d0b4108305
- SessionRecorder states classical_declared from the WAI2 envelopes it holds, readable or encrypted, as rule 9 counts them (claim_receipt::session::classical_length, which rule 9 now calls); SessionRecorder::classical is removed. A page that receives a WAI2 it does not decode seals its session. counting/ gains wai2-classical.wai and its encrypted twin; verify_measure seals a session over each in wasm32 and checks the length against the table, and the native and Python verifiers agree. - Rule 4: of an encrypted envelope, a rendition's sha256 is still required absent or well formed (RenditionMeta::digest_well_formed, digest_well_formed in Python): a malformed one matches no payload. jwp-receipts rule 3 narrowed to whether a payload matches a well-formed sha256. Twelve vectors, by rendition and by exploded hash; the differential's encrypted family mutates the named rendition's sha256. - Stale docs: rule 3 and rule 9 in session.rs, the Python docstring, the received doc, the corpus comments. Co-Authored-By: Claude Opus 5.5 <[email protected]>
2107 lines
126 KiB
Python
2107 lines
126 KiB
Python
#!/usr/bin/env python3
|
||
"""Differential check: wai-rs and claim_verify.py give the same verdict on inputs the corpus
|
||
does not hold.
|
||
|
||
The corpus pins one declared outcome per vector, which shows both verifiers agree on those
|
||
vectors and nothing more. This script generates inputs around the edges where two readings
|
||
of jwp-receipts §3.1-§3.2 could part — JSON numbers and strings at the limits of what a
|
||
parser accepts, nesting, calendar dates, URI authorities, acquisition classes with every
|
||
combination of evidence around energy-measurement §2.3-§2.5's rules, OER figures near a
|
||
half-microjoule and at the edges of oer-receipts §4's quantities, OER receipts whose sums
|
||
and floors sit at §6's tolerances, OER receipts holding numbers §4 forbids in a body
|
||
(`-0`, fractions, integers beyond I-JSON's range) in members of every kind, context
|
||
receipts that do not themselves verify, and bodies of every class with one defect, or two
|
||
at different positions, where the order §3.1 step 2 reads a body in decides the reason;
|
||
measurement reports whose brackets sit at the least declaration and one ppm below it, at a
|
||
half microjoule, at the ratio thresholds, with a flat idle bracket, with products far past
|
||
128 bits and with every pattern of liveness facts; operating-point claims with every
|
||
presence combination of their points and prediction; reports checked against figure views
|
||
that agree and disagree in each member; energy claims against labelled objects; stage
|
||
claims around every rule of §2.4.10 — points, the stage hash, the units, the objects'
|
||
order, statuses, evidence, reasons and bytes, and the delivered stage they give — and
|
||
delivery-gap claims against object receipts at their coordinates and elsewhere; and
|
||
reconstruct-session claims built from random presentation lists, around every rule of
|
||
§2.4.12 — units, leaves and paths off by one, switches and missed presentations at and past
|
||
their caps, a null deadline with misses, presences that disagree, decoder indices out of
|
||
range, leaves moved between entries, entries out of first-use order, every acquisition
|
||
class — with their outputs, re-decodes, objects, entry envelopes and stage claims that
|
||
reach their stages or miss them, and sessions holding a base's and its layers' envelopes
|
||
— WAI1 envelopes or a WAI2 envelope's components, a base named by its exploded
|
||
envelope or by its envelope and rendition, a base whose fallback is of another format,
|
||
some layers a sink cannot apply —
|
||
presented within and past the layers and units each envelope gives, through entries at
|
||
their base's first unit or another, with objects whose figures are each envelope's or
|
||
told otherwise —
|
||
writes them in the corpus's format with no declared outcome, runs both verifiers'
|
||
`verdicts` mode, and reports every input on which they differ.
|
||
|
||
Some inputs cannot share a file with the rest: a number whose nearest double is infinite,
|
||
`NaN`, an unpaired surrogate, a repeated member name or nesting past 64 levels makes a
|
||
whole file unreadable (oer-receipts §4 "The document"), for both verifiers or — if they
|
||
differ — for one. Each of those is written to a file of its own with a readable twin beside
|
||
it, and the two verifiers must either both refuse the file or give the same verdicts on it.
|
||
|
||
python3 differential.py <wai_claim_vectors binary> [--random N] [--seed S] [--keep FILE]
|
||
|
||
Keys are the corpus's fixed seeds, so every signed input is reproducible. Exit status 0
|
||
only if the two verdict lists are identical, every file of its own is read or refused
|
||
alike, and neither verifier printed `crash`; any difference, whatever step it is at,
|
||
exits 1.
