Skip to main content
  • Subscribe by RSS
  • Ex Libris Knowledge Center

    How to - Find Central Discovery Index (CDI) records for specific collections in Primo

    Created By: Stacey van Groll
    Created on: 6/02/2020

    Key troubleshooting tool - Primo PNX Bookmarklet, developed by Gabriele Hofler

    For details of the features of this bookmarklet and others, see also: How to – Useful bookmarklets for Primo New UI troubleshooting

    See also some details for PCI, although this will no longer be relevant soon as Primo Central is decommissioned and as features may develop differently in CDI: How to find Primo Central Index (PCI) records in Primo


    System Specs

    • This guide is relevant for Primo sites on CDI, regardless of deployment model aka it applies to both Primo on Back Office and Primo VE
    • The only variation for Primo on Back Office is the presence of TN_ (aka Third node) at the beginning of Record IDs, but this does not affect this guide as it is not used in CDI search
    • Note: Screenshots include a scope of "PCI" - this is shorthand for remote records in my testing view. I'll update images when I get a second, now that I've changed the label to CDI post-switchover


    Some options

    • From an Alma collection to Primo records by DBID - most effective with unique Link in Record type collections
    • From a Primo Available online record to an Alma collection by Link Resolver services - most effective with title-based Link Resolver type collections


    Steps - from an Alma collection to Primo records by DBID

    • Find the collection record in Alma, by Electronic Collection search

    • Click on Edit Collection, and then click and copy the Provider package code (DB id)

    • Go to Primo and search on the DBID
      • To increase the likelihood of success, make sure to authenticate and tick the box into expanded search to Include results not available online
      • Change the 'any' portion of the URL to dbid eg change this: search?query=any,contains,GOX&tab... to this: search?query=dbid,contains,GOX&tab...
      • Use the Primo PNX bookmarklet to check for matched Record IDs

    • If no luck by PNX bookmarklet in the first set of results to the correct collection, scroll down and see if the Source facet (aka Collection) has the collection

    • Select the facet as an active included filter, and then use the Primo PNX bookmarklet again to find the collection prepend

    • Lock the facet, copy the initial portion of the collection id, and search with a wildcard, for example cdi_arxiv_primary*

    • Crosscheck into Alma, and see the expected fairly close match by Number of Records (recognizing that Ex Libris is only updating these periodically)
    • Additional tips
      • Once you've figured out how to find a collection in Primo, add details to the CDI Local Notes, such as collection prepend and facet presence. Do this BEFORE you use the suppression option, as there is an issue with loss of the Source facet entry for many collections moved into expanded search by the checkbox for 'Do not show as Full Text available in CDI even if active in Alma'. This is the case even if the data is in the PNX facet section. This has been reported to Ex Libris
      • If there is no Source facet even in filtered search, use the Resource Types in Alma to try to narrow the search by Content Types facet in Primo. This is particularly effective if there are types which have lower number such as audio, datasets, or government documents, as that increases the chance you'll get a hit on the right collection in a result on the first page. There is an issue with many collections having either no Resources Types listed, and the types are the original Summon types rather than the mapped Primo types - both of these issues have been reported
      • Some collections have very broad collection prepends like cdi_proquest_miscellaneous.  In those cases, it can be helpful to keep the DBID in the search, for example: cdi_proquest_miscellaneous* PBHAV, and then also look for a sub-collection Source facet
      • If still no luck, try to hunt down an actual title of a resource and search that specifically
      • If all else fails, try to guess the collection prepend by initial key term or acronym from the collection eg Narcis: Open Access by cdi_narcis*

    Additional Notes

    • Records are expected to include all DBIDs for associated collections, regardless of your institutional rights, but there is an issue where many do not. This has been reported to Ex Libris


    Steps - from a Primo Available online record to an Alma collection by Link Resolver services

    • In PCI, using the PNX bookmarklet to reveal the recordid was a clear sign of record source, but this is often not the case on CDI, particularly for non-unique content
    • This is because CDI uses a vast pool of citation metadata, matched to your full text holdings to mark availability, and this is not necessarily easily visible by elements such as DBID
    • As such, the recordid may appear to be from a collection which is not active for search in your IZ, which is quite confusing, but the record will appear in your environment due to full text available holdings for the same content somewhere in your IZ
      • Be aware! The recordid may change, as Primo will chose one of the source recordids at random.  All of the IDs will work though, so a permalink saved one day with one ID, will still work the next day when the record appears to have a different ID

    • Click on the record to open the full record display and see the full text availability, eg via AIP Journals and PubMed

    • When checking the content in Alma by various means, such as ISSN, you'll see the expected Link Resolver collections

    • If checking your CDI Institutional Holdings File, you'll see the title being published from Alma to CDI, along with details of the active services via AIP Journals and PubMed, any embargos, coverage, and the matching DBIDs from the Electronic Collection through the holdings file




    Feedback and suggestions