use class from self.__class__ when serializing BackLinks with fetched data

[alexdlukens]

This is my proposed solution for #1189. Please find the minimal code example in the issue report.

This fixes serialization in the case when BackLink data has been fetched using fetch_links=True. In this situation, the BackLink field contains a Document object, which does not have the document_class attribute.

Thanks,

[CAPITAINMARVEL]

This is my proposed solution for #1189. Please find the minimal code example in the issue report.

This fixes serialization in the case when BackLink data has been fetched using fetch_links=True. In this situation, the BackLink field contains a Document object, which does not have the document_class attribute.

Thanks,

maybe a getattr(self, "document_class", self.class) instead ?

With attribute:
try/except: 0.0754515000153333
getattr: 0.08347969996975735

Without attribute:
try/except: 0.4134541000239551
getattr: 0.08493790001375601

this is what i got with a testing script (using timeit)

[github-actions]

This PR is stale because it has been open 45 days with no activity.

[github-actions]

This PR is stale because it has been open 45 days with no activity.

[github-actions]

This PR is stale because it has been open 45 days with no activity.

[roman-right]

The fix looks correct - using self.class instead of the hardcoded Document class when serializing BackLinks with fetched data makes sense. However, the mergeability is currently marked as unknown, likely needs a rebase. Could you rebase against main?

[roman-right]

Thanks for the fix and the tests, LGTM.