December 2003 - Inkscape Devel

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

08 Dec '03

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

1 0

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

08 Dec '03

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

6 7

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

08 Dec '03

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

5 6

Inkscape 0.36-pre1
by Ted Gould 08 Dec '03

08 Dec '03

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

1 0

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

08 Dec '03

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

3 5

Ruler issues
by Michael D. Seymour 08 Dec '03

08 Dec '03

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

1 0

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

08 Dec '03

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

1 0

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

08 Dec '03

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

1 0

node edit news
by bulia byak 08 Dec '03

08 Dec '03

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

3 3

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

08 Dec '03

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

2 2