[Bioperl-l] Looking good.

[Jason Stajich]

On Wed, 9 Jan 2002, Ewan Birney wrote:

>
> Things to do:
>
>
> Code
> ----
>
> (1) Jason - are we all smoothed off for the SearchIO system? Does anything
> else needed to be added?
>

Steve's psiblast does run and does the lazy parsing okay.  I would like
to add some more tests here but that can be done pre 1.0 but not
necessarily for 0.9.3.  Will want to merge psiblast.pm and blast.pm into a
single module before 1.0 though and probably remove Aaron's
Bio/Search/Processor directory and associated files.


> (2) Is there anything outstanding to go in? Kris - you had some check-ins
> you wanted to put into for PDB?
>

What about merging in the STRIDE stuff with Kris's stuff - I don't think
we want those to be part of the stable code a with a separate API?

>
>
> (3) What do we do about semi-defunct modules?
>
>   Bio/Tools/Blast.pm
>
> and
>
>   Bio/Tools/BPlite.pm
>
> and (potentialy)
>
>   Bio/Tools/HMMER.pm
>
>
> What I would like to suggest is that we have the following code in the
> "new" functions of these methods
>
>
> (In Bio::Root::RootI.pm ---
>
>     $WARN_ABOUT_DEPRECATED_MODULES = 0;
>
>
>
>  In the modules 'new'
>
>
>   if( $WARN_ABOUT_DEPRECATED_MODULES ) {
>      $self->deprecated("XXXXX has been deprecated for the new
> Bio::SearchIO framework. Please investigating porting to the new system");
>    }
>
> )
>

Yes this seems like the best way to go.

>
> The big global variable is so that people can move from 0.7.2 to 1.0 and
> then set that flag to 1 in their local copy to prevent endless painful
> messages arriving on their screen --- but they do have to actively switch
> off the warnings.
>
>
> Later on in the 1.0 series we might get more frantic.
>
>

Some small stuff includes integrating Boulder::Genbank's nice batch entrez
interface into Bio::DB::GenBank.  Roger had mentioned he would look into
it, but haven't heard anything.  This should just hopefully be a port of
raw socket communication to LWP and some queuing up of requests that are
too long.  It should all be done in NCBIHelper/WebDBSeqI, I guess I can
try and implement if you don't think it can get done in the next week
or two Roger.

> Testing
> -------
>
>
> We need broader testing against existing code bases. I will start to drum
> up larger test sites...
>

I've contacted a couple of groups who will undertake some testing for us.
Do we need to expand the test suite itsself in any way?  We probably need
to prune some things (WWW.t, HTML.t)

>
>
> Documentation
> -------------
>
> (1) POD stuff. Most of Heikki's POD errors are SteveC's modules - Steve -
> have you been using the wrong POD syntax for something? Any views?
>
>
> (2) Documentation on modules: We want every synopsis to be
> cut-and-paste-able as much as possible. Does anyone want to try that out?
> Volunteers? (go on...)
>
>
> (3) Top level documentation: We need to overhaul bioperl.pod,
> biostart.pod, bptutorial.pl, README to all work together. Peter I assumme
> will be doing bptutorial - Peter - do you still have time? Does anyone
> want to work on the other modules?
>
>
>
> External code
> -------------
>
>
> Someone should read all the examples/ code and check it is up to
> date (and at the very least, works). Again - any volunteers?
>
>
>
>
> This is one of those moments when lots of people can help (in particular
> with documentation!). So... it is time not to ask what bioperl can do for
> you, but what you can do for bioperl...
>
>
>
>
> ewan
>
>
>
>
> -----------------------------------------------------------------
> Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
> <birney@ebi.ac.uk>.
> -----------------------------------------------------------------
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
>

-- 
Jason Stajich
Duke University
jason@cgt.mc.duke.edu