- Inkscape Devel - Inkscape Mailman

Re: [Inkscape-devel] button size
by bulia byak 01 Dec '03

01 Dec '03

For now I would vote for: #define TOOL_BUTTON_SIZE 28 #define AUX_BUTTON_SIZE 22 but I think we need to try to add more stuff to the secondary toolbar to see how it will look. > Well, as it is the toolbar size forces the window to a fairly large > minimum height. I'm a little uncomfortable forcing a much larger > minimum size. We have a lot of room to grow :) Currently the default window is very small - 400x400 if without the toolbar strut. BTW maybe we should increase this a bit, say to 640x480. But ultimately the size and maximized state of the document window should be remembered from previous session. > Do you think it would be okay to permit the button bar to be cut off if > the window is resized smaller? I think the main toolbar should not be cut (and it's not likely to be cut anyway), but the secondary toolbar must be permitted to be cut off if the window is small. > Well, perhaps I should use the GTK detachable toolbar widget there, > though the plain GTK one is kind of weak. :/ What is wrong with it? Is there an alternative? Maybe in Gtkmm? It would be nice to have this done without coding it from scratch :) -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Warnings for unimplemented SVG styles
by Bryce Harrington 01 Dec '03

01 Dec '03

I've put in a change to the codebase for some warning messages. Previously, if Inkscape encountered style properties that were not implemented, it would issue a generic warning such as: Unimplemented style property id: 1042 I've created specific warnings for each feature, so now it will report: Unimplemented style property id SP_PROP_FONT_SIZE_ADJUST: value: -4 This won't affect the quantity of warnings issued, just their specificity. I figured as we strive to gain better SVG compliance, we'll want to know more precisely what features we're missing. The warnings are all in style.c. If you're interested in trying your hand at adding a new SVG style property, have a look-see there at the sp_style_merge_property function. Bryce

1 0

Re: [Inkscape-devel] button size
by bulia byak 01 Dec '03

01 Dec '03

Also, I think it makes sense to establish a certain difference in style between the primary and secondary buttons. The primaries may be more 3Dish, with shades and highlights; the primaries should be much simpler, with less colors, without any highlights or gradients. This contrast need not be too drastic, but still it may be helpful. Thus, from this viewpoint, the current node editor secondaries are OK, while for the selector, only the three last buttons (rotate and flip) are OK, others must be made simpler, flatter and more laconic. The same goes for zoom secondaries which are currently "too beautifil". -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0

Re: [Inkscape-devel] button size
by Robert Crosbie 01 Dec '03

01 Dec '03

IMHO, I think the top bar could use the larger buttons too. Rob. http://members.rogers.com/rcrosbie >From: MenTaLguY <mental@...3...> >To: Inkscape ML <inkscape-devel(a)lists.sourceforge.net> >Subject: [Inkscape-devel] button size >Date: Sun, 30 Nov 2003 18:12:50 -0500 > >I made the tool selection buttons along the side larger than the buttons >for various commands up top. > >While I'd experimented with this a while back and abandoned it, it's >since been suggested a number of times (most recently by Jon), and Xara >appears to do so as well, I decided to try it again. > >See what y'all think... > >-mental ><< signature.asc >> _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

[Fwd: Re: [Sodipodi-list] Win32 0.33 prerelease beta2]
by MenTaLguY 30 Nov '03

30 Nov '03

For those of you working on the Inkscape Win32 stuff... (you might want to hit the sodipodi list archives and follow the whole thread though) -----Forwarded Message----- From: Mitsuru Oka <oka326@...90...> To: Lauris Kaplinski <lauris@...39...> Cc: sodipodi-list(a)lists.sourceforge.net Subject: Re: [Sodipodi-list] Win32 0.33 prerelease beta2 Date: Mon, 01 Dec 2003 04:20:09 +0900 Lauris Kaplinski <1070186315.8512.36.camel@...91...> wrote: >Hello! > >I wonder, if gtk+ is built with multibyte charset, and sodipodi >uses Unicode, is the result functional at all? >We can switch back to multibyte, if it is needed, although this makes >things much uglier and error-prone. Yeah, It's reasonable. I found that dirty hack also works on win98: Index: nr-type-w32.c =================================================================== RCS file: /cvs/gnome/sodipodi/src/libnrtype/nr-type-w32.c,v retrieving revision 1.18 diff -u -r1.18 nr-type-w32.c --- nr-type-w32.c 29 Nov 2003 23:50:52 -0000 1.18 +++ nr-type-w32.c 30 Nov 2003 19:15:43 -0000 @@ -15,6 +15,8 @@ #include <stdio.h> #include <tchar.h> +#include <glib.h> + #include <libarikkei/arikkei-dict.h> #include <libarikkei/arikkei-strlib.h> @@ -565,11 +567,16 @@ nr_type_w32_typefaces_enum_proc (LOGFONT *lfp, TEXTMETRIC *metrics, DWORD fontType, LPARAM lParam) { if (fontType == TRUETYPE_FONTTYPE) { - LOGFONT lf; + GSList *list; + LOGFONT *lf; + + lf = nr_new (LOGFONT, 1); + *lf = *lfp; - lf = *lfp; + list = g_slist_append (*(GSList **)lParam, lf); + *(GSList **)lParam = list; - EnumFontFamiliesEx (hdc, &lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, lParam, 0); +/* EnumFontFamiliesEx (hdc, &lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, lParam, 0); */ } return 1; @@ -578,12 +585,14 @@ static void nr_type_w32_init (void) { + GSList *font_que = NULL; LOGFONT logfont; int pos, i; /* g_print ("Loading W32 type directory...\n"); */ - hdc = CreateDC (L"DISPLAY", NULL, NULL, NULL); + /* hdc = CreateDC (L"DISPLAY", NULL, NULL, NULL); */ + hdc = CreateDCA ("DISPLAY", NULL, NULL, NULL); arikkei_dict_setup_string (&familydict, 131); arikkei_dict_setup_string (&namedict, 537); @@ -591,7 +600,14 @@ /* read system font directory */ memset (&logfont, 0, sizeof (LOGFONT)); logfont.lfCharSet = DEFAULT_CHARSET; - EnumFontFamiliesEx (hdc, &logfont, (FONTENUMPROC) nr_type_w32_typefaces_enum_proc, 0, 0); + EnumFontFamiliesEx (hdc, &logfont, (FONTENUMPROC) nr_type_w32_typefaces_enum_proc, (LPARAM)&font_que, 0); + + while (font_que) { + LOGFONT *lf; + lf = (LOGFONT*)font_que->data; + EnumFontFamiliesEx (hdc, lf, (FONTENUMPROC) nr_type_w32_inner_enum_proc, 0, 0); + font_que = g_slist_remove (font_que, (gpointer)lf); + } /* Fill in lists */ NRW32Families.length = families_len; Best regards,

