Re: pgindent fixups
От | Robert Haas |
---|---|
Тема | Re: pgindent fixups |
Дата | |
Msg-id | CA+TgmoZ9nrVi_=JZ4K0xsQZLDbgsYfrVx+kuN-Fz5kCspzGn7A@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: pgindent fixups (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: pgindent fixups
|
Список | pgsql-hackers |
On Tue, May 3, 2016 at 9:40 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Robert Haas <robertmhaas@gmail.com> writes: >> 1. Is pgindent supposed to touch DATA() lines? Because it does. > > It always has. > >> 2. CustomPathMethods is not in the buildfarm's typedefs.list. Why not? > > Probably because there are no variables, parameters, or fields declared > with that typedef, so it doesn't get into the debugger symbol table of > any .o file. Grep says that the single use of the struct doesn't use > the typedef; it's > const struct CustomPathMethods *methods; > So you'd need to change that, or else tweak some code somewhere to have > a variable explicitly declared using the typedef. Ah. It's a minor issue, so probably not worth worrying about. >> - In src/backend/executor/execParallel.c, it dodges two cases where >> pgindent does stupid things with offsetof. > > Well, it's also pretty stupid about sizeof, or casts generally, so > I'm not really convinced you need to get exercised about these two > places in particular. But no objection to tweaking them if you > want to. OK, I committed this. Barring objections, I'll go ahead and pgindent the whole tree tomorrow. If we're going to revert anything big then we might want to hold off, but otherwise I think its better to get this done sooner rather than later. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
В списке pgsql-hackers по дате отправления: