HathiTrust Digital Library Scraper

Look up HathiTrust's 18M+ digitized volumes by ISBN, OCLC, LCCN, ISSN, HathiTrust item ID (HTID) or catalog record number — and get back clean, structured bibliographic data: titles, editions, publication dates, cross-linked identifiers, holding libraries, item-level HathiTrust IDs and US copyright rights status.

No API key. No login. No proxy required.

HathiTrust is one of the largest digital preservation repositories in the world — a shared collection of books digitized by Google, the Internet Archive and 60+ research libraries including Michigan, California, Harvard, Princeton and the Library of Congress. This actor turns its Bibliographic API into a dataset you can actually work with.

---

What you get

One row per HathiTrust catalog record, with all of its digitized items rolled up:

Field	Description
query	The identifier that was looked up (e.g. isbn:0226458075)
record_id	HathiTrust catalog record number
title	Primary title
all_titles	Every title variant on the record
isbns / oclcs / lccns / issns	All cross-linked identifiers — great for ID reconciliation
publish_dates	Publication year(s)
record_url	Link to the HathiTrust catalog record
item_count	How many digitized copies/volumes exist
htids	HathiTrust item IDs (up to 20)
item_urls	Direct page-turner links (up to 10)
rights_codes	Unique rights codes across items (pd, ic, und, pdus, …)
us_rights	Human-readable US access status (e.g. Full view, Limited (search-only))
source_libraries	Which institutions contributed the scans
last_update	Most recent item update, ISO-8601
marc_xml	Full MARC-XML record (only when Fetch full records is enabled)
source, scraped_at	Provenance

Sample output

{
  "query": "isbn:0226458075",
  "record_id": "003112209",
  "title": "The structure of scientific revolutions",
  "isbns": ["9780226458083", "9780226458076"],
  "oclcs": ["34548541"],
  "lccns": ["96013195"],
  "publish_dates": ["1996"],
  "record_url": "https://catalog.hathitrust.org/Record/003112209",
  "item_count": 1,
  "htids": ["uc1.31822031154305"],
  "item_urls": ["https://babel.hathitrust.org/cgi/pt?id=uc1.31822031154305"],
  "rights_codes": ["ic"],
  "us_rights": "Limited (search-only)",
  "source_libraries": ["University of California"],
  "last_update": "2013-08-10",
  "source": "hathitrust"
}

---

Input

Field	Type	Default	Description
mode	select	lookup	Resolve identifiers into volume records
idType	select	isbn	isbn, oclc, lccn, issn, htid or recordnumber
ids	array	—	The identifiers to look up (all of the chosen type)
useFullRecords	boolean	false	Fetch full records including MARC-XML
maxItems	integer	200	Cap on dataset rows (max 2000)

{
  "mode": "lookup",
  "idType": "isbn",
  "ids": ["9780195079630", "0226458075"],
  "useFullRecords": false,
  "maxItems": 200
}

---

Pricing

Roughly $0.5 per 1,000 volumes. The actor packs 20 identifiers into a single API request using HathiTrust's multi-lookup syntax, so a 1,000-ISBN list is just 50 HTTP calls — fast and cheap. Runs on 512 MB.

---

Use cases

• Library metadata enrichment — take a shelf list or acquisitions file of ISBNs and enrich it with authoritative catalog data, OCLC/LCCN cross-references and holdings counts.
• Book research — check which scholarly editions and printings exist for a work, when they were published, and which institutions hold them.
• Digitization coverage checks — before scanning a collection, find out which titles are already digitized in HathiTrust and by whom, so you don't duplicate effort.
• Rights clearance — bulk-check US copyright status (pd public domain vs ic in-copyright) across a catalog to determine what can be opened up to full view.

Also useful for ID reconciliation between ISBN, OCLC and LCCN namespaces, and for building union-catalog datasets.

---

Notes & limitations

• Coverage is strongest for academic, scholarly and older public-domain titles. Recent trade paperbacks are frequently absent — an identifier with no HathiTrust holdings returns an empty result and is skipped (and counted in the log).
• All identifiers in a single run must be of the same type. Run the actor twice if you have a mixed set.
• The actor reads bibliographic metadata only. It does not download page images or full text.
• The run fails loudly (RuntimeError) if no identifier matched anything — that usually means a wrong idType.

---

Related actors

• Open Library Book Scraper — editions, authors and covers from Open Library
• Google Books Scraper — Google Books search and volume metadata
• Project Gutenberg Scraper — 70k+ free public-domain ebooks
• Crossref Scraper — DOI and scholarly publication metadata

---

Legal

This actor reads HathiTrust's public Bibliographic API, which is provided for programmatic access to catalog metadata and requires no authentication. Only bibliographic metadata is collected — no copyrighted page content. Respect HathiTrust's terms of use and keep request volumes reasonable.