But that one is already written by me :) Thanks for reminding me, I just put it up on Wiki (replacng Sodipodi by Inkscape - it was written for Inkscape originally). Feel free to edit, or add more documentation :) -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Object and XML trees
by MenTaLguY 29 Nov '03

29 Nov '03

Just a quick note ... I think tomorrow (or later today, after I get up, depending on your timezone) I will start work on an object ("layers") view, and a reworked XML view. The plan is to have both of the new views derive from SPView and their corresponding widgets derive from SPViewWidget. -mental

1 0

ANNOUNCE: New Scribus Website
by Peter Linnell 29 Nov '03

29 Nov '03

Hi all, I wanted to announce a complete redesign of the Scribus website. Many thanks to Craig Bradney for helping set this up and nurturing my newbie php skills. We still have a few tweaks to make, but its up and running. www.scribus.net or www.scribus.org.uk There is also a nice surprise waiting on the front page... Scribus complements Inkscape by importing most Inkscape created SVG when saved as "plain SVG" and exporting it to print, PDF or EPS. Scribus also supports CUPS directly for enhanced printing support including the CUPS-Gimp-Print Drivers. Enjoy, Peter

1 0

Re: [Inkscape-devel] saving files
by Robert Crosbie 28 Nov '03

28 Nov '03

>On Thu, 2003-11-27 at 19:29, Robert Crosbie wrote: > > I have fixed the problem with the save dialog dropping back a directory >each > > time you "Save as", but it seems the system module is not saving the the > > docname, docbase and a lot of other attributes to the file. This means >that > > inkscape keeps prompting you for the file name to save as every time you > > save. > >The problem appears to be that the first time a file is saved, the "pure >SVG" output module is being used instead of the "SVG with extensions" >module (meaning no inkscape: or sodipodi: attributes are saved). > >I'm not sure why this is yet... it's causing all sorts of problems. > >Ted, any insights? > >-mental Yeah, I figured this out after my last post. Perhaps, even in "pure SVG" mode we could still store some important attributes, like docname and modified. Do we really need docbase? Couldn't it be determined from docname? docname always seems to have the complete path for me. I'm a big fan of having minimal, clean svg. Cheers, Rob. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

Re: [Inkscape-devel] DOM SVG Canvas
by Ian Grant 28 Nov '03

28 Nov '03

> Yes Inkscape does have such an internal structure. A few months ago > Mental had begun work on breaking this out into a separate SVG Canvas > but I don't think the work reached a completion point. Does this mean it will be C++ classes? -- Ian Grant, Computer Lab., William Gates Building, JJ Thomson Ave., Cambridge Phone: +44 1223 334420

3 2

[Fwd: Re: inkscape]
by MenTaLguY 28 Nov '03

28 Nov '03

-----Forwarded Message----- From: Marek Peteraj <marpet@...83...> To: MenTaLguY <mental@...3...> Subject: Re: inkscape Date: Fri, 28 Nov 2003 16:33:14 +0100 On Thu, 2003-11-27 at 08:52, MenTaLguY wrote: > On Tue, 2003-11-25 at 11:40, Marek Peteraj wrote: > > Hi Mental, > > > > i just learned about inkscape, seems to be a very cool project, and as > > i'm not subscribed to the mailing list yet, i'd like to make a few > > suggestions: > > (sorry for the belated reply; I've been buried in mail) np :) > May I have your permission to forward your message to the Inkscape > mailing list? Yes of course, you can also forward this one. > I hadn't seen a recent version of the GIMP for some time; > I wasn't very > impressed with it the last time I used it, but it looks like they've > made some major changes to the UI. I'll have to work with it some more > soon and get a feel for what's new. Check out the latest release(iirc 1.3.22, i'm still on .18) and take your time to organize and dock your dialogs and toy around with it to get the feel of the UI :) Also see my gimp screenshot. Also i've made an inkscape screenshot to demonstrate why the gimp UI is so efficient(i'm a heavy gimp user, it's my everyday tool so to speak, it's been like that since 1.0 versions) Everybody bashing the gimp UI please forget 1.2 :) and check out the latest release of the 1.3.x development cycle, toy around with docking and reorganizing dialogs. So here it is: http://www.pleasewipeyourfeet.com/webtest/pr/inkscape.jpg In apps such as gimp and inkscape, tools and tools options/settings need to be accesible immidiately. Gimp solves that issue by providing a combined dock/tabbed interface. You click on a tool in the toolbox and the appropriate options appear in the tool options dialog, so there are general tool options dialogs which is bound to the toolbox. http://www.pleasewipeyourfeet.com/webtest/pr/gimp.jpg see the brush tool selected in the toolbox on the left, the brush tool options bottomright. The options in inkscape are accesible only through a menu, and thery're not dockable in any way which leads to a mess (13 dialogs open!). Gimp 1.2 suffered from the same problem. Gimp tends to organize the tools and options very efficiently, it seems to me that the key for designing such UI would be to determine what a tool is, isolate it and set the options for it. See the transformation dialog or the text dialog in the screenshot. Can these be tools? Of course they can :) Can Fill/Stroke or align be a tool? I think it can. Then put all those tools in a static toolbox. (and ask jimmac to make some cool stock icons :) Settings should be put else where, see gimps preferences. The toolbox should offer only drawing tools or tools closely related to drawing(as i stated in my previous mail). The current tree-like menu is a misconception and should be removed from HiG IMO. The purpose of hiding the tools should be to save some extra space and to organize tools in a logical way. 1. More nodes will introduce even more space consuption so it's not a win at all to use it for that purpose, and there are lots of tools that are not related to each other so what happens is that you make a node called 'miscelaneous' and put all those unrelated tools in there. This doesn't solve the 'break into categories' issue at all. 2. Gimp 1.3.x has introduced a somewhat shuffled toolbox(new tools added etc), and i had some hard time getting used at that. Node-like organizing introduces continuous 'shuffling' in that you always end-up having different nodes opened and closed, the toolbox changes it's size each time you expand/collaps a node. So the purpose of organizing into categories gets totally broken. 3. In a professional environment, where you need to work fast and reliable, a static toolbox is the most efficient thing you'd find. Once you learn it, it never changes. The tools need to be accesible immidiatelly, expanding and collapsing will just slow down the user and distract him from his work. Besides why would you want to toy around with collapsing/expanding when you just need a tool to work with? 4. If i want to use a tool which i haven't used yet and i'm not quite sure what it does, the category won't help me to understand that, for example "touch up" category won't explain what a smudge tool does. Stock icons should be as self-explanatory as possible, and tooltips should be used. If you still don't know and haven't managed to figure by trial&error, the manual should handle the rest. Another important point is: The C-SDI interface is a perfect replacement for a WiW like environment(window-in-window, see the screenshots in previous mail) when combined with the dock/tabbed UI and should be *enforced* by HiG for apps like gimp and inkscape or DTP apps, not considered to be removed. Compare the gimp screenshot with the other screenshots i've posted. Also once the user learns how to use one app, it's easy to learn how the other one works, which is also the purpose of HiG i guess - to unify the UI experience. A good idea would be to talk to the gimp developers and work together on UI usability issues in the future, the results would be included in next revisions of HiG for similar types of apps. There are 3 major advantages in adopting the gimp UI where possible: 1. unifying the UI in the graphics-aps domain 2. avoid reinventing the wheel, gimp has been developed since 1997(afaik) and can already be considered as a mature professional raster-based app. 3. adopting the UI from an opensource app instead of its proprietary counterparts(no WiW, docking for example). Simply put, it's time for gimp to have a professional vector-based brother :) Marek

