- Updated pidman client to the current version of requests
- Removed Record Info section from Video edit page
- Making changes to celery backend
- Switching to a different RabbitMQ
- changing celery task to convert from wav to mp3
- Keep reindex errors ( premis metadata version )
- Manage command for pids that don't have matching mimetypes
- ModsrecordChangeDate is not updating when pid is updated
- mov objects are not calculating the correct mimetype
- Updated to a specific version of celery because 4.0 is failing
- Fixing Mods Dates errors
- Keep collection generation isn't working because the repository name for the Rose Library is incorrect.
- Testkeep is producing an error message for some audio objects.
- Bags with Iso disk images cannot be ingested into the Keep.
- INC02570212 - Keep configuration change: new codec creator
- Fixing indexing issue related to provenanceMetadata
- As a staff user, when I'm viewing a Rushdie object with a reassigned pid, I want to be able to see PREMIS metadata documenting this change so that I understand the provenance of the object
- As a Keep administrator, I want Rushdie pids mismatched in Pidman to be assigned to a new pid within the Rushdie allocation range in Pidman
- As a Keep administrator, I want a one-time update of pidman metadata for Rushdie objects (label/target) where objects exist in Fedora.
- bugfix: Fix Keep load_ead script (generate Keep collections from EAD) to work with latest code updates
- As a collection manager, I want to be able to create new collections from finding aids so that the metadata for collections is consistent
- Update to existdb 2.2
- Update to use migrated taskresult in eulcommon and remove eullocal dependency
- Set up testkeep fedora content and pidqas so pid updates can be tested
- As a Keep administrator, I want a one-time update of ARK records in the Pid Manager to match current titles in the Keep so that the metadata is consistent in the two systems
- As a collection curator, when I edit an object title in the Keep, I want the Pidman title to automatically update in order to have accurate/matching information in Pidman
- bugfix: The Pidman rest api does not find pids in subdomains when searching by domain (requires Pidman 1.0.3)
- Resolve an issue in which Rushdie pids in Fedora do not match Pidman - Report on unused PIDS and missmatches
- bugfix: researcher access permissions to view video revisited (updated to use videoperms instead of video)
- bugfix: Audio view now checks content models and will raise a 404 for non audio, rather than incorrectly displaying video content
- Require latest eulfedora (1.6) for debug filter and retries
- Enable new eulfedora debug filter to keep fedora auth credentials from being displayed in stack traces and debug emails
- Enable django-debug-toolbar for development with eulfedora panel
- Regenerate accounts migrations and remove eullocal.emory_ldap dependency
- Browser warning removed (removed UnsupportedBrowserMiddleware middleware, which check for supported browsers and displayed a banner; browser support is no longer an issue)
- Video Permissions bug
- Audio timeout error
- Improved documentation for adding top-level collections
- Documentation for queuing batch-conversion of audio
- Adds Law Libary as a top-level collection
- As a staff user, I want to be able to find objects with no collection or access status set so that I can identify and fix content that has been uploaded but not had any metadata added.
- Bugfix for error editing objects that include empty, non-existent xml datastreams (actual fix in eulfedora 1.5.1)
- Bugfix: catch and report problems with MediaInfo when getting video duration on ingest
- As a researcher, I want to be able to view video in full screen mode so that I can see the video more easily.
- Bugfix: Improve video playback on startup and skip (now using streaming response and Fedora range requests)
- As a repository administrator, I can migrate large AFF master disk images and add metadata that reflects these changes in order to update obsolete file formats in the repository.
- An administrator can migrate AD1 master files to TAR files generated by digital archives staff and add metadata that reflects this change in order to update obsolete file formats in the repository.
- An administrator can migrate the format of an AFF master composite file and add metadata that reflects this change in order to update obsolete file formats in the repository.
- An archivist can download a local copy of an image file in the lab for processing, arrangement, and description.
- As a staff user, when I'm viewing a migrated disk image object, I want to be able to access the original version (including metadata) that was migrated, so that I understand the provenance of the object.
- As a staff user, when I'm viewing the original version of a migrated disk image object, I want to be able to access the migrated version (including metadata) so that I can get to the most recent version.
- As a staff user, I want to filter search results by disk image, email, other born digital files, and format so that I can restrict results to a specific set of content.
- As a staff user, when my search results include migrated disk images, I want to see the original version grouped with the migrated version so that I understand that the files are related and I can tell which one is the most recent version.
- Bugfix: Primary target URI in pid manager for new Keep objects were being garbled.
- Bugfix: Fix video record download original file (now using streaming download)
- Bugfix: correct logic for creating ARKs via PID manager so that the placeholder is not url-encoded and gets replaced properly with the newly-minted noid.
- Support for running celery daemon on a separate server from the web application; modifies audio access copy conversion file handling on ingest to allow for celery and audio file conversions to run on a seprate server from the web UI where ingest is done.
- An archivist can upload .tar, E01, and .mbox files and associated metadata into the Keep as a part of a bag in order to ingest preservation-stable file formats for disk images and composite files.
- bugfix: prevent users from double-clicking ingest when uploading files and ingesting the same file twice.
- bugfix: make Keep frontend external dependencies configurable to support restricted access researcher machines.
- Upgrade to Django 1.8
- Migrate from eullocal to django-auth-ldap for LDAP login support
- Changes to work with fedora 3.8: work around file uri checksum bug.
- As a site user I will see a Site Down page when maintenance is being performed on the site or or other circumstances that will cause the site to be temporarily unavailable so that I will have a general idea of when I can use the site again.
- As a site user I will see a banner that displays an informative message on every page of the site so that I can be informed of future site maintenance or other events.
- As an application administrator, I want to generate a list of pids for testing so that I can verify the application works with a subset of representative but not sensitive real data.
- Added Health Sciences Library
- Migration scripts to migrate metadata and video files fromm DM to Keep
- Filter archive collections for search results
- Added library for ETD
- Fixed bug that prevented edit button for videos from displaying in collection view.
- Fixed problem when convert large Audio files to MP3
- Enabled researcher video search and playback
- Fixed issue date and creation date bug
- Fixed index bug when file size is too large
- Added ability for video ingest, search (by staff) and editing of metadata (by staff with appropriate permissions)
- Video objects include access copy in bagit package
- Larg File Upload SFTP server permissions and structure have been reworked to allow for different types of content
- Video objects are browseable in collection view
- Now All staff can search for all types of content but viewing and editing metadata are still restricted by perms
- added dc_cleanup manage command
- added link to Keep Manual on dashboard for staff members
- added ability to upload 64bit wave files
- bugfix release: workaround for Chrome v35 or later issue in recognizing MP3s as playable in HTML5
- Bug fix: cleaner jplayer syntax for specifying mp3 or m4a audio file, to avoid issues with some versions of Chrome
- As a user (researcher and archivist) of the Keep, I can see a note when I first access the application that informs me what browser I should use in order to avoid problems when playing audio incompatible with some browsers.
- As a staff user, I can see a visual indicator in the list views (search results and browse pages) that indicates whether an audio item is available to the public so that I can quickly see which items are inaccessible to non-staff users.
- As a researcher, I want to see a single audio item in a format consistent with the search results, so that I know where to look to find the same information.
- bugfix: error when LDAP fails is not obvious (generic 500?)
- bugfix: edit field sizes are too large for Sublocation, Tape Brand/Stock, and Part Note.
- bugfix: Keep does not support 32bit float wav files.
- bugfix: Django admin reassigns the staff flag if the user is an LDAP user even if an administrator removed the flag. (note: fixed in eullocal 0.21)
- As a Keep administrator, I can assign granular permissions to groups and individuals so that I can manage what users can view and do within the site.
- As a logged in staff user, I will be redirected to the dashboard page after saving a new or edited record, so that I can continue my work in the Keep.
- As a logged in archivist, I can see the item view page for items with any permissions status in order to see everything that is in the Keep.
