[Biopython-dev] [BioPython] Bio.Medline parser

[Peter]

On the main mailing list Michiel and I wrote:

>> That sounds sensible.  Maybe we should have an example
>> in the Tutorial of using Bio.Entrez to download some data
>> in the plain text MedLine format, and parsing it with
>> Bio.MedLine?  And perhaps also an equivalent using the
>> XML Medline format parsed using Bio.Entrez?
>
> Done -- see CVS.

Great - I've updated CVS slightly (missing line break in one example,
and I expanded on the section introduction to mention the rettype and
retmode paramters for efetch explicitly).

There is an older PubMed/Medline example in the cookbook section which
uses the Bio.PubMed.Dictionary class.  The Bio.PubMed.Dictionary
expects the old (now deprecated) parser from Bio.Medline, and also as
we discussed last month, these dictionary interfaces to the NCBI do
encourage the user to make a series of separate calls to Entrez
(rather than doing linked a search/retreive with the web-history).
Perhaps Bio.PubMed.Dictionary should also be deprecated?  Indeed could
the whole of Bio.PubMed be deprecated in favour of Bio.Entrez?

The Bio.GenBank.NCBIDictionary class is a similar case, but here the
underlying GenBank parser framework is still present.

Peter