A utility gem for working with Alma / Primo.
In your Gemfile:
gem 'berkeley_library-alma'
In your code:
require 'berkeley_library/alma'
The BerkeleyLibrary::Alma::Config
class includes the options below. These
can be read automatically from a Rails configuration
(Rails.application.config.alma_sru_host = "..."
, etc.) or set directly
(BerkeleyLibrary::Alma::Config.alma_sru_host = "..."
).
Alternatively, BerkeleyLibrary::Alma::Config.default!
will set the
options to either:
- the value of the corresponding environment variable, if set, or
- the default value for the UC Berkeley Library.
Option | Environment variable | Purpose | Berkeley default |
---|---|---|---|
alma_sru_host |
LIT_ALMA_SRU_HOST |
the Alma SRU hostname | berkeley.alma.exlibrisgroup.com |
alma_institution_code |
LIT_ALMA_INSTITUTION_CODE |
the Alma institution code | search.library.berkeley.edu |
alma_primo_host |
LIT_ALMA_PRIMO_HOST |
the Alma Primo host | 01UCS_BER |
alma_permalink_key |
LIT_ALMA_PERMALINK_KEY |
the Alma permalink key | iqob43 |
The SRU
module encapsulates Alma
SRU queries.
Retrieving MARC records:
reader = BerkeleyLibrary::Alma::SRU.get_marc_records('991005668209706532', '991005668359706532')
# => #<MARC::XMLReader:0x0000000135b940e8 @freeze=false, @handle=#<StringIO:0x0000000135b94160>...
Making arbitrary SRU queries:
BerkeleyLibrary::Alma::SRU.make_sru_query('alma.other_system_number=UCB-b123230470-01ucs_ber')
# => "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?><searchRetrieveResponse xmlns=...
In addition, individual records can be retrieved from an instance of the
RecordId
class, which encapsulates
an Alma MMS ID, Millennium bib number, or item barcode.
Alma MMS ID:
mms_id_str = '991054360089706532'
record_id = BerkeleyLibrary::Alma::RecordId.parse(mms_id_str)
# => #<BerkeleyLibrary::Alma::MMSID:0x0000000138949830 @institution="6532", @mms_id="991054360089706532", @type_prefix="99", @unique_part="105436008970">
Millennium bib number:
bib_number_str = 'b11082434'
record_id = BerkeleyLibrary::Alma::RecordId.parse(bib_number_str)
# => #<BerkeleyLibrary::Alma::BibNumber:0x0000000118815038 @check_str="9", @digit_str="11082434">
Item barcode:
barcode_str = 'C084093187'
barcode = BerkeleyLibrary::Alma::BarCode.new(barcode_str)
# => #<BerkeleyLibrary::Alma::BarCode:0x000000013fac4c08 @barcode="C084093187">
RecordId#parse
method; they must be instantiated
directly.
Given a RecordId
object, you can get the SRU query URI for the corresponding MARC record:
record_id.marc_uri
# => #<URI::HTTPS https://berkeley.alma.exlibrisgroup.com/view/sru/01UCS_BER?version=1.2&operation=searchRetrieve&query=alma.mms_id%3D991054360089706532>
Or the MARC XML as a string:
marc_xml_str = record_id.get_marc_xml
# => "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?><searchRetrieveResponse>...</searchRetrieveResponse>"
Or just a ruby-marc MARC::Record
object:
marc_record = record_id.get_marc_record
# => #<MARC::Record:0x0000000138b45490
You can also use RecordId
to verify or calculate the check digit for a Millennium bib.
(Check digit calculation is based on the algorithm described in "Recipe: Computing the Millennium checkdigit" by Lisa Wong of the McHenry Library, UC Santa Cruz.
