pgAudit 1.6.0 commit log

commit   : 6afeae52d8e4569235bf6088e983d95ec26f13b7    
  
author   : David Steele <[email protected]>    
date     : Thu, 5 Aug 2021 08:22:32 -0400    
  
committer: David Steele <[email protected]>    
date     : Thu, 5 Aug 2021 08:22:32 -0400    

Testing is entirely Docker-based now.  

commit   : bd6a261f72e08daa3a5f36b46aa4b1728d23709b    
  
author   : David Steele <[email protected]>    
date     : Thu, 5 Aug 2021 08:02:22 -0400    
  
committer: GitHub <[email protected]>    
date     : Thu, 5 Aug 2021 08:02:22 -0400    

This logic was submitted in PR #88 but there was some confusion on my part about what it was supposed to do. Since the title was "Suppress logging for internally generated foreign-key queries" I tried to make it do that, and broke SELECT FOR UPDATE logging, which unfortunately had no test.
  

  
Reading the PR again, it seems Peter's intention was only to correctly classify SELECT FOR UPDATE as SELECT. In any case that represents an improvement over what we have, even if it does not suppress logging for internally generated foreign-key queries, at least not in the case of SELECT.
  

  
So, revert the logic that suppressed the SELECT FOR UPDATE logging and use Peter's logic that correctly classifies them as SELECT.
  

  
Also add a SELECT FOR UPDATE test to prevent regressions.
  

  
Reported by Sergey Shinderuk (@shinderuk).
  
Reviewed by Sergey Shinderuk (@shinderuk), Stephen Frost (@sfrost).

commit   : bb816445df1c770a6f034d7bee0f428fc2265904    
  
author   : David Steele <[email protected]>    
date     : Tue, 3 Aug 2021 16:17:09 -0400    
  
committer: David Steele <[email protected]>    
date     : Tue, 3 Aug 2021 16:17:09 -0400    

commit   : e8cded51a4a1a7f503ed29dcce0c85447d6448c5    
  
author   : Mingchun Zhao <[email protected]>    
date     : Wed, 4 Aug 2021 04:06:45 +0900    
  
committer: GitHub <[email protected]>    
date     : Wed, 4 Aug 2021 04:06:45 +0900    

Specifies that audit logging should include the rows retrieved or affected by a statement. When enabled the rows field will be included after the parameter field.
  

  
The default is `off`.

commit   : ed6975c5222de085926221596c98bfb3531f1a97    
  
author   : David Steele <[email protected]>    
date     : Sat, 17 Jul 2021 09:26:28 -0400    
  
committer: David Steele <[email protected]>    
date     : Sat, 17 Jul 2021 09:26:28 -0400    

This allows the test to be run multiple times without leaving behind stopped containers.  

commit   : 70b30d43799edc106f9ebfaee178959043e33081    
  
author   : David Steele <[email protected]>    
date     : Mon, 12 Jul 2021 14:42:50 -0400    
  
committer: David Steele <[email protected]>    
date     : Mon, 12 Jul 2021 14:42:50 -0400    

Commit 7c337b6b in beta2 changed ProcessUtility_hook_type so update pgaudit to match.  

commit   : 002f2c3c3b88b367388e04932f9234c7792485cd    
  
author   : David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 16:51:51 -0400    
  
committer: GitHub <[email protected]>    
date     : Wed, 23 Jun 2021 16:51:51 -0400    

Since ExecCheckRTPerms() is no longer called for "create table as" and "create materialized view" the INSERT record will no longer be audit logged. However, the "create table as" and "create materialized view" records are still logged and they imply an insert.

commit   : e3d79b03ee03735cce3b94cd01d1f53d9b04f638    
  
author   : David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 15:19:17 -0400    
  
committer: David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 15:19:17 -0400    

This will prevent errors when switching branches and since there is only one file to compile there is not much to be gained by reusing prior objects.  

commit   : 6b56031e87221cab510494e00c410246d129e53c    
  
author   : David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 14:49:10 -0400    
  
committer: David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 14:49:10 -0400    

This is no longer needed since Docker is now used for testing.  

commit   : e2e5a69c4dc949d17f86b783c69c7c9b1c76f876    
  
author   : David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 14:40:00 -0400    
  
committer: David Steele <[email protected]>    
date     : Wed, 23 Jun 2021 14:40:00 -0400    

commit   : c6d958bb4d1b19cdb649d78f45e4ca501e081d8b    
  
author   : David Steele <[email protected]>    
date     : Tue, 22 Jun 2021 14:26:35 -0400    
  
committer: David Steele <[email protected]>    
date     : Tue, 22 Jun 2021 14:26:35 -0400    

This reverts commit b045fb9b90b7225b533502f733250d5a95b94450.  
  
Somehow a development version of this branch got committed, so revert until it can be fixed.  

commit   : b045fb9b90b7225b533502f733250d5a95b94450    
  
author   : David Steele <[email protected]>    
date     : Fri, 18 Jun 2021 09:00:26 -0400    
  
committer: GitHub <[email protected]>    
date     : Fri, 18 Jun 2021 09:00:26 -0400    

Since ExecCheckRTPerms() is no longer called for "create table as" and "create materialized view" we can't depend on that mechanism to log inserts into the newly created table / mat view.
  

  
Instead, capture the into relation in the utility hook and then use it to generate an insert log record at query start.

commit   : 5b0a3a6c1b436adb46d133c6c4e0a9ff306db094    
  
author   : David Steele <[email protected]>    
date     : Fri, 11 Jun 2021 11:06:48 -0400    
  
committer: David Steele <[email protected]>    
date     : Fri, 11 Jun 2021 11:06:48 -0400    

commit   : 8831cef691b081ce71a14c10a2980de5c46a88f7    
  
author   : Peter Eisentraut <[email protected]>    
date     : Thu, 10 Jun 2021 18:27:55 +0200    
  
committer: GitHub <[email protected]>    
date     : Thu, 10 Jun 2021 18:27:55 +0200    

This setting, when turned off (not default), turns off all logging of the full statement text and parameters. This is analogous to log_statement_once, except it never logs the statement.
  

  
Depending on requirements, the full statement text might not be required in the audit log. The combination of command tag and object can be enough. Omitting the full statement text makes the logs less verbose and can also prevent some accidental information leaks.

commit   : 28faa197d3a4c63738e9a2f74488f555dc9d95e5    
  
author   : David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 09:09:59 -0500    
  
committer: David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 09:09:59 -0500    

commit   : 94a2ae8c203a0a305edf59787a4af57651e59d2b    
  
author   : David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 09:05:18 -0500    
  
committer: David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 09:05:18 -0500    

commit   : fd4319f7c8e215a4f291975f48c931fd55f5b037    
  
author   : David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 08:55:42 -0500    
  
committer: David Steele <[email protected]>    
date     : Tue, 12 Jan 2021 08:55:42 -0500    

Adding PG_CONFIG to make suggested by @vkhvorostianyi.  

commit   : 7169e84e1aa6d68f5c59701655ef31f810d171f5    
  
author   : Michael Otte <[email protected]>    
date     : Tue, 12 Jan 2021 05:32:44 -0800    
  
committer: GitHub <[email protected]>    
date     : Tue, 12 Jan 2021 05:32:44 -0800    

This only works when pgaudit is located in a PostgreSQL source tree.