Re: [HACKERS] Bug in LIKE ?
От | Tom Lane |
---|---|
Тема | Re: [HACKERS] Bug in LIKE ? |
Дата | |
Msg-id | 23773.928763047@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: [HACKERS] Bug in LIKE ? (Daniele Orlandi <daniele@orlandi.com>) |
Ответы |
Re: [HACKERS] Bug in LIKE ?
Re: [HACKERS] Bug in LIKE ? |
Список | pgsql-hackers |
Daniele Orlandi <daniele@orlandi.com> writes: > Uhm.... I think the problem is a little worse: It's a real bug, and I see the problem: someone changed the handling of LIKE prefixes in gram.y, without understanding quite what they were doing. 6.4.2 has: if (n->val.val.str[pos] == '\\' || n->val.val.str[pos] == '%') pos++; where 6.5 has: if (n->val.val.str[pos] == '\\' || n->val.val.str[pos+1] == '%') pos++; The first one is right and the second is not. Unless we fix this, LIKE will be completely busted for any string containing non-leading %. Shall I ... ? regards, tom lane
В списке pgsql-hackers по дате отправления: