Releases: procrastinate-org/procrastinate
Releases · procrastinate-org/procrastinate
0.8.0: Concurrent tasks and administration shell
Breaking changes
- None \o/ (that we know of)
Migrations
Features
- Refactor builtin tasks to use pass_context #205
- Real concurrent asynchronous tasks #206
- Implement an administration prompt #204
Fixes
- Worker should instantiate Event lazily, and should use wait_for, not wait #202
- Insert event after defer job #184
Tests
- Activate test_lock #210
- Add a schema migration test #209
- Instrument test_lock in case it fails again #214
- Simplify migration tests #215
Docs
- Fix connector doc & worker doc (& probably other docs too) #208
- Add migration tests to contributing doc #211
- Fix typo in locks howto #216
Misc
- Remove extraneous spaces in license file #218
0.7.1: Fix migration script
Fixes
Change CREATE FUNCTION
to CREATE OR REPLACE FUNCTION
in the delta_0.6.0_001_fix_procrastinate_fetch_job.sql
migration script.
0.7.0: Cleaning Postgres Connector
Migrations
Features
Postgres Connector
Breaking compatibility
Now, to setup your postgres connector you can do:
connector = procrastinate.PostgresConnector(
dsn="postgres://user:password@host:port/dbname"
)
# or if you already have an aiopg pool:
connector = procrastinate.PostgresConnector()
connector.set_pool(my_pool)
Associated PRs:
- Remove PostgresJobStore (compatibility layer) #188
- Create aiopg pool lazily #185
- Use a pool instead of a single connection, and reorganize the worker around that #173
Docker
- Adding a first version of dockerized development environment. #144
- add POSTGRES_PASSWORD to docker-compose.yml and in contributing documentation #177
Logs
Migrations
- Renaming Migrator as SchemaManager #161
- Derive schema version from migration scripts #162
- Improve our migration story #167
Documentation
- Reword migration section of contributing guide #168
- Add a "Set database schema" how-to #169
- Add a "Use Pum for migrations" how-to #170
Misc., including process changes
0.6.0: Migrations
Migrations
- Add a "Use Pum for migrations" how-to #170
- Add a "Set database schema" how-to #169
- Reword migration section of contributing guide #168
- Improve our migration story #167
- Fix builds 164 #165
- Derive schema version from migration scripts #162
- (internal) Renaming Migrator as SchemaManager #161
The rest
0.5.0: Handling connection
0.4.0: From Job Store to Connector
Breaking changes
App
is now initialized with a PostgresConnector
object instead of a PostgresJobStore
.
Change your code from:
app = procrastinate.App(job_store=procrastinate.PostgresJobStore(dsn="..."))
to
app = procrastinate.App(connector=procrastinate.PostgresConnector(dsn="..."))
Changelog
- Change PostgresJobStore to Pos #147
- Retry on specific exception type 84 #135
- Add name attribute to workers and print it in the logs #132
- New main contributors #138
- First monitor and corresponding cli command #129
- Custom JSON encoders and decoders #131
- Typo fix and a missing sentance in the doc. #136
- Set enable_uuid to False when calling aiopg.connect #134
- Fix bug in get_connection #130
- Make two tests in test_testing async #128
- Use asyncio.TimeoutError #125
- Quick improvement of logic flow #121
- Remove the Task name/full_path warning #123
- Fix/improve quickstart #120
- Fix "docker-compose" in contributing doc #122
- Add @sophie-ulti to the codeowners #116
- add *.pyc to gitignore #115
0.3.0: We're going async !
- Don't overwrite psycopg2 adapters when creating the connection (#113)
- A pass on the contributing doc (#109)
- Procrastinate's core has become async, and it's now quite easier to support sync and async projects (#96)
- Documentation spell checker integrated in the project (#77)
- Fix race conditions (#104)
- Improve SQL schema (#107)
0.2.1: Metadata update
- Fixing long description on PyPI (#97)
0.2.0: Ground work
- Launch tasks when they're corountines (but still sequentially) (#93)
- Make it possible to launch tasks not defined in the project (#91)
- Add apt install of postgresql-client in the contributing page (#90)
- Fix multiple doc issues (#89)
- Query system. Now the connector is agnostic of the queries it runs (#83)
- Renaming things (#81)
- All sql queries in an SQL file, and a system to turn a query into a function (#71)
- Add Codeowners and PR template (#73)
- Add more granular tests to cli, and fix tests that actually didn't test anything (#69)
- Save job events in the DB (#70)
- Add linear and exponential backoff (#64)
- Merge pull request #63 from peopledoc/split-howto
- Wording, documentation & doc review, typos (#58, #72, #74, #75, #76, #78)
0.1.1: Documentation, packaging, licensing
- Documentation, packaging, licensing