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

Re: Requiring python sqlite3 module for devs

From: Hyrum K. Wright <hyrum_at_hyrumwright.org>
Date: Wed, 30 Sep 2009 12:18:25 -0400

On Sep 30, 2009, at 11:54 AM, Mark Phippard wrote:

> On Wed, Sep 30, 2009 at 10:49 AM, Hyrum K. Wright <hyrum_at_hyrumwright.org
> > wrote:
>> We currently have a couple of tests which conditionally run if the
>> sqlite3 python module is found. As we move more working copy stuff
>> into sqlite, the number of tests that will be needed to be skipped
>> will increase, and the transparency of *why* they need to be skipped
>> will decrease. For this reason, I'd like to propose that we require
>> the python sqlite3 module be installed when running the tests.
>> The sqlite3 modules ships as part of Python 2.5, but we only require
>> Python 2.4. Most distributions which currently ship Python 2.4
>> either
>> ship the sqlite3 module, or provide a way to trivially install it.
>> For Windows users using Python 2.4 (such as the buildbot), binaries
>> can be obtained from [1].
>> Any objections to requiring the python sqlite3 module for running
>> tests?
> Given how essential SQLite is in WC-NG I cannot see anything but a +1.
> In terms of anticipating problems, do you know if the Python module
> will pick up the SQLite library we build with? If I get Python from
> my distribution, it seems likely the module might pick up the library
> from my distribution and not use the amalgamation or whatever version
> I am building with. Given that we need a fairly new SQLite in SVN,
> this seems like a potential problem area.
> That said, maybe the way Python uses SQLite it does not matter if an
> old library is used?

Shouldn't matter[1], for a couple of reasons: the python and
Subversion process spaces are completely distinct, and therefore their
symbols shouldn't clobber each other. Also, the disk formats are
fully backwards compatible, so there shouldn't be an issue there.


[1] Of course, this is all theory, and I'm pretty sure it's sound, but
I haven't gone out of my way to verify that an old version of py-
sqlite and a new version of the sqlite-amalgamation compiled into
Subversion don't cause problems. However, I trust that the sqlite
devs posses sufficient paranoia that it won't.

Received on 2009-09-30 18:18:45 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.