1 0

Re: [Inkscape-devel] paste style
by bulia byak 30 Nov '03

30 Nov '03

> It seems like when copying to the clipboard, there needs to be a > mechanism to scan the selected objects for any referenced objects > and also copy the defs for those. The recipient of the paste > would then need to add those refs to its own set of defs. It's > possible that the receiving document could a) have a different def > by the same name, or b) have the same def by a different name; in > case (a) it would seem prudent to rename the incoming style, and > in case (b) perhaps it should normally merge by default but allow > the user an option to not merge when pasting. That sounds like a lot of code. Maybe we can simplify it a bit as follows: - a new clipboard is made for storing defs - whenever anything is cut or copied, the entire defs from the current document are placed on that defs clipboard - whenever clipboard objects (or style) are pasted, the contents of the defs clipboard are merged with the current document defs: -- same name, same content: disregard -- same name, different content: rename both defs and the pasted objects' references (this is the most tricky part) -- different name, same content: just add, no merge An advantage of this approach is that you don't need to scan your objects and select relevant defs when copying. However, you'll still have to scan and re-id what you are pasting in case of name conflicts (although these are likely to be rare). -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

2 1

paste style
by bulia byak 30 Nov '03

30 Nov '03

2003-11-29 bulia byak <bulia@...23...> * src/selection-ch emistry.c src/selection-chemistry.h src/verbs.c src/verbs.h src/shortcuts.c src/interface.c: new function, verb, shortcut (shift-ctrl-v), menu command: paste style (apply the style of copied object to selection). If more than one object is copied, the first one in the clipboard list is taken as the source of the style. One problem with both paste and the new paste style: gradients are not copied between documents. This is because the CSS refers to a gradient which is stored separately in <defs>, and it's not copied or pasted when you copy or paste an object using it. The same will apply to bitmap fills and masks when we implement them. Did anyone think about how to solve this in the most generic and elegant way? -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

