Fwd: New release of nikita. Version 0.6

Thomas Sødring tsodring at oslomet.no
Sat Jun 12 08:23:18 CEST 2021


Jeg ble tipset at kanskje noen på denne listen kunne være interessert i 
arbeidet vårt. Ved OsloMet har vi hatt et flerårig prosjekt for å 
realisere en fri programvare implementasjon av den tekniske varianten av 
Noark standarden 
(https://github.com/arkivverket/noark5-tjenestegrensesnitt-standard). Vi 
har kommet til versjon 0.6 og selv om produktet er fortsatt under 
utvikling har den flere gode bruksområder. Dersom du du liker å leke 
litt med teknologi  og har teknisk interesse i noen av følgende er 
nikita verdt en titt:

  * standardisering

  * arkiv / dokumentasjonsforvaltnings metadata

  * metadata søk med OData standarden

  * Nasjonale identifikatorer

  * utvikling av virksomhetsspesifikkeMetadata

  * interoperabilitet

  * fri programvare


I den forbindelse deler jeg kunngjøringen om siste versjon med dere. 
Dersom du er interessert er kildekoden et godt sted å starte 

  - Thomas

-------- Forwarded Message --------
Subject: 	New release of nikita. Version 0.6
Date: 	Thu, 10 Jun 2021 16:40:44 +0200
From: 	Thomas Sødring <tsodring at oslomet.no>
To: 	nikita-noark at nuug.no <nikita-noark at nuug.no>


It is with great pleasure that we can announce a new release of nikita. 
Version 0.6 (https://gitlab.com/OsloMet-ABI/nikita-noark5-core). This 
release makes new record keeping functionality available. This really is 
a maturity release. Both in terms of functionality but also code. 
Considerable effort has gone into refactoring the codebase and 
simplifying the code. Notable changes for this release include:

* Significantly improved OData parsing

* Support for business specific metadata and national identifiers

* Continued implementation of domain model and endpoints

* Improved testing

* Ability to export and import from arkivstruktur.xml

We are currently in the process of reaching an agreement with an archive 
institution to publish their picture archive using nikita with business 
specific metadata and we hope that we can share this with you soon. This 
is an interesting project as it allows the organisation to bring an 
older picture archive back to life while using the original metadata 
values stored as business specific metadata. Combined with OData means 
the scope and use of the archive is significantly increased and will 
showcase both the flexibility and power of Noark.

I really think we are approaching a version 1.0 of nikita, even though 
there is still a lot of work to be done. The notable work at the moment 
is to implement access-control and full text indexing of documents.

My sincere thanks to everyone who has contributed to this release!

  - Thomas

Release 0.6 2021-06-10 (d1ba5fc7e8bad0cfdce45ac20354b19d10ebbc7b)

* Refactor metadata entity search
* Remove redundant security configuration
* Make OpenAPI documentation work
* Change database structure / inheritance model to a more sensible approach
* Make it possible to move entities around the fonds structure
* Implemented a number of missing endpoints
* Make sure yml files are in sync
* Implemented/finalised storing and use of
     - Business Specific Metadata
     - Norwegian National Identifiers
     - Cross Reference
     - Keyword
     - StorageLocation
     - Author
     - Screening for relevant objects
     - ChangeLog
     - EventLog
* Make generation of updated docker image part of successful CI pipeline
* Implement pagination for all list requests
     - Refactor code to support lists
     - Refactor code for readability
     - Standardise the controller/service code
* Finalise File->CaseFile expansion and Record->registryEntry/recordNote 
* Improved Continuous Integration (CI) approach via gitlab
* Changed conversion approach to generate tagged PDF documents
* Updated dependencies
     - For security reasons
     - Brought codebase to spring-boot version 2.5.0
     - Remove import of necessary dependencies
     - Remove non-used metrics classes
* Added new analysis to CI including
* Implemented storing of Keyword
* Implemented storing of Screening and ScreeningMetadata
* Improved OData support
     - Better support for inheritance in queries where applicable
     - Brought in more OData tests
     - Improved OData/hibernate understanding of queries
     - Implement $count, $orderby
     - Finalise $top and $skip
     - Make sure & is used between query parameters
* Improved Testing in codebase
     - A new approach for integration tests to make test more readable
     - Introduce tests in parallel with code development for TDD approach
     - Remove test that required particular access to storage
* Implement case-handling process from received email to case-handler
     - Develop required GUI elements (digital postroom from email)
     - Introduced leader, quality control and postroom roles
* Make PUT requests return 200 OK not 201 CREATED
* Make DELETE requests return 204 NO CONTENT not 200 OK
* Replaced 'oppdatert*' with 'endret*' everywhere to match latest spec
* Upgrade Gitlab CI to use python > 3 for CI scripts
* Bug fixes
     - Fix missing ALLOW
     - Fix reading of objects from jar file during start-up
     - Reduce the number of warnings in the codebase
     - Fix delete problems
     - Make better use of cascade for "leaf" objects
     - Add missing annotations where relevant
     - Remove the use of ETAG for delete
     - Fix missing/wrong/broken rels discovered by runtest
     - Drop unofficial convertFil (konverterFil) end point
     - Fix regex problem for dateTime
     - Fix multiple static analysis issues discovered by coverity
     - Fix proxy problem when looking for object class names
     - Add many missing translated Norwegian to English (internal) 
attribute/entity names
     - Change UUID generation approach to allow code also set a value
     - Fix problem with Part/PartParson
     - Fix problem with empty OData search results
     - Fix metadata entity domain problem
* General Improvements
     - Makes future refactoring easier as coupling is reduced
     - Allow some constant variables to be set from property file
     - Refactor code to make reflection work better across codebase
     - Reduce the number of @Service layer classes used in @Controller 
     - Be more consistent on naming of similar variable types
     - Start printing rels/href if they are applicable
     - Cleaner / standardised approach to deleting objects
     - Avoid concatenation when using StringBuilder
     - Consolidate code to avoid duplication
     - Tidy formatting for a more consistent reading style across 
similar class files
     - Make throw a log.error message not an log.info message
     - Make throw print the log value rather than printing in multiple 
     - Add some missing pronom codes
     - Fix time formatting issue in Gitlab CI
     - Remove stale / unused code
     - Use only UUID datatype rather than combination String/UUID for 
     - Mark variables final and @NotNull where relevant to indicate 
* Change Date values to DateTime to maintain compliance with Noark 5 
* Domain model improvements using Hypersistence Optimizer
     - Move @Transactional from class to methods to avoid borrowing the 
JDBC Connection unnecessarily
     - Fix OneToOne performance issues
     - Fix ManyToMany performance issues
     - Add missing bidirectional synchronization support
     - Fix ManyToMany performance issue
* Make List<> and Set<> use final-keyword to avoid potential problems 
during update operations
* Changed internal URLs, replaced "hateoas-api" with "api".
* Implemented storing of Precedence.
* Corrected handling of screening.
* Corrected \_links collection returned for list of mixed entity types 
to match the specific entity.
* Improved several internal structures.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.kunnskapsallmenning.no/pipermail/biblioteknorge/attachments/20210612/ba6b1a61/attachment-0001.htm>

More information about the Biblioteknorge mailing list