[BioPython] QIO

[Andrew Dalke]

A few weeks ago I mentioned that one of the problems with Python
compared to Perl for bioinformatics analysis was the slower
performance when reading data files

The original email from me is:
http://www.biopython.org/pipermail/biopython/1999-September/000004.html

In the last few weeks on the python newsgroup there was a discussion
of how to fix that problem.  Someone (g.j. andruk <gerglery@usa.net>)
took the I/O parsing from INN and converted it for use as a Python
extension

http://members.xoom.com/_XOOM/meowing/python/index.html

I tried compiling on my machine, but it seems to require Python
1.5.2, or at least more time than I wanted to put into it for now.
Tim Peters (if I can't benchmark I can at least get numbers from
the same source) said:
http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=527209193&fmt=text

> Under Win95 it was between 2-3 times as fast as 1.5.2 Python doing
> a native readline() loop over an 11Mb text file.

Since his original statement was
> Line-at-a-time input on my platform is about 3 times faster in
> Perl than Python; 

then Perl and Python should be neck and neck for reading large
data files when the run-time is read bound.

Cool!

						Andrew
						dalke@acm.org