Repository navigation
Conversation
8d34237 to
c5ce03d
Compare
Documentation build overview
14 files changed ·
|
| result = await self.conn.execute(stmt) | ||
| return list(result.scalars().all()) | ||
|
|
||
| async def get_feeder_columns( |
There was a problem hiding this comment.
Can potentially be replaced by get_transformation(), from:
https://github.com/DIRACGrid/diracx/pull/1150/changes#diff-5a1e1a2b10303a0dc725d87e13772f68eb8f33b335591de07109bb83ec806fd9R95
It comes with some extra information, but I don't think it matters for the amount of transformations that would be retrieved. Thoughts?
c5ce03d to
ec25db6
Compare
|
GitHub was lagging behind with my last commit, so I closed/reopened it to try to resolve that (which it did). |
There was a problem hiding this comment.
To make it consistent with the other db/sql/<domain> directories, could you move this function to db.py please?
db.py might grow fast and become very large at some point.
When that would happen, if it happens, we should reconsider the structure and make it consistent across all the <domain> (auth, resource_system, transformation) we have I think.
| "were not Unassigned", | ||
| ) | ||
|
|
||
| async def list_feedable_transformation_ids(self) -> list[UUID]: |
There was a problem hiding this comment.
Shouldn't you reuse
There was a problem hiding this comment.
Same comment everywhere you are using UUID directly
There was a problem hiding this comment.
Haa, definitely. I will have it updated very soon. Thanks
|
one of the unit tests in pytest-integration failed, apparently for: I will try to retrigger that pipeline. |
a few more comments to add actually, sorry (coming soon)
Partially resolving #1082 (part 2/5)
Blocked by #1144
Superseds #1143
This pull request contains
diracx-db-only changes. In summary:Unassignedinputs intoInputstable_input_hash()function to hash LFN/Descriptor input content - as discussed in MM and in feat(core): provide hashing module for content-addressed and input objects #1154Question: I feel like the
diracx-db/src/diracx/db/sql/transformation/db.pymodule is becoming too large, with specifics of feeder, packer and probably others to come. I'd suggest breaking it similarly to how we did underdiracx-core, where packer and feeder would have their own python module. Any thoughts?