- Inkscape Devel - Inkscape Mailman

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