Doc patch, index search_path where it's used to secure functions

Lists: pgsql-hackers
From: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Doc patch, index search_path where it's used to secure functions
Date: 2012-09-28 16:28:39
Message-ID: 1348849719.13705.8@mofo
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

Hi,

Doc patch, index search_path where it's used to secure functions.
search_path-securing.patch

Against git head. (As are the previous doc patches.)

Karl <kop(at)meme(dot)com>
Free Software: "You don't pay back, you pay forward."
-- Robert A. Heinlein

Attachment Content-Type Size
search_path-securing.patch text/x-patch 642 bytes

From: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Doc patch, index search_path where it's used to secure functions
Date: 2012-09-28 17:17:12
Message-ID: 1348852632.13705.9@mofo
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On 09/28/2012 11:28:39 AM, Karl O. Pinc wrote:

> Doc patch, index search_path where it's used to secure functions.
> search_path-securing.patch

Second version. Should be indexing the concept, not the run-time
setting.

Karl <kop(at)meme(dot)com>
Free Software: "You don't pay back, you pay forward."
-- Robert A. Heinlein

Attachment Content-Type Size
search_path-securing_v2.patch text/x-patch 642 bytes

From: Peter Eisentraut <peter_e(at)gmx(dot)net>
To: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Doc patch, index search_path where it's used to secure functions
Date: 2012-11-14 02:46:19
Message-ID: 1352861179.26167.2.camel@vanquo.pezone.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On Fri, 2012-09-28 at 12:17 -0500, Karl O. Pinc wrote:
> On 09/28/2012 11:28:39 AM, Karl O. Pinc wrote:
>
> > Doc patch, index search_path where it's used to secure functions.
> > search_path-securing.patch
>
> Second version. Should be indexing the concept, not the run-time
> setting.

Well, I'm not sure. We currently have three index entries on the topic:

search path
search_path
search_path configuration parameter

I think I'd put them all under search_path.


From: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Doc patch, index search_path where it's used to secure functions
Date: 2012-11-14 06:21:20
Message-ID: 1352874080.4647.0@mofo
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On 11/13/2012 08:46:19 PM, Peter Eisentraut wrote:
>
> Well, I'm not sure. We currently have three index entries on the
> topic:
>
> search path
> search_path
> search_path configuration parameter
>
> I think I'd put them all under search_path.

Ok.

I think you are right that they need to all be
under one index entry.

It might make sense to have a "search term"
entry that says "see search_term", but since
the two entries would be right next to each
other in the index this seems overkill.

I'm going to send this in as a single patch
that fixes all the search path related
index entries:

search_path-index.patch

(replaces search_path-normalize.patch
and search_path-securing_v2.patch)

This new patch also changes the
search path index in doc/src/sgml/func.sgml.
Perhaps I just don't understand the logic
in the indexing in that file but the
search path index there seems to point to the
entire file/nowhere useful. Since there are 2 places
in the file that are concerned with
search path I've removed the "global"
index into the file and added new
index entries. This also moved the
"schema, current" index target.
(Since that too involves search path.)

I've also added a <seealso> entry to "schema, current"
to the top-level "search_path" index entry.

If you want this broken up into smaller
patches let me know.

Regards,

Karl <kop(at)meme(dot)com>
Free Software: "You don't pay back, you pay forward."
-- Robert A. Heinlein

Attachment Content-Type Size
search_path-index.patch text/x-patch 3.4 KB

From: Peter Eisentraut <peter_e(at)gmx(dot)net>
To: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Doc patch, index search_path where it's used to secure functions
Date: 2012-12-14 04:05:15
Message-ID: 1355457915.27717.1.camel@vanquo.pezone.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On Wed, 2012-11-14 at 00:21 -0600, Karl O. Pinc wrote:
> I'm going to send this in as a single patch
> that fixes all the search path related
> index entries:
>
> search_path-index.patch
>
> (replaces search_path-normalize.patch
> and search_path-securing_v2.patch)

The other configuration parameters are all indexed as "x_y_z
configuration parameter", so I've kept search_path aligned with that. I
have applied your other changes, so I think it's good now. Let me know
if you feel additional changes should be made.


From: "Karl O(dot) Pinc" <kop(at)meme(dot)com>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Doc patch, index search_path where it's used to secure functions
Date: 2012-12-14 15:52:48
Message-ID: 1355500368.28556.3@mofo
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On 12/13/2012 10:05:15 PM, Peter Eisentraut wrote:

> The other configuration parameters are all indexed as "x_y_z
> configuration parameter", so I've kept search_path aligned with that.
> I
> have applied your other changes, so I think it's good now. Let me
> know
> if you feel additional changes should be made.

I like the way it has come out. Thanks!

Karl <kop(at)meme(dot)com>
Free Software: "You don't pay back, you pay forward."
-- Robert A. Heinlein