Different behaviors on the same website.

[BGELEX]

Retrieve metadata from the same website (journal) behaves inconsistently. For most cases, the abstract field is missing. However, when it does appear, the content is corrupted by extra characters or formatting issues. Both web translate and search translate yield the same results.

Examples of sources:
DOI: 10.1103/PhysRevLett.133.016501 (corrupted abstract retrieved)
DOI: 10.1103/PhysRevLett.133.010401 (no abstract retrieved)

I built Docker images from the source and tested them on two networks: one at my institute and another at home.

what could cause the standalone server's inconsistent behavior? Is it possible that different translators are being applied randomly?

Retrieving the abstract correctly is my primary concern, but I’m also curious about the root cause of this issue. Any discussions or insights would be appreciated. Thank you for your work on this project!

EDIT: After running in the debug mode in Zotero Client, I realized that the client's search import has the same issue. Only the web imports are correct. This then is likely an issue of translators.