RSS/Atom feed Twitter
Site is read-only, email is disabled

Things left to do with gfig.

This discussion is connected to the gimp-developer-list.gnome.org mailing list which is provided by the GIMP developers and not related to gimpusers.com.

This is a read-only list on gimpusers.com so this discussion thread is read-only, too.

4 of 4 messages available
Toggle history

Please log in to manage your subscriptions.

Things left to do with gfig. David Odin 14 Nov 12:18
  Things left to do with gfig. Karine Proot 14 Nov 13:29
   Things left to do with gfig. David Odin 14 Nov 17:50
Things left to do with gfig. shaneyfelt@juno.com 15 Nov 02:23
David Odin
2004-11-14 12:18:43 UTC (over 19 years ago)

Things left to do with gfig.

Hi,

As you may know the gfig plug-in was in bad shape, and I've been working on it for a few weeks. Still, there are some issues. I've tried to make a list (ordered by severity):

1 Major. 1.1 deleting an object, by using the delete tool or even undo mess up the current_style variable and can cause a crash on the next object creation. This will be easier to fix when 3.2 and 3.3 will be done.

2 Normal. 2.1 The undo button should be moved to the toolbar. 2.2 The 'select object' tool seems useless, and might be removed. 2.3 Find out how the gradient fill is supposed to work (it is unimplemented for now), or remove it from the filling choice.

3 Cleanups. 3.1 gfig should use glist in some places instead of reimplementing its own structures (DAllObjs, DObjpoints), undos and styles would also benefit to use something else than a static array. 3.2 dobject is used as an abstract class for every kind of objects (arc, line, bezier, etc.) This should be redone using the gobject system.
3.3 the style class should also be derived from a gobject, so it can be easily refcounted.
3.4 use our coding standards all over the place.

4 Tests. As I focus only on some points, I certainly missed some important parts/bugs. For instance, I haven't try to test the saving and loading code.

As I said, 1.1 is the major problem for now. Depending on the time we have I would like to have 3.2 and 3.3 fixed before dealings with 1, but it isn't really necessary.

All the points in 2 are very easy to fix once it has been decided how to fix them.

I'm currently working on 3.1.

Any help is of course welcome, but please contact me here or on #gimp before coding anything, to avoid duplicate work.

Regards,

DindinX

Karine Proot
2004-11-14 13:29:58 UTC (over 19 years ago)

Things left to do with gfig.

David Odin wrote:

[...]

2 Normal.
2.1 The undo button should be moved to the toolbar. 2.2 The 'select object' tool seems useless, and might be removed. 2.3 Find out how the gradient fill is supposed to work (it is unimplemented for now), or remove it from the filling choice.

[...]

All the points in 2 are very easy to fix once it has been decided how to fix them.

I'd like to work on them as soon as things are decided.

Karine

David Odin
2004-11-14 17:50:40 UTC (over 19 years ago)

Things left to do with gfig.

On Sun, Nov 14, 2004 at 01:29:58PM +0100, Karine Proot wrote:

David Odin wrote:

[...]

2 Normal.
2.1 The undo button should be moved to the toolbar. 2.2 The 'select object' tool seems useless, and might be removed. 2.3 Find out how the gradient fill is supposed to work (it is unimplemented for now), or remove it from the filling choice.

[...]

All the points in 2 are very easy to fix once it has been decided how to fix them.

I'd like to work on them as soon as things are decided.

2.1 should really happen. (I would very much appreciate a patch for this).
2.2 should be fixed the right way, so the selected object can become the current one (it doesn't seems to work atm). So my point of view is that the select object tool should stay. 2.3 is only a matter of calling the right function in gfig-dialog.c:paint_layer_fill() to use a shapeburst gradient fill since it is the only gradient fill that doesn't need a direction. Then again, I would appreciate a patch or even a direct commit since I don't intend to change this function in a near future.

Regards,

DindinX

shaneyfelt@juno.com
2004-11-15 02:23:48 UTC (over 19 years ago)

Things left to do with gfig.

I've been playing with methods of implementing polar grids that are not congested in the center, related to Bug# 151092 (GFig's Polar Grid: Independent Sector Spacing Control)

Here's the URL of an applet I wrote to test and demo the concept. http://cs.uhh.hawaii.edu/cs/courses/cs200/grids/

I could update gfig-grid.c if you like, but I'd like some direction (from someone who's done it recently) setting up a build environment for it on Windows to test it. (Of course I have the Java source if you want to take and adapt it yourself, or I could let someone else compile and test under Linux if that's easier)

_-T

_________________