1 0

[Fwd: inkscape]
by MenTaLguY 28 Nov '03

28 Nov '03

-----Forwarded Message----- From: Marek Peteraj <marpet@...83...> To: mental@...58... Subject: inkscape Date: Tue, 25 Nov 2003 17:40:51 +0100 Hi Mental, i just learned about inkscape, seems to be a very cool project, and as i'm not subscribed to the mailing list yet, i'd like to make a few suggestions: It seems that inkscape could become the gimp of vector graphics, that's why i think it would be really cool to design the UI to be as close to gimp as possible, since the gimp is pretty mature to date and it can already be considered as a porfessional tool for raster image manipulation. I took a look at the HiG and it's suggesting the same type of toolbox as sodipodi/inkscape offers. Honestly i must say that i *hate* that kind of organizing. :) The gimp toolbox is pretty slick and offers only drawing tools or tools closely related to drawing such as selections. The idea to organize them in collapsible "trees" seems completely broken to me since actually hiding the tools which you want to see in a toolbox, if expanded it makes the toolbox bigger, and a user that doesn't understand what each button/tool means has to learn it whether it's organized or not, the stock icons are rather self-explanatory and describe the tools in an understandable manner. So things i would suggest to avoid: 1. remove all items that aren't drawing tools or tools related to drawing, such as save as, open etc 2. avoid organizing in trees such as stated in HiG(unfortunately) 3. keep all buttons directly in the toolbox, hide them in pop-up colums(such as the dynahand in sodipodi) This is basically what gimp does. Next thing i'd like to suggest is to organize features in dialogs that are either dockable or organized in tabs. I've collected a few screenshots when i wanted to look at how all apps manage their features and functions. If you take a look at all of them, you'll find that they all look pretty much the same, dialogs on the right, toolbox on the left. Except that *only* gimp doesn't do WiW(window in window). And *only* gimp does provide a reasonable way to manage a lot of dialogs, namely 1) docking them and 2) organizing them in tabs. Gimp does manage tool/functions very carefully, it provides a UI model that is perfectly customizable, and groups features and functions in a clever way. http://www.pleasewipeyourfeet.com/webtest/pr/gimp.jpg http://www.pleasewipeyourfeet.com/webtest/pr/scribus.jpg http://www.qmedia.co.za/image/screen32791.jpg http://www.activewin.com/reviews/software/graphics/adobe/illustrator10/imag… http://www.suse.de/en/private/products/suse_linux/winerack/images/photoshop… http://www.activewin.com/reviews/software/graphics/adobe/pagemaker7/images/… A very important thing to consider here is whether to design the UI of those apps to be as close as possible to each other. This is perfectly possible since only gimp is a mature app to date and it's an app which provides an even more clever ui than it's proprietary counterparts. If we had 2-3 powerful apps that would create a professional linux graphics studio they would probably be: * gimp for raster image editing, photo retouching * inkscape for vector image editing * passepartout for DTP (it's in early devel stage currently so i can't tell right now) Keeping their UIs as close as possible is the key to build a pro linux graphics studio since the user isn't forced to get use to another type of ui. :) Marek

1 0

Re: [Inkscape-devel] Re: Planning for 0.37 goal
by bulia byak 28 Nov '03

28 Nov '03

> >As for the grid dialog, why separate dialog? Is it so big and complex it won't fit into the Grid tab of the current ctrl-shift-d dialog? > > > > > It's not big, in fact it is simpler. It's just that my changes mean > that I have to restructure the grid representation and it would be > easier to get right using C++, which is our planned language anyway. > It's not a separate dialog, so in fact my first step will probably be to > port the whole ctrl-shift-d dialog to gtkmm to get a feel for how things > work. Great! In the DialogsReorganization plan, the former "document settings" dialog is to be merged with the ctrl-shift-d too - its content should go on the Page tab. Assuming nobody objects to this idea, if you implement it too along with gtkmmization, that would be fantastic. Also, in a past milestone, there's a mention of a new export dialog. What's its status? Can we do it in gtkmm too? > Btw, I don't have close boxes on my dialogs (see attached png) > > My window manager is sawfish (default). Hmm, is this a case of the lack-of-decoration-on-transient-windows that we discussed some time ago? Or can it be fixed simply by adding a hint when the window is created? I'll try to see what can be done about that. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

utest
by Bryce Harrington 28 Nov '03

28 Nov '03

utest.inc 0. Very straightforward and basic testing. 1. Needs a copyright/licensing statement 2. Needs a README 3. Add doxygen comments for each function 4. Assert for NULL const char* inputs 5. strncpy character strings (line 20) 6. Maybe consider 't_' as the prefix for testing functions instead of 'utest_"? Might make it more concise and eye catching? repr-action-test.c 1. Document what it's doing in each stanza so if people use this as an example, it'll explain what's being done and why. Looks like a good start all around. Nice straightforward system with the basic capabilities. We can build on it as we learn more about testing. Bryce

1 0

Inkscape Subsystem Block Diagram
by Bryce Harrington 28 Nov '03

28 Nov '03

Mental and I put together this top-level block diagram of the Inkscape application today. http://www.inkscape.org/doc/ It identifies the 10 main "pieces" that make up Inkscape, what files belong to each of them, and roughly how they dovetail together. The main motivation for drawing this is to provide the team with a set of groupings that we can use to take a 'divide and conquer' approach to large efforts such as unit tests, documentation, or code style alterations. The original SVG is checked into CVS in inkscape/doc/. Bryce

1 1

fontproblem 0.36cvs win32
by unknown＠example.com 28 Nov '03

28 Nov '03

Hi, I have a svg file that looks different in sodipodi and inkscape. The current inkscape doesn't show the correct font. Look at the pics in the attachement. This file was written in Webdraw and by hand. And yes I can change the font-attributes with the dialog. I have inkscape 0.36cvs on a win2K and gtk runtime 2.2.4.1. Thanks, Adib.

