- Inkscape Devel - Inkscape Mailman

Re: [Inkscape-devel] Release
by Ted Gould 24 Mar '04

24 Mar '04

On Wed, 2004-03-24 at 03:10, Lucas Vieites Fariña wrote: > I suppose a 'feature freeze' means I can still update the gui > translation, doesn't it? Yes, ofcourse! Most projects do a 'string freeze' where they don't change translatable strings to help translators. Now that we have more people starting to translate Inkscape should we start doing that? > How much time is there between a 'feature freeze' and the final > release date? Well, usually I would say a week or so. Because this release has bug targets - it is going to be until the bugs are fixed - so it is a touch special. --Ted

1 0

Re: Supershapes for dia?
by Alan Horkan 24 Mar '04

24 Mar '04

On Sun, 21 Mar 2004, W. Borgert wrote: > Date: Sun, 21 Mar 2004 22:27:41 +0000 > From: W. Borgert <debacle@...153...> > Reply-To: dia-list@...45... > To: dia-list@...45... > Subject: Supershapes for dia? > > This would make a nice addition to dia: http://astronomy.swin.edu.au/~pbourke/curves/supershape/ > (with C code exmample). Any takers? "The supershape equation is an extension of the both the equation of the sphere and ellipse" Very cool. Curved stars and polygons, a weird droplet shape. Ooh pretty! It is like Spirograph* all over again. I think for a complicated sphere/ellipse like this to be really practical it would require something like handles to allow interactive adjustment a feature that Dia doesn't have. I think it would be difficult to make this into a simple to use feature and I'm not even sure the markup would be all that easy generate either (a complex SVG path I'd guess) I'm not sure this is ideal for Dia. In my humble opinion, the ideal would be if other applications would provide the drawing tools Dia needs and Dia would focus on fast production of technical diagrams from premade shapes. A more generic way to plot equations as paths in Dia might be approrpriate to Dia. I have forwarded a copy of this mail to the Inkscape developers who I hope might be interested to implement an advanced Ellipse tool like this. Despite my reservations on how best to implement this it it would certainly be great if someone was interested enough to implement it. Sincerely Alan Horkan http://advogato.org/person/AlanHorkan/ * Obligatory Simpsons quote: "Dr. S: Wait: did you know that there's a direct correlation between the decline of Spirograph and the rise in gang activity? Think about it." PS Please dont crosspost in reply to this message. I have deliberately set the Reply-To to go to me. The Dia list will overwrite this and set the Reply-To as the Dia list and the Inkscape list will leave it alone, so please respond only to the relevant list.

3 7

tutorial news
by bulia byak 24 Mar '04

24 Mar '04

* The Inkscape tutorial is split into two parts, Basic and Advanced, and significantly expanded (new sections on offsets, simplification, compound paths, text tool, and more). Most new material is at the end of the Advanced part. Check it out. _________________________________________________________________ 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

freehand continuity
by bulia byak 23 Mar '04

23 Mar '04

Fred, Huge thanks for your bugfixes today. Having fixed the spoonshapes, maybe you can also look at the continuity problem? It's not exactly a bug, but each segment of a freehand line is approximated without taking into account previous segments, and as a result, all nodes on the path are cusp, to various degrees. It would be much nicer if the approximator would take into account a wider context to create smooth nodes most of the time (and only creating cusps when there are really sharp turns). Do you think this is possible/easy with the current code? _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN Premium http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

2 1

Re: [Inkscape-devel] Gradient Meshes
by bulia byak 22 Mar '04

22 Mar '04

>Now, the question is: Is this useful? Do you think it's just a >half-assed gradient mesh or is it something that extends the artistic >tools available in Inkscape? My IMHO is that while this is nice and cool, it would not be very useful in my design work. I think that the important thing is to get the essentials right, and then users can build their own complex things from these essentialas. With regard to gradients, here's what is still glaringly missing: - elliptic gradients (that can be rotated, with both axes adjustable) - non-linear gradient functions (currently all gradients are linear, which often looks too crude and artificial) - on-canvas editing of gradients When these things are done, I will be almost happy. It will be relatively easy then to emulate most meshes, fields, etc. by overlaying semitransparent elliptic gradients. What would satisfy 100% of my gradient needs is stroke patterns, i.e. the ability to paint a stroke with repeated copies of objects. I would then take a circle with circular transparency gradient and repeat it along a path, to receive a blurred stroke that can be given any shape and combined with other blurred strokes for any gradient-mesh-like effects. Besides stroke patterns, this also requires fixing this bug: http://sourceforge.net/tracker/index.php?func=detail&aid=859579&group_id=93… Still another alternative is a support of SVG filters, most importanly gaussian blur. If you can take arbitrary shapes and blur them into transparency, then again, it'll cover most uses of gradient meshes, and will potentially be more convenient. This is just an IMHO. Others may love to have the gradient meshes in Inkscape and will find unique uses for them, but I'm speaking from my design experience. _________________________________________________________________ MSN Premium: Up to 11 personalized e-mail addresses and 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

2 1

Gradient Meshes
by Ted Gould 22 Mar '04

22 Mar '04

Okay, Mental a few nights ago put out the question: Can we do gradient meshes with standard SVG? So I've been thinking about it. Here is my idea. (also, I went into Illustrator, made a gradient mesh and exported it as SVG. Illustrator just renders it as a bitmap in the SVG file. I'm seriously unimpressed with the Adobe folks, I expect more) Okay, so we can think of a gradient mesh as points of color that are adjustable in their length and path of the color. There can potentially be multiple colors all over the area of the object, each contributing to a complex color created within. I think if we look at each of the points, and then divide the whole object into squares based off of those colors and the next color. So if: ------------------- | | | c1 | | | | | | c2 | | | ------------------- we can look at like: ------------------- | | | | |---c1-------------| | | | | | | | | |------------c2----| | | | | ------------------- Then, make a square for each area with the color that effects it. So there would be 4 squares with c2 and four with c1. Each of these squares would have a radial gradient that goes from the color to transparent. I think that this could also be extended to not be square shapes, so that the color could be 'turned' and expanded in odd ways through transforms of the shapes. I haven't played with this though. Anyway, we take all these squares, and then define them as a pattern for the shape. So that there can be complex shapes without us going crazy, and they can be resized and the mesh gradient grows with the bounding box. Okay, that's the idea. Thoughts? Comments? Does this do what people use gradient meshes for? I haven't played with them alot. --Ted PS - I attached a hand coded SVG file kinda showing what I'm thinking.

1 1

Re: [Inkscape-devel] inkscape interface icons
by Peter Linnell 21 Mar '04

21 Mar '04

One more good reason for a single icon file: packaging. One of the things which has helped to make my life easier is the sane way inkscape is setup as a source tarball and a well done configure script. This made making decent rpms miles easier. If you have spent half a night chasing down missing files (sometimes icons) and fixing broken makefiles trying to build an rpm - you'll understand completely ;) Peter

7 12

[Fwd: [cairo] Cairo SDL Demo]
by Jonathan Phillips 20 Mar '04

20 Mar '04

I keep tabs on CAIRO's list. Totally far out that you can use CAIRO with SDL and SVG. That means that games and multimedia could use SVG now. This is far out and I hope that here soon we will discuss our future plans with CAIRO developers. Jon -----Forwarded Message----- > From: Eric Windisch <lists@...277...> > To: cairo@...278... > Subject: [cairo] Cairo SDL Demo > Date: Fri, 19 Mar 2004 09:23:34 -0500 > > I spoke with Carl on IRC and decided to build a demo for using Cairo > with SDL. I don't have CVS write permission, so I'm providing a > (temporary) link to it. Maybe we will start seeing more SDL programs > using SVG now :) > > http://www.grokthis.net/~raptor/cairo-sdl.tgz > > Feedback is welcome. -- Jon Phillips Graduate Researcher Visual Arts Department PO BOX 948667 LA JOLLA, CA USA cell.858.361.2811 jon@...247... http://www.jonphillips.info

2 1

Re: [Inkscape-devel] my win32 patch does not work, why (splivarot.cpp)
by bulia byak 20 Mar '04

