CEUR-WS metadata graph/tree procurement

[WolfgangFahl]

Input:

• index.html
• per Volume html
• partly PDF

see e.g. https://github.com/WolfgangFahl/ProceedingsTitleParser/blob/master/ptp/ceurws.py

[WolfgangFahl]

see

• http://rdf.greggkellogg.net/distiller?command=serialize&url=http:%2F%2Fceur-ws.org%2FVol-2549%2F
• https://github.com/alexstolz/rdflib-rdfa-serializer

[WolfgangFahl]

see https://github.com/WolfgangFahl/ProceedingsTitleParser/blob/4284bc33a29479eba6332e02c7108176425dadc1/ptp/ceurws.py#L92
and the rdfA parser code in
https://github.com/WolfgangFahl/ProceedingsTitleParser/blob/4284bc33a29479eba6332e02c7108176425dadc1/ptp/webscrape.py#L103

[WolfgangFahl]

Using py-3rdparty-mediawiki library code it should be possible to create pages for each volume in a systematic/way.
The template for the pages needs to be specified!

For copyright reasons we'll have to start the trial from the most current volumes and work our way backwards. We'll start with a few dozen pages.

CC0 is available for around Volume 15xx up

[WolfgangFahl]

A Jinja 2 template could describe the page

[WolfgangFahl]

Output should be like https://confident.dbis.rwth-aachen.de/ceur-ws/index.php?title=Vol-2801 taken from the html as shown in the talk page ...

[WolfgangFahl]

https://github.com/ailabitmo/sempubchallenge2014-task1 shows a solution done a few years ago which is IMHO way too complex but might have useful bits and pieces of code and background information.