Re: The future of XEmacs

On 2012-07-31, Stephen J. Turnbull <stephen(a)xemacs.org&gt; wrote: Can't really just post a diff, because I've done so much other mucking around. However, what I've done is this: For decoding: added int num_decoding_errors; to struct decoding_stream modified all the decoding functions (except CCL, where of course there's nothing I can do) to increment that whenever they meet an invalid byte, and then in decoding_closer I replace the final return Lstream_close (str->other_end); by int r = Lstream_close (str->other_end); if ( str->num_decoding_errors ) { warn_when_safe_lispobj(Qstream,Qwarning, emacs_doprnt_string_lisp_2( "%d invalid bytes were encountered while reading from file %s in coding system %s\nThis probably means the file is in a different coding system.\nYour data will be corrupted if you save this data.", Qnil,-1,3,make_int(str->num_decoding_errors),fname,CODING_SYSTEM_NAME(str->codesys))); } return r; For encoding, it's much the same, except that I also store the last unencoded character, and print that in the error message: if ( str->num_unencodable ) { warn_when_safe_lispobj(Qstream,Qwarning, emacs_doprnt_string_lisp_2( "%d characters could not be encoded while writing to file %s in coding system %s.\nThe last unencodable character was %s \nThis means your data has not been correctly written. Try a universal coding system (utf-8 or iso-2022-7).", Qnil,-1,4,make_int(str->num_unencodable),fname,CODING_SYSTEM_NAME(str->codesys),make_char(str->last_unencodable))); } This all fairly painful to encounter, but that's part of the point: it should be painful enough to make me find and fix the coding system errors! _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta