Mercurial > octave
changeset 25620:467729037209
Strip BOM from UTF-8 encoded .m files (bug #52574).
* input.cc: Strip byte order mark (BOM) from UTF-8 encoded .m files.
| author | Markus Mützel <markus.muetzel@gmx.de> |
|---|---|
| date | Sat, 14 Jul 2018 14:16:22 +0200 |
| parents | b3f6443f6b20 |
| children | 0fa7312e34bc |
| files | libinterp/corefcn/input.cc |
| diffstat | 1 files changed, 7 insertions(+), 1 deletions(-) [+] |
line wrap: on
line diff
--- a/libinterp/corefcn/input.cc Fri Jul 13 21:46:50 2018 -0700 +++ b/libinterp/corefcn/input.cc Sat Jul 14 14:16:22 2018 +0200 @@ -1115,7 +1115,13 @@ = (mfile_encoding.compare ("system") == 0 ? octave_locale_charset_wrapper () : mfile_encoding); - if (encoding.compare ("utf-8") != 0) + if (encoding.compare ("utf-8") == 0) + { + // Check for BOM and strip it + if (src_str.compare (0, 3, "\xef\xbb\xbf") == 0) + src_str.erase (0, 3); + } + else { // convert encoding to UTF-8 before returning string const char *src = src_str.c_str ();