5 6

Re: Planning for 0.37 goal
by bulia byak 28 Nov '03

28 Nov '03

One dialog we'll likely need soon is the Layers. Ask Mental when he thinks the code will be ready for creating a UI. As for the grid dialog, why separate dialog? Is it so big and complex it won't fit into the Grid tab of the current ctrl-shift-d dialog? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

5 4

saving files
by Robert Crosbie 28 Nov '03

28 Nov '03

I have fixed the problem with the save dialog dropping back a directory each time you "Save as", but it seems the system module is not saving the the docname, docbase and a lot of other attributes to the file. This means that inkscape keeps prompting you for the file name to save as every time you save. Opening a file that alrady has the sodipodi:docname tag doesn't seem to have the same problems. Perhaps someone who is familiar with the open/save modules could have a look at it. Ted maybe? Cheers, Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

Re: Inkscape and Glade? [was Re: [Inkscape-devel] keyboard issues]
by bulia byak 27 Nov '03

27 Nov '03

First, a disclaimer: http://www.inkscape.org/cgi-bin/wiki.pl?BuliaByak Now, I have no experience with either Gtkmm or Glade, and I don't know much of C++, but whatever makes the life of developer easier is great IMHO. From what I see on screenshots, Glade is a really nice thing, and I like the general approach of creating interfaces in a declarative manner (in XML) with code being generated automatically. Currently my experience with Inkscape's UI code is rather horrible, and whatever helps us to make it saner should be seriously considered. In fact, I'm afraid to even think about DialogsReorganization until we have some nicer tool for dialog layout. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Planning for 0.37 goal
by Bryce Harrington 27 Nov '03

27 Nov '03

Currently we have 'Convert to Gtkmm' as the goal for 0.37. However, I think this is a bit ambitious for us to do at present. I'd like to propose we reduce the scope of the Gtkmm work for the next release. Instead of converting the entire codebase, I'd like to suggest we focus on creating one Gtkmmifying dialog, such as Nathan's grid dialog or perhaps implementing Tom Ingham's new object properties mockup (opinions?) This would allow us to put more focus into doing the C++ conversion in an orderly fashion. I think if we start converting to C++ willy nilly we may get ourselves into a royal mess, but if we focus on subsystems serially, one at a time, we can achieve better quality. How does this sound? Bryce

1 0

DOM SVG Canvas
by Ian Grant 27 Nov '03

27 Nov '03

I want to write a mapping/GIS application. I was hoping to find an 'SVG canvas' library that I could use to render the graphics. Librsvg is not ideal because it simply renders the entire SVG document at once. Rather I am looking for something that watches a DOM tree and renders changes as they occur. Do Inkscape or sodipodi have any such internal structure, or could they be used as a basis for such a thing? Seems like people are going to be writing a lot of low-level SVG rendering code otherwise.

2 1

BORKEN merge
by MenTaLguY 27 Nov '03

27 Nov '03

Just a quick heads-up; I've merged the "BORKEN" UI branch to HEAD. It still needs a lot of refinements, but I think it's far along enough now to merge. Have a look at Bulia's recent post and on the Wiki for more stuff we can do. -mental

1 0

Re: [Inkscape-devel] keyboard issues
by bulia byak 27 Nov '03

27 Nov '03

> bb> 3. Several people suggested that making keyboard shortcuts user-modifiable and providing "profiles" imitating other vector editors. This is an attractive idea but there are some issues. I added a > bb> brief overview of this to the KeyboardShortcutsToDo. > it would be a great feature. OK, so whoever is not busy coding (and especially fixing bugs) at the moment could start working towards this. Please pick your favorite non-Inkscape vector app and make a list of its shortcuts and corresponding functions. Please also categorize each function into one of the following groups: 1. There is a corresponding function (with the same or different shortcut) in shortcuts.c (i.e. this is a global verb in Inkscape). 2. There is no such function in shortcuts.c but such a function is listed on the KeyboardShortcuts wiki page (which tracks all shortcuts currently implemented in Inkscape). 3. There's no such function in Inkscape, but it would be relatively straightforward to implement it. 4. There's no such function in Inkscape, and it's not clear how to implement it (i.e. we don't have such a tool, or palette, etc.). Please use Wiki for creating your lists. You can link them from under "Interface discussions" on the Wiki front page. > corel has something similar in its coreldraw > and corel photopaint - but you can switch not only shortcuts but also > a layout of toolbars to imitate illustrator or photoshop > (respectively). That's I don't see how to do. (Especially given that Illustrator and Photoshop do not have a fixed layout.) And in any case, right now the dialog layout is very rigid and messy in the code ("boilerplate"), so this has to wait until we switch to something more flexible (Glade?). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

4 3

big dialogs upgrade
by bulia byak 27 Nov '03

27 Nov '03

Today's a big day: finally I made all dialogs unsinkable, defocusable, closeable, and remembering position across sessions. Well, not exactly all, because I decided that a number of dialogs can be simply removed so I did not bother upgrading them. See the new wiki page: http://www.inkscape.org/cgi-bin/wiki.pl?DialogsReorganization for my plans on what to do with what dialogs. COmments are welcome as always. So, please test the latest CVS and report any problems. There are probably quite a few of them because there were lots of changes in lots of files. Other wiki pages created or updated today: http://www.inkscape.org/cgi-bin/wiki.pl?TextUsability - specific ideas for the text tool and dialog http://www.inkscape.org/cgi-bin/wiki.pl?SecondaryToolbar - this is what's brewing inside BORKEN, and it's cool! http://www.inkscape.org/cgi-bin/wiki.pl?PreferencesDialog - we need a big and impressive Preferences dialog, so add your ideas on what to put there http://www.inkscape.org/cgi-bin/wiki.pl?KeyboardProfiles - the discussion on keyboard profiles is now on a page of its own. Please be more active on the Wiki. All these pages and others will benefit from your input and ideas. Also, please test drag tolerance that I added yesterday. You may need to remove old ~/.inkscape/preferences file for the new ~/.inkscape/preferences.xml to work reliably (I could not find how the old file can break things, but looks like it does sometimes). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Re[2]: [Inkscape-devel] key bindings and documentation
by Yemu 26 Nov '03

26 Nov '03

JP> Okay, so I created some SVG keys. Man, it is amazing the bugs you find JP> when trying to design something simple like keys in Inkscape. JP> Anyway, please look over and make suggestions. Otherwise I will try and JP> use the wiki to add PNGs or GIFs of the keys to the keybinding page as JP> soon as possible. GULP. I've never tried that before, but hopefully can JP> conquer that task. hi jonathan! i made some changes to your keys. use them if you like (keys are little lighter and "less round") best regards yemu

