[BioPython] Problems with Bio.Application.generic_run()

[Brad Chapman]

Hi Steffi;

> using the Bio.Application package, I got some trouble.
> 
> When using generic_run() to run blast, which in this case produced a 
> --very-- large output and several warnings, the blast process fell asleep.
> Running this blast as standalone, there were no problems.

Hmmm. This is definitely a problem, and not something I know much at
all about. Thanks for reporting it.

> Looking around in the Python Reference etc. I found that there is 
> possibly a dead lock of parent and child processes.
> 
> For now I solved this problem using a recipe from the Python Cookbook as 
> follows:
[...]
> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52296

Thanks for looking into this. I definitely agree that we need a
solution to this type of problem. However, looking at the solution,
the following line looks very UNIX-specific:

>   errorlevel = os.system("( %s ) > %s 2> %s" % 
> (str(commandline),outfile,errfile)) >> 8

I know next to nothing about Windows and the DOS commandline, but I
don't know if it has redirects like the UNIX ones. Am I true in
thinking this? Can anyone that knows anything about Windows help me
out?

So yeah, I'd like to have a solution that isn't UNIX-specific if
that's possible. Does the same BLAST lock up when run using
'blastall' in Bio.Blast.NCBIStandalone? It looks like Jeff and I run
the programs in the same way, but one never knows.

I'm not sure if this rambling reply helped much. Yes, so I'd like to
fix it but am worried about your solution being UNIX-specific.

Hope this made some sense!
Brad