[Bioperl-l] bioperl-db with bioperl 1.0a

[Keith Allen]

Elia, I'm not prepared to admit what my bozo-level error
was, except to say it was pretty dumb, and you're on the
right track.

Once over that small hurdle the bioperl1.0a code working
together with the latest bioperl-db code successfully processed
all of the human data from EMBL.  It took a good eight hours
on a 450MHz PIII, but still, guys, this is outstanding!

In working through 3.6 gigs of text, the parsers hit a grand
total of four instances of unbalanced quotes (all of these were
occurances of "" -- ie, two double quotes next to each other
in a place where one double quote would be expected).  Other
than that there are 5 warnings where the start and stop numbers
do not make sense, and the parser sets the strand to -1.   Here's
an example of this warning:

-------------------- WARNING ---------------------
MSG: When building a location start (37807) is expected to be less than
end (34808), however it was not was not. Switching start and end and
setting strand to -1
---------------------------------------------------

So that's a grand total of nine minor bobbles in processing 3.6
gigs of error prone text, using the 1.0a code.  (we can ignore the
fact that the error mesage seems to be channeling the eighties
dance band "was not was").

Am I the only one here who thinks that this is truly outstanding
work?  Do you guys not rock?


-Keith Allen