volt
2556dbb0aa
Measured at 67f94cc: 19 of the 20 tracked tools answered a --json target inside a nonexistent directory with an uncaught FileNotFoundError at rc 1. Only queue_append_audit.py already refused (rc 2). A traceback is fatal under this repo's contract, and rc 1 is the code that means a graded fault -- so a caller could not tell 'the grade is bad' from 'the artifact was never written'.
Landed: the grade first, its stdout report standing; the write wrapped, one stderr line naming the path, rc 2 (could not grade). gate_blame.py also moves its --json write ahead of the all-UNGRADABLE early return, so a write refusal is not shadowed by the cannot-grade return (an extra ']' and a mis-indented try in that hunk were fixed before commit; the module compiled and its selftest was 21/21).
tools/write_probe.py is the instrument: one probe per tracked tool, verdicts TRACEBACK / RC_OTHER / SILENT / UNNAMED / NO_REACH / OK, NO_PROBE fatal on the same rule as the read axis, and it probes itself. It is leg 6 of verify.py, and selftest_sweep.py now carries both a read probe and a malformed probe for it.
Verification at this commit follows in the next commit with docs/TOOLS.md and the queue record.
377 lines
18 KiB
Python
377 lines
18 KiB
Python
#!/usr/bin/env python3
|
|
"""Measure the WRITE end of every tool in tools/: `--json PATH` at a target the
|
|
host cannot write.
|
|
|
|
WHY THIS EXISTS
|
|
`selftest_sweep.py`'s read axis grades what a tool does with an input it
|
|
cannot read. Nothing graded the other end. Measured at `67f94cc`, the last
|
|
revision before the guards landed, over the 20 tools then tracked: **18
|
|
answered a `--json` target inside a nonexistent directory with an uncaught
|
|
FileNotFoundError at rc 1**; `gate_blame.py` returned rc 2 from its own input
|
|
and never reached the write (the UNNAMED class below, ungraded rather than
|
|
passed); `queue_append_audit.py` already refused (rc 2, "cannot write json").
|
|
A traceback is what this repo's contract calls fatal, and rc 1 is the code
|
|
that means a graded fault, so those 18 made "the grade is bad"
|
|
indistinguishable from "the artifact was never written". Two earlier drafts
|
|
of this paragraph said "five" and "13 of 20"; neither is what the instrument
|
|
below prints, and the number that counts is the one a reader can reproduce:
|
|
`git worktree add --detach <dir> 67f94cc && cp tools/write_probe.py
|
|
<dir>/tools/ && python3 <dir>/tools/write_probe.py --repo <dir>`. The guards
|
|
landed that day; this instrument is what proves them and what a new tool has
|
|
to pass -- including this one, which probes itself.
|
|
|
|
THE CONTRACT IT GRADES
|
|
The grade runs first and its stdout report stands. The write is wrapped, so a
|
|
target this host cannot write prints one line naming the path on stderr and
|
|
returns 2 -- "could not grade": the artifact was asked for and not delivered.
|
|
rc 2 outranks the 0/1 the report may have earned, because a caller who asked
|
|
for a file and got a stack trace cannot know a grade happened at all.
|
|
|
|
VERDICTS, IN THIS ORDER (a traceback is a defect whatever the exit code)
|
|
TRACEBACK uncaught exception text in either stream
|
|
RC_OTHER clean stream, rc != 2 -- the write failed and the tool did not
|
|
say so in the one code that means "could not grade"
|
|
SILENT rc 2 with no message at all -- a refusal nobody can read
|
|
UNNAMED rc 2, a message, but the message never names the target
|
|
OK rc 2, one line naming the target, no traceback
|
|
NO_REACH the tool never got as far as the write on this probe: its own
|
|
rc 2 came from the input, so the write path is UNGRADED here
|
|
(counted, never passed -- an ungraded path is not a clean one)
|
|
NO_JSON no --json argument in the source: nothing to probe
|
|
|
|
FIXTURES
|
|
Every probe runs in a throwaway tree, so the grade each tool performs is over
|
|
a few bytes rather than over this repository: the write path is what is being
|
|
graded, not the tool's subject matter. `{target}` is the only path that does
|
|
not exist -- it must not.
|
|
|
|
READ-ONLY on the repo: the only files created are under a TemporaryDirectory.
