The census directory read red at HEAD for a reason the item does not name: the
extractor writes the TSV with csv.writer(delimiter='\t') (QUOTE_MINIMAL), and the
CHECK read it back with line.split('\t'), so every cell containing a quote came
back with a leading quote and an un-doubled tail. 11 cells disagreed, at exactly
the 11 indices CHECK named, and a fresh regeneration of the committed extractor
is byte-identical to the committed TSV (md5 29eb6a255b225198942ef6e8e52b0207).
check_workqueue_census.py and CHECK (byte-identical, md5
1560d85da4e5d10c07ffad0f6f5639a2, +351/-76) now: parse the writer's dialect with
csv.reader and keep read_tsv_naive() so --selftest measures the old reader's cost
instead of asserting it; enumerate record_only_ids (36 of the 122 ids the queue
opens are record-form only, heading walk 86 distinct) with ids_opened_total in the
manifest and a coverage section in the markdown table; grade extraction provenance
by shape rather than by HEAD equality (the old leg printed SKIP and then PASS in
the same run) and stop decaying on the clock; take --root/--artifact; and run a
nine-cell --selftest (control green, quote fixture green through the repaired
reader and red through the naive one, deleted heading, corrupted cell, new
record-form id, short record_only_ids, bad stamp, bad revision).
Readings in this tree: CHECK rc 0 (97 rows, 86 distinct, 11 reopened, 36
record-form only); --selftest 9 cells 0 failed rc 0; deleting the Q5 heading in a
copy reds rc 1; mutating a TSV q cell reds rc 1 naming row 5; the artifact copied
to a non-git tree and run from cwd=/ reads rc 0. Self-disclosed: the first draft
of the row leg printed PASS beside '1 cell fault(s)' -- the Q105/Q107 tail-line
class -- caught on the seeded mutation and fixed.
Also lands the Q119 DONE record, whose artifacts are the commit above.
The Q119 DONE record shipped at 2615c32 without its artifacts: classify.py and
verify.sh were modified and test_classify_fallback.py untracked in the worktree,
so every path the record names read MISSING at the tree about to be committed
(path-claims leg: 2 new rows). The three files land here byte-for-byte as the
record pins them -- classify.py md5 89a5b9d3476265069a0bf2694af9510e (162 lines),
verify.sh md5 4747faef183017c79620d3daf5d918c3 (105 lines),
test_classify_fallback.py md5 4bc2b53d4bcde155616d4741c708bc4a (159 lines) --
and bash wizard-work/sol/verify.sh reads CHECK OK, Ran 10 tests OK, rc 0 in this
tree. Not claimed: that anything in Q119's grading changed; this is its landing.
test_shell_lone_copy_reds_on_its_sibling asserted the fixture path it had just
built while verify_hermes_setup.sh reports the path it resolved with readlink -f.
Under TMPDIR=/root/.hermes/profiles/volt/cache/scratch_link a correct pair read
MISSING and the gate went rc 1 (measured before the repair: that cell was the
only red, 15 others ok). The fixture root is now canonicalised once in setUp
with os.path.realpath; two new cells pin the alias reading instead of assuming
it -- a lone copy reached by a directory symlink names the real sibling that is
absent (rc 1), and the same alias with the sibling present reads PASS (rc 0).
CHECK-env-verifiers now runs the suite twice, ambient TMPDIR and a directory
alias it builds, so the shape that produced the flake is graded every run.
16 cells; negative control (raw mkdtemp restored in a scratch copy) reds the
alias leg and only that leg, rc 1.
snippet.py: check_tools() imports hermes_tools and requires callable(getattr(...));
--require is the negative-control hook; HERMES_HOME/HERMES_PROFILE/--home/--profile
resolve every path; rc 0/1/2. verify_hermes_setup.sh: HERE from $0, sibling
snippet.py graded, --home/--root/--profile, VERDICT line, rc 0/1/2.
New test_env_verifiers.py (14 black-box cells, stub hermes_tools on PYTHONPATH,
synthetic trees) and CHECK-env-verifiers. Pre-repair blobs: 9 cells red for
snippet.py, 6 for the shell script.
beam
spark
ledger
sol
quill
volt
volt
zephyrus-sparkborn