I'm trying to learn about buildbot and the pybots project
(http://www.pybots.org/). I decided to try and set up a buildbot slave for
sqlalchemy on my Mac OSX G5. With a little effort I got pysqlite built on
the G5 and ran sqlalchemy's test suite. I got this one failure:
FAIL: testforupdate
this is a test.
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
Time to move to Google Groups? :)
On 10/3/06, Michael Bayer [EMAIL PROTECTED] wrote:
test
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
soon. im waiting for SF to fix this one, its taken them awhile to
realize theres a problem. two things i dont like about google
groups is that i cant have an archive of the existing SA mail
imported into it, and also im not sure if it has much of a method to
get an export of everything
test received
On 10/2/06, mike bayer [EMAIL PROTECTED] wrote:
this is a test.
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions
Hello,
Michael Bayer said the following on 05.10.2006 1:11:
soon. im waiting for SF to fix this one, its taken them awhile to
realize theres a problem. two things i dont like about google
groups is that i cant have an archive of the existing SA mail
imported into it,
Yes, but if
test
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT business topics through brief surveys -- and earn cash
It appears that the test suite depends on pysqlite2. Is this a hard
requirement? I only have postgres here and I'd like to be able to run
the test suite without modifying it. This is what I get when I try to
run the tests using a fresh checkout from SVN:
$ python test/alltests.py --db
all of the offending tests, if you add a decorator above them which
says:
@testbase.supported('sqlite')
it should cause those tests only to be run with --db sqlite or
default, not --db postgres. can you give that a try ?
also, with sqlsoup, i had a stub module in test/ext/ to
On Aug 21, 2006, at 12:43 PM, Michael Bayer wrote:
all of the offending tests, if you add a decorator above them which
says:
@testbase.supported('sqlite')
it should cause those tests only to be run with --db sqlite or
default, not --db postgres. can you give that a try ?
Sure,
what version of postgres are you running ?
Those two reflection tests are doing something tricky; they create
some tables, then under a new MetaData reflect them from the pg
system catalog. then they drop the tables, and try to re-create them
using the reflected table information. in
On Aug 21, 2006, at 2:54 PM, Michael Bayer wrote:
what version of postgres are you running ?
8.1.2
Those two reflection tests are doing something tricky; they create
some tables, then under a new MetaData reflect them from the pg
system catalog. then they drop the tables, and try to
also, i committed the tests you patched in 1814 except I just changed
pool.py to use a MockDBAPI() object instead of needing pysqlite at
all (but your patch served as inspiration so your efforts were not
unnoticed !).
On Aug 21, 2006, at 2:08 PM, dmiller wrote:
On Aug 21, 2006, at 12:43
patch is applied in rev 1691. also includes a workaround for the specific
issue with sqlite (im guessing its sqlite, but it could be pysqlite),
which I have posted to the sqlite site at:
http://www.sqlite.org/cvstrac/tktview?tn=1884
all unit tests now pass on my windows machine using pysqlite
Thanks!On 7/9/06, Michael Bayer [EMAIL PROTECTED] wrote:
patch is applied in rev 1691.also includes a workaround for the specificissue with sqlite (im guessing its sqlite, but it could be pysqlite),which I have posted to the sqlite site at:
http://www.sqlite.org/cvstrac/tktview?tn=1884all unit
pysqlite 2.3.2, python 2.3, whatever version of sqlite is bundled with their windows binary.I think the pysqlite version I'm using with python 2.4 is 2.2.2 but I'm not 100% sure on that.The bug happens with both versions.
-JonathanOn 7/8/06, Michael Bayer [EMAIL PROTECTED] wrote:
thats weird.what
so im hearing !it just takes one bad test to throw the whole thing off.in other news, why are so many python dev's running windows ? linux is free/better ?On Jul 7, 2006, at 3:53 PM, Jonathan Ellis wrote:From a brand-new checkout:Ran 302 tests in 5.844sFAILED (failures=11, errors=148)I take it
I haven't used a desktop as my primary machine for years, and linux laptop support has historically been sucky.I hear Ubuntu is changing that, but I don't get any thrill from being on the bleeding edge of OS adoption anymore.
On 7/7/06, Michael Bayer [EMAIL PROTECTED] wrote:
so im hearing !it just
I'm starting to think there are real problems on windows with sqlite.I fixed one that caused a test failure in proxy_engine; patch is attached.But I haven't been able to figure out this other one from cascade.py
. What is happening is that after the first test (the last test in the code, but the
test
___
Sqlalchemy-users mailing list
Sqlalchemy-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sqlalchemy-users
20 matches
Mail list logo