|
|
|
|
USAGE
|
|
python3 tools/write_probe.py [--repo DIR] [--only TOOL] [--timeout S]
|
|
[--json PATH] [--verbose]
|
|
python3 tools/write_probe.py --selftest # 10 cells, judge, table, fixture
|
|
EXIT
|
|
0 every probed tool is OK; 1 at least one is not (including NO_REACH and
|
|
NO_JSON: an ungraded write path may not be reported clean); 2 could not
|
|
grade -- no tracked tools discovered, or --json itself unwritable.
|
|
"""
|
|
import argparse
|
|
import json
|
|
import os
|
|
import re
|
|
import subprocess
|
|
import sys
|
|
import tempfile
|
|
|
|
JSON_FLAG = re.compile(r"add_argument\(\s*[\"']--json[\"']")
|
|
TRACEBACK = "Traceback (most recent call last)"
|
|
|
|
# (tool, argv) -- `{...}` placeholders are filled by fixture_paths(). One entry
|
|
# per tracked tool: a tool the table does not name is NO_PROBE (fatal), on the
|
|
# same rule as the read axis -- an ungraded path may not pass.
|
|
WRITE_PROBES = (
|
|
("append_only_audit.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("append_only_owners.py", ["{audit}", "--json", "{target}"]),
|
|
("census_row_delta.py", ["{census_a}", "{census_b}", "--json", "{target}"]),
|
|
("check_lint.py", ["--check", "echo ok", "--json", "{target}"]),
|
|
("checkin_check.py", ["--dir", "{checkins}", "--roster", "volt",
|
|
"--json", "{target}"]),
|
|
("citation_hash_class.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("citation_resolve_check.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("doc_census_check.py", ["--repo", "{tinyrepo}", "--no-run",
|
|
"--json", "{target}"]),
|
|
("doc_command_check.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("gate_blame.py", ["--repo", "{tinyrepo}", "--leg", "selftest-sweep",
|
|
"--range", "1", "--json", "{target}"]),
|
|
("landing_check.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("path_claim_check.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("queue_append_audit.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("queue_claim_check.py", ["--queue", "{queue}", "--json", "{target}"]),
|
|
("queue_index.py", ["--queue", "{queue}", "--json", "{target}"]),
|
|
("reply_check.py", ["--dir", "{replies}", "--json", "{target}"]),
|
|
("selftest_sweep.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("served_drift_check.py", ["--root", "{tinyrepo}", "--base-url", "{baseurl}",
|
|
"--json", "{target}"]),
|
|
("tool_map.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
("verify.py", ["--repo", "{tinyrepo}", "--fast", "--json", "{target}"]),
|
|
# It grades itself: run over the one-tool fixture repo, this probe reports
|
|
# that fixture tool as NO_PROBE (rc 1), then turns to its own --json write,
|
|
# which is the path being graded. Committing the instrument without this
|
|
# row would make the table's own NO_PROBE rule fail on its author.
|
|
("write_probe.py", ["--repo", "{tinyrepo}", "--json", "{target}"]),
|
|
)
|
|
|
|
|
|
def _write(path, text):
|
|
d = os.path.dirname(path)
|
|
if d and not os.path.isdir(d):
|
|
os.makedirs(d, exist_ok=True)
|
|
with open(path, "w", encoding="utf-8") as fh:
|
|
fh.write(text)
|
|
|
|
|
|
def build_fixtures(td, real_repo):
|
|
"""A one-commit git repo, one fake instrument, one reply, one check-in, one
|
|
log, one queue and two census files: enough for every tool in the table to
|
|
earn a 0 or a 1 and then turn to its write. Sizes are deliberately tiny."""