|
||
|
||
An energy claim that names an OER receipt reaches the OER verifier of its own language
|
||
(§2.4.5 check 2): wai-rs's `crate::oer`, and the reference `oer2_verify.py`. A difference
|
||
there is still a difference in the claim verdict, and is reported like any other; the
|
||
listing marks it, since its cause is in the two OER verifiers rather than the claim rules.
|
||
Those two are compared on their own by `../oer-conformance/fuzz_differential.py`.
|
||
"""
|
||
import argparse, base64, copy, hashlib, io, json, os, random, struct, subprocess, sys, tempfile
|
||
from decimal import Decimal, localcontext
|
||
|
||
HERE = os.path.dirname(os.path.abspath(__file__))
|
||
sys.path.insert(0, HERE)
|
||
import claim_verify as CV # noqa: E402
|
||
SV = CV.SV # the strict JSON, manifest and envelope readers claim_verify.py uses
|
||
|
||
from cryptography.hazmat.primitives import serialization # noqa: E402
|
||
from cryptography.hazmat.primitives.asymmetric import ed25519 # noqa: E402
|
||
|
||
OER = CV.OER
|
||
|
||
|
||
def key(seed):
|
||
return ed25519.Ed25519PrivateKey.from_private_bytes(bytes([seed] * 32))
|
||
|
||
|
||
def pk(k):
|
||
return k.public_key().public_bytes(serialization.Encoding.Raw, serialization.PublicFormat.Raw).hex()
|
||
|
||
|
||
PUBLISHER, METER, OPERATOR, VIEWER, RESOLVER, RELAY = (key(s) for s in (0x11, 0x22, 0x33, 0x44, 0x55, 0x66))
|
||
OER_SIGNER = "ed25519:claim-conformance-meter"
|
||
WM_ALG = "org.example.watermark.1"
|
||
|
||
|
||
def sign(c, k):
|
||
"""`c` with its signature made under `k` over §2.4.3's signing input."""
|
||
c = copy.deepcopy(c)
|
||
c.pop("sig", None)
|
||
c["signer_pubkey"] = pk(k)
|
||
c["sig"] = k.sign(CV.signing_input(c)).hex()
|
||
return c
|
||
|
||
|
||
RAW = {}
|
||
|
||
|
||
def raw(spelling):
|
||
"""A JSON value written into the input file exactly as `spelling` — `-0`, an integer
|
||
beyond 64 bits, a string with an unpaired surrogate escape — where json.dumps would
|
||
write another spelling or none. A placeholder string until text_of() writes it."""
|
||
k = f"\u0000raw{len(RAW)}\u0000"
|
||
RAW[k] = spelling
|
||
return k
|
||
|
||
|
||
def text_of(v, **kw):
|
||
"""json.dumps(v, ensure_ascii=True), every raw() placeholder written as its spelling."""
|
||
t = json.dumps(v, ensure_ascii=True, **kw)
|
||
if "\\u0000raw" in t:
|
||
for k, spelling in RAW.items():
|
||
t = t.replace(json.dumps(k), spelling)
|
||
return t
|
||
|
||
|
||
def seal_oer(template, joules_total, terms=None, floor=None, edit=None):
|
||
"""`template` recording `joules_total`, re-sealed and signed as oer-receipts §4-§5
|
||
define. The breakdown is `terms` (compute_j first) or, by default, `joules_total` as
|
||
compute_j with the other terms as the template has them (zero); `floor`, if given, is
|
||
(bits_committed, temperature_k, floor_j); `edit`, if given, changes the receipt last.
|
||
|
||
The content address is computed over the receipt as the file will be read: its text,
|
||
read as §4 reads a document (`OER.load`). A receipt whose body holds a number §4
|
||
forbids is sealed all the same; both verifiers must refuse it before its address."""
