Have the result of coding_character_tell() reflect str->convert_to, too. src/ChangeLog addition: 2014-01-22 Aidan Kehoe <kehoea@parhasard.net> * text.c (buffered_bytecount_to_charcount): New. Return the number of characters in a coding or lstream buffer, including partial characters at the beginning of the buffer, but not including those at the end. Refactored from Lstream_character_tell(). * text.h: * text.h (buffered_bytecount_to_charcount): Declare it. * lstream.c (Lstream_character_tell): Use the refactored buffered_bytecount_to_charcount () here, both for the unget buffer and in_buffer. * file-coding.c (coding_character_tell): Check the character count of the lstream buffer too, when passing back the character count from the coding stream.

These files are *not* part of XEmacs.

They are all available publicly from O'Reilly Associates or the
Unicode Consortium, and are collected here *unchanged* for the
convenience of the user/builder of XEmacs.  Redistribution is governed
by their own licensing notices.

See "ibm/COPYING", "mule-ucs/README" and "unicode-consortium/COPYING"
for details.