|
|
p = {}
|
|
|
|
tiny = os.path.join(td, "tiny")
|
|
os.makedirs(os.path.join(tiny, "docs"), exist_ok=True)
|
|
os.makedirs(os.path.join(tiny, "checkins"), exist_ok=True)
|
|
os.makedirs(os.path.join(tiny, "replies"), exist_ok=True)
|
|
os.makedirs(os.path.join(tiny, "tools"), exist_ok=True)
|
|
_write(os.path.join(tiny, "README.md"), "# tiny\n\nA fixture repo.\n")
|
|
_write(os.path.join(tiny, "WORK-QUEUE.md"),
|
|
"# Work queue\n\n**Q1 -- OPEN (pipe, 2026-01-01). One item.**\n\n"
|
|
"Body.\n")
|
|
_write(os.path.join(tiny, "docs", "queue-index.md"),
|
|
"# Index\n\n| id | state |\n| --- | --- |\n| Q1 | OPEN |\n")
|
|
_write(os.path.join(tiny, "tools", "fake_tool.py"),
|
|
'#!/usr/bin/env python3\n"""fake_tool.py -- a fake instrument.\n\n'
|
|
'USAGE:\n python3 fake_tool.py --selftest # 1 cells\n"""\n'
|
|
'import sys\nif "--selftest" in sys.argv:\n'
|
|
' print("selftest: 1 cells, 1 passed, 0 failed")\n'
|
|
' sys.exit(0)\nprint("fake")\n')
|
|
_write(os.path.join(tiny, "checkins", "2026-01-01-volt.md"),
|
|
"# volt check-in 2026-01-01\n\n"
|
|
"- Built: the fixture.\n- Next: nothing.\n")
|
|
_write(os.path.join(tiny, "replies", "2026-01-01-to-merlin-01.md"),
|
|
"# reply 01 to merlin\n\n## 1. Subject\n\nBody text, one finding.\n\n"
|
|
"-- volt\n")
|
|
for cmd in (["init", "-q", "-b", "main"], ["add", "-A"],
|
|
["-c", "user.email=v@example.invalid",
|
|
"-c", "user.name=volt", "commit", "-q", "-m", "fixture"]):
|
|
subprocess.run(["git", "-C", tiny] + cmd, capture_output=True, text=True)
|
|
p["tinyrepo"] = tiny
|
|
p["queue"] = os.path.join(tiny, "WORK-QUEUE.md")
|
|
p["index"] = os.path.join(tiny, "docs", "queue-index.md")
|
|
p["replies"] = os.path.join(tiny, "replies")
|
|
p["checkins"] = os.path.join(tiny, "checkins")
|
|
p["toolsdir"] = os.path.join(tiny, "tools")
|
|
|
|
p["audit"] = os.path.join(td, "audit.json")
|
|
_write(p["audit"], json.dumps({"rows": []}) + "\n")
|
|
p["log"] = os.path.join(td, "log.txt")
|
|
_write(p["log"], "checked 2026-01-01 Q1 clean\n")
|
|
for key in ("census_a", "census_b"):
|
|
p[key] = os.path.join(td, key + ".json")
|
|
_write(p[key], json.dumps({"rev": "abc1234", "repo": tiny,
|
|
"reports": [], "rows": []}) + "\n")
|
|
|
|
p["baseurl"] = "file://" + tiny
|
|
|
|
p["target"] = os.path.join(td, "no-such-dir", "out.json")
|
|
# The probe is meaningless if the target's parent exists: an existing
|
|
# directory would let the write succeed and grade the wrong thing.
|
|
assert not os.path.isdir(os.path.dirname(p["target"]))
|
|
p["repo"] = real_repo
|
|
return p
|
|
|
|
|
|
def judge(rc, out, err, target):
|
|
text = (out or "") + (err or "")
|
|
if TRACEBACK in text:
|
|
tail = [l for l in text.splitlines() if l.strip()][-1:]
|
|
return {"verdict": "TRACEBACK", "rc": rc,
|
|
"detail": (tail[0] if tail else "")[:110]}
|
|
if rc != 2:
|
|
first = next((l.strip() for l in (err or out or "").splitlines()
|
|
if l.strip()), "")
|
|
return {"verdict": "RC_OTHER", "rc": rc,
|
|
"detail": (first or "rc %s, no message" % rc)[:110]}
|
|
msg = [l for l in (err or "").splitlines() if l.strip()] or \
|
|
[l for l in (out or "").splitlines() if l.strip()]
|
|
if not msg:
|
|
return {"verdict": "SILENT", "rc": rc, "detail": "rc 2 with no message"}
|
|
if target not in text:
|
|
return {"verdict": "UNNAMED", "rc": rc,
|
|
"detail": "rc 2 but the message never names the target: "
|
|
+ msg[-1][:64]}
|
|
if "cannot write" not in text and "cannot be written" not in text:
|
|
# rc 2 and the path is named, but not as a write refusal: the tool
|
|
# refused its INPUT. The write path is therefore ungraded here.
|
|
return {"verdict": "NO_REACH", "rc": rc,
|
|
"detail": "rc 2 on the input, never reached the write: "
|
|
+ msg[-1][:64]}
|
|
return {"verdict": "OK", "rc": rc, "detail": msg[-1][:110]}
|
|
|
|
|
|
def run_probe(path, argv, timeout, cwd):
|
|
try:
|
|
p = subprocess.run([sys.executable, path] + argv, cwd=cwd,
|
|
capture_output=True, text=True, errors="replace",
|
|
timeout=timeout)
|
|
return p.returncode, p.stdout or "", p.stderr or ""
|
|
except subprocess.TimeoutExpired:
|
|
return None, "", "TIMEOUT after %ss" % timeout
|
|
except OSError as exc:
|
|
return None, "", "<exec failed: %s>" % exc
|
|
|
|
|
|
def tracked_tools(repo):
|
|
p = subprocess.run(["git", "-C", repo, "ls-tree", "-r", "--name-only",
|
|
"HEAD", "tools/"], capture_output=True, text=True)
|
|
if p.returncode != 0:
|
|
return None, (p.stderr or "git ls-tree failed").strip()
|
|
return [os.path.basename(x) for x in p.stdout.splitlines()
|
|
if x.endswith(".py")], "git ls-tree HEAD tools/"
|
|
|
|
|
|
def selftest(repo):
|
|
"""Ten cells: six for the judge's ordered verdicts, four for the table and
|
|
the fixture. The judge cells matter because the order is the contract -- a
|
|
traceback outranks a right code, and a rc-2 input refusal (NO_REACH) must
|
|
never read as a passed write path.
|
|
"""
|
|
bad = []
|
|
|
|
def cell(name, ok, why=""):
|
|
print((" ok " if ok else " FAIL ") + name
|
|
+ ("" if ok else " [%s]" % why))
|
|
if not ok:
|
|
bad.append(name)
|
|
|
|
t = os.path.join("no-such-dir", "out.json")
|
|
tb = TRACEBACK + "\n File \"x.py\", line 1\nFileNotFoundError: [Errno 2]"
|
|
cell("c1 a traceback is TRACEBACK whatever the exit code",
|
|
judge(2, "", tb, t)["verdict"] == "TRACEBACK")
|
|
cell("c2 a clean rc 1 is RC_OTHER, not a pass",
|
|
judge(1, "", "cannot write --json " + t + "\n", t)["verdict"] == "RC_OTHER")
|
|
cell("c3 rc 2 with no message is SILENT",
|
|
judge(2, "", "", t)["verdict"] == "SILENT")
|
|
cell("c4 rc 2 that never names the target is UNNAMED",
|
|
judge(2, "", "gate_blame: cannot grade: all 1 commit(s) UNGRADABLE\n",
|
|
t)["verdict"] == "UNNAMED")
|
|
cell("c5 rc 2 naming the target but refusing INPUT is NO_REACH (ungraded)",
|
|
judge(2, "", "cannot read " + t + ": no such file\n",
|
|
t)["verdict"] == "NO_REACH")
|
|
cell("c6 rc 2 naming the target as a WRITE refusal is OK",
|
|
judge(2, "", "cannot write --json " + t + ": [Errno 2]\n",
|
|
t)["verdict"] == "OK")
|
|
|
|
names, source = tracked_tools(repo)
|
|
table = dict(WRITE_PROBES)
|
|
dupes = len([n for n, _ in WRITE_PROBES]) - len(table)
|
|
cell("c7 the table names every tracked tool once, and no tool it cannot see",
|
|
names is not None and not dupes and set(table) == set(names),
|
|
"%s: table-only=%s tool-only=%s dupes=%s"
|
|
% (source, sorted(set(table) - set(names or [])),
|
|
sorted(set(names or []) - set(table)), dupes))
|
|
|
|
with tempfile.TemporaryDirectory(prefix="write-selftest-") as td:
|
|
fx = build_fixtures(td, repo)
|
|
cell("c8 the probe target's parent does not exist (or the write would succeed)",
|
|
not os.path.isdir(os.path.dirname(fx["target"])), fx["target"])
|
|
if names and "check_lint.py" in names:
|
|
rc, out, err = run_probe(os.path.join(repo, "tools", "check_lint.py"),
|
|
["--check", "echo ok", "--json", fx["target"]],
|
|
120, fx["tinyrepo"])
|
|
v = judge(rc, out, err, fx["target"])
|
|
cell("c9 a real tool, probed: rc 2, one line naming the target, no traceback",
|
|
v["verdict"] == "OK", str(v))
|
|
else:
|
|
cell("c9 a real tool, probed: rc 2, one line naming the target", False,
|
|
