| From: | Peter Eisentraut <peter(at)eisentraut(dot)org> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Michael Paquier <michael(at)paquier(dot)xyz>, Postgres hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: More CppAsString2() in psql's describe.c |
| Date: | 2024-12-02 11:26:47 |
| Message-ID: | 21ec3fa4-e30d-4941-b980-7636b9061be0@eisentraut.org |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 02.12.24 08:52, Tom Lane wrote:
>> (Moreover, the current structure assumes that the C character literal
>> syntax used by the PROKIND_* and other symbols happens to be the same as
>> the SQL string literal syntax required in those queries, which is just
>> an accident.)
> So? There isn't much about C syntax that isn't an accident.
> Neither literal syntax is going to change, so I don't see why
> it's problematic to rely on them being the same.
For example, if you write
#define RELKIND_RELATION '\x72'
then it won't work anymore.
I was also curious whether
#define FOO 'r'
#define RELKIND_RELATION FOO
would work. It appears it does. But this syntactic construction is
quite hard to completely understand currently.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Zhijie Hou (Fujitsu) | 2024-12-02 11:33:38 | RE: Conflict detection for update_deleted in logical replication |
| Previous Message | Peter Eisentraut | 2024-12-02 11:13:56 | Re: Proper object locking for GRANT/REVOKE |