[Bioperl-l] Re: Pipeline Input/Output refactoring plan

[Elia Stupka]

> Think so yes.  I'm still not clear what the difference is between the
> db_locator and the biodbname.

db_locator simply points to a mysql database biodbname (in databases like
compara and biosql) says which biological database to pull out,
e.g. genbank, ensembl human proteins (in compara), etc.

> I'm not convinced here that you gain flexibility - you still have to have
> code to fetch stuff somewhere.  Your way you have to specify the method
> name as well as the module name and in our way we just have one module per
> method name.   I'll go away and muse about this some more.

The main reason behind this was that unless you specify a method name you
end up having to make sure that all possible inputs in the world will
respect a certain interface (e.g. fetch_by_DbID). Adding this column means
we can be using get_Seq_by_id from a RandomAccessI object as well as
fetch_family_by_description as well as fetch_by_DbID.... oh, also you
could use things like fetch_above_score, and as long as your input ids are
fine it would work....

Elia

-- 
********************************
* http://www.fugu-sg.org/~elia *
* tel:    +65 874 1467         *
* mobile: +65 90307613         *
* fax:    +65 777 0402         *
********************************