Module parse 

BibTeX / BibLaTeX → BibEntry intermediate records.

Tolerant of both dialects:

• year = {2024} (BibTeX) and date = {2024-05} (BibLaTeX)
• archivePrefix + eprint (arXiv v1 convention) and eprinttype = {arxiv} + eprint (BibLaTeX)
• note = {...} / keywords = {...} / file = {...} (Zotero)

Field extraction is deliberately forgiving: a field that fails to parse (e.g., malformed date = {2024-may}) is dropped with its Err surfaced via [ParseWarning], not a hard error, so a single bad entry in a 5000-paper Zotero dump doesn’t sink the whole run.

Structs

BibEntry

Intermediate record extracted from one .bib entry. Owns its data — no lifetimes tied to the source buffer.

Enums

ParseError

Parse failure — surfaces upstream to scitadel bib import.

Functions

normalize_doi

Normalize a DOI for matching. Lowercase + strip the two common URL/CURIE prefixes that .bib files carry.

parse_bibtex

Parse .bib source into a list of BibEntry. Empty source is legal (returns empty vec), matching Zotero’s export-nothing edge.