20 Mar '04

Hi, we decided that the best way to unify serialization is via a C++ stream. I just checked in src/svg/stringstream.h which is a beginning of such a class that can be used already. If we use this class throughout, later it will be easy to change precision or number formatting globally for all SVG output, as well as to add custom serializers for our own classes. An example of using this class is in function sp_svg_write_path() in svg-path.cpp. You declare a variable (os) of this class and then feed it with the data you need to serialize using <<. To access the result, you call os.gcharp() which returns const gchar * of the accummulated string, and you'll probably want to also do a g_strdup on it (depending on whether it's going to be used immediately in the same function, or returned or sent elsewhere). So, if you're interested in helping out, you can start searching for printf functions that are used for serializing into SVG (in splivarot and in other places) and replace them with the stream stuff. Of course the first to replace are printfs that serialize floating point numbers. Please do one function at a time and test it thoroughly after the replacement. Do a test patch for one file, and if it works we can give you developer access to CVS to speed things up. _________________________________________________________________ MSN Premium includes powerful parental controls and get 2 months FREE* http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

5 4

inkscape crash on startup
by David Christian Berg 20 Mar '04

20 Mar '04

Yesterday I tried to open the attached document and it wouldn't, but cause inkscape to crash. Since I created it with inkscape I thought, well, get the newest cvs and it'll be working again. but having installed todays cvs inkscape crashes on startup without any error messages of use: ** (inkscape:11051): WARNING **: Emergency save activated Emergency save completed. Inkscape will close now. If you can reproduce this crash, please file a bug at www.inkscape.org with a detailed description of the steps leading to the crash, so we can fix it.Segmentation fault At least I this doesn't help _me_. I hope someone can help me... maybe find an error in the svg code, so that I could at least open it with sodipodi (which still starts up) or anything... I need to finish that flyer over till monday and it really bothers me that I can't open it right now. Thanx in advance David

4 5

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

Re: [Inkscape-devel] inkscape crash on startup
by bulia byak 17 Mar '04

17 Mar '04

> > Even if I try to start it without loading a file. It also > > produces a dialog saying "Inkscape has encountered an internal error and >will > > close now". > > -- > >That's what I meant, when saying on startup... Can you do a backtrace? _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN Premium http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=ht…

1 0

Re: [Inkscape-devel] inkscape crash on startup
by bulia byak 17 Mar '04

17 Mar '04

>I get precisely the same error, using an "inkscape-current" downloaded and >compiled today. Even if I try to start it without loading a file. It also >produces a dialog saying "Inkscape has encountered an internal error and >will >close now". Please file the bug to the tracker, with all relevant details. Especially useful would be a backtrace. _________________________________________________________________ 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

Fileselector
by Ted Gould 17 Mar '04

17 Mar '04

Hey, I was looking through the new fileselector code and found a couple of issues that I think should be worked out. Namely, with the inclusion of input and output formats that are done through extensions. Previously, the file dialog called the extensions code to get a menu that contained all the input/output formats that could be handled, and each one had a key attached to it that specified which extension should be used. I'm not saying that this is a good thing, there is too much GUI code in the extensions system with that solution. But, I still think that the extensions system needs to be queried on the matter, and get that key passed into the save functions. Probably the best thing would be a list of verbs, but I don't know if that is the best option as these aren't actions as much as selections. So, I'm proposing this: GSList * Inkscape::Extension::get_input_list (void); This will return a list of items that are: gchar * name gchar * extension gchar * key And then the GUI code can do what ever it wants with them, so long as it returns the key when it finishes (along with the filename ;) Thoughts? --Ted

3 3

RE: [Inkscape-devel] inkscape crash on startup
by bulia byak 17 Mar '04

17 Mar '04

>Yesterday I tried to open the attached document and it wouldn't, but >cause inkscape to crash. >Since I created it with inkscape I thought, well, get the newest cvs and >it'll be working again. but having installed todays cvs inkscape crashes >on startup without any error messages of use: I can open your attached file with no problem using today's CVS. Except that it warns it can't find the font AF PEPSI and the embedded .tif image is also missing. Maybe it would crash on that .tif or that font, but I can't check that. If you still can't open it, submit a bug and attach all relevant files to it. Crash bugs are usually fixed quickly. _________________________________________________________________ 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