3 2

g_free cleanup
by Peter Moulder 30 Nov '03

30 Nov '03

This is a low-priority cleanup that may be objectionable, so I'll post here first. According to the g_free documentation, g_free accepts a NULL argument (as a no-op), yet much code explicitly checks for NULL before calling g_free. Disadvantage of the transformation: - Involves a function call even in the NULL case (unless g_free is a macro or inline function that tests for NULL before passing to the "real" version). I've left a couple of places untransformed where it looks like NULL has a significant probability. If NULL is unlikely, then we're better off having more readable code [and saving a couple of bytes of object code]. - May make diff3 against sodipodi code harder. Strong candidates for applying the transformation are where the tested pointer is something long like `style->text->font_family.value', and src/module.c where it uses a IF_NOT_NULL_FREE macro. Btw, some (but not all) of the searching for this pattern was using a regexp to reduce the probability of accidentally transforming `if (foo) g_free (bar)' for distinct `foo' and `bar'. Any comments on where this transformation (removing the `if') should be applied? In absence of comments, I'll probably commit the patch in its current form (other than ChangeLog date). pjrm. Index: ChangeLog =================================================================== RCS file: /cvsroot/inkscape/inkscape/ChangeLog,v retrieving revision 1.56 diff -d -p -U5 -r1.56 ChangeLog --- ChangeLog 29 Nov 2003 09:20:28 -0000 1.56 +++ ChangeLog 30 Nov 2003 03:34:38 -0000 @@ -7,10 +7,33 @@ Make utest.c,utest.h formatting conform more to doc/coding-style.txt. Have utest.c built into libutest.a (see src/utest/Makefile.am comments). + * src/document.c (sp_document_new), file.c (file_open_ok, + sp_file_open_dialog, sp_file_save_dialog, sp_file_do_import), + main.c (sp_main_console), src/module.c (sp_module_finalize, + sp_module_input_finalize, sp_module_input_build, + sp_module_output_finalize, sp_module_output_build), + src/sp-anchor.c (sp_anchor_set), src/sp-image.c (sp_image_set), + src/sp-text.c (sp_string_release, sp_string_read_content, + sp_string_set_shape), src/sp-use.c (sp_use_release), src/style.c + (sp_style_unref, sp_style_merge_property, + sp_style_merge_from_parent, sp_text_style_unref, + sp_style_read_istring), src/bonobo/embeddable-document.c + (sp_bonobo_stream_read, sp_embeddable_document_ps_save), + src/dialogs/stroke-style.c (sp_stroke_style_scale_line), + src/dialogs/text-edit.c (sp_text_edit_dialog_text_changed), + src/helper/sodipodi-ctrl.c (sp_ctrl_build_cache), + src/modules/ps.c (sp_ps_print_image), + src/widgets/font-selector.c (sp_font_preview_set_phrase), + src/widgets/gradient-image.c (sp_gradient_image_size_allocate), + src/xml/repr-action.c (coalesce_chgattr, coalesce_chgcontent, + free_action), src/xml/repr.c (repr_finalize, sp_repr_merge): + Minor cleanup: replace `if (POINTER_EXPR) g_free (POINTER_EXPR)' + with the simpler (more readable) `g_free (POINTER_EXPR)'. + * src/xml/repr-action.c (reverse_log): Add doc comment. Rename some vars. * src/xml/repr-action.c (sp_repr_replay_log, sp_repr_undo_log), src/xml/repr-action.h, src/document-undo.c (sp_document_cancel, sp_document_redo, sp_document_undo): Remove the unused doc Index: src/document.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/document.c,v retrieving revision 1.9 diff -d -p -U5 -r1.9 document.c --- src/document.c 29 Nov 2003 21:55:57 -0000 1.9 +++ src/document.c 30 Nov 2003 03:34:38 -0000 @@ -334,12 +334,12 @@ sp_document_new (const gchar *uri, unsig name = g_strdup_printf (_("New document %d"), ++doc_count); } doc = sp_document_create (rdoc, uri, base, name, advertize, keepalive); - if (base) g_free (base); - if (name) g_free (name); + g_free (base); + g_free (name); return doc; } SPDocument * Index: src/file.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/file.c,v retrieving revision 1.10 diff -d -p -U5 -r1.10 file.c --- src/file.c 30 Nov 2003 03:29:01 -0000 1.10 +++ src/file.c 30 Nov 2003 03:34:38 -0000 @@ -99,11 +99,11 @@ file_open_ok (GtkWidget *widget, GtkFile gchar *filename; filename = g_strdup (gtk_file_selection_get_filename (fs)); if (filename && g_file_test (filename, G_FILE_TEST_IS_DIR)) { - if (open_path) g_free (open_path); + g_free (open_path); if (filename[strlen(filename) - 1] != G_DIR_SEPARATOR) { open_path = g_strconcat (filename, G_DIR_SEPARATOR_S, NULL); g_free (filename); } else { open_path = filename; @@ -112,11 +112,11 @@ file_open_ok (GtkWidget *widget, GtkFile return; } if (filename != NULL) { gchar* key; - if (open_path) g_free (open_path); + g_free (open_path); open_path = g_dirname (filename); if (open_path) open_path = g_strconcat (open_path, G_DIR_SEPARATOR_S, NULL); key = (gchar*)g_object_get_data (G_OBJECT (fs), "type-key"); sp_file_open (filename, key); g_free (filename); @@ -142,11 +142,11 @@ sp_file_open_dialog (gpointer object, gp { #ifdef WIN32 char *filename; filename = sp_win32_get_open_filename (open_path, "SVG files\0*.svg;*.svgz\0All files\0*\0", _("Select file to open")); if (filename) { - if (open_path) g_free (open_path); + g_free (open_path); open_path = g_dirname (filename); if (open_path) open_path = g_strconcat (open_path, G_DIR_SEPARATOR_S, NULL); sp_file_open (filename, NULL); g_free (filename); } @@ -199,11 +199,11 @@ sp_file_save_dialog (SPDocument *doc) char *filename; unsigned int spns; filename = sp_win32_get_save_filename (save_path, &spns); if (filename && *filename) { sp_file_do_save (doc, filename, (spns) ? SP_MODULE_KEY_OUTPUT_SVG_INKSCAPE : SP_MODULE_KEY_OUTPUT_SVG); - if (save_path) g_free (save_path); + g_free (save_path); save_path = g_dirname (filename); save_path = g_strdup (save_path); g_free (filename); } #else @@ -320,11 +320,11 @@ sp_file_do_import (SPDocument *doc, cons const gchar *e, *docbase, *relname; SPRepr * repr; SPReprDoc * rnewdoc; if (filename && g_file_test (filename, G_FILE_TEST_IS_DIR)) { - if (import_path) g_free (import_path); + g_free (import_path); if (filename[strlen(filename) - 1] != G_DIR_SEPARATOR) { import_path = g_strconcat (filename, G_DIR_SEPARATOR_S, NULL); } else { import_path = g_strdup (filename); } Index: src/main.