D089 - A reverse hash search takes its vocabulary by path

Status: decided Date: 2026-09-26

Finding a name for an import hash is examples/name_nid, which hashes every word of a vocabulary file the caller passes, searches a raw value in both byte orders, and prints the encoded form of each reading.

Why: the hash is one-way, so the vocabulary is the whole answer. The mined identifier corpus belongs to obSCEne; a tool carrying its own word list would be a second copy of it. Projects in the collection print the same identifier in both byte orders.

Rejected:

• A built-in vocabulary: a drifting copy of another project's corpus.
• Searching one byte order: silently misses half the callers.