Migrate to Python 3.11 following udata
dependencies upgrade #249
Switch harvest backend to new harvest sync system #251
Update constants imports from .models
to .constants
#250
Use datetime.utcnow
to make sure to handle utc datetimes #246
Remove valid_at
on deprecated GeoZones #248
Use id as remote_id ASAP for better error handling #239
Update mongoDB in CI #242
Refactor tests to mock CKAN results instead of starting a CKAN instance #245
Upgrade CKAN images used in docker-compose #244
Fix resource.published that is not needed anymore #243
⚠️ Breaking change Use harvest dynamic field introduced in udata 5 #227
Replace mongo legacy image in CI #219
Make revision_id optional to match latest CKAN versions #220
Migrate to python3 🐍 #110
Always fill extras.remote_url #103
Max out package_search rows limit #100
Fix filetype (always remote) #98
Support both inclusion and exclusion filters #42
Localization support #43
Test the minimum accepted CKAN dataset payload and make the extras
property optional #57
Improved error handling (support details in JSON responses, also handle raw quoted strings and HTML) #56
Only store url
field in remote_url
extra if this is an URL otherwise store it in ckan:source
#30
Properly handle geometry errors #31
Improve extras parsing #32 :
Skip empty extras
Parse update frequencies as RDF URI or udata frequency identifier
Parse spatial-text
matching a known zone name or slug
Store unknown spatial-uri
, spatial-text
and frequency
as ckan:spatial-uri
, ckan:spatial-text
and ckan:frequency
Test against a real CKAN instance #23
Allows to filter on Organizations and Tags #26
Register ckan:
prefixed extras #28
Fix packaging #2
Make use of udata pytest plugin #3
Enable the ckan
plugin in test (plugin needs to be enabled to use the harvester) #8
You can’t perform that action at this time.