NVA BRAGE migration

This repository contains code for extracting metadata from Brage bundles.

Documentation

Documentation of this application exits on Jira

If the dublin_core.xml contains dcvalue with element="identifier" qualifier="cristin", it will be skipped (and logged as error).

Run the CLI tool

Read the manual:

java -jar build/libs/nva-brage-migration-1.0-SNAPSHOT-all.jar -h

Sample import localy:

With specified zipfiles:

java -jar build/libs/nva-brage-migration-1.0-SNAPSHOT-all.jar zipfile1.zip zipfile2.zip -c custommer-id

With all zipfiles specified by the samlingsfil.txt:

java -jar build/libs/nva-brage-migration-1.0-SNAPSHOT-all.jar -c custommer-id

Sample import from brage-instance:

It is possible to run the .jar file by sending customer value only, but only when running .jar file on brage instance and archives are located in /brageexports/customer/ In this case archive will be unzipped at /brageexports/outputcustomer/

java -jar build/libs/nva-brage-migration-1.0-SNAPSHOT-all.jar -c custommer

Run the CLI tool from brage instance

Sample import pushing resources to default bucket:

.jar file has to be placed into the root directory, and we have to specify directory with .zip bundles

Have to specify location with your JRE since default JRE is Java 8

When migrating same collection many times in row, remember to delete directory with unzipped bundles, otherwise log files will be affected.

/etc/alternatives/jre_11/bin/java -jar nva-brage-migration-1.1-all.jar -c NVE -a -D /brage/nve/app/export

How to run Excel scraping:

-j flag can be set to "experimental" (sandbox, no lambda listening), "sandbox", "dev", "test", or "prod"

In build.gradle change main class to "no.sikt.nva.unis.ExcelScrapingCommand" and build.

application {
    mainClass = "no.sikt.nva.unis.ExcelScrapingCommand"
}

java -jar build/libs/nva-brage-migration-1.2-all.jar -e ~/Desktop/unis/Metadata.xlsx -j experimental