January 2004 - Inkscape Devel

RE: [Inkscape-devel] 4 signals sent from sliders
by bulia byak 11 Jan '04

11 Jan '04

>The color sliders emit four different signals: > >grabbed >dragged >released >changed > >However, comparable GTK+ widgets only emit a single signal: > >changed > >and follow up with a gboolean xxx_is_adjusting() function. I'm not sure, but the best way to test is to try disabling the extra signals and see if something will break. Commit the change and we'll try to find anything wrong with it. _________________________________________________________________ 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

new keys for the panel
by bulia byak 11 Jan '04

11 Jan '04

The panel The panel at the top of the document window displays a different set of fields and controls for each tool. * Alt+x jump to the first editable field in the secondary toolbar * Enter set the current field's new value (if changed) and leave the secondary toolbar * Esc cancel any changes in the current field and leave the secondary toolbar * Ctrl+z cancel any changes in the current field * Tab, Shift+Tab jump between fields in the secondary toolbar (the value in the field you leave, if changed, is set) * Up/Down arrows change the current value by 0.1 * PgUp/PgDn change the current value by 5.0 This is the basic set of keybindings that makes the panel (aka secondary toolbar, but I like the "panel" term better) usable. Some of this functionality was provided by GTK but had to be cleaned up, some was implemented from scratch. Not to start a flamewar, but my impression of GTK so far is not too favorable: messy, underdocumented, somewhat similar to GIMP itself in its "interface" (I mean the programmer's interface, not only GUI). I don't know if other toolkits are much better, though. Questions for everyone: - alt-x: I like this shortcut, but if anyone will propose a better one I will be glad to hear. - Is alt-x plus tab/shift-tab enough for navigating the toolbar, or do we want each widget to have its own shortcut? Or will this be too much, with potentially tens of widgets in all tools? Of course such shortcuts would use consistent modifiers and would correspond to the widget labels, e.g. for the selector toolbar: ctrl-alt-x, ctrl-alt-y, ctrl-alt-w, ctrl-alt-h. - Now alt-x is implemented very stupidly, through grab_focus of a specific widget. I tried to use GTK "accelerators" but was unable to make them work. Does anyone know if Inkscape uses GTK accelerators? If not, how are the global verbs implemented? If I manage to add an accelerator group to the top-level document window, will it conflict with verbs? Can one point me to a not-too-complex GTK app that uses accelerators so I could figure it out? (The documentation is very scarce.) P.S. I've tried to send this message several times. Guess what, sourceforge was rejecting it as spam because in the phrase "through grab_focus" I had "via" instead of "through" - which matched the regexp for some well-known substance name! Oh my. Hope it will pass through this time. _________________________________________________________________ Check out the new MSN 9 Dial-up � fast & reliable Internet access with prime features! http://join.msn.com/?pgmarket=en-us&page=dialup/home&ST=1

2 1

4 signals sent from sliders
by Jon A.Cruz 11 Jan '04

11 Jan '04

The color sliders emit four different signals: grabbed dragged released changed However, comparable GTK+ widgets only emit a single signal: changed and follow up with a gboolean xxx_is_adjusting() function. I was wondering if there is any reason for the 4-signal method, and whether or not we should keep or lose it.

2 1

test
by bulia byak 10 Jan '04

10 Jan '04

test -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Sodipodi flags 1.6
by Christian Schaller 10 Jan '04

10 Jan '04

Sodipodi flags 1.6 This is the fourth release of the Sodipodi flag collection. This collection currently consist of over 300 different international and local flags. The flags are donated by their authors into the public domain. The flags are in the scalable and free SVG format which today is supported by most major vector graphics programs. Most flags are done using the Sodipodi vector drawing application. In others words these are free flags in a free format made with a free application. The core collection of flags in this collection are meant to be as accurate as possible. If you discover flags in here that have wrong proportions or wrong colors then please update the flag with the correct colors or sizes and send it to us with an explanation. New flags are of course also welcomed, but remember we need a clear statement from you that you dedicate the flag to the public domain, preferably using the Creative Commons public domain dedication process found here: http://creativecommons.org/license/publicdomain-2 We do hope that many of you take some time to help us make this an evenmore complete collection; for instance the missing international flags per ISO 3166 are: bm, BERMUDA ky, CAYMAN ISLANDS fk, FALKLAND ISLANDS (MALVINAS) sh, SAINT HELENA tc, TURKS AND CAICOS ISLANDS vg, VIRGIN ISLANDS, BRITISH We hope that this collection can be useful to people in many different settings be that in relation to software, publishing, education and so on. The assembly of this huge collection has been made possible thanks to the hard work of many people. We big thank you goes to all our contributors listed in the CREDITS file distributed with this package. We would especially like to thank Tobias Jacobs who contributed most of the new flags for this release. You can see all of the currently contributed flags here: http://www.sodipodi.com/index.php3?section=clipart/flags The creation of these flags had not been possible without the hard work that free software coders have put into great tools like Sodipodi, Inkscape and librsvg. The projects have helped increase the viability of SVG in free and open source software a lot. Both these two tools and the new Inkscape project do need more coders in order to improve further. For instance thanks to the help of Caleb Moore so does the latest versions of Nautilus(librsvg) render all these flags perfectly. People interested in working on these and other issues are highly encouraged to join the project mailing-lists. You find the projects here: http://www.sodipodi.org http://librsvg.sf.net http://www.inkscape.org You can download the package in either tar.gz or zip format from the sourceforge file area: http://prdownloads.sourceforge.net/sodipodi/sodipodi-flags-1.6.tar.bz2?down… http://prdownloads.sourceforge.net/sodipodi/sodipodi-flags-1.6.zip?download