1 0

Re: [Inkscape-devel] bugs bugs bugs
by Robert Crosbie 26 Nov '03

26 Nov '03

Of course. See my Stability section in Wiki. Simply starting and quitting inkscape will leave over 3 megs of unfreed memory on my machine! Rob. http://members.rogers.com/rcrosbie >From: "Jon A. Cruz" <jon@...18...> >To: Robert Crosbie <swingincelt@...19...> >CC: inkscape-devel(a)lists.sourceforge.net >Subject: Re: [Inkscape-devel] bugs bugs bugs >Date: Wed, 26 Nov 2003 08:57:59 -0800 > >Robert Crosbie wrote: > >> >>Relating to hammering out the bugs, I tried to compile inkscape with >>ElectricFence last night, but couldn't get inkscape to compile with it. I >>didn't have time to debug it, so I gave up and reverted back to my >>original configuration when it failed. > > >As an alternative, you might try running with valgrind. > >http://valgrind.kde.org/ > > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Re: [Inkscape-devel] bugs bugs bugs
by Robert Crosbie 26 Nov '03

26 Nov '03

Relating to hammering out the bugs, I tried to compile inkscape with ElectricFence last night, but couldn't get inkscape to compile with it. I didn't have time to debug it, so I gave up and reverted back to my original configuration when it failed. If anyone else would like to try and get it to work, I'm sure it would help us in the long run. There definately are a lot of memory management issues. I simply borrowed some efence code from another projects configure.in, ran 'autogen.sh --with-efence' and tried to remake inkscape. Cheers, Rob. http://members.rogers.com/rcrosbie >From: Bryce Harrington <bryce@...1...> >To: <inkscape-devel(a)lists.sourceforge.net> >Subject: Re: [Inkscape-devel] bugs bugs bugs >Date: Tue, 25 Nov 2003 21:36:10 -0800 (PST) > >On Tue, 25 Nov 2003, bulia byak wrote: > > I just put Inscape to a little real-world usability test by drawing > > several diagrams. While there are many small interface improvements I > > could think up for myself, the biggest problem were bugs. More > > specifically, two kinds of bugs: the undo bugs and save bugs. You run > > into them very often. I don't think we can release 0.36 with them. I > > urge everyone to please go to the tracker and fix as many as you > > can. Some of them must be pretty simple. > >Yes, this is an issue we'll need to hammer down prior to making a >release. > >But it isn't cause for alarm yet; this is normal for the development >process we're using. Since so many of us are working directly on the >codebase, it's going to get messy. This is why tracking, feature >freezing and bug hunts are critical parts of the process. (This is also >why modularity and testing frameworks are very important.) > >It is also important that we always test out each other's code. It's >very easy for us to focus on our own area of interest in the code, but >if we're going to succeed as a team, we do best to take some time to >review each other's work. Review is good for morale, good for quality, >and good for knowledge-transfer. > >So for right now, when anyone finds a bug, make sure it is listed in the >bug tracker at an appropriate criticality ranking, so we can be sure we >are not releasing Inkscape 0.36 with any critical bugs. > >Bryce > > > > >------------------------------------------------------- >This SF.net email is sponsored by: SF.net Giveback Program. >Does SourceForge.net help you be more productive? Does it >help you create better code? SHARE THE LOVE, and help us help >YOU! Click Here: http://sourceforge.net/donate/ >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

2 1

more BORKEN comments
by bulia byak 26 Nov '03

26 Nov '03

Mental, thanks for moving the secondary toolbar to the top. Now, like I said, it would be nice to move all tool options there (and remove the "tool options") dialog. However, this is not to be done indiscriminately, and other dialogs' content is sometimes appropriate for this secondary toolbar. I started a Wiki page, SecondaryToolbar - please review and add your ideas or comments. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

BORKEN comments
by bulia byak 26 Nov '03

26 Nov '03

I finally upgraded my Mandrake so I could compile the BORKEN branch. Overall, very good work, thanks! I have only one suggestion: Can we please move the second toolbar with tool-specific buttons to the top of the window, making it a horizontal toolbar under the menu? Here are the advantages: - With two vertical toolbars side by side, it's not quite clear that the second is auxiliary to the first. The second one looks like a continuation of the first. When the toolbars are orthogonal, it's more intuitive IMHO. - With maximized window, two docked toolbars at left leave too much unused space beneath them. - If we make the auxiliary toolbar horizontal, we may not limit it to buttons only. In fact, the entire contents of the "tool options" dialog can be moved to that horizontal bar too. Then everything in that secondary toolbar, including both buttons and text fields, will change in concert upon tool switching, but the toolbar itself won't disappear or reappear (although it may be empty for some tools). I think it will be both convenient and intuitive. I did not invent this; what is described above is exactly the interface approach of Xara X. (Which is, by the way, despite its lack of SVG support, by far the most convenient and usable vector application I've ever seen. Way more convenient than Illustrator, and I'm not alone in that assessment. If we must imitate any one program, I'd choose Xara any day.) --bb -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

keyboard issues
by bulia byak 26 Nov '03

26 Nov '03

1. I fixed editability of text fields in the fill&stroke dialog (that was my error). I also added moving focus to the document when you press Enter in any of the fields (in that dialog only, for now). Now the dialog is completely usable by keyboard. Question: do we want to similarly move focus to the document upon activation of any other widgets, e.g. buttons? 2. This change gives me one more reason to not make dialog shortcuts toggles. This discussion is now on the KeyboardShortcutsToDo wiki page; please read and comment. Overall, I think the arguments against toggles are now more persuasive. 3. Several people suggested that making keyboard shortcuts user-modifiable and providing "profiles" imitating other vector editors. This is an attractive idea but there are some issues. I added a brief overview of this to the KeyboardShortcutsToDo. P.S. One of my messages seems to be lost - I don't see it in the list. Here it is again: There's a new release of Sodipodi. Looks like there are many changes there, and while most of them are in the internals (which I'm no expert in) some are in UI and I'd like to port at least some of them to Inkscape. Anyone else feeling that may be a good idea? Maybe we can add to the roadmap "review and merge good changes from Sodipodi"? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

3 2

bugs bugs bugs
by bulia byak 26 Nov '03

26 Nov '03

I just put Inscape to a little real-world usability test by drawing several diagrams. While there are many small interface improvements I could think up for myself, the biggest problem were bugs. More specifically, two kinds of bugs: the undo bugs and save bugs. You run into them very often. I don't think we can release 0.36 with them. I urge everyone to please go to the tracker and fix as many as you can. Some of them must be pretty simple. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

keys svg file
by Jonathan Phillips 26 Nov '03

26 Nov '03

Oops...sent without attachment...Here are the SVG keys. jon -- Jonathan Phillips <jon@...15...>

1 0

key bindings and documentation
by Nathan Hurst 26 Nov '03

26 Nov '03

Would someone with two artistic neurons to rub together put together an SVG of a keyboard key with a place to put the letter symbol and some control, shift etc keys? This would make bulia's keybindings page look very professional! just a thought, njh

2 1

key graphics
by Jonathan Phillips 26 Nov '03

26 Nov '03

I'm creating the key graphics. I'm sending this so that work is not duplicated. thx, jon -- Jon Phillips UCSD Graduate Researcher Computational Media PO BOX 948667 La Jolla, Ca 92037 858.361.2811 www.protofunk.org jon@...15...

