|
Boost-Build : |
Subject: Re: [Boost-build] meaning of "rev" in test report
From: Rene Rivera (grafikrobot_at_[hidden])
Date: 2016-10-27 11:46:02
On Thu, Oct 27, 2016 at 10:37 AM, Stefan Seefeld <stefan_at_[hidden]>
wrote:
> the test report matrix reports a "rev" attribute for each test run,
> presumably showing a git SHA. However, I'm somewhat doubtful about what
> repository that is referring to. For example I'd expect the revs in
> http://www.boost.org/development/tests/develop/developer/python.html to
> correspond to SHAs in the https://github.com/boostorg/python repo.
> However, I can't seem to find a single match, i.e. none of the SHAs I
> see in the matrix appears in my git log.
>
> If that's not the repo whose revision is being reported, which repo is
> being used ? And how can I map that to the revision of the module being
> tested (Boost.Python in this case) ?
>
> (The reason I'm asking is that now that the NumPy tests trickle in, I
> see a bunch of failures, but looking at the error I notice an issue I
> fixed a week (or so) ago, yet the test runs are more recent than that. I
> think it's crucial for the test report to give a clear indication of the
> tested module's SHA to make it possible to attempt to reproduce the
> tests locally.
The main rev you see in the reports is the SHA of the superproject <
https://github.com/boostorg/boost>. If you look at the info for each tester
there's a list of each of the submodules with a SHA of the submodule
version that was pulled in that superproject rev. For example in <
http://www.boost.org/development/tests/develop/teeks99-09e-win2012R2-64on64.html>
scroll down past the tester env info and you'll see a table of "Repo" and
"Revision" entries like this:
RepoRevision
\boost_bb
154667bb312be0cb62c9642689c566606eb7ed05
\boost_regression
35d42b5d50dff7008374807aa136ddf26a97a2fd
\boost_root
ee9cddd7ea9ada871bea6de6c116e41c83c4654a
\boost_root\libs\accumulators
70ec4c602be9d014c7246eb8ce50a6cb81f45885
The first entry if the Boost Build version used to drive the testing. The
second is the regression tools used. The third is the same as the one
reported in the main result pages (ie the superproject). And the rest are
the ones for submodule.
HTH.
-- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net -- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk