Re: Why does pg_checksums -r not have a long option? - Mailing list pgsql-hackers

From Fabien COELHO
Subject Re: Why does pg_checksums -r not have a long option?
Date
Msg-id alpine.DEB.2.21.1905271616270.24257@lancre
Whole thread Raw
In response to Re: Why does pg_checksums -r not have a long option?  (Michael Paquier <michael@paquier.xyz>)
Responses Re: Why does pg_checksums -r not have a long option?  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
Bonjour Michael,

> +     <varlistentry>
> +      <term><option>-f <replaceable>filenode</replaceable></option></term>
> +      <term><option>--filenode=<replaceable>filenode</replaceable></option></term>
> +      <listitem>
> +       <para>
> +        Only validate checksums in the relation with specified relation file node.
> +       </para>
> Two nits.  I would just have been careful about the number of
> characters in the line within the <para> markup.  And we use
> extensively "filenode" in the docs.

Ok.

> +       [ 'pg_checksums', '--enable', '-filenode', '1234', '-D', $pgdata ],
> This fails, but not for the reason it is written for.

Indeed. command_fails() is a little too simplistic, it should really check 
that the error message is there.

> It looks strange to not order --filenode alphabetically in --help.

Forgot, it stayed at the r position for no good reason.

> With all these issues cleaned up, I got the attached.  Does that look
> fine?  (I ran pgperltidy and pgindent on top of it.)

Works for me. Doc build is ok as well.

-- 
Fabien.



pgsql-hackers by date:

Previous
From: Antonin Houska
Date:
Subject: Re: Converting NOT IN to anti-joins during planning
Next
From: Sascha Kuhl
Date:
Subject: Re: Indexing - comparison of tree structures