At a glance
DocAccessible groups successfully retrieved PDFs using the SHA-256 digest of their bytes. Two different download URLs can therefore belong to one identical-file group. Similar wording, matching filenames or the same page appearance do not establish a duplicate under this method.
Why we compare file bytes
A URL identifies a retrieval location. A hash summarizes the bytes received from that location. Comparing bytes handles the common case where departments post the same file in several folders without requiring matching paths or names.
It also makes the rule inspectable: the report can retain the source URL, file size, digest and observed count. A later download can change even when its URL stays the same, so the group describes the retrieved revision at the time of the scan.
Some apparent duplicates are different files
A metadata edit, re-export, digital signature or different compression can produce different bytes while leaving much of the visible content unchanged. Those files remain separate under exact-byte deduplication. That conservative boundary avoids deciding that two legal or operational revisions are interchangeable.
If you want to consolidate near-duplicates, inspect dates, signatures, identifiers and intended uses. Keep a decision record and speak with the document owner before replacing a link. The inventory neither deletes files nor rewrites your website.
- Same name is not enough: compare the actual revision.
- Same page count is not enough: different content can have equal length.
- Unknown retrieval means unknown identity, not a new verified file.
Use duplicate groups without losing the link footprint
For a remediation estimate, count an identical source once if the same approved output can serve every intended use. Keep all public URLs so the publishing team can update each location after review. An identical source does not prove that surrounding link labels or publication contexts are usable.
The duplicate workload calculator lets you compare confirmed pages across URLs with confirmed unique pages. It measures repeated processing you might avoid under that assumption. It does not discover duplicates itself or promise a saving.
| Measure | Value | Use |
|---|---|---|
| Unique files | 1 | Source work items |
| Unique pages | 10 | Work once per byte-identical source |
| Pages across URLs | 40 | Public download footprint |
Scroll sideways to see all columns. Use arrow keys when the table is focused.
Sources and scope
Methods, worksheets and illustrative examples are DocAccessible editorial contributions. These references support the technical or product details described above. W3C techniques are informative guidance; they are not a certification scheme.
- DocAccessible PDF inventory
The counting workflow, report options and purchase boundary.
- Python hashlib documentation
SHA-256 hashing of file bytes.
Written by the DocAccessible team under our editorial policy. Suggest a correction if a source or product detail needs updating.
See the evidence in a working example
Explore the public sample inventory report to try its table, full-screen view and missing-count reasons. The downloadable counting fixture check records what the local counter returned for seven self-authored inputs, with hashes and explicit limits.