Module parser 

Hand-rolled recursive-descent parser for the minimal BibTeX subset.

The grammar is intentionally tiny:

bibtex := ws* (entry ws*)*
entry  := '@' type '{' key (',' fields)? '}'
fields := field (',' field)* ','?
field  := name '=' value
value  := '{' .. '}' | '"' .. '"' | bare

Entry types and field names are lowercased; citation keys are kept verbatim. Brace values balance nested {} by naive counting, so {The {LaTeX} Companion} is captured whole, but their contents are stored as raw text; no TeX decoding, no @string / @preamble macro expansion, no # concatenation, no name parsing.

Structs

ParsedEntry 🔒

ParsedKey 🔒

Parser 🔒

A byte cursor over the BibTeX source. All structural delimiters (@ { } " , =) and whitespace are ASCII, so scanning byte-by-byte never splits a multi-byte UTF-8 sequence and every recorded offset lands on a char boundary.

Functions

is_bare_value_byte 🔒

Bytes allowed in a bare (unquoted, unbraced) value.

is_identifier_byte 🔒

Bytes allowed in an entry type or field name.

is_key_byte 🔒

Bytes allowed in a citation key: anything but a structural delimiter or whitespace.

parse_bibtex

Parse input as a minimal BibTeX database.