Skip to content

Latest commit

 

History

History
178 lines (176 loc) · 12.3 KB

RELEASES.md

File metadata and controls

178 lines (176 loc) · 12.3 KB

Releases

  • 1.0.0: Released to production
  • 1.2.0: Changed <provider_checksum> node to regular node. Must be run in combination with assembly 1.2.0.
  • 1.2.2: Significant refactoring of cleanup scripts and movement of methods to a new Util class. Update integration tests.
  • 1.3.0: Add the ability to re-accession
  • 1.3.1: Bug fixes related to re-accessioning
  • 1.3.2: Bug fixes related to smpl content metadata processing
  • 1.3.3: Add the ability to leave the progress log filename as nil in the config file, and it will be created automatically.
  • 1.3.4: Add contentMetadata and resourceTypes of 'image' during pre-assembly if you specify :simple_image as project type.
  • 1.4.6: Add functionality to discovery_report to provide information on whether objects are registered and if they have APOs
  • 1.4.7: Allow "book_as_image" as new project content type.
  • 1.4.8: Fix workflow_status report so it doesn't crash if a workflow is not found
  • 1.4.9: Bug fix in cleanup method
  • 1.5.0: Add the ability to accession or re-accession only specific items via "except" or "only" parameters
  • 1.5.1: Updates to discovery report to only show objects that will be processed
  • 1.5.2: Add some utility methods to quickly update datastreams for a list of DOR objects
  • 1.5.3: Add more options to the discovery_report to make it useful when running manifest style projects
  • 1.5.4: bug fixes to discovery report
  • 1.5.5: Add some more reporting info to the discovery_report (to tell you how many problems were found when running)
  • 1.5.6: Added additional check to discovery report -- check for any files that are present in the bundle directory but not referenced in the manifest
  • 1.5.7: Add the ability to check the uniqueness of source IDs to the discovery report
  • 1.5.8: bug fixes, update to fix tests
  • 1.5.9: update generate_collection_report to add more columns
  • 1.6.0: rename generate_collection_report to project_tag_report and add a new report that uses the progress YAML file to only report on successfully accessioned druids
  • 1.6.1: allow users to specify a manifest file for descriptive metadata purposes even if should_register=false; allow users to bypass image validation
  • 1.7.0: allow for 'joined' method of content metadata generation; allow user to specify publish/preserve/shelve attributes by mime-type
  • 1.7.1: add further validation checks to ensure only compatible values are set in YAML before proceeding, some refactoring to keep bundle class smaller, remove 'use_druid_minter' parameter and set it to a new style of getting druids, show warnings if you set development only parameters
  • 1.7.2: set completion_report to check actual workflow states in DOR instead of relying on SOLR, which can be out of date
  • 1.7.3: small bug fix when checking writable status of directories and files; try and find druids from barcodes when running discovery report
  • 1.7.4: more validation fixes, to indicate that container barcode projects must have the APO DRUID set
  • 1.7.5: add another check to discovery_report which confirms readable file permissions on object files
  • 1.7.6: bug fix in discovery_report for barcode projects
  • 1.7.7: move the Utils class into it's own gem so it can be used elsewhere ... methods that were PreAssembly::Utils are now Assembly::Utils; refactoring to account for new DruidTools gem usage
  • 1.7.8: bug fixes; update how configuration is done to make it part of assembly-utils gem
  • 1.7.9: add time stamps to progress display on screen (or into nohup.out file)
  • 1.8.0: allow the web service call to retry 5 times, sleeping between each call in order to avoid pre-assembly crashing if the dor web service is temporarily unavailable
  • 1.8.1: refactor: adjust object_file class to subclass from Assembly::ObjectFile so we don't need to redeclare convenience methods
  • 1.9.0: use the content metadata generation methods in the updated Assembly::ObjectFile gem, which expands the possible types of metadata that can be generated, update templates and documentation
  • 1.9.1: use the new DruidTools gem to generate staged content in the druid tree format: /oo/000/oo/0001/oo000oo0001/content
  • 1.9.2: update how set relationships are stored in an object -- add both an isMemberOf and an isMemberOfRelationship
  • 1.9.3: small updates to force new usage of assembly-objectfile gem for latest contentmetadata generation; additional output at end of preassembly
  • 1.9.4: update discovery_report to include checks for APO existence and existence of assemblyWF workflow definition
  • 1.9.5: more additions to discovery_report to provide total size and count by mimetype of discovered files; eliminate need to directly compute md5; this is now done via the assembly-objectfile gem
  • 1.9.6: add a new configuration parameter that determines if items are staged using the new style of druid trees or the old style of druid trees in the dor workspace
  • 1.9.7: provide the option to have no contentMetadata.xml file generated during pre-assembly. Useful if you have a current custom created one ready to stage.
  • 1.9.8: update web service call to remove "v1" from URL
  • 1.9.9: allow the content metadata creation style to be discovered from already registered objects by inspecting the 'Process : Content Type' tag. The default is used if not found or mapping incomplete.
  • 1.10.0: added support for the dir_validator gem
  • 1.10.1: updated discovery report to fetch druids; add files needed for mjf
  • 1.10.2: update discovery report to optionally show all staged files; add new parameter to stageable discovery to optionally stage files only
  • 1.10.3: show detailed exception messages if initiate_accessioning_workflow step fails after several attempts (useful for debugging)
  • 1.10.4: update discovery report
  • 1.10.5: update README to indicate new convenience methods available in assembly-utils, update required version of assembly-utils
  • 1.10.6: Update service_root configuration to allow different service URLs (e.g. with or without v1) for test and production
  • 1.10.7: Allow the dor register call to retry 5 (configurable) times, sleeping between each call in order to avoid pre-assembly crashing if the dor web service is temporarily unavailable
  • 1.10.8: Add a new parameter to allow pre-assembly to be throttled (add a sleep time between objects)
  • 1.10.9: Catch errors on register_object_in_set method call, and allow for retries before raising an exception
  • 1.10.10: Add spaces in filenames checks to discovery report
  • 1.10.11: Catch errors on get_pid_from_suri method call, and allow for retries before raising an exception
  • 1.10.12: Delete assemblyWF and accessionWF workflows from objects when re-accessioning using updated assembly-utils gem
  • 1.11.1: Bump up to latest versions of assembly gems; fix bug in publish attributing adding during content metadata generation
  • 1.11.2: If user had a trailing slash on the bundle_dir parameter, object discovery might fail. Remove any trailing slashes during setup.
  • 1.11.3: Make SMPL contentMetadata generation more robust so it doesn't crash if the checksum is missing from the preContentMetadata file.
  • 1.11.4: Changed the object type and resource type from 'file' to 'media' for SMPL conntent metadata.
  • 2.0.0: Allow user to stage items via symlink instead of copy -- useful when source bundle directory contains large files. New YAML config parameter is called "staging_style" and can be set to "copy" or "symlink" (default is "copy").
  • 2.0.1: Attempt cleanup if registration step fails -- try to delete object and purge from solr before trying again
  • 2.0.2: Add some more intelligence to the validate_usage step to better indicate errors in YAML configuration; default to new druid tree
  • 2.0.3: Verify that pid starts with druid: when initializing the assembly workflow
  • 2.0.4: Restructure project specific folder to be more generalizable. Add new "revs" project specific file that does LC term lookups when generating MODs.
  • 2.0.5: Add new "revs" project specific files with country and state parsing for location nodes on MODs.
  • 2.0.6: Add some new "revs" project specific methods to check format field
  • 2.1.0: Add a basic remediation framework for mass remediating objects.
  • 2.1.1 - 2.1.3: More updates to remediation framework; revs metadata and image updating specific code
  • 2.1.4: Allow DOR items to be registered even when a label is not provided in the manifest, by using a default label instead
  • 2.1.5: If registration fails, try to delete object by source ID as well as by druid
  • 2.1.6: Update remediation to allow you to skip workflow checks for in accessioning and versioning required when testing in development and production.
  • 2.1.7: Update remediation script to allow you to pass in a pre-assembly log file as well as a CSV with druids
  • 2.1.8: Show total number of objects and remaining in remediation output
  • 2.2.0: Allow the user to specify multiple sets/collection objects that should be associated with an item when registering
  • 2.2.1: Adjust how validate_files works to also check for color profile since that check is no longer provided with the existing method
  • 2.2.2: Change default mapping for flipbook registered object to simple_book content metadata; don't override yaml content metedata type with registered object content tag unless user explicitly indicates this should be done
  • 2.2.3: Move SMPL preContentMetadata generation methods into a class and allow it to be run automatically whenever SMPL accessioning occurs
  • 2.2.4: Make SMPL content metadata generation per object so that you do not need to run an extra step ahead of pre-assembly
  • 2.2.5: Make SMPL searching of checksums in MD5 file case insensitive for the role folder
  • 2.2.6: Add some additional info to the discovery report for SMPL manifest projects
  • 2.2.7 - 2.2.8: Enable automated combination of all techmd.xml files for SMPL projects into a single technical metadata datastream; cleanup old methods
  • 2.3.1: Remove revs specific shared methods from revs project file, include their functionality from the new revs-utils gem
  • 2.3.2: Small changes to discovery report around checking for APOs and objects; do not fail if label column is missing from manifest for should_register=false
  • 2.3.6: Use latest version of assembly-image and assembly-objectfile gems
  • 2.3.7: Use latest version of revs-utils. Allow manifests to have UTF-8 characters (switch manifest loading from csv-mapper to standard CSV library)
  • 2.3.8: Fixes to manifest loading and access to always use hash style references for columns (instead of allowing for struct like syntax that came from csv-mapper). Fix rdf gem to 1.0.9 until activefedora is updated to v6
  • 2.3.9: Send default title to registration service if no label is provided in the manifest
  • 2.4.0: Added a new MODs generation report so you can test your MODs desc md template before you actually run accessioning.
  • 2.4.1: Update a revs related mods test.
  • 2.4.2: Remove dependency on lyber-core, update other gems (including dor-services)
  • 2.4.3: Allow user to specify a tag to be applied to each object when registering
  • 2.4.4: Add some additional scripts to help prep Revs content. Allow desc_metadata_templates to be specified with an absolute path as well as a relative path.
  • 2.4.5: Add in retry logic around validate_files since it still seems to fail sporadically. Update to the revs_create_yaml script. Use new version of assembly-objectfile which gets mimetype from file extension.
  • 2.4.6: Add garbage collection each X objects pre-assembled; add optional memory profiling
  • 2.4.7: Do not add descmd xml template to every digital object when processing to lower memory usage. Add a new script for easier batch processing.

To check which version is running, cd into the home directory of the project and look at the VERSION file:

cd /home/lyberadmin/pre-assembly/current
more VERSION