find a proper way to encode reconstructed (?) publisher name and location

buda-base / xmltoldmigration

App to migrate from TBRC XML files to BDRC RDF LD

Apache License 2.0

0 stars 2 forks source link

find a proper way to encode reconstructed (?) publisher name and location #158

Closed eroux closed 3 years ago

eroux commented 3 years ago

The way some information is currently encoded is using some square brackets around a string to denote (something?...). The problem is that square brackets have a meaning in ewts and this means that the text is not really analyzed as Tibetan and can't be found.

see https://www.tbrc.org/xmldoc?rid=MW29191

Also, just for future references: https://www.ifla.org/files/assets/cataloguing/isbd/isbd-cons_20110321.pdf

eroux commented 3 years ago

According to the cataloging guidelines:

If known without being printed, it is entered in square brackets

eroux commented 3 years ago

possible solutions:

replace with a star at the beginning
make the inside of the square brackets Tibetan
adjust the Lucene analyzer, which would not really be great