SWDE — Structured Web Data Extraction
Original: https://archive.codeplex.com/?p=swde
Source mirror: https://github.com/woailaosang/swde/tree/e9b60dbbcb899d70e813e8b899b84e3540d540da
This is a fixed 800-page subset (10 pages per website), not the full 124,291-page corpus.
Original HTML bytes are unchanged. Project license is retained in LICENSE.txt; captured content retains its publishers' rights.