1 0

drag tolerance
by bulia byak 26 Nov '03

26 Nov '03

To complement teh cursor tolerance that affects grabbing, I added another option, drag tolerance. Currently it only affects mouse drags in selector. Any mouse drag which is smaller than the tolerance (in screen pixels) is ignored and is interpreted as a click. To me, this makes it much easier to select objects with mouse and especially tablet, as well as switching handles from scale to rotation, by eliminating the small accidental object drags that are very annoying. The default value for this option is 4 pixels. Please test the latest CVS and report if you like the change, and if so, what value is the best for you. There's no UI for changing the value yet but you can edit it in preferences.xml: <group id="dragtolerance" value="4.0"/> and then restart Inkscape. By the way I also wrote most of the HandlingPreferences tutorial on Wiki. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

0.36 release notes
by bulia byak 25 Nov '03

25 Nov '03

I started a ReleaseNotes document on Wiki. Let's use it to describe all our achievments and failings in this version in a way accessible both to an average non-technical user and to a potential contributor who hasn't followed our discussions and wants to get a scoop quickly. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

3 2

0.36 release soon?
by Bryce Harrington 25 Nov '03

25 Nov '03

Hi all, Looking at the planned tasks for this release it seems we're quite close. Plus, a lot of work has been done, so I think the codebase is getting ripe for a release within a week or two. For those working on things to go into 0.36, could you reply and let us know your status? Thanks, Bryce

8 17

bonobo
by Robert Crosbie 25 Nov '03

25 Nov '03

Are the inkscape bonobo libraties being used? They don't seem to get compiled. Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

4 4

nudge distance, cursor tolerance
by bulia byak 24 Nov '03

24 Nov '03

1. Nudge distance is now stored in preferences. Default is 1 mm, let me know if you think this value needs to be changed. This affects arrows and shift-arrows in selector and node edit. No GUI to change that value yet, but you can edit your preferences.xml to test. 2. Cursor tolerance (from display options) is now stored in preferences too. I changed default from 1.0 to 2.0, but perhaps it should be made even bigger. This affects how close you need to be to an object to grab it; bigger values make the tools much more usable, especially with a tablet. Please test and report what value works best for you (I like it at 8.0), then we'll take the average and make it the default :) -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

4 3

SF out of disk space
by Bryce Harrington 24 Nov '03

24 Nov '03

bash-2.05b$ df -h . Filesystem Size Used Avail Use% Mounted on pr-fs-groups-i:/home/groups/i 54G 51G 0 100% /home/groups/i I noticed this because I was updating the DevelopmentRoadmap and got a "Bad page version" error. Also it deleted the DevelopmentRoadmap and its change history. There's a backup version but I'm unsure how old it is. Bryce

2 4

new keys in node editor
by bulia byak 24 Nov '03

24 Nov '03

Please test! New keys in node editor: * esc deselects and cancels rubberband and node dragging * ctrl-a selects all nodes * tab and shift-tab cycle through nodes of the current shape All these keys work across all subpaths of a shape. Note that I had to change the lists of nodes, subpaths, and selected nodes from single-linked to double-linked. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: [Sodipodi-list] Comments and questions from a newbie (fwd)
by Bryce Harrington 24 Nov '03

24 Nov '03

---------- Forwarded message ---------- Date: Mon, 24 Nov 2003 09:02:39 -0500 (EST) From: Radical <radicalgfx@...78...> To: Bryce Harrington <bryce@...1...> Subject: Re: [Sodipodi-list] Comments and questions from a newbie > On Sun, 23 Nov 2003, Radical wrote: > > that available. It doesn't seem to save the setting though. > > I passed word along to Mental (the developer of the feature); he's going > to poke around at it. He asked what you mean by save the setting? I meant that the inkscape:groupmode attribute didn't seem to get saved in the SVG file. So in my case where I have have a lot of groups I have to go back and set that parameter on each one every time I load the drawing. R

1 0

Call for help: Doxygen headers
by Ted Gould 24 Nov '03

24 Nov '03

Hey, I was working on the Actions/Verbs stuff for the modules and learning about the implementation in Inkscape. So, as I was doing this I was writing documentation in the Doxygen format for the functions where I figured out what they do. This made me think: wow, this would be a good way for someone to help out who wants to get more familiar with the code base. No seriously, you don't have to learn how absolutely everything works to write some good descriptions of functions. Probably starting in one of the subdirectories would be best (most of the time these aren't heavily dependent on things outside of them) and writing the Doxygen headers. Doxygen is pretty easy. You can do all kinds of fancy stuff with it, but I'd just like to see descriptions of the functions and the parameters - which doesn't take that much knowledge of Doxygen either. I don't know if anyone is interested, but I'd love to see patches with this info in them. --Ted

1 0

bug transfer from sodi to Inkscape
by Jonathan Phillips 24 Nov '03

24 Nov '03

Hello, I've decided to go through Sodi's patch tracker and transfer what is buggy in Inkscape to Inkscape's cvs bug tracker. I'm trying to find some way to help and this seems like a good way. Could I get an update on progress of this? I see on the wiki that ted and bryce are listed as taking on this task. Otherwise, I'm going to start translating the bugs over to inkscape. thx Jon -- Jon Phillips UCSD Graduate Researcher Computational Media PO BOX 948667 La Jolla, Ca 92037 858.361.2811 www.protofunk.org jon@...15...

3 3

The End of the Illustrator Thread
by T Ingham 24 Nov '03

24 Nov '03

All, As both a professional designer and my previous work in the print industry ( not publishing, but actual printing ) I can tell you without a doubt that Adobe Illustrator is the de facto standard for production of any kind of high-resolution, device independent output. There isn't a print shop or pre-flight house in the usa that doesn't have at least one license for Illustrator. There are reasons that Illustrator is so popular that have nothing to do with quality, more to the point in fact it's just really old. With this, the reasons that Illustrator is popular, we need to focus on why it is that this fork started. Our goal is to open this application up to a wider user-base yes? If that's the case then I should think it would be in our best interests to attempt to provide some sort of mechanism for allowing users the ability to customize their own shortcuts. This sort of system would take the form of what Macromedia did when they released Freehand 7. In the preferences for that application, there was a simple checkbox "Emulate Illustrator Shortcuts". Whence checked, the application behaved nearly identically to Illustrator. While some have said that providing for a customization scheme for Inkscape, for some commands would prove difficult, I plead with you to find a way to make it happen. Issues in regards to the tools available in this application will mean nothing if people have to learn an entirely new set of commands to make it perform the way their used to. They won't, they'll walk. It's the subversion game, and we're playing with all balls. -Thomas

2 1

bugs
by bulia byak 24 Nov '03

24 Nov '03

1. I fixed 842565 long ago; can anyone please test and if ok, close the bug? 2. I'd like to nominate the new 848355 as critical for 0.36. Along with the other undo bugs it's very annoying. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Mental: Toolbar changes
by T Ingham 24 Nov '03

24 Nov '03

Mental, First off, I freaking LOVE YOU!!! The convenience of having those toolbars contextually show / hide is fantastic. However there is one minor inconsistency that needs attention. When you select a tool which has no sub-toolkit, could you have the "Select" subtoolkit display, but be disabled. This would prevent the panel from resizing, and throwing the document all over the place. Alternatively, we could adjust the current location of the artboard within the canvas window, by and inverse amount to the growth / shrinkage of the panel ( but that seems entirely more complicated and no more beneficial ) Spatiality Rules! -t

1 0

Don't Panic, or "New UI code getting merged in"
by Bryce Harrington 24 Nov '03

24 Nov '03

For those of you who have not been following Mental's work, be aware that he's going to be merging it in soon - probably this evening. The new SDI-based interface feels *quite* *different* from Sodipodi. Confusingly so. Things are in very different locations than you're used to. But this is all by intent. We have discussed the CSDI / SDI technicalities, the importance of HIG compliance, the desire to work like this app or that, etc. etc. This work has been going on in a branch for the past week, with screenshots on the website for about as long, so there's been some time for early testing. We are ONLY going to know whether all this theory will work in reality if we each TRY IT OUT. Some will ask why we are making these significant changes, and that is a good question. Stated briefly there are several reasons: 1. It significantly distinguishes Inkscape from Sodipodi 2. It is in keeping with GNOME HIG standards (the Sodipodi approach isn't). 3. It has been requested by a number of Inkscape users. 4. It forces us to really rethink the UI layout, including little things that we have tended to take for granted but that need to be re-examined. 5. It's better to make the change now while the project is still young and won't disrupt a large existing user base (other than ourselves!) Regarding testing, some tips. Well thought out, fairly worded observations are what we're looking for, moreso than gut reactions. Things are still fluid at this point so well thought out comments can have an important long-term impact. Take some time with your evaluation and think about how we can make it more powerful for skilled users, and intuitive enough to help novices find their way to what they need. You will be able to check it out from developer CVS soonish, or anonymous CVS within a day or so. We also hope to cut a pre-release and RPM's to make testing easier for folks. Bryce

1 0

Re: [Inkscape-devel] Testing feedback
by Robert Crosbie 24 Nov '03

24 Nov '03

----8<---Snip--->8---- > "The undo also seems to have some problems. I was messing around with > the layer selection stuff and moved an item which I then tried to undo > but it didn't work. It was kinda strange because some things could be > undone but not others. I kept pressing undo a whole bunch of times > and it crashed." > >Robert - Sounds like this one feeds into your work; would you mind >taking this one? > >Bryce Yeah, This is the problem I was looking into Saturdy Evening. I'll be looking at it again on Tuesday or Wednesday. If anyone else wants to have a go at it, It seems to be around sp_document_undo or sp_document_maybe_done. It seems that if you undo when there is nothing to undo, the undo list doesn't get updated when you create new object. Problem is kdbg was crashing on me when I tried to set a breakpoint inside sp_document_maybe_done to see what was happening there. Cheers, Rob. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

Testing feedback
by Bryce Harrington 24 Nov '03

24 Nov '03

Some feedback on Inkscape from CVS from a user: "I tried out Inkscape a bit. I was able to set the inkscape:groupmode parameter that allows me to select grouped items. I really like having that available. It doesn't seem to save the setting though." Mental - want to take a look at this one? I've asked for clarification on what he means by "save the setting". "I went to set the stroke width on a path and I couldn't type in the number. It seems to do the hotkeys instead of letting me type into the dialog (focus problem?)." Bulia - I verified this. Looks like the new number-key keyboard shortcuts are stealing focus. Could you take a look? "The undo also seems to have some problems. I was messing around with the layer selection stuff and moved an item which I then tried to undo but it didn't work. It was kinda strange because some things could be undone but not others. I kept pressing undo a whole bunch of times and it crashed." Robert - Sounds like this one feeds into your work; would you mind taking this one? Bryce

1 0

(no subject)
by bulia byak 24 Nov '03

24 Nov '03

> I'm torn between wanting to shamelessly pander to Illustrator users or > decide if doing it differently would be better enought to mitigate the > inconsistancey! Illustrator is far from being the standard. Xara, CorelDraw, Freehand, Canvas all have sizeable user communities, and as a rule they don't imitate Illustrator keybindings but do most things their own way. Besides, there's GIMP and other linux graphic soft that we must take into account. So I think in each case, we should review the existing approaches and either select one of them (not necessarily Illustrator) as the best, or devise our own. And by the way, at least global keybindings should ultimately be user configurable (it's difficult to do that for all keybindings, but at least those in shortcuts.c must be taken from the config file). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

4 8

Re: [Inkscape-devel] misc
by bulia byak 24 Nov '03

24 Nov '03

> > - ctrl-` and ctrl-shift-` opens the toolbox > > If only one instance of the toolbox is allowed, and will always be on > top, then might it worth having ctrl-` work as an on/off toggle? The same applies to all other dialogs. Do we want to make all ctrl-shift shorcuts toggles? Advantages of toggles: - you can switch off a dialog without first moving focus to it - Illustrator behaves the same way (toggles) Disadvantages: - may be counterintuitive: I see a dialog and I want to move focus to it to use some of its widgets. Now I press e.g. ctrl-shift-f and focus is moved to the dialog. With toggles, I will have to press ctrl-shift-f twice to do the same (once to switch it off and then on again). The same situation is when one dialog obscures another: now I press the shorcut of the bottom one and it jumps to top at once, with toggles I will have to press it twice. Any comments? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Re: [Inkscape-devel] 0.36
by bulia byak 23 Nov '03

