view admin/notes/years @ 107998:531d454c3a99

Implement GUI display of R2L lines, fix TTY display of R2L lines. xdisp.c [HAVE_WINDOW_SYSTEM]: Add prototype for append_stretch_glyph. (set_cursor_from_row) <cursor_x>: Remove unused variable. Fix off-by-one error in computing x at end of text in the row. (append_stretch_glyph): In reversed row, prepend the glyph rather than append it. Set resolved_level and bidi_type of the glyph. (extend_face_to_end_of_line): If the row is reversed, prepend a stretch glyph whose width is such that the rightmost glyph will be drawn at the right margin of the window. Fix off-by-one error on TTY frames in testing whether a line needs face extension. Fix face extension at ZV. If this is the last glyph row, use DEFAULT_FACE_ID, to avoid painting the rest of the window with the region face. (set_cursor_from_row, display_line): Use MATRIX_ROW_CONTINUATION_LINE_P instead of testing value of row->continuation_lines_width. (next_element_from_buffer): Don't call bidi_paragraph_init if we are at ZV. Fixes a crash when reseated to ZV by try_window_reusing_current_matrix. (display_and_set_cursor, erase_phys_cursor): Handle negative HPOS, which happens with R2L glyph rows. Fixes a crash when inserting a character at end of an R2L line. (set_cursor_from_row): Don't be fooled by truncated rows: don't treat them as having zero-width characters. Improve comments. Don't reverse pos_before and pos_after for reversed glyph rows. Set cursor.x to negative value when the cursor might be on the left fringe. (IT_OVERFLOW_NEWLINE_INTO_FRINGE): For R2L lines, consider the left fringe, not the right one. (notice_overwritten_cursor, draw_phys_cursor_glyph) (erase_phys_cursor): For reversed cursor_row, support cursor on the left fringe. fringe.c (update_window_fringes): For R2L rows, swap the bitmaps of continuation indicators on the fringes. (draw_fringe_bitmap): For reversed glyph rows, allow cursor on the left fringe. w32term.c (w32_draw_window_cursor): For reversed glyph rows, draw cursor on the left fringe. xterm.c (x_draw_window_cursor): For reversed glyph rows, draw cursor on the left fringe. dispnew.c (update_text_area): Handle reversed desired rows when the cursor is on the left fringe. (set_window_cursor_after_update): Limit cursor's hpos by -1 from below, not by 0, for when the cursor is on the left fringe. xdisp.c (unproduce_glyphs): New function. (display_line): Use it when produced glyphs are discarded from R2L glyph rows. (append_composite_glyph): In R2L rows, prepend the glyph rather than appending it. term.c (append_composite_glyph): In R2L rows, prepend the glyph rather than append it. Set up the resolved_level and bidi_type attributes of the appended glyph. (produce_special_glyphs): Mirror the backslash continuation character in R2L lines.
author Eli Zaretskii <eliz@gnu.org>
date Tue, 20 Apr 2010 16:31:28 +0300
parents 0138ae02cb0d
children
line wrap: on
line source

HOW TO MAINTAIN COPYRIGHT YEARS FOR GNU EMACS

Maintaining copyright years is now very simple: every time a new year
rolls around, add that year to every FSF (and AIST) copyright notice.

There's no need to worry about whether an individual file has changed
in a given year - it's sufficient that Emacs as a whole has changed.

For more detailed information on maintaining copyright, see the file
"copyright" in this directory.

The previous policy was more complex, but is now only of historical
interest (see versions of this file from before 2009).

The refcards in etc/refcards can print only the latest copyright year,
but should keep the full list in a comment in the source.


"Our lawyer says it is ok if we add, to each file that has been in Emacs
 since Emacs 21 came out in 2001, all the subsequent years[1].  We don't
 need to check whether *that file* was changed in those years.
 It's sufficient that *Emacs* was changed in those years (and it was!).
  
 For those files that have been added since then, we should add
 the year it was added to Emacs, and all subsequent years."
  
 --RMS, 2005-07-13

[1] Note that this includes 2001 - see
<http://lists.gnu.org/archive/html/emacs-pretest-bug/2006-12/msg00119.html>