"check_lint.py not discovered under %s" % repo)
|
|
|
|
rows = [{"tool": "x", "verdict": "OK"}, {"tool": "y", "verdict": "NO_PROBE"}]
|
|
cell("c10 NO_PROBE counts as a fault, so an ungraded path cannot pass",
|
|
len([r for r in rows if r["verdict"] != "OK"]) == 1)
|
|
|
|
print("selftest: 10 cells, %d passed, %d failed" % (10 - len(bad), len(bad)))
|
|
return 1 if bad else 0
|
|
|
|
|
|
def main(argv=None):
|
|
ap = argparse.ArgumentParser(description=__doc__.splitlines()[0])
|
|
ap.add_argument("--repo", default="/root/troupe-standup")
|
|
ap.add_argument("--only", default=None, metavar="TOOL")
|
|
ap.add_argument("--timeout", type=int, default=180, metavar="S")
|
|
ap.add_argument("--json", default=None, metavar="PATH")
|
|
ap.add_argument("--selftest", action="store_true",
|
|
help="prove the judge and the table, then exit")
|
|
ap.add_argument("--verbose", action="store_true")
|
|
a = ap.parse_args(argv)
|
|
|
|
repo = os.path.abspath(a.repo)
|
|
if a.selftest:
|
|
return selftest(repo)
|
|
names, source = tracked_tools(repo)
|
|
if names is None:
|
|
print("cannot read: " + source, file=sys.stderr)
|
|
return 2
|
|
if not names:
|
|
print("cannot read: no tracked tools under " + repo, file=sys.stderr)
|
|
return 2
|
|
|
|
table = dict(WRITE_PROBES)
|
|
rows = []
|
|
with tempfile.TemporaryDirectory(prefix="write-probe-") as td:
|
|
fx = build_fixtures(td, repo)
|
|
for name in names:
|
|
if a.only and name != a.only:
|
|
continue
|
|
path = os.path.join(repo, "tools", name)
|
|
try:
|
|
with open(path, encoding="utf-8") as fh:
|
|
src = fh.read()
|
|
except OSError as exc:
|
|
rows.append({"tool": name, "verdict": "UNREADABLE", "rc": None,
|
|
"detail": str(exc)[:110]})
|
|
continue
|
|
if name not in table:
|
|
rows.append({"tool": name, "verdict": "NO_PROBE", "rc": None,
|
|
"detail": "no write probe in the table -- ungraded"})
|
|
continue
|
|
if not JSON_FLAG.search(src):
|
|
rows.append({"tool": name, "verdict": "NO_JSON", "rc": None,
|
|
"detail": "no --json argument in source"})
|
|
continue
|
|
argv = [re.sub(r"\{(\w+)\}", lambda m: fx[m.group(1)], x)
|
|
for x in table[name]]
|
|
rc, out, err = run_probe(path, argv, a.timeout, fx["tinyrepo"])
|
|
rec = {"tool": name, "probe": table[name]}
|
|
rec.update(judge(rc, out, err, fx["target"]))
|
|
rows.append(rec)
|
|
if a.verbose:
|
|
print(" . %-24s %-9s rc=%s" % (name, rec["verdict"], rc))
|
|
if not rows:
|
|
print("cannot read: no tool matched --only " + str(a.only),
|
|
file=sys.stderr)
|
|
return 2
|
|
|
|
bad = [r for r in rows if r["verdict"] != "OK"]
|
|
width = max([len(r["tool"]) for r in rows] + [4])
|
|
print("write probe -- %d of %d tracked tool(s) probed under %s (%s)"
|
|
% (len(rows), len(names), os.path.join(repo, "tools"), source))
|
|
print()
|
|
print("%-*s %-9s %4s detail" % (width, "tool", "verdict", "rc"))
|
|
for r in rows:
|
|
print("%-*s %-9s %4s %s"
|
|
% (width, r["tool"], r["verdict"],
|
|
"-" if r.get("rc") is None else r["rc"], r["detail"]))
|
|
print()
|
|
print("write axis: %d probed, %d ok, %d fault(s)"
|
|
% (len(rows), len(rows) - len(bad), len(bad))
|
|
+ ("" if not bad else " (" + ", ".join(sorted({r["verdict"]
|
|
for r in bad})) + ")"))
|
|
|
|
if a.json:
|
|
payload = {"repo": repo, "discovery": source, "probed": len(rows),
|
|
"ok": len(rows) - len(bad), "faults": len(bad),
|
|
"tools": rows}
|
|
try:
|
|
with open(a.json, "w", encoding="utf-8") as fh:
|
|
json.dump(payload, fh, indent=1, sort_keys=True)
|
|
fh.write("\n")
|
|
except OSError as exc:
|
|
print("cannot write --json " + a.json + ": " + str(exc),
|
|
file=sys.stderr)
|
|
return 2
|
|
return 1 if bad else 0
|
|
|
|
|
|
if __name__ == "__main__":
|
|
sys.exit(main())
|