[PATCH 0/2] hashserv: tighten unihash handling

This series tightens unihash handling so hashserv accepts and stores only
valid lowercase 64-character hexadecimal unihash values, and BitBake treats
BB_UNIHASH as literal hash data when reading it from the datastore.

The first patch updates existing hashserv test fixtures to use valid
64-character unihash-shaped values.

The second patch adds unihash validation and focused coverage for hashserv
report paths and BitBake siggen handling. Invalid unihashes reported to
hashserv, or returned by hashserv to BitBake, are rejected before use.

Validation:
  - py_compile on touched Python files
  - focused hashserv/siggen tests
  - broader hashserv and siggen regression tests

Anders Heimer (2):
  hashserv/tests: use valid 64-character unihashes
  hashserv: validate unihash values

 lib/bb/siggen.py         |   9 +-
 lib/bb/tests/siggen.py   |  47 ++++++++++
 lib/hashserv/__init__.py |   7 ++
 lib/hashserv/server.py   |  27 +++++-
 lib/hashserv/tests.py    | 194 ++++++++++++++++++++++++++++-----------
 5 files changed, 224 insertions(+), 60 deletions(-)

-- 
2.34.1