9bac0f3009
- Add validate-corpus target to Makefile (calls scripts/validate-corpus.sh) - Regenerate corpus manifest with 1000 synthetic PDF entries - Validate corpus integrity: 1000 files, 10,590 pages, ~6.9 MB - Update tests/fixtures/grep-corpus/README.md with complete workflow documentation - Document Makefile targets (validate-corpus, download-grep-corpus) - Add manual usage instructions for advanced users - Verify complete end-to-end workflow: generation → manifest → validation Acceptance criteria: - ✓ make validate-corpus runs successfully - ✓ Manifest contains 1000 entries with all required fields - ✓ Validation confirms corpus meets size/count targets - ✓ README.md documents workflow - ✓ All child beads' outputs integrated Closes bf-2kre2
123 KiB
123 KiB
| 1 | # grep-corpus manifest |
|---|---|
| 2 | # |
| 3 | # This file documents the metadata and provenance of each PDF in the corpus. |
| 4 | # Used by the benchmark to validate corpus integrity and track sources. |
| 5 | # |
| 6 | # Fields: |
| 7 | # - source_url: URL where the PDF was downloaded from |
| 8 | # - page_count: Number of pages in the PDF |
| 9 | # - file_size: File size in bytes |
| 10 | # - checksum: SHA256 hash of the file contents |
| 11 | # |