About Features Downloads Getting Started Documentation Events Support GitHub

Love VuFind®? Consider becoming a financial supporter. Your support helps build a better VuFind®!

Site Tools


Warning: This page has not been updated in over over a year and may be outdated or deprecated.
configuration:record_versions

This is an old revision of the document!


Record Versions (FRBR)

:!: This feature was added in VuFind 7.1

Configuration

Versions support can be toggled using the display_versions settings in search configuration (e.g. searches.ini). If RecordTabs.ini is customized, make sure that the Versions tab and TabScripts section is present. For versions to display, the Solr index must be updated or recreated with VuFind 7.1 or later.

If you want to always display versions as a search result, you can disable the Versions tab in RecordTabs.ini. The links in search results and record screen are automatically adjusted to support both scenarios.

User Interface

Record versions support brings two elements to VuFind: Search results include a link for displaying other versions (including count) and a record tab that displays a set of versions for the record. All versions can also be displayed as a search result list.

It is an intentional choice to not introduce FRBR specific language in the user interface. The primary reason for this is that it can be a complicated model for a non-librarian to understand and even more complex to actually handle in the UI and between diverse data sources. This means that in practice, and depending on what the index actually contains, versions can include both FRBR expressions and manifestations.

Underlying Technology

Identifying versions of a record is based on so-called work identification keys. Any records that share keys are considered versions of each other. Keys are created from author + title combinations and uniform titles. It is possible that a version of a record has also other different versions since it may have additional keys not present in the initial record. Keys have been designed to be general enough that they can be created from MARC records as well as other formats. There are several rules for how the keys are created by default, notably:

  • For fields with non-filing characters or leading articles, two keys are created; one with non-filing characters and one without.
  • Author + title fields are created for all permutations, but never for an author or title alone.
  • Keys are transliterated to latin.
  • Everything is lowercased.
  • Everything but a-z and 0-9 is filtered out.
  • Strings are normalized to NFKC form.

For default MARC fields used, please see work_keys_str_mv in import/marc.properties. import/marc_local.properties includes a commented-out example that keeps also some additional characters. For other formats, see the corresponding properties file, e.g. import/dspace.properties.

configuration/record_versions.1606807737.txt.gz · Last modified: 2020/12/01 07:28 by emaijala