DRIVERS-2975 Fix logic for populating BulkWriteException.partialResult
#1665
Conversation
isabelatkinson
changed the title
BulkWriteException.partialResult
| newDocument: &newDocument { _id: 2, x: 22 } | ||
|
|
||
| tests: | ||
| - description: "partialResult is unset when first operation fails during an ordered bulk write (verbose)" |
There was a problem hiding this comment.
Verbose and summary are both tested here to make sure that drivers aren't determining the index of the write error by its position in the cursor rather than the idx field. The type of result is not relevant for the unordered logic, so I skipped testing both for the equivalent unordered test.
| document: *newDocument | ||
| ordered: true | ||
| verboseResults: true | ||
| expectError: |
There was a problem hiding this comment.
The other pieces of BulkWriteException are tested elsewhere, so I skipped including them here.
isabelatkinson
requested review from
nbbeeken,
kevinAlbs and
stIncMale
and removed request for
a team and
nbbeeken
There was a problem hiding this comment.
LGTM with a suggested change to test assertions. With suggested changes, verified tests pass in C driver locally.
| insertedId: 2 | ||
| updateResults: {} | ||
| deleteResults: {} | ||
| - description: "partialResult is set when first operation fails during an unordered bulk write (summary)" |
There was a problem hiding this comment.
@isabelatkinson: FYI, this is a duplicate test name (also used earlier on line 183). Assuming that was unintentional, what should this test actually be called?
Asking as this came up in mongodb/mongo-php-library#1429 /cc @alcaeus
There was a problem hiding this comment.
bad copy/paste, fixed here
Please complete the following before merging:
clusters, and serverless).