my win32 patch does not work, why (splivarot.cpp)
by Adib Taraben 17 Mar '04

17 Mar '04

Hello, I tried to start to fix those (win32) locale problems with the float numbers. In the file splivarot.cpp which is the "simplify path" command I replaced the printf's with g_ascii_formatd. This function is locale independent. Unfortunately it does not solve the problem. Anyhow what is the desired method generating locale undependent number output? Am I on the right way? Thanks, Adib. --- splivarot_org.cpp 2004-03-14 17:09:02.000000000 +0100 +++ splivarot.cpp 2004-03-16 14:57:26.000000000 +0100 @@ -1542,34 +1542,45 @@ if (typ == descr_moveto) { Path::path_descr_moveto* nData=(Path::path_descr_moveto*)(path->descr_data+theD.dStart); - g_string_sprintfa (result, "M %lf %lf ", nData->p[0], nData->p[1]); + gchar cX[33], cY[33]; + g_string_sprintfa (result, "M %s %s ", g_ascii_formatd(cX, 32, "%f", nData->p[0]), g_ascii_formatd(cY, 32, "%f", nData->p[1])); } else if (typ == descr_lineto) { Path::path_descr_lineto* nData=(Path::path_descr_lineto*)(path->descr_data+theD.dStart); - g_string_sprintfa (result, "L %lf %lf ", nData->p[0], nData->p[1]); + gchar cX[33], cY[33]; + g_string_sprintfa (result, "L %s %s ", g_ascii_formatd(cX, 32, "%f", nData->p[0]), g_ascii_formatd(cY, 32, "%f", nData->p[1])); } else if (typ == descr_cubicto) { Path::path_descr_cubicto* nData=(Path::path_descr_cubicto*)(path->descr_data+theD.dStart); float lastX, lastY; + gchar cX1[33], cY1[33], cX2[33], cY2[33], cX[33], cY[33]; { NR::Point tmp = path->PrevPoint (i - 1); lastX=tmp[0]; lastY=tmp[1]; } - g_string_sprintfa (result, "C %lf %lf %lf %lf %lf %lf ", - lastX + nData->stD[0] / 3, - lastY + nData->stD[1] / 3, - nData->p[0] - nData->enD[0] / 3, - nData->p[1] - nData->enD[1] / 3, nData->p[0],nData->p[1]); + g_string_sprintfa (result, "C %s %s %s %s %s %s ", + g_ascii_formatd(cX1, 32, "%f", lastX + nData->stD[0] / 3), + g_ascii_formatd(cY1, 32, "%f", lastY + nData->stD[1] / 3), + g_ascii_formatd(cX2, 32, "%f", nData->p[0] - nData->enD[0] / 3), + g_ascii_formatd(cY2, 32, "%f", nData->p[1] - nData->enD[1] / 3), + g_ascii_formatd(cX, 32, "%f", nData->p[0]), + g_ascii_formatd(cY, 32, "%f", nData->p[1])); } else if (typ == descr_arcto) { Path::path_descr_arcto* nData=(Path::path_descr_arcto*)(path->descr_data+theD.dStart); + gchar cRX[33], cRY[33], cA[33], cSX[33], cSY[33]; // g_string_sprintfa (result, "L %lf %lf ",theD.d.a.x,theD.d.a.y); - g_string_sprintfa (result, "A %g %g %g %i %i %g %g ", nData->rx,nData->ry, nData->angle, - (nData->large) ? 1 : 0,(nData->clockwise) ? 0 : 1, nData->p[NR::X],nData->p[NR::Y]); + g_string_sprintfa (result, "A %s %s %s %i %i %s %s ", + g_ascii_formatd(cRX, 32, "%f", nData->rx), + g_ascii_formatd(cRY, 32, "%f", nData->ry), + g_ascii_formatd(cA, 32, "%f", nData->angle), + (nData->large) ? 1 : 0, (nData->clockwise) ? 0 : 1, + g_ascii_formatd(cSX, 32, "%f", nData->p[NR::X]), + g_ascii_formatd(cSY, 32, "%f", nData->p[NR::Y])); } else if (typ == descr_close) {

2 1