|
|
Boost Testing : |
From: John Maddock (john_at_[hidden])
Date: 2005-04-10 04:33:09
> And why exactly do you think it has something to do with our
> "postprocessing"? The page clearly states that report has been
> generated on "Sat, 09 Apr 2005 00:27:16 +0000", while the results'
> time for compiler output is "2005-02-04 15:01:16 UTC". It takes
> exactly 3 minutes to download and unpack the latest Martin's XML
> from "ftp://fx.meta-comm.com/boost-regression/CVS-HEAD/" and see for
> yourself that the output shown on the results page is precisely the
> output recorded in the XML file Martin uploads to us.
Apologies if my last post was intemperate, what I'm really trying to say is:
We have tests that are known to pass that are shown as failing (cause
unknown).
We have tests that are known to fail that are shown as passing (cause
unknown).
Until we can track down the cause, it's hard to have confidence in the
currently displayed results.
Now: if the XML that Martin was uploading was at fault, then that helps a
little - but - and here's the thing, aren't the old style results and the
new style ones both generated from the same XML ?
How about the python script that collects the XML for upload, is that a
possible culprit?
Victor: are you able to check to see if the XML uploaded matched what you
expected for the range test program failures (your current results show them
as passing today, so I'm assuming that the source has been fixed, making the
issue disappear for now).
Just throwing ideas around in the hope that it clicks with someone...
Thanks