Stefan Seefeld wrote:
Roland Schwarz wrote:
Arkadiy Vertleyb wrote:
What happened? I thought we were down to 6 failures at some point... Are all these new failures just indication of a problem in some low-level library?
I am not absolutely sure, but as far as I understood the errors were always present, but masked due to a bug in the process_jam_log scripts.
What exactly does 'always' mean here ? Are we sure those newly exposed errors really are regressions ?
They are not. I *think* was happened is that process_jam_log failed to include some parts in XML and reporting scripts did not care. Then, they started to care, producing huge number of errors which did not corresponded to real errors. When process_jam_log was fixed. Then, for as yet unknown reason, the number of reported errors did not went down. One problem might be: http://article.gmane.org/gmane.comp.lib.boost.testing/3895 Another problem is: http://article.gmane.org/gmane.comp.lib.boost.testing/3902 Yet another one problem might be that new test runners don't have things configured completely correctly yet. - Volodya