Re: Status bibref gatherer (Was: Tasks pages (close to) fixed; Bibref does not seem to be updated automatically)
On Fri, Mar 09, 2012 at 09:06:14AM +0900, Charles Plessy wrote:
> On my side I made a first push to the collab-qa Subversion repository.
>
> http://anonscm.debian.org/viewvc/collab-qa/packages-metadata/
Great.
> Still, I am disappointed that you ignore the rest of my work. I spent a lot of
> days for writing it, and took care of writing documentation. Let's talk about
> problems and solve them instead of writing competing implementations. I
> understand well the shortcomings you pointed in my work, but for the moment
> your approach does not seem to solve them either.
Ups, seems there is some misscommunication? I simply assumed it would
have been the agreement that you would provide the upstream files that
way and I volunteered to parse them for UDD. Could you please be more
verbose about what I did ignored? Regarding solving a problem I think
we are very close now to be able to strip the references from the tasks
files. Moreover I consider exporting a BibTeX file of all those
references as a by-product of the parser (I'm even now creating a
bibtexkey which equals to package name if there is only one reference
and is prepended by year for more than one citation) and push it to UDD
as well to enable flexible generation of BibTeX entries by querying UDD.
So I think we are close to some helpful solutions thanks to your effort
to implement the upstream files in the first place.
> For adding a rank to the references, this may be a good idea, but how do you
> propose to implement this in YAML ?
We have an implementation in disulfinder. The current debian/upstream
file will end up in UDD like this:
udd=# SELECT * from bibref_new where package='disulfinder' ;
package | key | value | rank
-------------+---------+-------------------------------------------------------------------------------------------------+------
disulfinder | title | DISULFIND: a disulfide bonding state and cysteine connectivity prediction server. | 0
disulfinder | url | http://www.ncbi.nlm.nih.gov/pubmed/16844986 | 0
disulfinder | journal | Nucleic Acids Res | 0
disulfinder | author | A. Ceroni and A. Passerini and A. Vullo and P. Frasconi | 0
disulfinder | number | Web Server issue | 0
disulfinder | volume | 34 | 0
disulfinder | year | 2006 | 0
disulfinder | pmid | 16844986 | 0
disulfinder | pages | W177-81 | 0
disulfinder | bibtex | disulfinder | 0
disulfinder | comment | For the disulphide connectivity predictor | 1
disulfinder | title | Disulfide connectivity prediction using recursive neural networks and evolutionary information. | 1
disulfinder | journal | Bioinformatics | 1
disulfinder | author | Vullo, Alessandro, and Frasconi, Paolo | 1
disulfinder | number | 5 | 1
disulfinder | month | March | 1
disulfinder | volume | 20 | 1
disulfinder | year | 2004 | 1
disulfinder | pmid | 15033872 | 1
disulfinder | pages | 653-9 | 1
disulfinder | bibtex | disulfinder2004 | 1
disulfinder | comment | For the cystein bonding state predictor | 2
disulfinder | volume | 35 | 2
disulfinder | year | 2003 | 2
disulfinder | title | Predicting the Disulfide Bonding State of Cysteines with Combinations of Kernel Machines | 2
disulfinder | journal | Journal of VLSI Signal Processing | 2
disulfinder | author | A. Ceroni, P. Frasconi, A. Passerini and A. Vullo | 2
disulfinder | pages | 287-295 | 2
disulfinder | bibtex | disulfinder2003 | 2
disulfinder | comment | For the cystein bonding state predictor | 3
disulfinder | year | 2002 | 3
disulfinder | title | A Two-Stage SVM Architecture for Predicting the Disulfide Bonding State of Cysteines | 3
disulfinder | journal | Proc. IEEE Workshop on Neural Networks for Signal Processing | 3
disulfinder | author | P. Frasconi, A. Passerini and A. Vullo | 3
disulfinder | pages | 25-34 | 3
disulfinder | bibtex | disulfinder2002 | 3
The Python yaml parser returns a list of dictionaries which can be easily parsed.
> Perhaps it would be simpler to keep
> a single reference broken in YAML fields as it is now, and dump the rest in
> BiBTeX format in a separate field ?
In this case we would need to parse a BibTeX expression in addition to
YAML which is more complicated (and not yet solved.
I'm afraid I'll have not much time over the weekend and next week might
also be a bit dense. I hope to get the upstream content straight to the
tasks pages after Chemnitzer Linuxtage (=10 days).
Kind regards and thanks for all your hard work on gathering upstream data
Andreas.
--
http://fam-tille.de
Reply to: