[Biopython-dev] Re: [BioPython] tests failing

[Brad Chapman]

Jeff:
> I think this may be a problem in test_prodoc.py rather than the 
> regression testing framework.  This output is generated in a function 
> called print_references:
[...]
> It prints some text out 80 characters at a time.  Perhaps this 
> boundary is falling on different characters depending on the OS' line 
> breaking convention.  To make things more difficult, the text file 
> itself has different types of line breaks.
[...]
> I don't have a reproducible here, so could someone with a Windows 
> machine take a look at it?

I just tested it out (after much swearing while attempting to get CVS
working on Windows. Grrrrrr...) and it turns out, as I've always
suspected, that you are a genius. No more failing for test_prodoc, 
yippeee! Excellent deduction, Jeff.

My current test status is:

==> Windows 98 w/ Python 2.1

test_MultiProc is failing with a complaint about os.fork not existing
on Windows. I guess there is not much we can do about this.

test_GenBank is failing with a parse error. I'll investigate this
further once I manage to get CVS working properly.

test_SubsMat is failing with the -0.00 and 0.00 thing.

==> UNIX (my NetBSD machine) w/ Python 2.1

test_SubsMat is failing with the -0.00 and 0.00 thing.

test_interpro was failing with:

IOError: [Errno 2] No such file or directory: 'InterPro/ipr001064.htm'

which seems to be due to the fact the test files are named
IPR001064.htm (ie. capital IPR). This probably didn't fail on Windows,
but does on my machine. I just checked in a fix for this to
test_interpro, so it's taken care of.


So that's where I'm at. Thanks again Jeff for the prodoc fix!
Brad