Releasing solrpy 0.9.4?

83 views
Skip to first unread message

Friedrich Lindenberg

unread,
Feb 11, 2011, 8:40:33 AM2/11/11
to sol...@googlegroups.com
Hi all,

over the last months, I have submitted two changesets to solrpy [1].
The first one allows multiple values for the same query parameter, in
order to submit things like facet.field=tags&facet.field=creator - a
prerequisite for really using faceting and similar advanced functions
of solr. The other changeset addresses the handling of unicode in
queries - which was erroneous before (i.e. solrpy 0.9.3 doesn't do
non-ascii queries).

I think these two would merit the release of a new version to pypi -
also because it would be benficial to the release process of apps
based on this (in my case, CKAN, WhereDoesMyMoneyGo, Adhocracy and
publicdata.eu). So - whats the procedure? And would it be possible to
get involved?

Best regards,

Friedrich

[1] http://code.google.com/p/solrpy/source/list

Ed Summers

unread,
Feb 12, 2011, 6:12:20 PM2/12/11
to sol...@googlegroups.com
Hi Frederich,

Sorry for the lag in pushing out these changes. Do you see this when
running the test suite?

======================================================================
ERROR: Test for unicode support in subsequent paginator queries
----------------------------------------------------------------------
Traceback (most recent call last):
File "/home/ed/Projects/solrpy/tests/test_all.py", line 1388, in
test_unicode_query
result = self.query(self.conn, chinese_data.encode('utf-8'))
File "/home/ed/Projects/solrpy/tests/test_all.py", line 68, in query
return conn.query(*args, **params)
File "/home/ed/Projects/solrpy/solr/core.py", line 703, in query
return self.select(*args, **params)
File "/home/ed/Projects/solrpy/solr/core.py", line 797, in __call__
xml = self.raw(**params)
File "/home/ed/Projects/solrpy/solr/core.py", line 815, in raw
query.append((key, to_str(value)))
File "/home/ed/Projects/solrpy/solr/core.py", line 809, in <lambda>
to_str = lambda s: unicode(s).encode('utf-8')
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe6 in position
0: ordinal not in range(128)

//Ed

> --
> You received this message because you are subscribed to the Google Groups "solrpy" group.
> To post to this group, send email to sol...@googlegroups.com.
> To unsubscribe from this group, send email to solrpy+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/solrpy?hl=en.
>
>

Friedrich Lindenberg

unread,
Feb 15, 2011, 10:22:31 AM2/15/11
to sol...@googlegroups.com
Hi Ed,

On Sun, Feb 13, 2011 at 12:12 AM, Ed Summers <e...@pobox.com> wrote:
> Sorry for the lag in pushing out these changes. Do you see this when
> running the test suite?

terribly sorry about the test breakage, I've now submitted a patch
that fixes it.

Thanks,

Friedrich

Ed Summers

unread,
Feb 15, 2011, 10:49:21 AM2/15/11
to sol...@googlegroups.com
Hey, no worries. I'm not convinced that your commit actually caused
that test to fail ... it appeared to be present even after rolling
back your recent change. So thanks for the fix!

I just pushed v0.9.4 up to PyPI.

//Ed

Friedrich Lindenberg

unread,
Feb 15, 2011, 1:29:18 PM2/15/11
to sol...@googlegroups.com
Thanks for this, it's very helpful to us!

- Friedrich

Reply all
Reply to author
Forward
0 new messages