vila, it's reverse: pycurl of version 7.19.0 cause more problems then pycurl 7.16.4 (with the same curl version installed). And speaking about 1.9 (full output in attachment) it looks like same tests fail (even more, but I suppose some bugs are fixed and some tests were added):
About disabling tests, I mean, I need selectively disable tests which are known to fail. This helps me to work with our arch teams since they'll avoid test which are known to fail. And well, that's said, I'll move back to older py/curl to see how tests pass there...
vila, it's reverse: pycurl of version 7.19.0 cause more problems then pycurl 7.16.4 (with the same curl version installed). And speaking about 1.9 (full output in attachment) it looks like same tests fail (even more, but I suppose some bugs are fixed and some tests were added):
ERROR: blackbox. test_check. ChrootedCheckTe sts.test_ check_missing_ repository test_outside_ wt.TestOutsideW T.test_ url_log implementations .test_branch. ChrootedTests. test_open_ containing( BzrBranchFormat 5) implementations .test_branch. ChrootedTests. test_open_ containing( RemoteBranchFor mat-default) ChrootedTests. test_open_ containing ChrootedTests. test_open_ containing_ from_transport ChrootedTests. test_open_ containing_ tree_branch_ or_repository_ empty TestWallServer. test_http_ get(pycurl, HTTP/1. 1) TestWallServer. test_http_ has(pycurl, HTTP/1. 1)
ERROR: blackbox.
ERROR: branch_
ERROR: branch_
ERROR: test_bzrdir.
ERROR: test_bzrdir.
ERROR: test_bzrdir.
ERROR: test_http.
ERROR: test_http.
About disabling tests, I mean, I need selectively disable tests which are known to fail. This helps me to work with our arch teams since they'll avoid test which are known to fail. And well, that's said, I'll move back to older py/curl to see how tests pass there...