The staff catalogue detail page called $biblio->normalized_upc,
normalized_ean, and normalized_oclc unconditionally. These methods
go through metadata_extractor which calls $biblio->metadata->record
internally, re-fetching the (corrupt) metadata from the DB and
ignoring the already-sanitized $marc_record.
When a record has invalid MARCXML, this causes an unhandled exception
instead of displaying the 'biberror' warning.
Fix: use the Extractor directly with the already-recovered
$marc_record (which was sanitized via StripNonXmlChars), and guard
the block with a check that $marc_record is defined.
Test plan:
1. Apply patch
2. Run:
$ ktd --proxy --selenium up -d
$ ktd --wait-ready 120
$ ktd --shell
k$ prove t/db_dependent/selenium/regressions.t
=> SUCCESS: Tests pass (including 'detail page must not explode')
3. Sign off :-D
Signed-off-by: Pedro Amorim <pedro.amorim@openfifth.co.uk>
1839189b1c