[executor-preview] Fix the executor and test it #2525
Conversation
yaelbh
added
the
executor-preview
| if isinstance(mode, (Session, Batch)): | ||
| self._session = mode | ||
| self._backend = self._session._backend | ||
| self._service = self._session.service | ||
|
|
||
| elif open_session := get_cm_session(): | ||
| if open_session != mode: | ||
| if open_session._backend != mode: | ||
| raise ValueError( | ||
| "The backend passed in to the primitive is different from the session " | ||
| "backend. Please check which backend you intend to use or leave the mode " | ||
| "parameter empty to use the session backend." | ||
| ) | ||
| logger.warning( | ||
| "A backend was passed in as the mode but a session context manager " | ||
| "is open so this job will run inside this session/batch " | ||
| "instead of in job mode." | ||
| ) | ||
| self._session = open_session | ||
| self._backend = self._session._backend | ||
| self._service = self._session.service | ||
|
|
||
| elif isinstance(mode, IBMBackend): | ||
| self._backend = mode | ||
| self._service = self._backend.service | ||
|
|
||
| else: | ||
| raise ValueError( | ||
| "A backend or session/batch must be specified, or a session/batch must be open." | ||
| ) |
There was a problem hiding this comment.
Is this equal to what's inside get_mode_service_backend, or are there differences?
There was a problem hiding this comment.
There is one difference: if get_cm_session() is not None and mode is None, previously there was an error, now it will take the backend from the session.
SamFerracin
left a comment
SamFerracin
left a comment
There was a problem hiding this comment.
LGTM, just want to make sure that I am not missing anything in. the lines of code that you removed
| ) | ||
|
|
||
|
|
||
| class TestGetModeServiceBackend(IBMTestCase): |
There was a problem hiding this comment.
Should this be in test_base_primitive.py?
There was a problem hiding this comment.
Yes, because the function that it tests resides in base_primitive.py. I tried to move around the functions and its test in several ways, by starting new files, moving to existing utils files, etc. Everything that I did ended in circular imports that required substantial modifications to unresolve, resulting in effort and amount of modifications that were not justified.
Co-authored-by: Samuele Ferracin <[email protected]>
2 participants
Part of #2479.
This PR copies code fixes and tests not-yet-approved from the open PR #2498. I suggest to first review #2498, let me fix here in parallel according to the comments of #2498, and only then review this PR.