Skip to main content

Search and replace for subject fields in Alma records.

Project description

Almar · Travis Codecov

Almar (formerly Lokar) is a script for batch editing and removing controlled classification and subject heading fields (084/648/650/651/655) in bibliographic records in Alma using the Alma APIs. Tested with Python 2.7 and Python 3.5+.

It will use an SRU service to search for records, fetch and modify the MARCXML records and use the Alma Bibs API to write the modified records back to Alma.

The script will only work with fields having a vocabulary code defined in $2. Since the Alma SRU service does not provide search indexes for specific vocabularies, almar instead starts by searching using the alma.subjects + the alma.authority_vocabulary indices. This returns all records having a subject field A with the given term and a subject field B with the given vocabulary code, but where A is not necessarily equal to B, so almar filters the result list to find the records where A is actually the same as B.


Installation and configuration

  1. Run pip install -e . to install almar and its dependencies.
  2. Create a configuration file. Almar will first look for almar.yml in the current directory, then for lokar.yml (legacy) and finally for .almar.yml in your home directory.

Here's a minimal configuration file to start with:



default_env: prod

  - name: prod
    api_key: INSERT API KEY HERE
    api_region: eu
    sru_url: INSERT SRU URL HERE
  1. Replace INSERT MARC VOCABULARY CODE HERE with the vocabulary code of your vocabulary (the $2 value). The script uses this value as a filter, to ensure it only edits subject fields from the specified vocabulary.
  2. Replace INSERT API KEY HERE with the API key of your Alma instance. If you'r connected to a network zone, you should probably use a network zone key. Otherwise the edits will be stored as local edits in the institution zone.
  3. Optionally: Change api_region to 'na' (North America) or 'ap' (Asia Pacific).
  4. Replace INSERT SRU URL HERE with the URL to your SRU endpoint. Again: use the network zone endpoint if you're connected to a network zone. For Bibsys institutions, use

Note: In the file above, we've configured a single Alma environment called "prod". It's possible to add multiple environments (for instance a sandbox and a production environment) and switch between them using the -e command line option. Here's an example:

default_vocabulary: noubomn

  - marc_code: noubomn

default_env: nz_prod

  - name: nz_sandbox
    api_key: API KEY HERE
    api_region: eu
  - name: nz_prod
    api_key: API KEY HERE
    api_region: eu

For all configuration options, see configuration options.


Before using the tool, make sure you have set the vocabulary code (vocabulary.marc_code) for the vocabulary you want to work with in the configuration file. The tool will only make changes to fields having a $2 value that matches the vocabulary.marc_code code set in your configuration file.

Getting help:

  • almar -h to show a list of command and general command line options
  • almar replace -h to show help for the "replace" subcommand

Replace a subject heading

To replace "Term" with "New term" in 650 fields:

almar replace '650 Term' 'New term'

or, since 650 is defined as the default field, you can also use the shorthand:

almar replace 'Term' 'New term'

To work with any other field than the 650 field, the field number must be explicit:

almar replace '655 Term' 'New term'`

Supported fields are 084, 648, 650, 651 and 655.

Diffs and dry run

To see the changes made to each catalog record, add the --diffs flag. Combined with the --dry_run flag (or -d), you will see the changes that would be made to the records without actually touching any records:

almar replace --diffs --dry_run 'Term' 'New term'

This way, you can easily get a feel for how the tool works.

Moving a subject to another MARC tag

To move a subject heading from 650 to 651:

almar replace '650 Term' '651 Term'

or you can use the shorthand

almar replace '650 Term' '651'

if the term itself is the same. You can also move and change a heading in one operation:

almar replace '650 Term' '651 New term'

Removing a subject heading

To remove all 650 fields having either $a Term or $x Term:

almar remove '650 Term'

or, since 650 is the default field, the shorthand:

almar remove 'Term'

Listing documents

If you just want a list of documents without making any changes, use almar list:

almar list '650 Term'

Optionally with titles:

almar list '650 Term' --titles

Interactive replace (splitting)

If you need to split a concept into two or more concepts, you can use almar interactive mode. Example: to replace "Kretser" with "Integrerte kretser" on some documents, but with "Elektriske kretser" on other, run:

almar --diffs interactive 'Kretser' 'Integrerte kretser' 'Elektriske kretser'

For each record, Almar will print the title and subject headings and ask you which of the two headings to include on the record. Use the arrow keys and space to check one or the other, both or none of the headings, then press Enter to confirm the selection and save the record.