1 0

Call for MS Windows peoples
by Jon A.Cruz 09 Jan '04

09 Jan '04

There are a few things for win32 builds that I have been starting to get hammered out. However, I'd probably need some help with it (as I don't think just editing and checking in my changes from Linux would be the best way) So... I'd like to make a call for all present and potential Windows users. If you can compile Inkscape yourself or not, doesn't matter. I'd just like to find who can help maybe just run it on different setups. Windows 95/98/ME users are especially needed. Of people who can compile Inkscape for Win32, we'd like to know your exact setup. It would be nice to be able to try it on a few different build pathways. If you want to just sound off, you can reply directly to me at jon@...155...

3 2

RE: [Inkscape-devel] Warning / bug in nodepath.cpp
by bulia byak 09 Jan '04

09 Jan '04

>Anyway, there is a warning that 'xn' and 'xp' are >used unitialized in nodepath.cpp at line 2081. >There are several places in SP/Ink code where apparently >the author expects an uninitialized double to be 0.0, >and I thought this was one of them. On closer >inspection, the troublesome code is the following. At >the indicated line, if xn and xp are 0.0, that would >make the condition always false. It would be, and >probably is, the same as an '#IF 0'. I doubt if >that is the intention. What should we do? Thanks for the catch! I think it's my oversight. I will now fix this. _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/photos&pgmarket=en-ca&RU=http%3a%2f%2fjo…

2 1

RE: [Inkscape-devel] Win32 & SHGetSpecialFolderPath()
by bulia byak 09 Jan '04

09 Jan '04

>Creates > >\Windows\Profiles\%USER%\Application Data\Inkscape I have windows directory named c:\win98se. Is this covered? _________________________________________________________________ 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…

2 1

Warning / bug in nodepath.cpp
by Bob Jamison 09 Jan '04

09 Jan '04

I was going through some files, and cleaning up the minor warnings we get with -Wall. However, some are not minor! (-Wall is good. -Wall is our friend!) Anyway, there is a warning that 'xn' and 'xp' are used unitialized in nodepath.cpp at line 2081. There are several places in SP/Ink code where apparently the author expects an uninitialized double to be 0.0, and I thought this was one of them. On closer inspection, the troublesome code is the following. At the indicated line, if xn and xp are 0.0, that would make the condition always false. It would be, and probably is, the same as an '#IF 0'. I doubt if that is the intention. What should we do? Bob node_request (SPKnot *knot, NRPoint *p, guint state, gpointer data) { SPPathNode * n; double yn, xn, yp, xp; double an, ap, na, pa; double d_an, d_ap, d_na, d_pa; gboolean collinear = FALSE; NRPoint c; NRPoint pr; n = (SPPathNode *) data; if (state & GDK_CONTROL_MASK) { // constrained motion ##### THIS LINE #######: if (state & GDK_MOD1_MASK && !(xn == 0 && xp == 0)) { // sliding on handles, only if at least one of the... // calculate relative distances of control points yn = n->n.pos.y - n->pos.y; xn = n->n.pos.x - n->pos.x; if (xn < 0) { xn = -xn; yn = -yn; } // limit the ....

1 0

Win32 & SHGetSpecialFolderPath()
by Bob Jamison 09 Jan '04

09 Jan '04

I tested it on a Win98 box, and it works nicely. So I think we have the Win9x case covered by this function, and the NT+ case covered by the glib call. Creates \Windows\Profiles\%USER%\Application Data\Inkscape if it doesn't exist, and puts preferences.xml in it. So I -think- we are done with this one. Bob

1 0