Mercurial > hg > xemacs-beta

comparison src/toolbar.c @ 5090:0ca81354c4c7

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Further frame-geometry cleanups -------------------- ChangeLog entries follow: -------------------- man/ChangeLog addition: 2010-03-03 Ben Wing <ben@xemacs.org> * internals/internals.texi (Intro to Window and Frame Geometry): * internals/internals.texi (The Paned Area): * internals/internals.texi (The Displayable Area): Update to make note of e.g. the fact that the bottom gutter is actually above the minibuffer. src/ChangeLog addition: 2010-03-03 Ben Wing <ben@xemacs.org> * emacs.c: * emacs.c (assert_equal_failed): * lisp.h: * lisp.h (assert_equal): New fun assert_equal, asserting that two values == each other, and printing out both values upon failure. * frame-gtk.c (gtk_initialize_frame_size): * frame-impl.h: * frame-impl.h (FRAME_TOP_INTERNAL_BORDER_START): * frame-impl.h (FRAME_BOTTOM_INTERNAL_BORDER_START): * frame-impl.h (FRAME_LEFT_INTERNAL_BORDER_START): * frame-impl.h (FRAME_PANED_TOP_EDGE): * frame-impl.h (FRAME_NONPANED_SIZE): * frame-x.c (x_initialize_frame_size): * frame.c: * gutter.c (get_gutter_coords): * gutter.c (calculate_gutter_size): * gutter.h: * gutter.h (WINDOW_REAL_TOP_GUTTER_BOUNDS): * gutter.h (FRAME_TOP_GUTTER_BOUNDS): * input-method-xlib.c: * input-method-xlib.c (XIM_SetGeometry): * redisplay-output.c (clear_left_border): * redisplay-output.c (clear_right_border): * redisplay-output.c (redisplay_output_pixmap): * redisplay-output.c (redisplay_clear_region): * redisplay-output.c (redisplay_clear_top_of_window): * redisplay-output.c (redisplay_clear_to_window_end): * redisplay-xlike-inc.c (XLIKE_clear_frame): * redisplay.c: * redisplay.c (UPDATE_CACHE_RETURN): * redisplay.c (pixel_to_glyph_translation): * toolbar.c (update_frame_toolbars_geometry): * window.c (Fwindow_pixel_edges): Get rid of some redundant macros. Consistently use the FRAME_TOP_*_START, FRAME_RIGHT_*_END, etc. format. Rename FRAME_*_BORDER_* to FRAME_*_INTERNAL_BORDER_*. Comment out FRAME_BOTTOM_* for gutters and the paned area due to the uncertainty over where the paned area actually begins. (Eventually we should probably move the gutters outside the minibuffer so that the paned area is contiguous.) Use FRAME_PANED_* more often in the code to make things clearer. Update the diagram to show that the bottom gutter is inside the minibuffer (!) and that there are "junk boxes" when you have left and/or right gutters (dead boxes that are mistakenly left uncleared, unlike the corresponding scrollbar dead boxes). Update the text appropriately to cover the bottom gutter position, etc. Rewrite gutter-geometry code to use the FRAME_*_GUTTER_* in place of equivalent expressions referencing other frame elements, to make the code more portable in case we move around the gutter location. Cleanup FRAME_*_GUTTER_BOUNDS() in gutter.h. Add some #### GEOM! comments where I think code is incorrect -- typically, it wasn't fixed up properly when the gutter was added. Some cosmetic changes.
author Ben Wing <ben@xemacs.org>
date Wed, 03 Mar 2010 05:07:47 -0600
parents d372b17f63ce
children 7be849cb8828
comparison
equal deleted inserted replaced
5080:5502045ec510 5090:0ca81354c4c7
775 /* Removed the check for the minibuffer here. We handle this 775 /* Removed the check for the minibuffer here. We handle this
776 more correctly now by consistently using 776 more correctly now by consistently using
777 FRAME_LAST_NONMINIBUF_WINDOW instead of FRAME_SELECTED_WINDOW 777 FRAME_LAST_NONMINIBUF_WINDOW instead of FRAME_SELECTED_WINDOW
778 throughout the toolbar code. */ 778 throughout the toolbar code. */
779 compute_frame_toolbars_data (f); 779 compute_frame_toolbars_data (f);
780
781 /* #### GEOM! Turning the toolbar on and off repeatedly causes the
782 frame to steadily shrink. Basically, turning it on doesn't
783 increase the frame size, while turning it off does reduce the
784 frame size. The cause has something to do with the combination
785 of this maybe questionable code here, plus the fact that toolbars
786 are included in the displayable area, and the difference between
787 real and theoretical toolbar sizes, and exactly when the various
788 computations happen w.r.t. the specifiers or whatever that control
789 whether toolbars are visible and hence whether their thickness is
790 greater than zero. --ben */
780 791
781 if (frame_size_changed) 792 if (frame_size_changed)
782 { 793 {
783 int width, height; 794 int width, height;
784 pixel_to_frame_unit_size (f, FRAME_PIXWIDTH (f), FRAME_PIXHEIGHT (f), 795 pixel_to_frame_unit_size (f, FRAME_PIXWIDTH (f), FRAME_PIXHEIGHT (f),