23 Nov '03

> > 2a. Recent reports and testing them under windows show that the > > transiency code is not perfect. Maybe we'll have to switch "unsinkable > > dialogs" off on Windows because it behaves erratic there when you try > > to minimize a document window (maybe this is GTK+ issue). I lean > > towards accepting the current unsinkable dialogs as temporary solution > > for the 0.36 release (but not for windows build), but after that I > > want to explore other options such as GIMP's dockable > > palettes. Dockability is a good thing in any case, and writing it from > > scratch is too much work; maybe they will take care of staying on top > > as well. > > That sounds like a good approach. Can you configure it to not use the > unsinkable dialogs feature for Windows compiles? Also good idea to > investigate dockability; maybe you could add that to the roadmap for > 0.37 or 0.38? Perhaps you and Mental can talk about that. OK > I compiled Inkscape from CVS and played with opening and closing > Inkscape and its dialogs after moving dialogs around. A few dialogs > seemed to remember their screen location when closed and reopened, but > not all. Sure, only toolbox and ctrl-shift-f, as I explained before > None seemed to remember their location between Inkscape > invocations. That's strange - these two should remember position and size, they do on my system. Is ~/inkscape/preferences.xml created/updated after you exit? > Also I noticed many dialogs first popped up in a corner, > whereas it probably would be better for them to pop up either > auto-positioned (if possible) or centered in the screen. OK I'll think about centering those dialogs that don't yet have saved position, but this is minor importance because this position will only be used once - after you drag the dialog to any other place it will remember it and never again pop in the center. > > 2c. I still want someone to test one small change in the transient > > behavior though. It's simple: Please compile the latest CVS/tarball, > > open two document windows, maximize both, and switch between > > them. Does toolbox stay on top? Please report also your window > > manager. Thanks! > > I've tested this on KDE (Mandrake 9.1), the toolbox does not stay on top > of the canvas windows when switching via Alt-tab. If I am switching > between a single canvas and the toolbar, though, it does. That's the problem. Here's the dillemma: - either we have it as it is now, with the problem of dialogs not staying on top when switching maximized windows (at least on KDE), - or I can fix this problem, but then we'll have another: as reported by Tom von Schwerdtner, "the document window is auto-raising itself when it gets the focus. To me this is is counter-intuitive, as every other gnome application I use follows the window manager settings of not auto-raising windows when they get the focus." I don't use raise-on-click so I did not notice this myself; with raise-on-focus everything seems to be OK. So we have to choose which of these two problems is more serious. I can think of no way to fix them both. Maybe some magic dockable windows will solve both, but the current transient windows have to choose. I don't want to make this user-configurable because that would be a rather obscure and confusing option. What do you think? > I don't have an opinion on this one and am not familiar with how it's > done in other applications, but if I were going to try to guess what the > mappings were, I'd probably try ctrl-pgup/dn, then ctrl-up/dn, then > (maybe) home/end, and finally if it wasn't any of those I'd look at the > keyboard shortcuts list. So my vote is for Home/End for top/bottom (as it is now) and Ctrl-Home/Ctrl-end for one layer up/down (instead of current pgup/pgdn). > I'll add 'make this decision' to the roadmap. Who do you think would be > a good person to make the call? (I also don't have a mouse wheel and > also don't have an opinion.) Alan, can you summarize what is the current consensus (if any)? --bb -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Re: Menu layout feedback [Re: [Inkscape-devel] Snap to grid]
by Robert Crosbie 23 Nov '03

23 Nov '03

> >I'd add a Help menu, even if you only have the About dialog to put there. > In the Help menu we could put a link to the sourceforge site as well as the About dialog. Rob. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 3

[Fwd: [Sodipodi-list] Comments and questions from a newbie]
by Nathan Hurst 23 Nov '03

23 Nov '03

some thoughts for checking against 0.36.

1 0

dialogs memory
by bulia byak 23 Nov '03

23 Nov '03

With the preferences restored, I've implemented remembering size and position for dialogs. Only two dialogs are affected for now, the toolbox (only position) and fill/stroke (size and position). Please test and report any problems. Finally, after this change is tested, I will copy the entire code over to all other dialogs. Preferences are now saved in ~/.inkscape/preferences.xml, instead of simply ~/.inkscape/preferences. I think it's more logical this way. A new signal was added, "shut_down". It is sent by the inkscape object before shutting down inkscape and before saving preferences, so that dialogs have the chance to write their status into the preferences. If anyone else needs to do some housekeeping before program shuts down, they may use this signal. New shortcut: * for the star tool. By the way, the number of star rays was written in prefs, and this now works. Lots of other parameters need to be similarly written and restored. I will soon update the wiki page with instructions on how to handle this. -- __________________________________________________________ Sign-up for your own personalized E-mail at Mail.com http://www.mail.com/?sr=signup Search Smarter - get the new eXact Search Bar for free! http://www.exactsearchbar.com/

2 1

misc
by bulia byak 23 Nov '03

23 Nov '03

Have anyone tested the unsinkable dialogs? Any comments? Just to remind you, the two dialogs I've been working on - toolbox and fill/stroke - are now not only transient but unsinkable (stay on top of the topmost document window when there are many). Also they remember shape and position within session so you can turn them off and then on again without losing your layout. (What remains to be done is an all-dialogs-toggle; I will do this shortly.) Inserting the same into all other dialogs is a boring work, and I want to be sure that these new features work perfectly with these two dialogs before I spread them everywhere. So please test heavily and report any problems or concerns. By the way, be prepared for a possible flame from Gnome developers. Here's what Havoc Pennington writes in his book: " A transient dialog is one that appears and is dismissed relatively quickly. (GnomeDialog is really meant for transient dialogs.) Some "dialogs" are just small windows, such as the tool palette in the Gimp. These persistent ("floating") dialogs should be minimizable without minimizing the parent, and they should not be forced to stay above the parent window." Which IMHO is very wrong: it does not matter how long a dialog stays open, it only matters what you do with it. A tool palette makes no sense without a document to apply these tools to, and therefore must be "transient" to stay on top. Unfortunately someone chose this wrong term ("transient" meaning "temporary") and this seems to affect many programmers who underutilize transient windows making their interfaces almost unusable (GIMP is not the only example). Also checked in today: - space switches to selector temporarily; another space switches back. This required writing a new set of functions to switch tools (tools-switch.c) but now it's much easier to do - ctrl-` and ctrl-shift-` opens the toolbox - misc fixes I may be absent for some days - need to reboot to windows to do a project in Xara/Illustrator :) Feel free to leave comments on wiki. --bb -- __________________________________________________________ Sign-up for your own personalized E-mail at Mail.com http://www.mail.com/?sr=signup Search Smarter - get the new eXact Search Bar for free! http://www.exactsearchbar.com/