|
||
r = copy.deepcopy(template)
|
||
r.pop("chain", None)
|
||
r.pop("receipt_id", None)
|
||
r["energy"]["joules_total"] = joules_total
|
||
b = r["energy"]["breakdown"]
|
||
for k, t in zip(list(b), terms if terms is not None else [joules_total]):
|
||
b[k] = t
|
||
if floor is not None:
|
||
r["floor"]["bits_committed"], r["energy"]["temperature_k"], r["floor"]["floor_j"] = floor
|
||
if edit is not None:
|
||
edit(r)
|
||
body = OER.canonical_bytes(OER.load(io.StringIO(text_of(r))))
|
||
prev = "sha256:" + "0" * 64
|
||
rh = OER.h("sha-256", prev.encode() + body)
|
||
r["chain"] = {"prev_hash": prev, "receipt_hash": rh, "signer": OER_SIGNER,
|
||
"signature": METER.sign(rh.encode()).hex()}
|
||
r["receipt_id"] = OER.h("sha-256", body)
|
||
return r
|
||
|
||
|
||
class Inputs:
|
||
def __init__(self, corpus):
|
||
self.by_name = {v["name"]: v for v in corpus["vectors"]}
|
||
self.out = []
|
||
|
||
def receipt(self, name):
|
||
return json.loads(self.by_name[name]["receipt"])
|
||
|
||
def wire(self, name):
|
||
return self.by_name[name]["receipt"]
|
||
|
||
def add(self, name, receipt=None, context=None, procedure="claim"):
|
||
v = {"name": f"{len(self.out):05d}_{name}", "procedure": procedure}
|
||
if receipt is not None:
|
||
v["receipt"] = receipt
|
||
if context:
|
||
v["context"] = context
|
||
self.out.append(v)
|
||
|
||
|
||
# ------------------------------------------------------------------ input families
|
||
|
||
NUMBER_TOKENS = [
|
||
"0", "-0", "1", "-1", "9007199254740991", "9007199254740992", "18446744073709551615",
|
||
"18446744073709551616", "1" + "0" * 30, "1" + "0" * 308, "1" + "0" * 309, "1" + "0" * 400,
|
||
"-" + "1" + "0" * 400, "0.0", "-0.0", "1.0", "1e0", "1E2", "1e-2", "1e308", "1e309", "1e400",
|
||
"-1e400", "1E400", "1e-400", "1.7976931348623157e308", "1.7976931348623158e308",
|
||
"1.7976931348623159e308", "1.8e308", "2e308", "1e99999999999999999999",
|
||
"179769313486231580793728971405303415079934132710037826936173778980444968292764750946649017977587207096330286416692887910946555547851940402630657488671505820681908902000708383676273854845817711531764475730270069855571366959622842914819860834936475292719074168444365510704342711559699508093042880177904174497791",
|
||
"01", "+1", ".5", "1.", "1e", "1e+", "--1", "0x10", "NaN", "Infinity", "-Infinity", "1_0",
|
||
"１", "١",
|
||
]
|
||
|
||
STRING_BODIES = [
|
||
"\\ud800", "\\udc00", "\\udbff\\udfff", "\\ud800\\u0041", "\\ud83d\\ude00", "\\ud83d", "a\\u0000b",
|
||
"\\u00e9", "\\u00E9", "\\/", "\x01", "\x7f", "
", "", "", "", "\\ud800x",
|
||
"\\uDFFF", "\\u", "\\u12", "\\x41", "\\'", "tab\there",
|
||
]
|
||
|
||
|
||
def numbers(ins):
|
||
wire = ins.wire("energy_accept")
|
||
for t in NUMBER_TOKENS:
|
||
ins.add("seq_" + t[:24], wire.replace('"seq":0', '"seq":' + t, 1))
|
||
ins.add("work_units_" + t[:24], wire.replace('"work_units":49152', '"work_units":' + t, 1))
|
||
|
||
|
||
def strings(ins):
|
||
wire = ins.wire("energy_accept")
|
||
for s in STRING_BODIES:
|
||
ins.add("signer_id_" + s.encode("unicode_escape").decode()[:24],
|
||
wire.replace('"did:example:meter"', '"' + s + '"', 1))
|
||
ins.add("member_name_" + s.encode("unicode_escape").decode()[:24],
|
||
wire.replace('{"body":', '{"' + s + '":1,"body":', 1))
