Mercurial > octave
diff libinterp/corefcn/file-io.cc @ 26706:ccea3574f36b
Support encoding of file streams in textscan (bug #55452).
* oct-stream.cc (do_textscan): Pass encoding in constructor for textscan object.
(textscan): Store encoding in object. Convert strings from encoding.
* file-io.cc (textscan): Add BIST.
* io.tst (fopen): Use code page identifier that better works cross-platform.
| author | Markus Mützel <markus.muetzel@gmx.de> |
|---|---|
| date | Sat, 09 Feb 2019 20:05:47 +0100 |
| parents | c13143821eef |
| children | 7506884fc158 |
line wrap: on
line diff
--- a/libinterp/corefcn/file-io.cc Sat Feb 09 17:41:48 2019 +0100 +++ b/libinterp/corefcn/file-io.cc Sat Feb 09 20:05:47 2019 +0100 @@ -2267,6 +2267,19 @@ %! obs = textscan (str, "%f", "delimiter", {",",";","$"}); %! assert (obs, { [0; 1; NaN; 2; 3] }); +## file stream with encoding +%!test +%! f = tempname (); +%! fid = fopen (f, "w+", "n", "iso-8859-1"); +%! unwind_protect +%! fprintf (fid, "abc,äöü\n"); +%! fseek (fid, 0, "bof"); +%! obs = textscan (fid, "%s", "delimiter", ","); +%! fclose (fid); +%! assert (obs, { {"abc"; "äöü"} }); +%! unwind_protect_cleanup +%! unlink (f); +%! end_unwind_protect */ // These tests have end-comment sequences, so can't just be in a comment