2 1

Status
by Bryce Harrington 23 Nov '03

23 Nov '03

Below are the tasks for our current milestone. For those 20 or so of you just joining us, we're going to try to organize efforts into near-term sets of goals, that we can reasonably achieve every week or so. What we mainly need right now is to get the codebase to pass a 'make distcheck'. In addition to being able to identify any remaining things needing finished for the C++-ification and renaming, we had noticed when working on Sodipodi-hydra that distcheck wasn't passing. If anyone's interested in helping on this, it's straightforward: Get the code from CVS, compile it (with either CC=gcc or CC=g++), run `make distcheck`, and report on what needs to be done to get it to pass. Once it passes distcheck we will cut release 0.35 and create tarballs and rpms. Bryce === Milestone 1 === Inkscape Production * (DONE) Get inkscape to build with a C++ compiler [team] * (DONE) Fix CDATA issue [mental] * Get codebase to successfully complete `make distcheck` * Cut a release of Inkscape 0.35 * Upload release tarball of Inkscape 0.35 to Inkscape SF file manager Inkscape Experimental * Build a new grid options dialog using gtkmm [njh] Inkscape Website * Create logo for Inkscape [mental] * Create look and feel graphical design prototype for website [mental] * Create 3+ screenshots of Inkscape Production * (DONE) Flesh out website more fully [bryce]

6 17

SInce your forking might as well get rid of csdi
by David Adam Bordoley 23 Nov '03

23 Nov '03

One nice interface improvement you could make for inkspace would be to crap the csdi interface in favor of a sdi interface in which toolboxes and pallettes are shared in between windows. There is a gnome HIG bug in which both seth and greg merchan have agreed that csdi should be removed from the HIG http://bugzilla.gnome.org/show_bug.cgi?id=116599. Just a suggestion. dave

3 6

0.36
by bulia byak 23 Nov '03

23 Nov '03

1. My current status: working on the remaining keys I wanted to enable for 0.36 (esc in the node editor, etc.) 2. Dialog code is finished, but: 2a. Recent reports and testing them under windows show that the transiency code is not perfect. Maybe we'll have to switch "unsinkable dialogs" off on Windows because it behaves erratic there when you try to minimize a document window (maybe this is GTK+ issue). I lean towards accepting the current unsinkable dialogs as temporary solution for the 0.36 release (but not for windows build), but after that I want to explore other options such as GIMP's dockable palettes. Dockability is a good thing in any case, and writing it from scratch is too much work; maybe they will take care of staying on top as well. 2b. The rest of dialog improvements seem to work OK (keys, titles, remembering position) although nobody yet commented. I think I'll go on and copy the code to all dialogs. 2c. I still want someone to test one small change in the transient behavior though. It's simple: Please compile the latest CVS/tarball, open two document windows, maximize both, and switch between them. Does toolbox stay on top? Please report also your window manager. Thanks! 3. Thanks to Robert Crosbie for fixing the save crash, but I think the remaining problems he reported are critical and must be fixed before 0.36: > - autodetect fails if no extension is added to filename (.svg) > - on quit, still says I have unsaved changes, even though I saved > - can't open compressed svg unless it has .svgz extension. especially the second one. Please add them to the tracker and/or roadmap. 4. When a consensus is reached on which keys to use for moving to front/back, please add this to the KeyboardShortcuts TODO and I'll implement that. My opinion: f/b with ctrl or shift are clumsy (although this is what Xara uses), and please don't use [] because I plan to use them for rotation (this is consistent with gqview and I think other image viewers). Maybe Home/End with ctrl will work for moving one layer up/down? They appear intuitive for this task. 5. I have no opinion on the mouse scroll modifiers (I have no mouse wheel), but again, when you agree on something please add that to KeyboardShortcuts. 6. I also plan to finish my HandlingPreferences tutorial and comment on some other new pages on Wiki when I have a little time. 7. I noticed I forgot to update Changelog with my changes all this time - sorry! I've prepared a list of everything I've done and will add it now. --bb -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

Re: [Inkscape-devel] transformation innaccuracies
by Bryce Harrington 23 Nov '03

23 Nov '03

Hi Robert, Could you submit this as a bug report? Thanks, Bryce ---------- Forwarded message ---------- Date: Thu, 20 Nov 2003 20:28:21 -0500 From: Robert Crosbie <swingincelt@...19...> To: inkscape-devel(a)lists.sourceforge.net Subject: [Inkscape-devel] transformation innaccuracies I noticed the transformations applied to objects is quite inaccurate: <rect width="200.000000" height="200.000000" x="100.000000" y="100.000000" style="font-size:12;fill:#808080;fill-rule:evenodd;stroke-width:1pt;" id="rect562" /> Opening the transformations dialog and applying 10 horizontal gives this: <rect width="200.000000" height="200.000000" x="112.500000" y="99.9999695" style="font-size:12;fill:#808080;fill-rule:evenodd;stroke-width:1pt;" id="rect562" /> Also, If I take that original square and rotate it 10, I get this: <rect width="200.000000" height="200.000000" x="62.2319183" y="131.691177" transform="matrix(0.984808,-0.173648,0.173648,0.984808,0.000000,0.000000)" style="font-size:12;fill:#808080;fill-rule:evenodd;stroke-width:1pt;" id="rect562" /> Why have my x and y changed? Shouldn't the matrix handle all aspects of the transformation and leave x and y as they were? Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Inkscape-devel mailing list Inkscape-devel(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel

1 0

mouse wheel
by Yemu 23 Nov '03

23 Nov '03

i've just updated my copy of inkscape and i have one suggestion concerning mouse wheel support. inkscape now uses wheel to move vertically, ctrl+wheel to move horizontally and shift+wheel to zoom in/out. in other applications (mozilla, opera, abiword, openoffice, corel draw) ctrl+wheel zooms (i know, i know in gimp shift+wheel zooms :-). so maybe it would be possible to have ctrl+wheel zoom? or both ctlr+wheel or shift+whell zoom, and for example alt +wheel move horizontally. one more think i noticed is a fact that when scrolling, some black horizontal stripes blink during refreshing of a document. this doesn't happen when zooming. best regards yemu

3 5

Re: [Inkscape-cvs] inkscape
by Bryce Harrington 23 Nov '03

23 Nov '03

Thanks Bulia! On Sat, 22 Nov 2003, bulia byak wrote: > Update of /cvsroot/inkscape/inkscape > In directory sc8-pr-cvs1:/tmp/cvs-serv21746 > > Modified Files: > ChangeLog > Log Message: > list of changes I did not log in the last couple weeks > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Inkscape-cvs mailing list > Inkscape-cvs(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/inkscape-cvs >

1 0