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/main.c,v retrieving revision 1.9 diff -d -p -U5 -r1.9 main.c --- src/main.c 27 Nov 2003 06:14:27 -0000 1.9 +++ src/main.c 30 Nov 2003 03:34:38 -0000 @@ -372,11 +372,11 @@ sp_main_console (int argc, const char ** } } fl = g_slist_remove (fl, fl->data); } - if (printer) g_free (printer); + g_free (printer); inkscape_unref (); return 0; } Index: src/module.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/module.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 module.c --- src/module.c 16 Nov 2003 07:09:42 -0000 1.5 +++ src/module.c 30 Nov 2003 03:34:38 -0000 @@ -27,20 +27,10 @@ #include "document.h" #include "module.h" #include "modules/db.h" -/** - \def IF_NOT_NULL_FREE - - This is a quick little macro to check if a value is null, and - if it is not, then free the data. It's used in most of the - finalize functions, and then also when updating values. It - doesn't do that much, but it makes the code easier to read -*/ -#define IF_NOT_NULL_FREE(x) if ((x) != NULL) g_free((x)) - /* SPModule */ static void sp_module_class_init (SPModuleClass *klass); static void sp_module_init (SPModule *module); static void sp_module_finalize (GObject *object); @@ -52,11 +42,11 @@ static SPModule * sp_module_new (GType t /** \var module_parent_class This is the parent class for the modules. It should be - GObject - but no promises */ + GObject - but no promises */ static GObjectClass * module_parent_class; /** \return The type value for a SP Module \brief A quick way to get the type value for a SPModule @@ -157,12 +147,12 @@ sp_module_finalize (GObject *object) sp_module_db_unregister (module); if (module->repr) sp_repr_unref (module->repr); - IF_NOT_NULL_FREE(module->name); - IF_NOT_NULL_FREE(module->id); + g_free (module->name); + g_free (module->id); G_OBJECT_CLASS (module_parent_class)->finalize (object); return; } @@ -285,11 +275,11 @@ sp_module_unload_default (SPModule * mod should be created. \param repr The XML structure of the module definition. This function is used to build an SPModule (really a subclass of that though) from an XML description stored in a SPRepr tree. If the class - has it's own build function, then that function is used. Otherwise + has its own build function, then that function is used. Otherwise the generic 'sp_module_private_build' function is used. A reference to the SPRepr structure is added in this function so it should not need to be added by each individual subclass. */ @@ -424,14 +414,14 @@ sp_module_input_finalize (GObject *objec { SPModuleInput *imod; imod = (SPModuleInput *) object; - IF_NOT_NULL_FREE(imod->mimetype); - IF_NOT_NULL_FREE(imod->extension); - IF_NOT_NULL_FREE(imod->filetypename); - IF_NOT_NULL_FREE(imod->filetypetooltip); + g_free (imod->mimetype); + g_free (imod->extension); + g_free (imod->filetypename); + g_free (imod->filetypetooltip); G_OBJECT_CLASS (input_parent_class)->finalize (object); return; } @@ -470,23 +460,23 @@ sp_module_input_build (SPModule *module, while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "input")) { child_repr = sp_repr_children(child_repr); while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "extension")) { - IF_NOT_NULL_FREE(imod->extension); + g_free (imod->extension); imod->extension = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "mimetype")) { - IF_NOT_NULL_FREE(imod->mimetype); + g_free (imod->mimetype); imod->mimetype = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypename")) { - IF_NOT_NULL_FREE(imod->filetypename); + g_free (imod->filetypename); imod->filetypename = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypetooltip")) { - IF_NOT_NULL_FREE(imod->filetypetooltip); + g_free (imod->filetypetooltip); imod->filetypetooltip = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } child_repr = sp_repr_next(child_repr); } @@ -623,14 +613,14 @@ sp_module_output_finalize (GObject *obje { SPModuleOutput *omod; omod = (SPModuleOutput *) object; - IF_NOT_NULL_FREE(omod->mimetype); - IF_NOT_NULL_FREE(omod->extension); - IF_NOT_NULL_FREE(omod->filetypename); - IF_NOT_NULL_FREE(omod->filetypetooltip); + g_free (omod->mimetype); + g_free (omod->extension); + g_free (omod->filetypename); + g_free (omod->filetypetooltip); G_OBJECT_CLASS (output_parent_class)->finalize (object); } /** @@ -667,23 +657,23 @@ sp_module_output_build (SPModule *module while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "output")) { child_repr = sp_repr_children(child_repr); while (child_repr != NULL) { if (!strcmp(sp_repr_name(child_repr), "extension")) { - IF_NOT_NULL_FREE(omod->extension); + g_free (omod->extension); omod->extension = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "mimetype")) { - IF_NOT_NULL_FREE(omod->mimetype); + g_free (omod->mimetype); omod->mimetype = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypename")) { - IF_NOT_NULL_FREE(omod->filetypename); + g_free (omod->filetypename); omod->filetypename = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } if (!strcmp(sp_repr_name(child_repr), "filetypetooltip")) { - IF_NOT_NULL_FREE(omod->filetypetooltip); + g_free (omod->filetypetooltip); omod->filetypetooltip = g_strdup(sp_repr_content(sp_repr_children(child_repr))); } child_repr = sp_repr_next(child_repr); } Index: src/sp-anchor.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-anchor.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-anchor.c --- src/sp-anchor.c 2 Nov 2003 00:46:23 -0000 1.2 +++ src/sp-anchor.c 30 Nov 2003 03:34:38 -0000 @@ -130,11 +130,11 @@ sp_anchor_set (SPObject *object, unsigne anchor = SP_ANCHOR (object); switch (key) { case SP_ATTR_XLINK_HREF: - if (anchor->href) g_free (anchor->href); + g_free (anchor->href); anchor->href = g_strdup (value); sp_object_request_modified (object, SP_OBJECT_MODIFIED_FLAG); break; case SP_ATTR_XLINK_TYPE: case SP_ATTR_XLINK_ROLE: Index: src/sp-image.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-image.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-image.c --- src/sp-image.c 31 Oct 2003 07:05:50 -0000 1.2 +++ src/sp-image.c 30 Nov 2003 03:34:38 -0000 @@ -171,11 +171,11 @@ sp_image_set (SPObject *object, unsigned image = SP_IMAGE (object); switch (key) { case SP_ATTR_XLINK_HREF: - if (image->href) g_free (image->href); + g_free (image->href); image->href = (value) ? g_strdup (value) : NULL; sp_object_request_update (object, SP_OBJECT_MODIFIED_FLAG | SP_IMAGE_HREF_MODIFIED_FLAG); break; case SP_ATTR_X: if (sp_svg_length_read_lff (value, &unit, &image->x.value, &image->x.computed) && Index: src/sp-text.