[Biophp-dev] Fasta filetype parser updated

[S Clark]

On Tuesday 06 May 2003 09:44 pm, Serge Gregorio wrote:
> Pretty impressive work... btw, if I don't appear to be writing
> too much, it's because of the difference in our time zomes...
> By the time I read your postings, questions, etc. you've most
> likely resolved or answered the issues already...  =)

I figured being several timezones away from the timezones we're in over here
was having an effect.  Obviously the solution is that you should sleep during
the day and deal with the BioPHP email during the night, right? I'm sure any
reasonable employer won't mind accomodating that. :-)

> However, just so you know, I've made a mental note of Sean's suggestion of
> having "aliases" for some of the attributes or properties of
> classes/objects (e.g. id = label = name).  I
> realize the dilemma of making the user memorize all of them.
> However, I'm holding my comments till I've formed an opinion
> on it.

I wouldn't worry too much about having them in the core seq object - I was 
thinking more in terms of having the seq_factory do some simple "translation" 
for some of the terms that might have other common synonyms or 
abbreviations (to the "official" names in the seq object).  When working
directly with seq objects, some "easily remembered" interface methods for the
attributes should be all that anyone really needs, or at least, that's my
opinion.

I would advocate having an "annotations" or "otherAttributes" (or whatever
name) array for putting less-commonly-used (or unique) attributes, as a 
place to store odd information that may come up in special applications,
though.

> My WinCVS still isn't set up for bioinfo.org... in the mean-
> time, I notice my code is still without the "banner" Andres
> was mentioning...
>
>    /* **********************************************
>    seqdb.inc.php
>
>    Description - etc.
>    Written by: Serge Gregorio
>    Date: March/April 2003
>    ********************************************** */
>
> Could someone (Sean?) do something about this?  I don't know
> what needs to be done, if it means recreating an existing
> directory, etc. but it seems only proper to do so.  I can
> send the *.inc.php files with the "banners" tomorrow.

I'm beginning to wonder how hard it is to set up webdav, and maybe
a cron job to update CVS from a webdav-accessible copy of the CVS structure.

(Not sure how much it'd help, but if CVS for windows isn't working it
may be an alternative.)