Re: resowner.c $PostgreSQL$
От | Tom Lane |
---|---|
Тема | Re: resowner.c $PostgreSQL$ |
Дата | |
Msg-id | 14449.1090775672@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | resowner.c $PostgreSQL$ (Alvaro Herrera <alvherre@dcc.uchile.cl>) |
Список | pgsql-hackers |
Alvaro Herrera <alvherre@dcc.uchile.cl> writes: > Is it just me or the resowner.c file does not have $PostgreSQL$ > expanded? I see * $PostgreSQL: pgsql-server/src/backend/utils/resowner/resowner.c,v 1.1 2004/07/17 03:30:10 tgl Exp $ > May it be related to the fact that I'm using CVSup? Probably. It sure sounds like *something* in your setup doesn't know that that's a keyword. > Also I see that in all files, the $PostgreSQL$ tag shows a version > number that's one less than the actual version number of the file (shown > with cvs log). Can this be a local config error or something strange is > happenning? This is exactly what I'd expect with a setup unaware of the keyword. AIUI, the keyword substitution is done during checkout. When you check in, you are supplying a file with what is about to become the former value of the keyword string, and since cvs commit doesn't really treat keywords specially, that value is what is actually in the cvs database's new version of the file. You can see similar things happening in our cvsweb interface, since Marc still hasn't fixed it to know about $PostgreSQL$ :-(. For instance go to http://developer.postgresql.org/cvsweb.cgi/pgsql-server/GNUmakefile.in.diff?r1=1.38&r2=1.39 and take a close look at the reported diff in $PostgreSQL$ lines. regards, tom lane
В списке pgsql-hackers по дате отправления: