Re: Multiline-statement and multi-statement for pgbench custom script.

On Fri, Aug 28, 2015 at 4:33 AM, Kyotaro HORIGUCHI
<horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp> wrote:
> Hi, this is a spin-off patch from Fabien COELHO's
> backslash-continuations.
>
> The major concept of this patch is making usage of psql's scanner
> to get rid of home-grown scanner of pgbench to make
> multi-statement feature available for pgbench custom scripts.
>
> This patch does the following things.
>
> - Modify psqlscan.l so that unnecessary functions of it can be
> masked when used in other modules like pgbench.
>
> - Modify pgbench to use psqlscan.l so that the following features
> available in pgbench.
>
> - multi-statement in custom scripts.
> - natural continuation of SQL statements in costom scripts.
> - backslash-continuation for pgbench metacommands in costom scripts.
>
> The patch consists of following files.
>
> - 0001-Prepare-to-share-psqlscan-with-pgbench.patch
> Modifies psqlscan.l in psql as the preparation.
>
> - 0002-Make-use-of-psqlscan-for-parsing-of-custom-script.patch
> Modifies pgbench to use psqlscan.l.
>
> - 0003-Change-MSVC-Build-script.patch
> Modify MSVC build script.
>
> - 0004-Change-the-way-to-hold-command-list.patch
> Get rid of double-format of internal command list.
> This changes the way of holding command list to linked list
> totally.

Has anyone reviewed this patch? Does it need to be reviewed?

This CommitFest entry looks related:

http://www.postgresql.org/message-id/flat/alpine(dot)DEB(dot)2(dot)10(dot)1505141510130(dot)1821(at)sto#alpine(dot)DEB(dot)2(dot)10(dot)1505141510130(dot)1821@sto

But I'm not sure how it relates to this thread.

(And let me say parenthetically that the web site's insistence on
adding #THE-SAME-MESSAGE-ID-THAT-IS-ALREADY-IN-THE-URL to everything
is pretty annoying ... it makes the links really long for no good
reason.)

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company