[Biojava-l] deprecating some SymbolList methods

David Waring dwaring@u.washington.edu
Mon, 15 Jul 2002 17:49:42 -0700

I would be strongly against it. Deprecating methods will always break
people's code. I'm sure it would break some of mine. I find that I am often
going between a String and a SymbolList since 1) I work with other peoples
non-biojavaized code 2) I work with a database that stores its sequence as
varchars or clobs - getting my sequence in requires putting it into a String
3) Sometimes I want to output my sequence in a form that is readable, and I
find that a String is a generally useful means of displaying it ;)

I'm not sure why you would want to deprecate methods that are as obvious to
have on a SymbolList.  I agree that the Tokenization code is a great way to
handle the intricacies of multiple alphabets. Yes it is possible to use this
whenever you want to get a String, after all that is what AbstractSymbolList
does! I would say that seqString() is as obviously necessary to an object
representing a DNA or protein sequence, as toString() is to an Object.


> -----Original Message-----
> From: biojava-l-admin@biojava.org [mailto:biojava-l-admin@biojava.org]On
> Behalf Of Matthew Pocock
> Sent: Monday, July 15, 2002 8:53 AM
> To: biojava-l@biojava.org
> Subject: [Biojava-l] deprecating some SymbolList methods
> Hi All,
> I hope you are all enjoying your summer holidays. I look forward to
> seeing some of you at ISMB soon.
> I was wondering if anybody would object if the subStr and seqString
> methods of SymbolList were deprecated with a message saying that you
> should be using Thomas' excelent symbol tokenization code? I'm more than
>   willing to go with popular opinion on this one.
> Matthew
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l@biojava.org
> http://biojava.org/mailman/listinfo/biojava-l