Thread: [PATCH] log_statement docs
Patch amends description of log_statement to include definition of "all" option. gabrielle
Attachment
gabrielle <gorthx@gmail.com> writes:
> Patch amends description of log_statement to include definition of "all" option.
Shouldn't it just say "all means all", if we think that needs to be
explained?
regards, tom lane
On Wed, Feb 24, 2010 at 7:14 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > gabrielle <gorthx@gmail.com> writes: >> Patch amends description of log_statement to include definition of "all" option. > > Shouldn't it just say "all means all", if we think that needs to be > explained? I think it should say something about the "all" option. I read & re-read that section trying to figure out what the difference between "mod" and "all" was, and finally asked on the irc channel. As it stands, it essentially says "there are three options to this command, but we're only going to tell you about two of them." :) Thanks! gabrielle
gabrielle wrote: > On Wed, Feb 24, 2010 at 7:14 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > gabrielle <gorthx@gmail.com> writes: > >> Patch amends description of log_statement to include definition of "all" option. > > > > Shouldn't it just say "all means all", if we think that needs to be > > explained? > > I think it should say something about the "all" option. I read & > re-read that section trying to figure out what the difference between > "mod" and "all" was, and finally asked on the irc channel. As it > stands, it essentially says "there are three options to this command, > but we're only going to tell you about two of them." :) Well, if you take a look at utility.c:GetCommandLogLevel() you will see that ALL includes a lot more commands than just SELECT. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com PG East: http://www.enterprisedb.com/community/nav-pg-east-2010.do + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian <bruce@momjian.us> writes:
> gabrielle wrote:
>> On Wed, Feb 24, 2010 at 7:14 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> Shouldn't it just say "all means all", if we think that needs to be
>>> explained?
>>
>> I think it should say something about the "all" option. I read &
>> re-read that section trying to figure out what the difference between
>> "mod" and "all" was, and finally asked on the irc channel. As it
>> stands, it essentially says "there are three options to this command,
>> but we're only going to tell you about two of them." :)
> Well, if you take a look at utility.c:GetCommandLogLevel() you will see
> that ALL includes a lot more commands than just SELECT.
Yeah. My objection was not to documenting ALL, it was to documenting it
with exactly that sentence, which seems both needlessly complicated and
subject to errors of omission.
regards, tom lane
Tom Lane wrote: > Bruce Momjian <bruce@momjian.us> writes: > > gabrielle wrote: > >> On Wed, Feb 24, 2010 at 7:14 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > >>> Shouldn't it just say "all means all", if we think that needs to be > >>> explained? > >> > >> I think it should say something about the "all" option. I read & > >> re-read that section trying to figure out what the difference between > >> "mod" and "all" was, and finally asked on the irc channel. As it > >> stands, it essentially says "there are three options to this command, > >> but we're only going to tell you about two of them." :) > > > Well, if you take a look at utility.c:GetCommandLogLevel() you will see > > that ALL includes a lot more commands than just SELECT. > > Yeah. My objection was not to documenting ALL, it was to documenting it > with exactly that sentence, which seems both needlessly complicated and > subject to errors of omission. I have applied the following patch with adds "off" and "all statements" identifiers. This is probaby the right level of detail for this. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com PG East: http://www.enterprisedb.com/community/nav-pg-east-2010.do + If your life is a hard drive, Christ can be your backup. + Index: doc/src/sgml/config.sgml =================================================================== RCS file: /cvsroot/pgsql/doc/src/sgml/config.sgml,v retrieving revision 1.253 diff -c -c -r1.253 config.sgml *** doc/src/sgml/config.sgml 25 Feb 2010 13:26:15 -0000 1.253 --- doc/src/sgml/config.sgml 25 Feb 2010 19:19:14 -0000 *************** *** 3392,3399 **** <listitem> <para> Controls which SQL statements are logged. Valid values are ! <literal>none</>, <literal>ddl</>, <literal>mod</>, and ! <literal>all</>. <literal>ddl</> logs all data definition statements, such as <command>CREATE</>, <command>ALTER</>, and <command>DROP</> statements. <literal>mod</> logs all <literal>ddl</> statements, plus data-modifying statements --- 3392,3399 ---- <listitem> <para> Controls which SQL statements are logged. Valid values are ! <literal>none</> (off), <literal>ddl</>, <literal>mod</>, and ! <literal>all</> (all statements). <literal>ddl</> logs all data definition statements, such as <command>CREATE</>, <command>ALTER</>, and <command>DROP</> statements. <literal>mod</> logs all <literal>ddl</> statements, plus data-modifying statements
On Thu, Feb 25, 2010 at 11:20 AM, Bruce Momjian <bruce@momjian.us> wrote: > I have applied the following patch with adds "off" and "all statements" > identifiers. This is probaby the right level of detail for this. Thank you! gabrielle