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-text.c,v retrieving revision 1.6 diff -d -p -U5 -r1.6 sp-text.c --- src/sp-text.c 16 Nov 2003 23:37:19 -0000 1.6 +++ src/sp-text.c 30 Nov 2003 03:34:39 -0000 @@ -132,12 +132,12 @@ sp_string_release (SPObject *object) { SPString *string; string = SP_STRING (object); - if (string->p) g_free (string->p); - if (string->text) g_free (string->text); + g_free (string->p); + g_free (string->text); if (((SPObjectClass *) string_parent_class)->release) ((SPObjectClass *) string_parent_class)->release (object); } @@ -149,13 +149,13 @@ sp_string_read_content (SPObject *object SPString *string; const gchar *t; string = SP_STRING (object); - if (string->p) g_free (string->p); + g_free (string->p); string->p = NULL; - if (string->text) g_free (string->text); + g_free (string->text); t = sp_repr_content (object->repr); string->text = (t) ? g_strdup (t) : NULL; /* Is this correct? I think so (Lauris) */ /* Virtual method will be invoked BEFORE signal, so we can update there */ @@ -295,11 +295,11 @@ sp_string_set_shape (SPString *string, S sp_chars_clear (chars); if (!string->text || !*string->text) return; len = g_utf8_strlen (string->text, -1); if (!len) return; - if (string->p) g_free (string->p); + g_free (string->p); string->p = g_new (NRPointF, len + 1); /* fixme: Adjusted value (Lauris) */ size = style->font_size.computed; face = nr_type_directory_lookup_fuzzy (style->text->font_family.value, sp_text_font_style_to_lookup (style)); Index: src/sp-use.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/sp-use.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 sp-use.c --- src/sp-use.c 2 Nov 2003 00:30:53 -0000 1.2 +++ src/sp-use.c 30 Nov 2003 03:34:39 -0000 @@ -147,11 +147,11 @@ sp_use_release (SPObject *object) if (use->child) { use->child = sp_object_detach_unref (SP_OBJECT (object), use->child); } - if (use->href) g_free (use->href); + g_free (use->href); if (((SPObjectClass *) parent_class)->release) ((SPObjectClass *) parent_class)->release (object); } Index: src/style.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/style.c,v retrieving revision 1.4 diff -d -p -U5 -r1.4 style.c --- src/style.c 16 Nov 2003 23:37:19 -0000 1.4 +++ src/style.c 30 Nov 2003 03:34:39 -0000 @@ -259,13 +259,11 @@ sp_style_unref (SPStyle *style) /* if (style->object) gtk_signal_disconnect_by_data (GTK_OBJECT (style->object), style); */ if (style->object) g_signal_handlers_disconnect_matched (G_OBJECT(style->object), G_SIGNAL_MATCH_DATA, 0, 0, NULL, NULL, style); if (style->text) sp_text_style_unref (style->text); sp_style_paint_clear (style, &style->fill, TRUE, FALSE); sp_style_paint_clear (style, &style->stroke, TRUE, FALSE); - if (style->stroke_dash.dash) { - g_free (style->stroke_dash.dash); - } + g_free (style->stroke_dash.dash); g_free (style); } return NULL; } @@ -472,11 +470,11 @@ sp_style_merge_property (SPStyle *style, SPS_READ_IENUM_IF_UNSET (&style->font_stretch, val, enum_font_stretch, TRUE); break; case SP_PROP_FONT: if (!style->text_private) sp_style_privatize_text (style); if (!style->text->font.set) { - if (style->text->font.value) g_free (style->text->font.value); + g_free (style->text->font.value); style->text->font.value = g_strdup (val); style->text->font.set = TRUE; style->text->font.inherit = (val && !strcmp (val, "inherit")); } break; @@ -812,11 +810,11 @@ sp_style_merge_from_parent (SPStyle *sty style->writing_mode.computed = parent->writing_mode.computed; } if (style->text && parent->text) { if (!style->text->font_family.set || style->text->font_family.inherit) { - if (style->text->font_family.value) g_free (style->text->font_family.value); + g_free (style->text->font_family.value); style->text->font_family.value = g_strdup (parent->text->font_family.value); } } } @@ -1222,12 +1220,12 @@ static SPTextStyle * sp_text_style_unref (SPTextStyle *st) { st->refcount -= 1; if (st->refcount < 1) { - if (st->font.value) g_free (st->font.value); - if (st->font_family.value) g_free (st->font_family.value); + g_free (st->font.value); + g_free (st->font_family.value); g_free (st); } return NULL; } @@ -1316,11 +1314,11 @@ sp_style_read_ienum (SPIEnum *val, const } static void sp_style_read_istring (SPIString *val, const gchar *str) { - if (val->value) g_free (val->value); + g_free (val->value); if (!strcmp (str, "inherit")) { val->set = TRUE; val->inherit = TRUE; val->value = NULL; Index: src/bonobo/embeddable-document.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/bonobo/embeddable-document.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 embeddable-document.c --- src/bonobo/embeddable-document.c 30 Oct 2003 05:52:06 -0000 1.2 +++ src/bonobo/embeddable-document.c 30 Nov 2003 03:34:39 -0000 @@ -132,11 +132,11 @@ sp_bonobo_stream_read (Bonobo_Stream str do { Bonobo_Stream_read (stream, STREAM_CHUNK_SIZE, &iobuf, &ev); if (ev._major != CORBA_NO_EXCEPTION) { - if (* buffer != NULL) g_free (* buffer); + g_free (* buffer); len = -1; break; } * buffer = g_realloc (* buffer, len + iobuf->_length); @@ -204,11 +204,11 @@ sp_embeddable_document_ps_save (BonoboPe document = SP_EMBEDDABLE_DOCUMENT (closure); /* FIXME: super dirty but functional */ do { - if (filename) g_free (filename); + g_free (filename); filename = g_strdup ("inkscapeXXXXXX"); fd = mkstemp (filename); } while (fd == -1); Index: src/dialogs/stroke-style.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/dialogs/stroke-style.c,v retrieving revision 1.20 diff -d -p -U5 -r1.20 stroke-style.c --- src/dialogs/stroke-style.c 29 Nov 2003 22:03:24 -0000 1.20 +++ src/dialogs/stroke-style.c 30 Nov 2003 03:34:39 -0000 @@ -1101,11 +1101,11 @@ sp_stroke_style_scale_line (SPWidget *sp g_snprintf (c, 32, "%g", dist); sp_repr_css_set_property (css, "stroke-width", c); /* Set dash */ sp_stroke_style_set_scaled_dash (css, ndash, dash, offset, dist); sp_repr_css_change_recursive (SP_OBJECT_REPR (i->data), css, "style"); - if (dash) g_free (dash); + g_free (dash); } } else { for (r = reprs; r != NULL; r = r->next) { double length; double *dash, offset; @@ -1115,11 +1115,11 @@ sp_stroke_style_scale_line (SPWidget *sp sp_convert_distance (&length, sp_unit_selector_get_unit (us), SP_PS_UNIT); g_snprintf (c, 32, "%g", length * 1.25); sp_repr_css_set_property (css, "stroke-width", c); sp_stroke_style_set_scaled_dash (css, ndash, dash, offset, length); sp_repr_css_change_recursive ((SPRepr *) r->data, css, "style"); - if (dash) g_free (dash); + g_free (dash); } } sp_repr_css_attr_unref (css); if (spw->inkscape) sp_document_done (SP_WIDGET_DOCUMENT (spw)); Index: src/dialogs/text-edit.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/dialogs/text-edit.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 text-edit.c --- src/dialogs/text-edit.c 27 Nov 2003 00:55:38 -0000 1.5 +++ src/dialogs/text-edit.c 30 Nov 2003 03:34:39 -0000 @@ -637,11 +637,11 @@ sp_text_edit_dialog_text_changed (GtkTex if (str && *str) { sp_font_preview_set_phrase (SP_FONT_PREVIEW (preview), str); } else { sp_font_preview_set_phrase (SP_FONT_PREVIEW (preview), NULL); } - if (str) g_free (str); + g_free (str); if (text) { gtk_widget_set_sensitive (apply, TRUE); } gtk_widget_set_sensitive (def, TRUE); Index: src/helper/sodipodi-ctrl.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/helper/sodipodi-ctrl.c,v retrieving revision 1.3 diff -d -p -U5 -r1.3 sodipodi-ctrl.c --- src/helper/sodipodi-ctrl.c 30 Oct 2003 11:33:16 -0000 1.3 +++ src/helper/sodipodi-ctrl.c 30 Nov 2003 03:34:39 -0000 @@ -298,11 +298,11 @@ sp_ctrl_build_cache (SPCtrl *ctrl) } else { sr = fr; sg = fg; sb = fb; sa = fa; } side = (ctrl->span * 2 +1); c = ctrl->span ; - if (ctrl->cache) g_free (ctrl->cache); + g_free (ctrl->cache); size = (side) * (side) * 4; ctrl->cache = (guchar*)g_malloc (size); if (side < 2) return; switch (ctrl->shape) { Index: src/modules/ps.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/modules/ps.c,v retrieving revision 1.8 diff -d -p -U5 -r1.8 ps.c --- src/modules/ps.c 16 Nov 2003 23:37:19 -0000 1.8 +++ src/modules/ps.c 30 Nov 2003 03:34:39 -0000 @@ -854,12 +854,12 @@ sp_ps_print_image (FILE *ofp, guchar *px #if 0 fprintf (ofp, "showpage\n"); g_free (data); #endif - if (packb != NULL) g_free (packb); - if (plane != NULL) g_free (plane); + g_free (packb); + g_free (plane); #if 0 if (ferror (ofp)) { g_message (_("write error occured")); Index: src/widgets/font-selector.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/widgets/font-selector.c,v retrieving revision 1.3 diff -d -p -U5 -r1.3 font-selector.c --- src/widgets/font-selector.c 16 Nov 2003 23:37:19 -0000 1.3 +++ src/widgets/font-selector.c 30 Nov 2003 03:34:39 -0000 @@ -617,11 +617,11 @@ sp_font_preview_set_rgba32 (SPFontPrevie } void sp_font_preview_set_phrase (SPFontPreview *fprev, const gchar *phrase) { - if (fprev->phrase) g_free (fprev->phrase); + g_free (fprev->phrase); if (phrase) { fprev->phrase = g_strdup (phrase); } else { fprev->phrase = NULL; } Index: src/widgets/gradient-image.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/widgets/gradient-image.c,v retrieving revision 1.2 diff -d -p -U5 -r1.2 gradient-image.c --- src/widgets/gradient-image.c 30 Oct 2003 20:48:05 -0000 1.2 +++ src/widgets/gradient-image.c 30 Nov 2003 03:34:39 -0000 @@ -149,11 +149,11 @@ sp_gradient_image_size_allocate (GtkWidg image = SP_GRADIENT_IMAGE (widget); widget->allocation = *allocation; if (GTK_WIDGET_REALIZED (widget)) { - if (image->px) g_free (image->px); + g_free (image->px); image->px = g_new (guchar, 3 * VBLOCK * allocation->width); } sp_gradient_image_update (image); } Index: src/xml/repr-action.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/xml/repr-action.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 repr-action.c --- src/xml/repr-action.c 29 Nov 2003 02:55:37 -0000 1.5 +++ src/xml/repr-action.c 30 Nov 2003 03:34:39 -0000 @@ -228,12 +228,11 @@ coalesce_chgattr (SPReprAction *action) { /* ensure changes are continuous */ if (strcmp(action->act.chgattr.oldval, iter->act.chgattr.newval)) break; - if (action->act.chgattr.oldval) - g_free (action->act.chgattr.oldval); + g_free (action->act.chgattr.oldval); action->act.chgattr.oldval = iter->act.chgattr.oldval; iter->act.chgattr.oldval = NULL; free_action (iter); prev->next = next; @@ -262,12 +261,11 @@ coalesce_chgcontent (SPReprAction *actio if ( action->repr == iter->repr ) { /* ensure changes are continuous */ if (strcmp(action->act.chgcontent.oldval, iter->act.chgcontent.newval)) break; - if (action->act.chgcontent.oldval) - g_free (action->act.chgcontent.oldval); + g_free (action->act.chgcontent.oldval); action->act.chgcontent.oldval = iter->act.chgcontent.oldval; iter->act.chgcontent.oldval = NULL; @@ -429,20 +427,16 @@ free_action (SPReprAction *action) sp_repr_unref (action->act.del.child); if (action->act.del.ref) sp_repr_unref (action->act.del.ref); break; case SP_REPR_ACTION_CHGATTR: - if (action->act.chgattr.oldval) - g_free (action->act.chgattr.oldval); - if (action->act.chgattr.newval) - g_free (action->act.chgattr.newval); + g_free (action->act.chgattr.oldval); + g_free (action->act.chgattr.newval); break; case SP_REPR_ACTION_CHGCONTENT: - if (action->act.chgcontent.oldval) - g_free (action->act.chgcontent.oldval); - if (action->act.chgcontent.newval) - g_free (action->act.chgcontent.newval); + g_free (action->act.chgcontent.oldval); + g_free (action->act.chgcontent.newval); break; case SP_REPR_ACTION_CHGORDER: sp_repr_unref (action->act.chgorder.child); if (action->act.chgorder.oldref) sp_repr_unref (action->act.chgorder.oldref); Index: src/xml/repr.c =================================================================== RCS file: /cvsroot/inkscape/inkscape/src/xml/repr.c,v retrieving revision 1.5 diff -d -p -U5 -r1.5 repr.c --- src/xml/repr.c 16 Nov 2003 23:37:20 -0000 1.5 +++ src/xml/repr.c 30 Nov 2003 03:34:39 -0000 @@ -173,11 +173,11 @@ repr_finalize (SPRepr *repr) for (rl = repr->listeners; rl; rl = rl->next) { if (rl->vector->destroy) (* rl->vector->destroy) (repr, rl->data); } while (repr->children) sp_repr_remove_child (repr, repr->children); while (repr->attributes) sp_repr_remove_attribute (repr, repr->attributes); - if (repr->content) g_free (repr->content); + g_free (repr->content); while (repr->listeners) sp_repr_remove_listener (repr, repr->listeners); } static void repr_doc_finalize (SPRepr *repr) @@ -859,14 +859,14 @@ sp_repr_merge (SPRepr *repr, const SPRep g_return_val_if_fail (repr != NULL, FALSE); g_return_val_if_fail (src != NULL, FALSE); g_return_val_if_fail (key != NULL, FALSE); if (src->content) { - if (repr->content) g_free (repr->content); + g_free (repr->content); repr->content = g_strdup (src->content); } else { - if (repr->content) g_free (repr->content); + g_free (repr->content); repr->content = NULL; } for (child = src->children; child != NULL; child = child->next) { SPRepr *rch;

3 2

New Inkscape man page
by Bryce Harrington 30 Nov '03

30 Nov '03

Here's my first draft of a new man page for Inkscape. The old one as mostly just a list of options, most of which were invalid. I created this as POD 'inkscape.pod' and generated the man page 'inkscape.1.in' by running pod2man and changing "...Perl Documentation..." to "GNOME Documentation". This is set up to be installed when you run 'make install', so after installation you can view it via 'man inkscape'. Bryce INKSCAPE(1) GNOME Documentation INKSCAPE(1) NAME Inkscape - a 2D vector graphics editing program. SYNOPSIS "inkscape [options] [filename ...]" options: -?, --help -a, --export-area=x0:y0:x1:y1 -b, --export-background=COLOR -d, --export-dpi=DPI -e, --export-png=FILENAME -f, --file=FILENAME -h, --export-height=HEIGHT -i, --bitmap-icons -p, --print=FILENAME -s, --slideshow -w, --export-width=WIDTH -x, --with-gui -z, --without-gui --export-svg=FILENAME --usage DESCRIPTION Inkscape is a GUI editor for Scalable Vector Graphics (SVG) format drawing files, with capabilities similar to Illustrator, CorelDraw, Visio, etc. Supported SVG features include basic shapes, paths, text, alpha blending, transforms, gradients, node editing, svg-to-png export, grouping, and more. . The interface is designed to be comfortable and efficient for skilled users, while remaining conformant to GNOME stan- dards so that users familiar with other GNOME applications can learn it's interface rapidly. SVG is a W3C standard XML format for 2D vector drawing, that allows defining objects in the drawing using points, paths, and primitive shapes. Colors, fonts, line thicknesses, and so forth are specified as 'style' attributes to these objects. The intent is that since SVG is a standard, and since its files are text/xml instead of binary, it is be possible to use SVG files in a sizeable number of programs and for a wide range of uses. Inkscape uses SVG as its native document format, and has the goal of becoming the most fully compliant drawing program for SVG files avail- able in the Open Source community. OPTIONS -?, --help Show help message -a x0:y0:x1:y1, --export-area=x0:y0:x1:y1 Exported area in millimeters (default is full document, 0,0 is lower-left corner) -b COLOR, --export-background=COLOR Background color of exported bitmap (any SVG supported color string) -d DPI, --export-dpi=DPI The resolution used for converting SVG into bitmap (default 72.0) -e FILENAME, --export-png=FILENAME Export document to png file -f FILENAME, --file=FILENAME Open specified document(s) (option string may be excluded) -h HEIGHT, --export-height=HEIGHT The height of generated bitmap in pixels (overwrites dpi) -i, --bitmap-icons Prefer bitmap (xpm) icons to SVG ones -p FILENAME, --print=FILENAME Print document(s) to specified output file (use '| program' for pipe) -s, --slideshow Show given files one-by-one, switch to next on any key/mouse event -w WIDTH, --export-width=WIDTH The width of generated bitmap in pixels (overwrites dpi) -x, --with-gui Try to use X server even if $DISPLAY is not set) -z, --without-gui Do not open GUI and do not use X server (only process files from console). Useful for printing SVG documents. --export-svg=FILENAME Export document to plain SVG file (no "xmlns:sodipodi" names- pace) --usage Display brief usage message KEYBINDINGS See keybindings.txt for a listing of keybindings. DIAGNOSTICS The program returns zero on success or non-zero on failure. A variety of error messages and warnings are printed to STDERR or STD- OUT. If the program behaves erratically with a particular SVG file or crashes, it is sometimes useful to look at this output for clues. For example, it may indicate if a given SVG file contains elements or attributes that are not yet supported by the codebase. EXAMPLES While obviously Inkscape is primarily intended as a GUI application, it can be used for doing SVG processing on the commandline as well. (Note though that other SVG tools and libraries such as librsvg may be better suited for production SVG processing.) Printing an SVG file directly: inkscape filename.svg -p | lpr Converting SVG file to a png: inkscape --export-png=filename.png --export-height=400 --export-width=600 Convert an SVG document using items in the sodipodi namespace to plain SVG: inkscape filename1.svg e --export-svg=filename2.svg ENVIRONMENT DISPLAY to get the default host and display number. FILES $HOME/.inkscape/preferences.xml - The user's preference settings. $HOME/.inkscape/extensions.xml - The filter programs to be used in the application. OTHER INFO The canonical place to find Inkscape info is at http://www.inkscape.org/. The website includes links to other relevant documentation, tutorials, user manual, examples, mailing list archives, the latest released version of the program, and more. SEE ALSO gimp(1) NOTES Ctrl & Shift modify dragging behaviour in some modes Shift click/rubberband usually toggles selection Snapping to guidelines/grid work only for certain modes To break line (or add node) select its endpoints It should import png, bmp, jpg, xpm & gif bitmaps and svg vector images. Open can handle only svg graphics BUGS When importing svg generated with ill2svg, all paths are usually grouped. Ungroup these manually. Many bugs are known; please refer to the website for reviewing the reported ones and to report newly found issues. AUTHORS This codebase owes its existance to a large number of contributors throughout its various incarnations. The following list is certainly incomplete, but serves to recognize the many shoulders on which this application sits: John Bintz, Hans Breuer, Bulia Byak, Chema Celorio, Zbigniew Chyla, Kees Cook, Robert Crosbie, Jon Cruz, Danilo Egan, Frank Felfe, Ted Gould, Bryce Harrington, Nathan Hurst, Thomas Ingham, Bob Jamison, Lau- ris Kaplinski, Lynn Kerby, Petr Kovar, Raph Levien, Vitaly Lipatov, Dmitry G. Mastrukov, Michael Meeks, Frederico Mena, MenTaLguY, Yukihiro Nakai, Christian Neumair, Mitsuru Oka, Christian Schaller, Tom von Schwerdtner, Daniel Yacob, Masatake Yamato, This man page was put together by Bryce Harrington <brycehar@...88... rington.com> HISTORY The codebase that would become Inkscape began life in 1999 as the pro- gram 'GILL', the GNOME Illustrator application, created by Raph Levien. The stated objective for Gill was to eventually support all of SVG. Raph implemented the PostScript bezier imaging model, including stroking and filling, line cap style, line join style, text, etc. A Bonobo Raph's Gill page is at http://www.levien.com/svg/. Work on Gill appears to have slowed or ceased in 2000, perhaps in part due to frus- trations dealing with implementing DOM and SVG. The next incarnation of the codebase was to become the highly popular program Sodipodi, led by Lauris Kaplinkski. The codebase was turned into a very usable illustration program over the course of several year's work, adding several new features, multi-lingual support, port- ing to Windows and other operating systems, and eliminating dependen- cies. Inkscape was formed in 2003 by a number of active Sodipodi developers wanting to take a different direction with the codebase in terms of focus on SVG-compliance, interface look-and-feel, and a desire to open development opportunities to more participants. In a way, this direc- tion marks a return to the codebase's Gill roots. COPYRIGHT AND LICENSE Copyright (C) 1999, 2000, 2001, 2002, 2003 by Authors. Inkscape is free software; you can redistribute it and/or modify it under the terms of the GPL. Inkscape 0.36cvs 2003-11-29 INKSCAPE(1)

2 2

Re: GIMP user interface discussion
by bulia byak 30 Nov '03

30 Nov '03

> > Of course it would be better! So, are you doing it already? When can we expect to see the unsinkability option for utility windows in, say, KDE? > > Well, you're using KDE not me ;) Like I said, Gimp has a much higher profile than us currently. If you push the KDE people they're more likely to listen. We have plans to push for that too, but our main priority is to roll out a program that WORKS NOW. This includes unsinkable dialogs, too. > > > The A4 rectangle in the canvas window is horribly disturbing. > > > > It helps me, but you can switch it off easily. In a next version this preference will be saved for future sessions. > > How about choosing "new" and set A4 to open up an image window, the > edges of which would determine the size of the image, that is A4? The > rulers would then copy the image size. Have you ever worked with any commercial vector editor? They ALL have paper size shown by edges on the canvas, and they all allow to draw both within the paper and outside. That's standard. > > No, WiW is secondary. The biggest problem with Gimp is that dialogs sink. They never do in any Windows or Mac program I've EVER seen. > > :) > > No it's not. The problem with sinking dialogs comes from the > WiW/separate issue. See below. What I am trying to do is to show you that sinking/unsinking dialogs are orthogonal to the WiW. And I'm using Inkscape to demonstrate that. Inkscape has no WiW but its dialogs are unsinkable. > > Besides, what's the purpose of having the main toolbox on the left > while having the other one on the top? It makes no sense. > > > > It makes perfect sence as they are the primary and secondary toolbars, directly linked (change tool on the primary and get all relevant controls on the secondary). > > This is the approach taken by (at least) Photoshop and Xara, and it's very convenient. > > Please compare the following images. > > http://www.suse.de/en/private/products/suse_linux/winerack/images/photoshop… > http://www.pleasewipeyourfeet.com/webtest/pr/gimp.jpg > > Do you really think that having a subtoolbox above the primary toolbox > makes any sense? Have you ever seen trees organized bottom-up? It may be surprising for you, but it's very convenient. Photoshop users were raging when this secondary toolbox appeared (in version 6 I think). Xara is considered one of the most convenient and usable vector app by most people who have ever used it, and it had such a secondary toolbox from the start. > See how many options gimp provides for the brushtool. > How would you fit them in the top toolbar? Plenty of options is not necessarily an advantage. They may be distracting. As for your example: - First, some of the brush options are very unlikely to be needed every time you use brush, and therefore should be removed into an options dialog tab for this tool. - Second, I do indeed think that even all of the Gimp brush tools on your screenshot can be put into a horizontal toolbar without problems. - Third, this actually does not matter - your Tool options is worse NOT because it's not horizontal but because it sinks and grabs focus and clutters taskbar. If it didn't (i.e. if it was transient as are dialogs in Inkscape, or dockable into a document window), the difference between a tool options dialog and our secondary toolbar would not amount to much because both switch their content when you change tools (which is the main point of having such a dialog). > Regarding dialog sinking which you've never seen on any windows/mac app, > to be honest i haven't used PS for a *long* time, just guessing from the > screenshot - the toolbox and docks, how do they behave if you maximise > the image window? Are they always on top? OF COURSE!!! Moreover, MOST people work in Photoshop with their images maximized. This would be impossible if palettes didn't stay on top. And it's pretty much impossible in Gimp for the same reason. > Can you hide them so that all > space on screen belongs to the image window? Yes, easily, just press Tab to toggle all palettes on or off. In Inkscape, we plan to use the ` key for the same because Tab is taken. > It's all the same thing except - if you minimise PS you minimise all, All including dialogs. Believe me, it's a huge difference in usability. > but you end up having a mess on the bottom of the main window when > minimising separate images. > Subtle differences between WiW and non-WiW. I repeat, I don't care about WiW. In fact I like documents in separate windows better, and Inkscape is going to stay that way. What we're speaking about is the behavior of dialogs, not document windows. > BTW Seems like Xara isn't available anymore, just checked the corel > site, or is there any other company providing the sw? See a screenshot: http://www.xara.com/press/graphics/xarax/screenshots/firestorm.jpg This is default interface, pretty cluttered, but in daily work you can easily remove all the toolbars under the menu except the bottom one. That's the secondary toolbar whose content switches as you switch tools. As you can see, quite a lot of options there (shown for the selector tool), all very compact and convenient. > > > As for the transients - as long as 1.) the toolbox and the docks all behave like one group(for the purpose of minimising, > > > bringing forward, always on top etc), > > > and making the image window active switches the images/layers/channels/paths/ dock to that image, it will do a much better job > > > than WiW has ever done. the images should always be kept as separate objects/windows. > > > > Exactly! This is _precisely_ what we're doing in Inkscape and what Gimp refuses to do. > > I don't think so. :) What are you denying? That your description precisely fits the current Inkscape interface? DOwnload the CVS and see for yourself. That Gimp flatly refuses to use transients in the same way? Ask Sven and you'll hear for yourself :) > > I ran the new Gimp and saw severe usability problems at once. You say they can be fixed by changing preferences? Great! But first, not everything can be fixed, and the biggest problem - sinking dialogs - is unfixable. > > Bulia, the main problem is that you're approaching it with destructive > criticism: > > "I ran gimp for the *first* time and *immidiatelly saw* problems A, B, C > which are unsolvable" Hey, I did not call them unsolvable until Sven confirmed that :) > Awwww, c'mon! > > the problem with sinking dialogs or transients or whatever doesn't lie > in the gimp domain, it lies in the freedesktop domain, the freedesktop > proposal should enforce certain rules on Wm for apps like gimp, which > are not WiW. If that it isn't implemented yet, is that a real problem? Of course it is. Users need usable programs NOW. Until there are other widely supported ways to achieve the same goal, we'll use transients. -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm

1 0