Mica2 Upgrade to Java 21

[kazoompa]

Mica Java 21 Upgarde

The port is completed but needs a thorough QA some listed below. There are some dependencies that are minor but also upgraded to Java 21:

• Mica search plugins
• Datasource plugins
• obiba-commons
• magma

Test Cases

Mica Seeding

• Seed mica via Obiba-Home (java21 branch)

Authentication/Authorization

• ini realm authentication
• agate realm authentication (signin/signup)
• oidc realm authentication (signin/signup)
• role based authorizations
• acl based authorizations
• public vs. secured REST resources (authz annotations)

Mica Documents

• Perform CRUD operations on all types of Mica documents
  • Study / Population /DCE
  • Dataset
  • Initiative
  • Protocol
  • Network
• Revision flow (Under Review, Publish, Unpublish)
• Revision view and restoring
• View published versions in the public pages
  • Study
  • Datasets
  • Inititive
  • Protocols
  • Network
• Import Studies / Initiatives from another Mica
• Change the order of Study populations and DCEs
• Annotate Study (datasets and its study must be indexed in that order)

Made a change so only editors and reviewers can annotate)

• Comment any of the documents

No Opal

• Create a triad of Study-Dataset-Variables based
  • XLSX file
  • SAV file
• For the above example add taxonomies from a file

Document Permissions

• Use one of agate test user1/2/3 and grant them different permissions

Need to fix navigation to admin UI when user has the proper permissions

Document Files

• Upload files
• CRUD (add label and description)
• Publish and make sure the public pages show the files
• Add/Remove permissions

Search

• Listing pages with stats
• Search Pages (Individuals / Initiatives)
• Coverage (Study / DCE / Dataset)
• Downloads
• Add to Cart or Lists
• Graphics (special attention to table values and order)
• Navigations by clicking on stats from public detail or listing pages
• Both mica-search-es and mica-search-es7 plugins

Cart / Lists

• Creating a list from a list
• Deletion
• Downloads and reports
• Searching

Data Access Requests

• Perform CRUD operations on all types of requests (enable all)
• Work flow (Submit, Review, ...)
• Comments
• Private comments
• Actions
• Download
• Adding variable using Cart (do this from Cart page)

Persons (Study/Network)

• Add/Remove contacts
• Change contact order
• Persons Page
• Perform CRUD operations in the Person view

Admin Page

• Go over the pages and change configs and make sure all works (ex. currently logs page does not show the labels!!)

Cache

• Integrity (need to view content to make sure everything is properly serialized)
• Deletion (make sure the entries are gone)

Notes

• We can update OBiBa Protobuf provider to ignore unknownFields. It looks like proto3 is more picky and does not like when there are unknown fields. This case can happen when the client code adds fields to an entity JSON for UI purposes and does not cleanup the DTO before sending it back to the server. For now, the Mica client (angularjs) has been tested and augnmented for cleanups.

• The cache configuration is not completely ported, there are a few settings I could not port completely... yet!

• Test Mica-Python-Client

[rwissa]

Had to update spring library to 3.3 because previous version was already at end of life.
implication: need to use latest elasticsearch version 8. implementation done, now being tester RHA.
Need ok from YM to do release.

[rwissa]

release candidate created.
Il va y avoir des changements dans le API de Mica.
Tant que Opal n'est pas fini et released, on peut pas release Mica.

[rwissa]

dépendance elastic search, YM veut remplacer (affecte pas MR)