Re: Misleading docs or implementation of file:read/2 and friends
Dan Gudmundsson <[email protected]>
| Newsgroups | gmane.comp.lang.erlang.bugs |
|---|---|
| Message-ID | <CANX4uuORnN_WOvv2mrkmqaYZcdkurRef2x3NvRtX=DkdZ0-hdQ@mail.gmail.com> |
It does for latin1, i.e. it can convert latin1 chars to utf8|utf16|utf32 encoding on file and vice verse. That is no change from what file:read and write do. And the doc is the same for read_line as for read. Anything above 256 gives badarg no_translation. I left 'pread' as it is, it doesn't do any conversion today, i.e. ignores encoding. I don't know if it is a bug or intentionally ignored since it is strange to jump around at byte lengths when content is not byte oriented? On Sat, Mar 21, 2015 at 1:38 PM, Anthony Ramine <[email protected]> wrote: > Le 16 mars 2015 à 08:22, Dan Gudmundsson <[email protected]> a écrit : > > > > https://github.com/erlang/otp/commit/7efa93c9dd4ce25c5754c1b72ec804e8172c0099 > > So in the end, the file module too supports unicode reading? > > > _______________________________________________ erlang-bugs mailing list [email protected] http://erlang.org/mailman/listinfo/erlang-bugs