- bugfix: permissions error redirects user to login page even if already logged in
- bugfix: dashboard facets should be filtered by user permissions
- bugfix: disk image objects are listed in search results for audio curation users
- As a researcher, I want to access and search the metadata for digitized audio recordings in The Keep in order to find materials relevant to my research.
- As a researcher viewing detailed metadata about an item, I want to be able to listen to the associated audio recording in order to conduct research.
- As a researcher viewing a list of search results, I can choose any item to view more detailed metadata about that item in order to find materials relevant to my research.
- As a researcher, I want to filter results based on collection name or number in order to find research materials that most closely relate to my research.
- As a researcher, I want to browse a paginated list of collections by owning repository in order to see groups of content.
- As a researcher, I can search across the collections by repository and collection number in order to quickly find a collection when I know exactly what I'm looking for.
- As a logged in archivist with edit permissions, I can move from the view page to the edit page in order to make updates.
- As a researcher, I want to browse a paginated list of materials by collection from a given repository in order to access materials related to my research.
- As a researcher, I want to filter results based on origin date by a single date, before or after a given date, or a specific date range in order to restrict results to a specific time period.
- As a researcher searching items, I will only find collections that include items I am allowed to use in order to avoid wasting time browsing collections for materials I'm unable to access.
- As a researcher, I can only access materials when in the MARBL Reading Room in order to protect MARBL copyright.
- As a researcher, I want to filter results based on owning library in order to find materials that most closely relate to my research.
- As a researcher, when I search using the advanced search filters, the filters will be displayed by default on the results page so I can see the filters that are active.
- As a researcher, I am unable to search or see digital objects that MARBL administrators have not made available to me.
- As a researcher when viewing search results, I can see the most up to date list of materials available to me in order to view the most updated and accurate materials.
- As a staff user, I can see a message that indicates when a audio item is inaccessible to patrons along with the rights code/override that governs this in order to distinguish between objects available to researchers and those available only to staff.
- A repository administrator can configure a script to periodically check content checksums in order to identify integrity issues so that they can be dealt with. (implemented in eulfedora)
- A repository administrator will receive an email notification if the system encounters bad or missing checksums so that they can then resolve any integrity issues. (implemented in eulfedora)
- A repository admin can view fixity check results for individual objects in the premis data stream (for objects where premis exists) in order to view a more detailed result and the history. (implemented in eulfedora)
- As a Keep user, when I log in I want to see on the home page a count of any objects that have failed a fixity check in the last 30 days and be able to view metadata records for those objects so that I can follow up and/or document as needed.
- An archivist can ingest an .iso as a disk image for preservation, storage, and data management when content can't be converted to AFF or AD1.
- As an archivist, I can view and edit necessary metadata fields within disk image Keep records so that I can manage digital assets.
- Only processing archivists in digital archives and university archives can view and edit metadata records to ensure the security of unprocessed digital archives materials.
- bugfix: disk image ingest returns a 500 error if bagit data filenames contain whitespace
- bugfix: streamline disk image ingest to avoid timeout errors (now requires bagit input with both MD5 and SHA1 checksums; no checksums are calculated by the Django app during disk image ingest)
- An archivist can upload a large file and its checksum into a staging area, so it can be selected for ingest into the Repository without having to upload large files in the web interface.
- An archivist can select a file uploaded to the staging area for ingest, so that content too large for web upload can be ingested into the repository.
- When archivist selects an uploaded file for ingest, any supplemental files in bagit will be ingested and associated with disk image object in order to allow staff to assess and document contents of image.
- Archivist can view the supplemental file name (as file title) in the metadata record for the disk image and view or download the supplemental file, in order to review the content in the file.
- Archivist can add supplemental files (and edit file names) associated with an ingested composite file in order to maintain security, chain of custody, and appraisal information about the digital object.
- When a data curator attempts to ingest content via the Keep, they will receive an alert if the file is already present in the repository, so that staff can avoid duplicating digital objects in the Keep.
