March 2004 - Inkscape Devel - Inkscape Mailman

Passpartout and Inkscape
by Bluefuture 20 Mar '04

20 Mar '04

It could be useful a collaboration between Passpartout (a GTK DTP projects like scribus) and Inkscape in particular for this todo point: * Simple graphics directly in passepartout, probably as a subset of SVG, as well as being able to import SVG graphics. Passpartout home page: http://www.stacken.kth.se/project/pptout/ For contact Passpartout developer managers: Fredrik Arnerup <e97_far@...270...> and Rasmus Kaj <kaj@...270...>. Regars. Blue

2 1

Re: [Inkscape-devel] one more :)
by bulia byak 19 Mar '04

19 Mar '04

>mmmhh, maybe. one of the "speedup" i put was to delay >uncrossing of polygons until they appear in the >visible part of the canvas. the reason was that i add >to open some huge svg files (thousands of elements, >lots of text; automatically generated) and at high >zooms it was still extremely slow, because it was >computing all the polygons, even the ones outside of >the visible portion of the canvas. of course, delaying >the computation means you have a slowdown each time a >new polygon enters the visible area, instead of a >unique slowdown at zoom time. if that behavior bugs >you, comment the line in nr_arena_shape_update() where >shape->delayed_shp is set to true; that way you'll >deactivate delaying. I certainly do not want to desactivate any speedups you're working on :) Instead, I worked around this by increasing the delay which governs accelerated scrolling from 1/8 to 1/4 sec. Previously if some redraw took more than 1/8 sec, it dropped the accumulated acceleration. Now it scrolls smoothly again (at least for keys.svg in fullscreen, which is a rather complex file). _________________________________________________________________ Add photos to your messages with MSN Premium. Get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

small gradient editing improvements
by bulia byak 19 Mar '04

19 Mar '04

In the linear and radial gradient panes of the Fill and Stroke dialog: * You can press Ctrl while dragging handles to snap the linear gradient angle to 15 degrees increments (makes it easy to create strictly horizontal/vertical gradients). * The handles, bounding box, and gradient axes now use inversion for better visibility. * The default linear gradient now goes through the center of the object. * The radial gradient displays a radius handle for adjusting the gradient radius. _________________________________________________________________ Add photos to your e-mail with MSN Premium. Get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

inkscape interface icons
by Jonathan Phillips 19 Mar '04

19 Mar '04

Hello All, I'm now well versed in how to implement new icons in Inkscape. If you design icons and would like to help develop an excellent icon theme for Inkscape, then either A.) jump on chat and I can show you how to edit icons, and/or B.) Make your icons and forward them to me and I will implement them. There are plenty of icons left to designed. Don't forget that artists are developers as well... I'm moving all icons in the application to icons.svg which is by default at: /usr/local/lib/inkscape/icons/icons.svg If you open that file up and copy how those were laid out, and then name the group of the file icon uniquely, that is the best way to understand how to add/edit icons. Forward me any changes and I'll add them into the app. Jon -- Jonathan Phillips <jon@...15...>

6 11

and more...
by bulia byak 18 Mar '04

18 Mar '04

Here's another bad one: https://sourceforge.net/tracker/index.php?func=detail&aid=918478&group_id=9… _________________________________________________________________ http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

RE: [Inkscape-devel] Re: dialog delete handlers
by bulia byak 18 Mar '04

18 Mar '04

>Can someone else have a look and confirm my analysis? I'm not so >familiar with glib/gtk signals, perhaps I've missed something simple. I second that. The difference between data and dlg does seem strange, but I have no idea of what may be happening here. _________________________________________________________________ STOP MORE SPAM with the MSN Premium and get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

one more :)
by bulia byak 18 Mar '04

18 Mar '04

The bug I wrote you about earlier today is fixed, thanks :) Now there's another: https://sourceforge.net/tracker/index.php?func=detail&aid=918458&group_id=9… Besides the scrolling by Ctrl+arrows is now rather uneven - it used to accelerate monotonously but now it moves jerkily. Do you think this is something related to rendering too (e.g. slowdown)? _________________________________________________________________ http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

dialog delete handlers
by bulia byak 18 Mar '04

18 Mar '04

Peter, in align.cpp and transformation.cpp, in delete signal handlers, you wrote static gboolean sp_transformation_dialog_delete(GtkObject *, GdkEvent *, gpointer data) { if ( data != static_cast<gpointer>(dlg) ) { g_warning("possible bug: dlg differs from passed data"); /* E.g. maybe we shouldn't update the global x,y,w,h, or maybe we should use dlg instead of dead_dlg below. */ } GtkWindow &dead_dlg = *static_cast<GtkWindow *>(data); gtk_window_get_position(&dead_dlg, &x, &y); gtk_window_get_size(&dead_dlg, &w, &h); instead of the previous code which simply used dlg, ignoring data. I don't quite understand the reason for this change. Besides: - Why only these two files? The same code is used in all dialogs. - Attempting to close Inkscape with any of these two dialogs open will crash (bug 917122). So for now I restored the old code to fix the crash, but I'm just curious why you deemed this change necessary. _________________________________________________________________ Add photos to your messages with MSN Premium. Get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

2 1

rendering bug
by bulia byak 17 Mar '04

17 Mar '04

Hi, I think it's your recent caching changes that caused this: http://sourceforge.net/tracker/?atid=604306&group_id=93438&func=browse Can you please have a look? _________________________________________________________________ MSN Premium with Virus Guard and Firewall* from McAfee� Security : 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

853858 patch comments
by bulia byak 17 Mar '04

17 Mar '04

1. - GSList *sl; - sl = g_slist_copy ((GSList *) sp_selection_repr_list (selection)); - sl = g_slist_sort (sl, (GCompareFunc) sp_repr_compare_position); + const GSList *items = sp_selection_item_list (SP_DT_SELECTION (desktop)); I.e. you removed sorting the copied list of items. Why? The result is that the copied items now do not preserve z-order when pasted; they're pasted in the order in which they were selected, which is wrong (try it). And you cannot sort the list returned by sp_selectiion_{repr_}list, only a copy of it, which is what was done in the original code. 2. You only copy gradients applied to the fill, not to the stroke, so gradient-filled stroke is lost when copying between documents. 3. Here you check that the fill is either pattern or gradient, and then, without any further check, cast it into a gradient: + if (style && style->fill.type == SP_PAINT_TYPE_PAINTSERVER) { /* Object has pattern or gradient fill*/ + SPGradient *gradient = SP_GRADIENT (SP_OBJECT_STYLE_FILL_SERVER (items->data)); The result is quite predictable - it crashes when you try to copy an object with pattern fill. (Probably with bitmap fill also.) 4. When you paste copied objects with gradients within the same document, it still adds a copy of the gradient (with different id), although the pasted object still refers to the old gradient. This will lead to an unchecked growth of defs each time you do copy/paste. This should be prevented. E.g. check if the current defs contain an exact copy of the gradient you're about to add (not only same id but also same content), and if so, skip adding it. 5. Same thing when you copy an object with gradient to another document and that document already contains a different gradient with the same id. The result is that the pasted object uses that other document's gradient (different from its original gradient), while the correct gradient is added to the defs with a different id (and thus left unused). Fixing this is more complex: if the document you're pasting to contains a gradient with the same id but different content, you must change the id both of the gradient you're inserting to the defs _and_ the reference in the object style. The difficulty being, to check for identical content, you must check not only the content of the repr, but all other reprs it is linked to, and their links, and so on until the tree is exhausted. _________________________________________________________________ MSN Premium helps eliminate e-mail viruses. Get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

2 1