> Date: Sun, 23 Oct 2022 17:06:55 +0200 > From: pertu...@free.fr > Cc: gavinsmith0...@gmail.com, bug-texinfo@gnu.org > > On Sun, Oct 23, 2022 at 05:30:24PM +0300, Eli Zaretskii wrote: > > > Date: Sun, 23 Oct 2022 14:44:33 +0200 > > > From: pertu...@free.fr > > > Cc: gavinsmith0...@gmail.com, bug-texinfo@gnu.org > > > > > Thanks, I've done that, and now at least the first part of the test > > suite, in test_scripts/, does run. Surprisingly, > > input_file_names_recoded_stamp.txt has "OK" written to it. > > > > Shouldn't the file have had "FAILED" in it? > > It depends whether a file with latin1 name was copied or not. If the > file was copied, it is normal to have OK. So, could you please check > that there is a file included_latîn1.texi in tp/tests, with the î > encoded in latin1? On my Debian GNU/Linux it shows as: > > 'included_lat'$'\356''n1.texi'
The file exists, but its name is included_latמn1.texi. That's how Windows interpreted that non-ASCII character in my locale. > > Out of 144 tests in test_scripts/, 105 pass, 31 are skipped, and 8 > > fail. The failed ones are all about non-ASCII characters, it seems. > > Could you please check whether the following two tests which depend on > included_latîn1.texi are ok or fail: > formatting/manual_include_accented_file_name_latin1 > formatting/manual_include_accented_file_name_latin1_use_locale_encoding They fail when input_file_names_recoded_stamp.txt has "OK" and are skipped when it has "FAILED".
