Re: Infinite Interval - Mailing list pgsql-hackers

On Sat, Mar 18, 2023 at 3:55 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
>    Joseph Koshakow <koshy44@gmail.com> writes:
>    > On Sat, Mar 18, 2023 at 3:08 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>    >> More specifically, those are from running pg_indent with an obsolete
>    >> typedefs list.
>
>    > I must be doing something wrong because even after doing that I get the
>    > same strange formatting. Specifically from the root directory I ran
>
>    Hmm, I dunno what's going on there.  When I do this:
>
>    >   curl https://buildfarm.postgresql.org/cgi-bin/typedefs.pl -o
>    > src/tools/pgindent/typedefs.list
>
>    I end up with a plausible set of updates, notably
>
>    $ git diff
>    diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
>    index 097f42e1b3..667f8e13ed 100644
>    --- a/src/tools/pgindent/typedefs.list
>    +++ b/src/tools/pgindent/typedefs.list
>    ...
>    @@ -545,10 +548,12 @@ DataDumperPtr
>     DataPageDeleteStack
>     DatabaseInfo
>     DateADT
>    +DateTimeErrorExtra
>     Datum
>     DatumTupleFields
>     DbInfo
>     DbInfoArr
>    +DbLocaleInfo
>     DeClonePtrType
>     DeadLockState
>     DeallocateStmt
>
>    so it sure ought to know DateTimeErrorExtra is a typedef.
>    I then tried pgindent'ing datetime.c and timestamp.c,
>    and it did not want to change either file.  I do get
>    diffs like

>     DecodeDateTime(char **field, int *ftype, int nf,
>                   int *dtype, struct pg_tm *tm, fsec_t *fsec, int *tzp,
>    -              DateTimeErrorExtra *extra)
>    +              DateTimeErrorExtra * extra)
>     {
>        int         fmask = 0,
>
>    if I try to pgindent datetime.c with typedefs.list as it
>    stands in HEAD.  That's pretty much pgindent's normal
>    behavior when it doesn't recognize a name as a typedef.

I must have been doing something wrong because I tried again today and
it worked fine. However, I go get a lot of changes like the following:

  -               if TIMESTAMP_IS_NOBEGIN(dt2)
  -                       ereport(ERROR,
  -                                       (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
  -                                        errmsg("timestamp out of range")));
  +               if TIMESTAMP_IS_NOBEGIN
  +                       (dt2)
  +                               ereport(ERROR,
  +                                               (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
  +                                                errmsg("timestamp out of range")));