- update to 392:
  * improve input decoding for non-Latin1 character sets by preserving the sense of GL/GR.
  * add resource preferLatin1 to simplify UPSS configuration (Gentoo #932154).
  * build-fix for --disable-boxchars; patch #390 reuses that feature's code to draw the part of the DEC Technical character set which has no Unicode equivalent.
  * improve definitions used in clock_gettime logic in graphics_sixel.c, as well as updating comments (patch by Ben Wong).
  * amend allowC1Printable changes from patch #391, restoring a special case which caused C1 characters to be ignored (report/testcase by Dmytro Bagrii).
- update to 391:
  * improve integration between UTF-8 and allowC1Printable.
  * alter SIXEL HLS computation to make blue 0 degrees (patch by Ben Wong).
  * disable SIXEL aspect-ratio, pending a rewrite (patch by Ben Wong).
  * add resource incrementalGraphics (prompted by patch by Ben Wong).
  * reorganize graphics_sixel.c as a step toward eliminating graphics buffer-size (integrated patch by Ben Wong).
  * fix regression in patch #390: non-UTF8 text pasted as a XA_UTF8_STRING was not decoded (report by Petri Kaukasoina).
- update to 390:
  * improve typography of control-sequences page (patches by Branden Robinson).
  * amend UPSS change from patch #389, fixing a regression in VT100/VT220 character sets.
  * modify ANSI conformance per ECMA-43 and DEC 070:
    * set ANSI conformance level to 3 for DEC levels 2 and up.
    * disable locking shifts in level 1, e.g., VT100.
    * disable single-shifts from G2/G3 in level 1.
    * use UPSS for G2/G3 in levels 2 and up.
  * modify invisible-character attribute to permit DECRQCRA to report a consistent checksum (report/testcase by Thomas Wolff).
  * align terminfo building blocks in terminfo to correspond to recent refactoring/trimming in ncurses.
  * set flag in regexec call for onNClicks to handle “^” anchor (patch by Matthew Martin).
  * add line-drawing data for the characters in DEC Technical which have no Unicode equivalents.
  * remove a special case in encode_scs which made xterm report Latin-1 when encoding is set to UTF-8, where ASCII is the selected character set. This fixes a regression in vttest for the DECRSPS cursor test.
  * correct values for Ps of DECAUPSS in ctlseqs.ms (report by James Holderness).
  * simplify/correct expression for checking Unicode non-characters (report by Thomas Wolff).
  * correct loop for trimming environment variables (report/patch by Casper Dik).

• Files Changed
• Browse Source