|
||
# A validly signed claim whose signer_id holds characters outside the BMP and a control
|
||
# character, sent with every such character escaped: canonicalisation undoes escapes.
|
||
c = sign(dict(ins.receipt("energy_accept"), signer_id="did:example:m\U0001F600é\u001f"), METER)
|
||
escaped = json.dumps(c, ensure_ascii=True, separators=(", ", ": "))
|
||
ins.add("escaped_astral_signer_id", escaped)
|
||
ins.add("canonical_astral_signer_id", CV.jcs(c))
|
||
ins.add("bom_prefixed", "" + wire)
|
||
ins.add("trailing_garbage", wire + " x")
|
||
ins.add("trailing_whitespace", wire + " \n\t\r")
|
||
ins.add("nbsp_whitespace", " " + wire)
|
||
ins.add("empty", "")
|
||
|
||
|
||
def nesting(ins):
|
||
wire = ins.wire("energy_accept")
|
||
for d in list(range(1, 40)) + [63, 64, 65, 126, 127, 128, 129, 200, 999, 1001, 5000]:
|
||
ins.add(f"nesting_{d}", wire.replace('"did:example:meter"', "[" * d + "]" * d, 1))
|
||
ins.add("nesting_obj_40", wire.replace('"did:example:meter"', '{"a":' * 40 + "1" + "}" * 40, 1))
|
||
|
||
|
||
def dates(ins, rng, n):
|
||
base = ins.receipt("energy_accept")
|
||
fixed = ["2026-02-28T00:00:00Z", "2026-02-29T00:00:00Z", "2028-02-29T00:00:00Z", "2100-02-29T00:00:00Z",
|
||
"2000-02-29T00:00:00Z", "1900-02-29T00:00:00Z", "0000-02-29T00:00:00Z", "0004-02-29T00:00:00Z",
|
||
"2026-04-31T00:00:00Z", "2026-04-30T23:59:60Z", "2026-12-31T23:59:60+14:00", "2026-01-00T00:00:00Z",
|
||
"2026-00-10T00:00:00Z", "2026-13-01T00:00:00Z", "2026-09-27T24:00:00Z", "2026-09-27T12:60:00Z",
|
||
"2026-09-27T12:00:61Z", "2026-09-27T12:00:00.Z", "2026-09-27T12:00:00.000000001Z",
|
||
"2026-09-27t12:00:00z", "2026-09-27T12:00:00-00:00", "2026-09-27T12:00:00+24:00",
|
||
"2026-09-27T12:00:00+23:60", "2026-09-27 12:00:00Z", "2026-9-27T12:00:00Z",
|
||
"２０２６-09-27T12:00:00Z", "2026-09-2٧T12:00:00Z", "2026-09-27T12:00:00Ｚ",
|
||
"+2026-09-27T12:00:00Z", "2026-09-27T12:00:00Z\n", "2026-09-31T12:00:00Z", "2026-06-31T12:00:00Z",
|
||
"2026-11-31T12:00:00Z", "2026-01-31T12:00:00Z", "2026-03-31T12:00:00Z"]
|
||
cases = list(fixed)
|
||
for _ in range(n):
|
||
y, mo, d = rng.randrange(0, 10000), rng.randrange(0, 14), rng.randrange(0, 33)
|
||
cases.append(f"{y:04d}-{mo:02d}-{d:02d}T{rng.randrange(0, 25):02d}:{rng.randrange(0, 61):02d}:"
|
||
f"{rng.randrange(0, 62):02d}{rng.choice(['Z', '+05:30', '-12:00', '.5Z', 'z'])}")
|
||
for t in cases:
|
||
ins.add("issued_at_" + t.encode("unicode_escape").decode()[:32], CV.jcs(sign(dict(base, issued_at=t), METER)))
|
||
|
||
|
||
def endpoints(ins, rng, n):
|
||
base = ins.receipt("soft_binding_accept")
|
||
fixed = ["https://resolver.example.org/matches", "https://:443/m", "https://:/m", "https://resolver.example.org:/m",
|
||
"https://resolver.example.org:8443/m", "https://resolver.example.org:x/m", "https://resolver.example.org:1:2/m",
|
||
"https://[2001:db8::1]:8443/m", "https://[2001:db8::1]/m", "https://[]/m", "https://[/m", "https://[::1/m",
|
||
"https://::1]/m", "https://[::1]x/m", "https://[::1]:/m", "https://[::1]:x/m", "https://h]:1/m",
|
||
"https://[email protected]/m", "https://@/m", "https://resolver.example.org",
|
||
"https://resolver.example.org?q", "https://resolver.example.org#f", "https://?q", "https://#f", "https://",
|
||
"https:///m", "HTTPS://resolver.example.org/m", "http://resolver.example.org/m", "https:resolver.example.org",
|
||
"https://résolver.example.org/m", "https://resolver.example.org:８/m", "https://1.2.3.4:5/m",
|
||
"https://%72esolver.example.org/m", "https://a/b@c"]
|
||
alphabet = "ab:[]@/?#1%."
