build remaining Flex files standalone

Starting a new thread to control clutter. [was: Re: [RFC] building
postgres with meson]

motivation: https://www.postgresql.org/message-id/20220810171935.7k5zgnjwqzalzmtm%40awork3.anarazel.de

On Thu, Aug 11, 2022 at 11:07 AM Andres Freund <andres(at)anarazel(dot)de> wrote:
> I think we should consider compiling it separately from guc.c. guc.c already
> compiles quite slowly (iirc beat only by ecpg and main grammar), and it's a
> relatively commonly changed source file.

Done in the attached, and will do the rest in time. It seemed most
straightforward to put ProcessConfigFileInternal() in guc.c since
that's where most of its callers are, and it relies on some vars and
types declared there. There are a couple new extern declarations in
guc.h that are only for guc.c and guc-file.c:

+/* functions shared between guc.c and guc-file.l */
+extern int guc_name_compare(const char *namea, const char *nameb);
+extern ConfigVariable *ProcessConfigFileInternal(GucContext context,
+ bool applySettings, int elevel);
+extern void record_config_file_error(const char *errmsg,
+ const char *config_file,
+ int lineno,
+ ConfigVariable **head_p,
+ ConfigVariable **tail_p);

These might be better placed in a new guc_internal.h. Thoughts?

> It might even be a good idea to split guc.c so it only contains the settings
> arrays + direct dependencies...

Perhaps this can be a TODO item, one which falls under "[E] marks
items that are easier to implement". I've been slacking on removing
the old/intractable cruft from the TODO list, but we should also be
sticking small nice-but-not-necessary things in there. That said, if
this idea has any bearing on the guc_internal.h idea, it might be
better dealt with now.

--
John Naylor
EDB: http://www.enterprisedb.com