- <li>correct <code>manlinks.sed</code> script introduced in ncurses 5.1
- to avoid using ERE "\+", which is not understood by standard versions
- of <code>sed</code>. This happens to work with GNU <code>sed</code>,
- but is not portable, and was the initial motivation for this release.
-
- <li>remove "hpux10.*" case from CF_SHARED_OPTS configure script macro.
- This differed from the "hpux*" case by using reversed symbolic
- links, which made the 5.1 version not match the configuration of
- 5.0 shared libraries.
-
- <li>guard against corrupt terminfo data:
- <ul>
- <li>modify <code>tparm()</code> to disallow arithmetic on strings,
- analyze the varargs list to read strings as strings and numbers as
- numbers.
-
- <li>modify <code>tparm()</code>'s internal function
- <code>spop()</code> to treat a null pointer as an empty string.
-
- <li>modify <code>parse_format()</code> in lib_tparm.c to ignore
- precision if it is longer than 10000.
-
- <li>rewrote limit checks in lib_mvcur.c using new functions
- <code>_nc_safe_strcat()</code>, etc. Made other related changes to
- check lengths used for <code>strcat()</code> and
- <code>strcpy()</code>.
- </ul>
-
- <li>corrections to screen optimization:
- <ul>
- <li>added special case in lib_vidattr.c to reset underline and
- standout for devices that have no sgr0 defined.
-
- <li>change handling of <code>non_dest_scroll_region</code> in
- tty_update.c to clear text after it is shifted in rather than before
- shifting out. Also correct row computation.
-
- <li>modify <code>rs2</code> capability in xterm-r6 and similar entries
- where cursor save/restore bracketed the sequence for resetting video
- attributes. The cursor restore would undo that.
- </ul>
-
- <li>UTF-8 support:
- <ul>
- <li>when checking LC_ALL, LC_CTYPE, and LANG environment variables
- for UTF-8 locale, ignore those which are set to an empty value, as
- per SUSV2.
-
- <li>encode 0xFFFD in UTF-8 with 3 bytes, not 2.
-
- <li>modify <code>_nc_utf8_outch()</code> to avoid sign-extension when
- checking for out-of-range value.
- </ul>
-
- <li>other library fixes:
- <ul>
- <li>added checks for an empty <code>$HOME</code> environment
- variable, to avoid retrieving terminfo descriptions from
- <code>./.terminfo</code> .
-
- <li>change functions <code>_nc_parse_entry()</code> and
- <code>postprocess_termcap()</code> to avoid using
- <code>strtok()</code>, because it is non-reentrant.
-
- <li>initialize <code>fds[]</code> array to 0's in
- <code>_nc_timed_wait()</code>; apparently <code>poll()</code> only
- sets the <code>revents</code> members of that array when there is
- activity corresponding to the related file.
-
- <li>add a check for null pointer in <code>Make_Enum_Type()</code>.
-
- <li>fix a heap problem with the c++ binding.
-
- <li>correct missing includes for <string.h> in several places,
- including the C++ binding. This is not noted by gcc unless we use
- the <code>-fno-builtin</code> option.
- </ul>
-
- <li>several fixes for tic:
- <ul>
- <li>add a check for empty buffers returned by <code>fgets()</code> in
- comp_scan.c <code>next_char()</code> function, in case
- <code>tic</code> is run on a non-text file (fixes a core dump).
-
- <li>modify <code>tic</code> to verify that its inputs are really files,
- in case someone tries to read a directory (or
- <code>/dev/zero</code>).
-
- <li>correct an uninitialized parameter to <code>open_tempfile()</code>
- in tic.c which made "tic -I" give an ambiguous error message about
- <code>tmpnam</code>.
-
- <li>correct logic in <code>adjust_cancels()</code>, which did not check
- both alternatives when reclassifying an extended name between
- boolean, number and string, causing an infinite loop in
- <code>tic</code>.
- </ul>
-
- <li>using new checks in <code>tic</code> for parameter counts in
- capability strings, found/fixed several errors both in the
- terminfo database and in the include/Caps file.
- <ul>
- <li>modified several terminfo capability strings, including the
- definitions for setaf, setab, in include/Caps to indicate that the
- entries are parameterized. This information is used to tell which
- strings are translated when converting to termcap. This fixes a
- problem where the generated termcap would contain a spurious "%p1"
- for the terminfo "%p1%d".
-
- <li>correct parameter counts in include/Caps for dclk as well as some
- printer-specific capabilities: csnm, defc, scs, scsd, smgtp, smglp.
- </ul>
-
- <li>various fixes for install scripts used to support configure
- <code>--srcdir</code> and <code>--with-install-prefix</code>.
-
- <li>correct several mismatches between manpage filename and ".TH"
- directives, renaming dft_fgbg.3x to default_colors.3x and
- menu_attribs.3x to menu_attributes.3x.
+
+ <li>improve logic in tgetent() which adjusts the termcap "me" string
+ to work with ISO-2022 string used in xterm-new.
+ This is a feature that was incompletely implemented in ncurses 5.3.
+ ncurses attempts to provide termcap clients with the portion of
+ the sgr0 (termcap "me") string that does not reset line-drawing.
+
+ <li>cells in the WINDOW which are continuations of a multicolumn
+ character are encoded differently, making repainting more
+ reliable.
+
+ <li>amend change to setupterm() in ncurses 5.4 (20030405) which would
+ reuse the value of cur_term if the same output was selected. This
+ now reuses it only when setupterm() is called from tgetent(),
+ which has no notion of separate SCREENs. Note that tgetent() must
+ be called after initscr() or newterm() to use this feature.
+
+ <li>make setcchar() now works when its wchar_t* parameter is
+ pointing to a string which contains more data than can be converted.
+
+ <li>win_wchnstr() now works for more than one cell.
+
+ <li>resizeterm() now processes all levels of window hierarchy.
+
+ <li>disable GPM mouse support when $TERM happens to be prefixed with
+ "xterm". Gpm_Open() would otherwise assert that it can deal with
+ mouse events in this case.
+
+ <li>add SP->_screen_acs_map[], used to ensure that mapping of missing
+ line-drawing characters is handled properly. For example, ACS_DARROW
+ is absent from xterm-new, and it was coincidentally displayed the
+ same as ACS_BTEE.
+