Working with a custom document set

By default, almar will check all the documents returned from the following CQL query: alma.subjects = "{term}" AND alma.authority_vocabulary = "{vocabulary}", but you can use the --cql argument to specify a different query if you only want to work with a subset of the documents. For instance,

lokar --cql 'alma.all_for_ui = "999707921404702201"' --diffs replace 'Some subject' 'Some other subject'

The variables {term} and {vocabulary} can be used in the query string.


  • For terms consisting of more than one word, you must add quotation marks (single or double) around the term, as in the examples above. For single word terms, this is optional.
  • In search, the first letter is case insensitive. If you search for "old term", both "old term" and "Old term" will be replaced (but not "old Term").


Identifiers ($0) are added/updated if you configure a ID lookup service URL (id_service) in your configuration file. The service should accept a GET request with the parameters vocabulary, term and tag and return the identifier of the matched concept as a JSON object. See this page for more details.

For an example service using Skosmos, see code and demo.

Limited support for subject strings

Four kinds of string operations are currently supported:

  • almar remove 'Aaa : Bbb' deletes occurances of $a Aaa $x Bbb
  • almar replace 'Aaa : Bbb' 'Ccc : Ddd' replaces $a Aaa $x Bbb with $a Ccc $x Ddd
  • almar replace 'Aaa : Bbb' 'Ccc' replaces $a Aaa $x Bbb with $a Ccc (replacing subfield $a and removing subfield $x)
  • almar replace 'Aaa' 'Bbb : Ccc' replaces $a Aaa with $a Bbb $x $Ccc (replacing subfield $a and adding subfield $x)

Note: A term is only recognized as a string if there is space before and after colon (:).

More complex replacements

To make more complex replacements, we can use the advanced MARC syntax, where each argument is a complete MARC field using double $s as subfield delimiters.

Let's start by listing documents having the subject "Advanced Composition Explorer" in our default vocabulary using the simple syntax:

almar list 'Advanced Composition Explorer'

To get the same list using the advanced syntax, we would write:

almar list '650 #7 $$a Advanced Composition Explorer $$2 noubomn'

Notice that the quotation encapsulates the entire MARC field. And that we have explicitly specified the vocabulary. This means we can make inter-vocabulary replacements. To move the term to the "bare" vocabulary:

almar replace '650 #7 $$a Advanced Composition Explorer $$2 noubomn' '610 27 $$a The Advanced Composition Explorer $$2 noubomn'

We also changed the Marc tag and the field indicators in the same process. We could also include more subfields in the process:

almar replace '650 #7 $$a Advanced Composition Explorer $$2 noubomn' '610 27 $$a The Advanced Composition Explorer $$2 noubomn $$0 (NO-TrBIB)99023187'

Note that unlike simple search and replace, the order of the subfields does not matter when matching. Extra subfields do matter, however, except for $0 and $9. To match any value (including no value) for some subfield, use the value {ANY_VALUE}. Example:

almar list --subjects '650 #7 $$a Sekvenseringsmetoder $$x {ANY_VALUE} $$2 noubomn'

Using it as a Python library

from almar import SruClient, Alma

api_region = 'eu'
api_key = 'SECRET'
sru_url = ''

sru = SruClient(sru_url)
alma = Alma(api_region, api_key)

query = 'alma.authority_vocabulary="noubomn"'
for record in
    for subject in record.subjects(vocabulary='noubomn'):
        if not subject.find('subfield[@code="0"]'):
            sa = subject.findtext('subfield[@code="a"]')
            sx = subject.findtext('subfield[@code="x"]')


To run tests:

pip install -r test-requirements.txt

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

almar-0.7.6.tar.gz (25.5 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page