The check digit will be appended to bib numbers without one,
or with the wildcard check digit a
.
bib_number_str = 'b11082434'
record_id = BerkeleyLibrary::Alma::RecordId.parse(bib_number_str)
# => #<BerkeleyLibrary::Alma::BibNumber:0x0000000118815038 @check_str="9", @digit_str="11082434">
record_id.check_str
# => "9"
record_id.full_bib
# => "b110824349"
bib_number_str = 'b11082434a'
record_id = BerkeleyLibrary::Alma::RecordId.parse(bib_number_str)
# => #<BerkeleyLibrary::Alma::BibNumber:0x0000000118815038 @check_str="9", @digit_str="11082434">
The RecordId#parse
method will raise an error if passed a bib with a bad
check digit.
bib_number_str = 'b110824341' # wrong check digit; should be 9, not 2
# bib_number.rb:78:in `ensure_check_digit': 110824341 check digit invalid: expected 9, got 1 (ArgumentError)
# from /Users/david/Work/BerkeleyLibrary/alma/lib/berkeley_library/alma/bib_number.rb:68:in `split_bib'
# from /Users/david/Work/BerkeleyLibrary/alma/lib/berkeley_library/alma/bib_number.rb:27:in `initialize'
# from /Users/david/Work/BerkeleyLibrary/alma/lib/berkeley_library/alma/record_id.rb:35:in `new'
# from /Users/david/Work/BerkeleyLibrary/alma/lib/berkeley_library/alma/record_id.rb:35:in `parse'
The alma-mms-lookup
script takes one or more record IDs (either Millennium bib
numbers or Alma MMS IDs) and attempts to read the corresponding MARC records via
SRU and extract
the canonical MMS ID for the record from the 001 control field.
If you clone this repository and run bundle install
from the project root,
you can run the bin/alma-mms-lookup
script directly. Note that this requires
Ruby 3.x to be installed. (Alternatively, you can use Docker; see below.)
To look up a single ID:
echo b11082434 | bin/alma-mms-lookup
To look up a list of IDs given in a file record-ids.txt
:
bin/alma-mms-lookup < record-ids.txt
To look up a single ID:
echo b11082434 | docker compose run gem bin/alma-mms-lookup
To look up a list of IDs given in a file record-ids.txt
:
docker compose run gem bin/alma-mms-lookup < record-ids.txt
The output is tab-separated, in the form
<original bib number> <bib number with check digit> <MMS ID from 001>
or
<orignal MMS id> <original MMS ID again> <MMS ID from 001>
e.g.:
b11082434 b110824349 991038544199706532
991038544199706532 991038544199706532 991038544199706532
Any IDs that cannot be retrieved or parsed are left blank:
echo 'b11082434
b1234
991038544199706532
9912348245906531
b110824349' | bin/alma-mms-lookup
produces:
b11082434 b110824349 991038544199706532
b1234
991038544199706532 991038544199706532 991038544199706532
[2022-01-03T14:17:09.060-08:00] WARN: GET https://berkeley.alma.exlibrisgroup.com/view/sru/01UCS_BER?version=1.2&operation=searchRetrieve&query=alma.mms_id%3D9912348245906531 did not return a MARC record
9912348245906531 9912348245906531
b110824349 b110824349 991038544199706532
Note that warning messages are written to STDOUT, so you may need to filter them
with grep
:
echo 'b11082434
b1234
991038544199706532
9912348245906531
b110824349' | bin/alma-mms-lookup | grep -v WARN
produces:
b11082434 b110824349 991038544199706532
b1234
991038544199706532 991038544199706532 991038544199706532
9912348245906531 9912348245906531
b110824349 b110824349 991038544199706532
The alma-oclc-lookup
script takes one ore more record IDs (either Millennium bib
numbers or Alma MMS IDs) and attempts to read the corresponding MARC records via
SRU and extract
the OCLC number(s) from the 035$a field.
The output is tab-separated, in the form
<record ID> <OCLC number(s)…>
E.g.
$ echo '991051353589706532' | bin/alma-oclc-lookup ✹ ✚main ‹ruby-2.7.5›
991051353589706532 1097551039