On Fri, Nov 01, 2024 at 03:45:43PM +0200, Eli Zaretskii wrote: > So we now encourage not to use @documentencoding at all if it's UTF-8?
Indeed. https://www.gnu.org/software/texinfo/manual/texinfo/texinfo.html#g_t_0040documentencoding "In the default case, the input and output document encoding are assumed to be UTF-8, the vast global character encoding, expressed in 8-bit bytes. UTF-8 is compatible with 7-bit ASCII. It is recommended to use UTF-8 encoding for Texinfo manuals." ... "UTF-8 should always be the best choice for the encoding. Texinfo still supports additional encodings, mainly for compatibility with older manuals" > Is that wise? I think so. I see no particular reason to use another encoding. For Info, it is probably needed to use UTF-8 for cross-references/dir file. I remember a discussions where someone even proposed to allow only UTF-8 as input encoding. I think that it is better to leave open the possibility to use different encodings. The main issue with using different encodings is some slowdown because we need to encode/decode, although it is also often a way to check the validity of the encoded text, in particular if UTF-8 encoded. -- Pat
