#1 2011-02-18 06:36:24

pajdus
Member
Registered: 2007-03-08
Posts: 22

New book plugin (Czech National Library)

Initial version of book plugin parsing http://aleph.nkp.cz (Czech National Library database only).

Though not yet tested much. It now parses 24 fields but there is probably more for books I didn't test. If someone wants to test it, I would appreciate any feedback (which books contain fields that are not parsed correctly or fields that are not parsed at all because I do not know abut them tongue).

Known issues/bugs: 
* Single list fields: values are stored as comma delimited string as the plugin hangs otherwise. I have serious problem with strings containing diacritics - I cannot store them as array items into single/double list without plugin hang.  Storing values as comma delimited string works but suffers from unwanted splitting items that already contain commas (i.e. authors). I didn't find out how to escape them so I replaced them with --. sad ... This is ugly.
* Czech translation of some fields in model is probably wrong (edition vs. náklad, format vs. rozměr).
* While searching for the title, results are limited to the first page as I do not know how to handle more pages. The pages count is not fix so I cannot probably use getnumberpasses and nextUrl as it seems to be dedicated only for fix number of passes. But I din't try.
* Most of parsed fields are not supported by the model, so I stored them in the description field. Though they are ready if the model will support them. I want to create a new model designated for rather complete library informations.
* Link button to the source page and maybe refresh will probably not work flawlessly as the session may expire, I must check this again.
* No images.  I will make a plugin for www.obalkyknih.cz that is simple parseable and contains images and TOC for books. So user will need to use both of these  plugins and combine results per fields or I will update this plugin to make two passes (second on www.obalkyknih.cz) -- but I don't know if it is good idea.

TODOs:
* Make enhanced model
* Parse MARC records instead of standard HTML view.  This would be great. Maybe I can use perl MARC library.
* http://aleph.nkp.cz contains more database, i.e. comprehensive catalog

Cheers,
Petr Gajdůšek

Last edited by pajdus (2011-02-18 07:26:17)


Attachments:
Attachment Icon GCNKC.pm, Size: 22,366 bytes, Downloads: 353

Offline

 

#2 2011-04-12 01:18:49

Halibutt
Member
From: Warsaw
Registered: 2007-05-12
Posts: 19
Website

Re: New book plugin (Czech National Library)

On a friendly note, I digitalised my entire book collection (mostly in Polish) using two plugins: one for merlin.pl and one for the Polish National Library (sadly both of them do not work any more, but anyway). During the tedious work of downloading all the info from national library and then the pics from the online bookstore, I always wondered why wasn't it possible to combine the two: just parse info from one site and pics from the other.
Regards


Why oh why is writing plugins for GCStar so complicated...

Offline

 

#3 2013-01-22 01:34:17

dr.micro
New member
Registered: 2013-01-22
Posts: 1

Re: New book plugin (Czech National Library)

I am still getiing the below error message when trying to download GCNKC.pm, although I have registered to this forum:

Info
Bad request. The link you followed is incorrect or outdated.
Go back

Last edited by dr.micro (2013-01-22 02:22:15)

Offline

 



Should you have a problem using GCstar, you can open a bug report or request some support on GCstar forums.