- Updated to Django 1.5
- An archivist can upload a disk image file via the Keep web interface to ingest it into the repository, in order to secure and preserve the data and to provide archival access for triage and processing.
- After uploading a disk image file (or batch of disk image files), an archivist should see a list of original file names and corresponding checksums so that they can verify data authenticity.
- When an archivist uploads a disk image file via the Keep, record identifiers and other standardized fields will be automatically stored in the metadata record, so that an archivist does not have to enter them manually.
- Archivist can search and discover metadata records for ingested disk images so that they can view, download and/or edit disk images as part of processing.
- An archivist can enter minimal required metadata after upload when ingesting a disk image file, in order to document custodial history.
- An archivist can add and edit entries in the controlled list of imaging software used for disk image metadata, so that new systems can be added as needed.
- Archivist can view status of process after changing the status of a "simple collection" in the Keep, so they can tell whether or not all items in the collection were updated. (correction to previous implementation)
- bugfix: marking an object as processed via "simple collection" Keep edit will now leave an audit trail message
- Django 1.4 cleanup: settings & templates, cache configuration example
in
localsettings.py.dist
- Added a new Codec Creator entry for audio
- A logged in user can see a checksum for binary file content on the record detail page (currently edit page), in order to verify file authenticity.
- A logged in user can search file content by checksum, in order to match ingested content with original metadata.
- A logged in user can identify email records and see high-level email information in Keep search results, so that email can be distinguished from other types of files.
- When a user clicks on an email record in the search results, they are taken to a brief view page so that they can see information about the email instead of loading the default arrangement edit form which does not entirely apply to email.
- A system administrator or developer can run a script to import verdict and series information (in a CSV file) for the processed 5300c files into the repository, so that verdict and series decisions do not have to be entered one at a time.
- A system administrator or developer can run a script to ingest 5300c email messages into the repository, so that email verdicts can be imported and email content can eventually be made accessible to researchers.
- A system administrator or developer can run a script to add item level content for non-email files to 5300c metadata-only records in the repository, so that processed file content can be made accessible to users for research purposes.
- A system administrator or developer can run a script to import verdict information from a CSV file for 5300c email messages, so that verdicts from messages sorted in the emulation can be applied to repository items.
- Recently added items on site home page now includes a list of collections with items added in the past 30 days.
- Administrative users receive notification when the number of available iTunes feeds changes, so that they can update the researcher kiosk.
- Bug fixes:
- Corrected item level search results link to parent collection
- Search audio/arrangement items (old search) by collection (broken after change to auto-complete collection input)
- Corrected sorting for recently added items by day (on site home page)
- Make Archive required in the collection create and edit forms.
- Prevent the creation of collections with duplicate Source Id in the same Archive.
- Correct a bug in the access link URL for downloading MP4/M4A version of audio items in the new combined search.
- Users can view the duration for audio items in the search display in HH:MM:SS format rather than in total seconds, so that duration can be easily understood.
- A logged in user can perform a simple keyword search to quickly find any records in the repository that contain the relevant keywords, so that any type of item can be found in a single search.
- A logged in user can search for records by the user who uploaded them, in order to easily find records they created or items uploaded by a specific user.
- A logged in user can search for records by creation date, in order to easily find recent items or items uploaded on a specific date.
- A user can filter keyword search results by type (collection/audio/born-digital), collection, rights status, or upload user so that they can easily narrow a large result set to the items they are interested in.
- When ingesting a composite file or batch of files, an archivist or curator can enter an optional log message for auditing purposes.
- An archivist can make an optional comment when making metadata changes using any edit form, so that the audit trail will be a more useful record of changes made to an object.
- Logged in users can quickly select a collection on edit, upload and audo/arrangement search form by typing any part of the collection name or number and choosing from matching suggestions.
- An archivist or curator uploading files for ingest must associate them with an archival collection, so that the content is linked to the appropriate collection from the point of ingest.
- An archivist can view a human-readable version of the Fedora audit trail for an item in order to review the history of all actions on a file since ingest.
- When a user attempts to upload a file without choosing a collection they see an error message, so that they don't waste time uploading files without setting a required field.
- Only authorized users, within the born-digital archives group, can access born-digital archival objects and metadata via the combined search.
- Updated to use Django staticfiles app to manage static site content.
- Added support for a beta warning to be displayend in test/staging sites; turn on via ENABLE_BETA_WARNING setting.
- A user browsing search results can navigate through paginated results by just a few pagination links, so that they can access all results without being overwhelmed by pagination links.
- A logged in user can use the item search form to select specific fields and an output mode (html or csv), in order to dynamically generate a report on a specific group of items.
- Updated to Django 1.3.1, Solr 3.3, httplib2 0.7.4
- 'old-dm' migration code has been removed.
- Update to the MD5 javascript implementation: now correctly calculates checksums for files >2GB.
- Require eulfedora 0.18.1 to avoid missing checksums on XML and RDF datastreams.
- Branched Keep code and added arrangement app
- Added load_arrangement management command
- Added SimpleCollection in collection.models
- Moved Rights and supporting models from audio app to common app
- Added FileMasterTech class in common app
- Added function to update all ArrangementObjects based on SimpleCollection status
- Added content modle fixtures AccessAllowed.xml and AccessRestricted.xml in arrangement/initial_objects
- Added migrate_rushdie manage command
- Added ability to ItemSearch to search for Arrangements and Audio
- Added ability to Item Search to fiter by Format (contenet_model)
- Added ability to Item Search to fiter by SimpleCollection
- Updated and added xacml policies
- Updated permission checking for existing keep views and new Arrangement views
- Added settings.json.dist example fab settings file
- Changed project to use django logging insted of logging.conf style
Fix several bugs in 1.1.0:
- Sort collections by archive in collection browse.
- Remove archive objects from collection browse.
- Include appropriate collection data in audio feed entries.
- Use eullocal templates for task package.
Migrate all metadata for audio files from the existing Digital Masters database to the new system so that existing users can transition to use the new system entirely in place of the old one for audio metadata.
- A system administrator can run a script that reads metadata from audio records in the old database and generates clear logs of items and metadata in the old system. Stakeholders can review this log to identify changes needed in either the source data or the processing.
- A system administrator can run the metadata log script to transform and migrate legacy system data into reposited metadata in the new system. (Logging “Dry run” functionality will still be available with a command-line option.)
- Metadata experts and archivists can view all migrated fields in the audio edit interface so that they can verify it and copy it to new fields.
- Users will be able to view and find migrated Audio File metadata by association with a Collection object, based on either the collection number (if location is MARBL and if an MSS collection number is assigned) or the location (unnumbered collections for MARBL, EU Archives or Oxford) in the legacy system data.
- A user can designate "trash" records in the old database that should not be reposited in the new system, by including the word "delete" (not case sensitive) in the title field. The migration script will not create an object to migrate metadata to, and the transaction log will record a special "delete" error condition.
- Users will see legacy system filename identifiers on the audio item edit page for migrated records (where audio files stored in the repository would normally be available for listening), so that they will be able to locate the audio files that are not yet available in the repository. (Note that future milestones will migrate this audio into the repository.)
- Archivists can use the web interface to search for migrated metadata using newly migrated critical file metadata.
- Researchers at the MARBL A/V kiosk see migrated records once (and only once) in the kiosk interface.
- A curator can select "Vendor" and "Unknown" (non-LDAP identities) from the list of Transfer Engineer choices.
- A curator can record correct speed metadata for digitized microcassettes.
- Librarians and preservation specialists can see in the legacy Digital Masters interface whether any particular record's metadata has been migrated to The Keep and into what PID for easy verification.
Additionally, this milestone includes updates to take advantage of EULindexer functionality and use Solr for searching:
- Users who browse or search for collections receive their results in under 1 second for a faster user experience and more efficient workflow.
