Re: extensible options syntax for replication parser? - Mailing list pgsql-hackers

From Fujii Masao
Subject Re: extensible options syntax for replication parser?
Date
Msg-id 38533959-ab69-7d47-6f3a-eb900f2fbc3f@oss.nttdata.com
Whole thread Raw
In response to Re: extensible options syntax for replication parser?  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: extensible options syntax for replication parser?  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers

On 2021/09/24 0:05, Robert Haas wrote:
> On Thu, Sep 23, 2021 at 2:55 AM tushar <tushar.ahuja@enterprisedb.com> wrote:
>> l checked and look like the issue is still not fixed against v7-* patches -
>>
>> postgres=#  create subscription test CONNECTION 'host=127.0.0.1 user=edb dbname=postgres' PUBLICATION p with
(create_slot= true);
 
>> ERROR:  could not create replication slot "test": ERROR:  syntax error
> 
> Thanks. Looks like that version had some stupid mistakes. Here's a new one.

-     <indexterm><primary>BASE_BACKUP</primary></indexterm>
+    <term><literal>BASE_BACKUP</literal> [ ( <replaceable class="parameter">option</replaceable> [, ...] ) ]

You seem to accidentally remove the index term for BASE_BACKUP.

+ident_or_keyword:
+            IDENT                            { $$ = $1; }

ident_or_keyword seems to be used only for generic options,
but it includes the keywords for legacy options like "FAST".
Isn't it better to remove the keywords for legacy options from
ident_or_keyword?

OTOH, the keywords for newly-introduced generic options like
CHECKPOINT should be defined in repl_scanner.l and repl_gram.y?


Regards,

-- 
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION



pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: proposal: possibility to read dumped table's name from file
Next
From: Greg Nancarrow
Date:
Subject: Re: Skipping logical replication transactions on subscriber side