mirror of
https://github.com/postgres/postgres.git
synced 2025-05-05 09:19:17 +03:00
The t_iseq() macro does not need to be guarded by a character length check (at least when the comparison value is an ASCII character, as its documentation requires). Some portions of contrib/ltree hadn't read that memo, so simplify them. The last change in gettoken_query, - else if (charlen == 1 && !t_iseq(state->buf, ' ')) + else if (!t_iseq(state->buf, ' ')) looks like it's actually a bug fix: I doubt that the intention was to silently ignore multibyte characters as if they were whitespace. I'm not tempted to back-patch though, because this will have the effect of tightening what is allowed in ltxtquery strings. Discussion: https://postgr.es/m/2548310.1664999615@sss.pgh.pa.us