Re: 7.0.2 cuts off attribute name
От | Tom Lane |
---|---|
Тема | Re: 7.0.2 cuts off attribute name |
Дата | |
Msg-id | 10801.960951566@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | 7.0.2 cuts off attribute name ("G. Anthony Reina" <reina@nsi.edu>) |
Список | pgsql-hackers |
"G. Anthony Reina" <reina@nsi.edu> writes: > "significance_of_anova_for_spike_rates". (I know, I know. I get laughed > at all the time for using prepositions in my variable names). In 6.5.1 > and before, Postgres took the whole name as the identifier. With 7.0.2 > it seems to truncate to "significance_of_anova_for_spike" (31 > characters). What? The default length limit has been 31 characters for a long time, certainly long before 6.5.*. 7.0 has a new behavior of *telling* you that it's truncating overlength identifiers, but the system has always truncated 'em. If you're simply complaining about the fact that it emits a notice, I agree with you 100%: that notice is one of the most nonstandard, useless, annoying bits of pointless pedantry I've seen in many years. I argued against it to start with but was outvoted. Maybe we can have a revote now that people have had some practical experience with it: who still thinks it's a good idea? > Is there any reason that this has changed? Anyway to get a larger length > (say 40 or 50 characters)? You could recompile with a larger NAMEDATALEN, but unless you did so in your 6.5.* installation, that's not what's bugging you. Look for the elog(NOTICE,...) call in src/backend/parser/scan.l and dike that out, instead. regards, tom lane
В списке pgsql-hackers по дате отправления: