- Inkscape Devel - Inkscape Mailman

RE: [Inkscape-devel] PathEstimation
by bulia byak 23 Dec '03

23 Dec '03

>you can see the feature on >http://inkscape.org/cgi-bin/wiki.pl?PathEstimation > >( When will the PathEstimation feature be available in Inkscape? For me, >it is the most important feature (for the design of logotypes, >arabesques... ) ) Outlining a stroke and insetting/outsetting a path are already in CVS. You can download the CVS and use it, or wait for the next release. _________________________________________________________________ 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

save confirm dialog
by bulia byak 23 Dec '03

23 Dec '03

I committed your new save confirm dialog. Some comments: >See Bug #863856 for details about the save confirmation >dialog. This is my attempt to make it 99% compliant to the HIG and >I believe I succeeded. In fact, this might be the _first_ dialog >ever to (almost) completely match the HIG suggestion. :) >Considering the amount of work this required, I can also very well >understand now, why this is the case... It's good to be HIG compliant, but the new dialog is HUGE. Still, since it's not going to stay on screen all the time, I'm OK with that. Being wordy also won't hurt for novices. >Additionally, I commented out the non-standard dialog placing code >(placing the dialog near the mouse pointer), as dialogs should >really be placed by the window manager (at least that's what GNOME >users will expect). I only commented the appropriate code out, so >if you disagree about this, it would be easy to change again. I still think popping a dialog at the mouse is better for usability, but again, this dialog is not so frequently needed so I think it's better to go the simple (and standards-compliant) route. Popping the fialog in the center of the document window seems to be conventional, so let it be that way. >There was also an issue when the user clicked "Save", but then >canceled the dialog. This situation wasn't detected and so the >document was closed, which is bad. To solve this, I made a few >functions return a gboolean and this also solved a TODO item on >the way. :) Excellent, thanks! >Other stuff that might still be done: >- I noticed that overwriting of files wasn't handled at all, but >didn't want to get too much distracted... It would be very nice if someone could finally take care of that. Recently Ted Gould introduced a module-based save system, and I saw some TODOs for this in his code, so you should perhaps discuss this with him first. >- Of course it would be very nice to show in the dialog, how much >time has passed since the document was last saved. This shouldn't >be very complicated, but I didn't want to get too deep into the >code (yet), because I'm very unexperienced. :) Not sure about the time (although why not?), but one thing we absolutely need is an indication of whether the sodipodi:modified is on or not. On windows, it's conventional to add * to the filename in the title bar, don't know what HIG says about this. >I hope you take a look at the patch and tell me if I did anything >wrong, because this is in fact the first patch I ever made. Not that _I_ am a a very experienced programmer, but I did not see any problems with your patch. Everything seems neat. Are you interested in doing more Inkscape work? If so we could give you developer access to the CVS. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

3 2

abs paths
by bulia byak 23 Dec '03

23 Dec '03

I found these functions for converting between absolute and relative pathnames: http://mail.gnome.org/archives/gtk-devel-list/2002-April/msg00142.html They were apparently proposed for glib but not accepted. I copied g_rel2abs() into document.cpp and used it for calculating the absolute pathname. It seems to work well for paths with any number of "." or "..", and the list of recent files now contains absolute pathnames (this fixes https://sourceforge.net/tracker/index.php?func=detail&aid=860202&group_id=9…). Please test. If any other file will need this function we'll probably need to separate it into a file of its own. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Re: [Inkscape-devel] questions
by bulia byak 23 Dec '03

23 Dec '03

> > Mine is 3.3.1. Do you think it's worth the trouble to protect 3.3.1 >users > > from this crash (which is admittedly rare, you'll need to scale a text > > string to microscopic sizes to get it)? > >Can you upgrade to 3.3.2 and try to replicate the bug, so we can rule >out factors other than the compiler? Just upgraded to 3.3.2 - alas, the bug is still there. What else can be done, aside from the "magic fix"? Here's my g++ -v: Reading specs from /usr/lib/gcc-lib/i586-mandrake-linux-gnu/3.3.2/specs Configured with: ../configure --prefix=/usr --libdir=/usr/lib --with-slibdir=/lib --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --enable-long-long --enable-__cxa_atexit --enable-languages=c,c++,ada,f77,objc,java,pascal --host=i586-mandrake-linux-gnu --with-system-zlib Thread model: posix gcc version 3.3.2 (Mandrake Linux 10.0 3.3.2-3mdk) _________________________________________________________________ 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

RE: dnd [was: Re: export patches]
by bulia byak 23 Dec '03

23 Dec '03

>On Mo, Dez 22, 2003 at 09:15:50 +0000, bulia byak wrote: > > Hi, > > > > I tested your dnd patch but I don't seem to be able to make it work. If >I > > type in some SVG in your test program and drag it into inkscape, no >object > > appears on the canvas, but there is a new empty <g> in the XML editor. >If I > > type <text>abc</text>, after dragging it becomed <g id="123">abc</g>. > >Try typing/pasting a whole svg document, that's what The Gimp will send, >when their path dnd support is ready :) Still does not work. Now the problem is namespaces. If I drag a document containing one <rect>, it appears as <svg:rect> within inkscape and is not displayed. (Even though in original SVG, there was no prefix.) This is because internally inkscape, simplistically, stores names with prefixes, so a mismatch in a prefix makes an element unrecognizable. I'm not sure what is the best/simplest to fix this: make it so that dragged SVG is inserted without namespace prefixes, or make it so that Inkscape can handle full qualified names (with namespace URI and prefix for each name). I guess the first way is simpler. Mental, can you comment? _________________________________________________________________ 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

RE: [Inkscape-devel] patch to add point class to libnr
by bulia byak 23 Dec '03

23 Dec '03

Nathan, recently I added a few small functions to nodepath.cpp that I needed for various path operations. They do things like finding the point closest to a given point on a given line (the line is assumed to go through the coordinate center and thus is given only by its tangent), distance from point to line, as well as converting from x/y coordinates to radial and back. I realize that these functions probably belong in a different file, and maybe even in your point class (or maybe they are already implemented somewhere, although I could not find them). However I do not know much of C++ and thus cannot move them there by myself. If you could do that it would be nice, and reading your converted code might be my first tutorial in C++ :) _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Re: [Inkscape-devel] Proposal for New Subsystem Architecture(andNamespaces)
by bulia byak 23 Dec '03

23 Dec '03

>My algorithm does text in shapes properly. In every other way the two are >equivalent (although not the same). TeX also provides Glue, which is easy >to add, but I never needed it for my work. It should go into inkscape, >however. I think mine is easier to understand in any case. I've put it in >experimental/njh/dyn-text/ Just make it. run ./dyn-text. I also don't see if it can do hyphenation. Can it? In any case, it would be fantastic to get this into Inkscape. But I foresee a lot of work with the UI. _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] [Fwd: Re: [Sodipodi-list] How do people use Sodipodi?]
by bulia byak 22 Dec '03

22 Dec '03

>I like the idea of making the text stuff more immediate. Of course, most text controls will be in the secondary toolbar at the top, when I (or anyone else, volunteers welcome) have time to put it there. It was discussed and put on the Wiki long ago. _________________________________________________________________ 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

[Fwd: Re: [Sodipodi-list] How do people use Sodipodi?]
by Nathan Hurst 22 Dec '03

22 Dec '03

I like the idea of making the text stuff more immediate. njh

1 0

Coding conventions
by unknown＠example.com 22 Dec '03

22 Dec '03

Greetings, So, what sort of coding conventions do people want to adopt? I'm not really a C++ coder so I wont offer up too many suggestions, but I do intend on doing what I can to 'help' developers follow the decided upon conventions by creating Emacs and Vim modelines for the source files...unless someone has a good argument against doing that. At any rate, the only thing that I think is important is 'no tabs'...anyone care to suggest/discuss/argue-without-starting-a-war other conventions (this would include filename schemes for C++ as mentioned in the booleans thread)? *g* -Tom

4 5