At 12:56 30/09/99 +0200, Michael Natterer wrote: >Federico Mena Quintero wrote: >> >> Fantastic! >> >> Before the GIMP maintainers apply this patch, I would appreciate it if >> they did the following. >> >> 1. This patch removes the control_sel_l and control_sel_r >> fields from the gradient_editor_t structure. If the >> selection is to be an attribute of a gradient, instead of >> the editor's, then please move them to the gradient_t >> structure instead of making them global variables. >> The selection is not an attribute of a gradient, because there's only one selection at a time and its in the active gradient. It would be a waste of memory to put them inside the gradient structure, and it makes sense that they should be global. >Applying Shlomi's patch manually shouldn't be too hard with the indented >version of gradient.c as it's mostly >s/g_editor->control_sel_[lr]/control_sel_[lr]/g and adding new functions, >however the PDB stuff is done by pdbgen now. I'd like to volunteer but >unfortunately I still have no idea of perl, sorry :-( > There are some pitfalls: 1. In two or three GUI functions I replaced their body with a call to cpopup_perform_something. (e.g: cpopup_perform_flip_selection()) and placed the main work code in cpopup_perform_something. 2. Whenever a gradient is switched from a PDB procedure, one has to reset the current selection to its leftmost segment. 3. At initialization of the Gimp, the control_sel_l and control_sel_r has to be set. I know perl quite well and if I can do anything to help I'd like to. Shlomi Fish ---------------------------------------------------------------------- Shlomi Fish shlomif@xxxxxxx Home Page: http://t2.technion.ac.il/~shlomif/ I don't believe in fairies. Oops! A fairy died. I don't believe in fairies. Oops! Another fairy died.
