Manuscript Index Builder

How to use Manuscript Index Builder

Prepare a subject index from references you have already verified in a manuscript. Main entries and one level of subentries keep their source-record evidence; overlapping or adjacent page intervals merge without expanding long ranges into arrays. Exact cross-references are checked before the tool produces printable backmatter.

1. Supply main/subterms, canonical decimal page references and exact existing main-term cross-reference targets.
2. Choose first-occurrence order or the current runtime English collation.
3. Review resolved references and source-record evidence, then download and verify the index after final pagination.

Good to know

• At most 100,000 CSV characters, 500 source records and 200 main terms. Main/subterms and reference targets at most 100 well-formed characters with no outer whitespace. Exactly term,subterm,pages,see,see_also headers are required. Downloaded document text refuses control characters other than tabs and line endings.
• Page lists use positive canonical decimal pages 1–999,999 or inclusive a-b ranges separated by commas; at most 30 intervals and 600 characters per record. No Roman numerals, prefixes, omitted range digits, page extraction or automatic pagination is inferred.
• Repeated exact main/subentry records merge page intervals and retain all record numbers. One subentry level is supported. Subentries need pages and cannot carry references; a missing parent row is created as a heading for its children.
• A see redirect cannot contain pages, see-also targets or subentries. Targets must be declared main terms; missing/self references and see cycles are refused. See-also cells are JSON string arrays. English collation uses runtime Intl data with an exact fallback, not a certified publishing style.
• Processing is local. Output is limited to 1,000,000 text characters unless a stricter file limit is stated.