- Users who browse or search for audio items receive their results in under 1 second for a faster user experience and more efficient workflow.
- When MARBL Kiosk requests audio items feeds, it receives results in under 1 second, so that iTunes can harvest available items without timing out.
Date: May 2011
- Update to Javascript MD5 checksum uploader code to work with the most recent versions of Firefox 4 and Google Chrome due to changes in HTML5 Blob.slice specification.
- Update to a newer version of eulcore for revised default pid logic; include the object label as a pid name when generating a new ARK for an object pid.
- Added a log message when ingesting new audio objects so that an audit trail mesage will be saved in Fedora.
- Convenience short-cut search links on the home page to fixnd items uploaded today, yesterday, and in the current month.
- Update to Fedora XACML policy for production environment.
Date: March 2011
Changes and fixes to Min items I-III that were needed for an initial production release.
- A curator will see a message, “Changes not saved,” if they attempt to save changes to a metadata record but the changes are not saved because of validation errors.
- When editing an audio file, drop-down selectors should default to blank, even when a nonblank selection is required.
- When a user (staff or public) downloads a file, the filename given to the downloaded file (copy of archival master, or derived use copy format) will be based on the "noid" part of the PID number.
- When a user searches for audio items, the “Rights” field should search the numeric rt:accessStatus/@code and should be called “Rights code.”
- A curator has an easy way (within a single-click or so) to find the most recently created items (new uploads) so that they can easily perform initial metadata on them. (Reversing the current item default search, which returns all items in chronological order, is expected to be an easy way to provide this.)
- The Collection metadata edit form will make field labels and field contents visually distinct to the user by use of different fonts, in the same way that the Audio File edit form does.
- Users creating or editing Collection metadata will not be required to add a Name element. If a Name element is created, adding a Role and/or Role term will be optional as well.
- A curator will have sufficient room to input and view a fairly long title for an item.
- A curator or archivist can select from revised rights access status codes (numeric), displayed with mnemonic abbreviations.
- A curator or archivist can add an "IP Note" field to Rights metadata.
- An archivist can over-ride an access status code that would grant access to the public to digitized file through the MARBL Kiosk, by selecting a checkbox in the form. Checking the box will cause "Deny access" to appear in red letters next to the field.
- A curator can choose from revised selections for the Source Tech Housing field; new choices are jewel case; plastic container; paper sleeve; cardboard sleeve; cardboard box; other; none.
- A curator can choose from 2 additional selections for Source Technical - Reel Size: "not applicable", and " 4" " (four inches) and the field will no longer be required.
- A curator can choose from revised selections for Source Technical - Recording Speed: aspect term "cylinder disk" should be changed to "phono cylinder" in the drop-down list and in the metadata.
- A curator will no longer be required to enter metadata in the Source Technical Sublocation field.
- A curator will no longer be required to enter metadata in the Digital Technical Digitization Purpose field.
- The Digital Technical Transfer Engineer field will no longer be a required field (until non-LDAP choices are available).
- Change label for Collection search result column from MSS# to Col. No., to better reflect meaning for users.
- Change label for Collection search result column from "Collection" to "Repository" to better reflect its revised meaning.
- In Collection search results, if there is no mods:title element (or it is empty) in a record retrieved by the search, users will see “(no title present)” as a hyperlink to the record, so that they can access the record to add a title.
- Metadata specialists and archivists will have access to links to view MODS, DC and RELS-EXT datastreams at the top of the Collection metadata editing form so that they can view the XML for Collection objects.
- When a user creates a new Collection object, the object will be available within a short time (less than 2 minutes) in drop-down selections for the Audio Files Search by Collection and Audio Files Edit, Collection choice.
- A user can search for Audio Files by Date Uploaded in order to enable date-based report generation. The Audio Files input box for Date Captured should no longer appear to users, as it is no longer needed.
Date: February 2011
NOT FOR PRODUCTION RELEASE
Digital technical metadata and rights metadata for audio items; automated access-copy audio file generation; support for batch upload of large files.
