commit/XEmacs: kehoea: Reduce test errors under no-Mule, mostly from case-tests.el

1 new commit in XEmacs: https://bitbucket.org/xemacs/xemacs/commits/9764d52c3fd9/ Changeset: 9764d52c3fd9 User: kehoea Date: 2017-11-13 23:19:01+00:00 Summary: Reduce test errors under no-Mule, mostly from case-tests.el tests/ChangeLog addition: 2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/case-tests.el: Change the encoding of this to iso-8859-1, so non-mule doesn't choke on old-format case tables. * automated/case-tests.el (downcase1): Create strings using non-Latin-1 using (format "...%c..." (decode-char 'ucs CODE) at runtime, rather than simply confusing a non-mule build at read time. * automated/case-tests.el (uni-mappings): Use #'decode-char at runtime, don't assume non-Mule can read a long list of mostly non-ASCII characters. * automated/extent-tests.el (string): Check whether CURRENT-LANGUAGE-ENVIRONMENT is bound (only if Mule) before dereferencing it. * automated/format-tests.el (Assert): Be better about not expecting the Lisp reader to understand non-Latin-1, use decode-char at runtime. Mark a few tests with Implementation-Incomplete-Expect-Failure. src/ChangeLog addition: 2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; * bytecode.c (bytecode_arithcompare): Rename a couple of local variables to quieten the compiler about shadowing. * doprnt.c (rewrite_floating_spec): Mark OBJ as used if HAVE_BIGFLOAT is not defined, to quieten the compiler. lisp/ChangeLog addition: 2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode.el (decode-char): Return nil instead of erroring when CODE cannot be encoded, as did this function pre-unicode-internal and as does (did) GNU's version. Affected #: 9 files diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 lisp/ChangeLog --- a/lisp/ChangeLog +++ b/lisp/ChangeLog ＠＠ -1,3 +1,10 ＠＠ +2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; + + * unicode.el (decode-char): + Return nil instead of erroring when CODE cannot be encoded, as did + this function pre-unicode-internal and as does (did) GNU's + version. + 2017-11-09 Aidan Kehoe <kehoea(a)parhasard.net&gt; Ben's approach in lib-src/make-case-conv.el didn't work, there are diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 lisp/unicode.el --- a/lisp/unicode.el +++ b/lisp/unicode.el ＠＠ -87,7 +87,7 ＠＠ (check-argument-range code #x0 #x10FFFF) (assert (eq quote-ucs 'ucs) t "Sorry, decode-char doesn't yet support anything but the UCS. ") - (unicode-to-char code)) + (unicode-to-char code nil 'fail)) (defun encode-char (char quote-ucs &optional restriction) "FSF compatibility--return the Unicode code point of CHAR. diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 src/ChangeLog --- a/src/ChangeLog +++ b/src/ChangeLog ＠＠ -1,3 +1,12 ＠＠ +2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; + + * bytecode.c (bytecode_arithcompare): + Rename a couple of local variables to quieten the compiler about + shadowing. + * doprnt.c (rewrite_floating_spec): + Mark OBJ as used if HAVE_BIGFLOAT is not defined, to quieten the + compiler. + 2017-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * data.c: diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 src/bytecode.c --- a/src/bytecode.c +++ b/src/bytecode.c ＠＠ -333,9 +333,9 ＠＠ if (MARKERP (obj2) && (XMARKER (obj1)->buffer == XMARKER (obj2)->buffer)) { - Bytebpos ival1 = byte_marker_position (obj1); - Bytebpos ival2 = byte_marker_position (obj2); - return ival1 < ival2 ? -1 : ival1 > ival2 ? 1 : 0; + Bytebpos bpval1 = byte_marker_position (obj1); + Bytebpos bpval2 = byte_marker_position (obj2); + return bpval1 < bpval2 ? -1 : bpval1 > bpval2 ? 1 : 0; } ival1 = marker_position (obj1); diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 src/doprnt.c --- a/src/doprnt.c +++ b/src/doprnt.c ＠＠ -1768,6 +1768,8 ＠＠ /* Leave obj alone. */ return ch; } +#else + USED (obj); #endif /* The float spec code calls extract_float(), this is a far less complex diff -r d79c7749a1011bfb27c1209ff365758f7e95ae51 -r 9764d52c3fd9829a3a53f090f971033c0cd34f37 tests/ChangeLog --- a/tests/ChangeLog +++ b/tests/ChangeLog ＠＠ -1,3 +1,23 ＠＠ +2017-11-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; + + * automated/case-tests.el: + Change the encoding of this to iso-8859-1, so non-mule doesn't + choke on old-format case tables. + * automated/case-tests.el (downcase1): + Create strings using non-Latin-1 using (format "...%c..." + (decode-char 'ucs CODE) at runtime, rather than simply confusing a + non-mule build at read time. + * automated/case-tests.el (uni-mappings): + Use #'decode-char at runtime, don't assume non-Mule can read a + long list of mostly non-ASCII characters. + * automated/extent-tests.el (string): + Check whether CURRENT-LANGUAGE-ENVIRONMENT is bound (only if Mule) + before dereferencing it. + * automated/format-tests.el (Assert): + Be better about not expecting the Lisp reader to understand + non-Latin-1, use decode-char at runtime. + Mark a few tests with Implementation-Incomplete-Expect-Failure. + 2017-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/lisp-tests.el: This diff is so big that we needed to truncate the remainder. Repository URL: https://bitbucket.org/xemacs/xemacs/ -- This is a commit notification from bitbucket.org. You are receiving this because you have the service enabled, addressing the recipient of this email.