WAS Metadata Extractor

Java code used to extract metadata from web archiving ARC and WARC files, used by WAS crawl preassembly robot WF.

Deployment of the jar is part of deployment tasks in was_robot_suite, via an artifact created at jenkinsqa. The deployed was_robot_suite houses the WASMetadataExtractor.jar in the jar directory.

Build

mvn install

Configuration

WAS Metadata Extractor depends on config/extractor.yml which describes the file types and the expected headers in each file.

Format of extractor.yml

ARC:
 id:
  type: Headers
  path: "ARC-Filename"

• Line 1. File type
  • Line 2. properties name
    • Line 3. the type of how to get this value
    • Line 4. the location of this value based on the properties type

Other types can have different properties.

Usage

List usage command

java -jar WASMetadataExtractor-0.0.2-SNAPSHOT-jar-with-dependencies.jar  -h
usage: WASMetadataExtractor
 -c,--config <file>                 YAML configuration file to extract the
                                    metadata.
    --collectionId <collectionId>   Collection id as it should appear in
                                    metadata files. It shouldn't have
                                    space.
    --crawlId <crawlId>             The crawl id as it should appear in
                                    metadata file.
 -d,--inputdir <directory>          The direcotyry of ARC/WARC file to be
                                    processed.
 -f,--format <XML|TXT>              The output format, TXT is the default.
 -h,--help                          Display help
 -i,--inputfile <file>              An input file that has a list of
                                    ARC/WARC files to be processed.
 -o,--outputFile <file>             The output file.

Run Tests

mvn test -B