- An authenticated user can log out of the Euterpe interface from any screen.
- A curator can use a web form to associate a digitized audio file with basic “stub” digital technical metadata.
- A curator can use a web form to associate a digitized audio file with basic “stub” rights metadata.
- When a new audio file is uploaded, the system automatically generates access copies in mp3 format to support kiosk access.
- Web users can listen to uploaded audio, linked from both the metadata view/edit views and search results.
- A researcher can use the MARBL A/V kiosk to search metadata in the system and listen to the audio.
- A curator can upload large files via drag & drop batch upload.
- Project rebranded as "The Keep".
Date: December 2010
NOT FOR PRODUCTION RELEASE
Minor enhancement to search functionality, use of ARKs for Fedora object pids, and audio items now contain source technical metadata.
- A user searching for collections by fields other than Manuscript Number will not have the default “MSS” in that box interfere with their search.
- The search interface contains a tool tip with documentation for case sensitive and wildcard searching for both collection and item search.
- Web users can search for stub records by keyword, associated manuscript collection, and date created.
- Web users can select file records from search results to view or edit file metadata.
- A curator can use a web form to associate a digitized audio file with basic “stub” source technical metadata.
- Web users can identify files by ARK in both the metadata view/edit views and search results so that they can easily reference these ARKs in external systems.
- Web users searching for files can see a count of matching records for simple report generation.
- Web users navigating to the collection browse page see the page load in under 5 seconds.
Date: December 2010
NOT FOR PRODUCTION RELEASE
Support for audio file uploads and basic descriptive metadata for for newly digitized sound recordings.
- A curator can ingest a batch of digitized audio files so that he doesn’t have to pause his workflow for several minutes for each one to upload individually.
- When a curator ingests audio files, additional metadata is generated from the file’s content and stored in reposited metadata to maintain accurate records.
- A curator can use a web form to associate a digitized audio file with basic “stub” descriptive metadata.
- A system administrator deploying the application can run a script to create a pre-selected list of collection objects based on the corresponding Finding Aids EAD XML for those collections.
Date: October 2010
NOT FOR RELEASE TO PRODUCTION
Support for basic, production-ready user interface for adding and managing the Fedora digital collection objects that will ultimately contain digital master items.
- An editor can create a collection object, associate it with a top-level collection, and enter basic initial metadata, so that the collection can be described and assigned objects.
- An editor can edit complete initial metadata so that the collection MODS can describe the full range of available metadata.
- An editor can update descriptive metadata for a collection to keep collection metadata up-to-date.
- An editor can search for a collection by title, manuscript number, creator, and top-level collection to locate one for editing or to check if a particular collection exists before creating it.
- An editor can view a hierarchical list of collections to locate one for editing or to understand the organization of collections.
- When any user creates or modifies a collection, the repository permanently associates that action with the user for preservation and auditing.
- When an editor saves changes on a collection, they can choose to continue editing or return to the default view.
Date: September 2010
NOT FOR RELEASE TO PRODUCTION
Prototype system that includes the simplest implementation of a metadata editor interface and content ingest. This includes a simple content model, ingest of a single sound file, and simple indexing, and uses LDAP authentication for library staff.
- Users can log into the application with their Emory User ID so they can be authorized if appropriate.
- Admins can assign roles and permissions to users to maintain security and workflow in the application. (built-in Django functionality)
- Editors can upload and ingest a sound file in wave format so metadata can be created for the ingest item.
- Editors can add or edit metadata to an ingested sound file from a selected mods subset to describe the sound file.
- Editors can search ingested content by PID or Title so they can find an item to modify or create metadata.
- Editors can download ingested audio files for review to assist them in creating metadata.
- Editors receive error messages pertaining to metadata validation when editing records to ensure data quality and consistency.
- The application logs and displays error messages related to interaction with Fedora for troubleshooting and communication.
- Developers can create django forms related to XML objects to ease the development of editing interfaces.
- Developers can link an XML Object with an XML schema so objects can be validated.