HANTransformations

Transformations (Catmandu, XSLT) and basic workflow scripts to prepare HAN-data for indexing in swissbib. For further documentation see http://www.swissbib.org/wiki/index.php?title=Staff:Indexierung_Archivdaten

Installation

The following scripts need to be saved in the HANTransformations root directory:

• workflow.HAN.sh: shell script that controls the main workflow. It calls the three scripts consecutively:
• transform.seq2hanmarc.pl (step 1)
• transform.han2sbmarc.sh (step 2)
• transform.into.1.line.sh (step 3)
• transform.into.1.line.pl: perl script to flatten the records into one line (called by step 3)

The Perl-script transform.seq2hanmarc.pl need Perl modules of the Catmandu project (can be installed using cpan).

The xslt-script HAN.Bestand.xslt needs to be saved in: HANTransformations/xslt

An XML parser needs to be saved in: HANTransformations/libs

We use saxon, e.g. saxon9pe.jar, along with the license

Directories

• Input directory step 1: HANTransformations/raw.hanseq; contains data in Aleph sequential format (usually dsv05.seq)
• Output directory step 1: HANTransformations/raw.hanmarc
• Input directory step 2: HANTransformations/raw.hanmarc; contains data in MarcXML format
• Output directory step 2: HANTransformations/out.swissbib-MARC
• Input directory step 3: HANTransformations/out.swissbib-MARC
• Output directory step 3: HANTransformations/out.swissbib-MARC-1line