[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn commit: r1672295 - in /subversion/trunk/subversion: libsvn_wc/wc-metadata.sql tests/libsvn_wc/wc-queries-test.c

From: Bert Huijben <bert_at_qqmail.nl>
Date: Tue, 14 Apr 2015 17:10:18 +0000

We probably should as I don’t think there is a unique constraint in the current sqlite schema for this system table (The or constraint doesn’t do much, but might one time avoid an error case). I’m not able to look at this now, but Given your comments I’m assuming it might allow adding this same row many times.


Bert






Sent from Windows Mail

Gight



From: Philip Martin
Sent: ‎Tuesday‎, ‎April‎ ‎14‎, ‎2015 ‎6‎:‎52‎ ‎PM
To: Bert Huijben
Cc: 'Subversion Development'





rhuijben_at_apache.org writes:

> Author: rhuijben
> Date: Thu Apr 9 11:31:12 2015
> New Revision: 1672295
  
> +INSERT OR REPLACE INTO sqlite_stat1(tbl, idx, stat) VALUES
> + ('EXTERNALS','sqlite_autoindex_EXTERNALS_1', '100 100 1');
> +INSERT OR REPLACE INTO sqlite_stat1(tbl, idx, stat) VALUES
> + ('EXTERNALS','I_EXTERNALS_DEFINED', '100 100 3 1');
> +
> /* sqlite_autoindex_WORK_QUEUE_1 doesn't exist because WORK_QUEUE is
> a INTEGER PRIMARY KEY AUTOINCREMENT table */

Do we need to delete the EXTERNALS rows first? That is what we do for
the other tables, a deliberate change introduced by r1571214. Given
that we do delete the rows first perhaps all the INSERT OR REPLACE
should be changed to just INSERT?

--
Philip Martin | Subversion Committer
WANdisco // *Non-Stop Data*
Received on 2015-04-14 19:13:36 CEST

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.