|
||
cases = list(fixed) + ["https://" + "".join(rng.choice(alphabet) for _ in range(rng.randrange(0, 9))) for _ in range(n)]
|
||
for e in cases:
|
||
body = dict(base["body"], resolution_endpoint=e)
|
||
ins.add("endpoint_" + e.encode("unicode_escape").decode()[:32], CV.jcs(sign(dict(base, body=body), PUBLISHER)))
|
||
|
||
|
||
FIGURES = ["0.0120005", "0.012000500000000000000000000000000", "0.0120004999999999999999999999999999",
|
||
"0.0120005000000000000000000000000000", "0.012000500000000000000000000000000000000000000",
|
||
"0.0120004999999999999999999999999999999999999999", "0.012000", "0.0120004", "0.0120006", "1.20005e-2",
|
||
"120005e-7", "1.20005E-2", "1.20005e-02", "1.20005e-002", "1.20005e-0002", "12000.5e-6",
|
||
"0.0000000000000000000000000000000000000000000000120005e43",
|
||
"+0.0120005", " 0.0120005", "0.0120005 ", "00.0120005", ".0120005", "0.0120005e", "0_0120005", "1_2",
|
||
"０.0120005", "NaN", "Infinity", "-0.0120005", "0", "-0", "0e0", "1e-999", "1e999", "1e-1000",
|
||
"12000500000000000000000000000000000000000000000e-51", "0.00000050", "0.0000005", "0.0000015", "0.0000025",
|
||
"18446744073709.5516155", "18446744073709.551615", "9007199254.740991", "9007199254.7409915"]
|
||
# JSON numbers: §4 admits integers within I-JSON's range in a body and nothing else, so every
|
||
# fraction, `-0` and integer beyond the range here must make the receipt malformed. Among
|
||
# them, doubles half-way between two shortest decimals, which a formatter could write two
|
||
# ways — no receipt body may carry one, so no content address may depend on it.
|
||
NUMERIC_FIGURES = [0.0120005, 0.012, 0.0120004, 0.0120006, 0.0000025, 1.5e-6, 2.5e-6, 12, 0, -0.0120005, 1e-300,
|
||
9007199254.740991, 1.2000500000000001e-2, True, False, None, -0.0,
|
||
1e37, 9.999999999999999e37, 1e38, 1e-38, 9.9e-39, 1.7976931348623157e308, 5e-324,
|
||
2 ** 53 - 1, 2 ** 53, -(2 ** 53 - 1), -(2 ** 53), 2 ** 63 - 1, 2 ** 64 - 1, 2 ** 64, 10 ** 30,
|
||
75.036163330078125, 2365.15167236328125, 5.9604644775390625e-8, 983123238.6575775]
|
||
# §4's decimal strings at their limits: 34 significand digits and a three-digit exponent.
|
||
RANGE_FIGURES = ["9" * 34, "9" * 35, "0." + "0" * 32 + "1", "0." + "0" * 33 + "1", "-" + "9" * 34, "1e999", "1e-999",
|
||
"9.999999999999999999999999999999999e999", "9.9999999999999999999999999999999999e999",
|
||
"0.0120004" + "9" * 26, "0.0120004" + "9" * 27, "0.0120005" + "0" * 26, "0.0120005" + "0" * 26 + "1"]
|
||
|
||
|
||
LABELS = ["OnChipCounter", "CalibratedInstrument", "ModelBased", "Estimator", "HwShunt", "onchipcounter",
|
||
"HWSHUNT", "Measured", "", " OnChipCounter", "OnChipCounter\u200b"]
|
||
UNCERTAINTIES = [None, {"relative_ppm": 20000, "window_us": 1000000}, {"relative_ppm": 0, "window_us": 1},
|
||
{"relative_ppm": 1, "window_us": 0}, {"relative_ppm": 2 ** 32 - 1, "window_us": 1},
|
||
{"relative_ppm": 2 ** 32, "window_us": 1}, {"relative_ppm": 2 ** 53 - 1, "window_us": 2 ** 53 - 1},
|
||
{"relative_ppm": 288676, "window_us": 1}, {"relative_ppm": 288675, "window_us": 1},
|
||
{"relative_ppm": 144338, "window_us": 1}, {"relative_ppm": 144337, "window_us": 1},
|
||
{"relative_ppm": 20000}, {"window_us": 1}, {}, {"relative_ppm": 20000, "window_us": 1, "k": 2},
|
||
{"relative_ppm": "20000", "window_us": 1}, [20000, 1], "20000", 20000, True]
|
||
FILTERINGS = [None, "On", "Off", "Unknown", "unknown", "", "Maybe", 0, False]
|
||
# A calibration reference is blank when it is White_Space only (energy-measurement §2.5, as
|
||
# the reference implementation trims it); U+001C to U+001F, U+200B and U+FEFF are not.
|
||
CALIBRATION_REFS = [None, "cert 7", "", " ", "\t", "\u3000", "\u0085", "\u00a0", "\u2028", "\u001c", "\u001f",
|
||
"\u200b", "\ufeff", " x ", "https://calibration.example.org/certificates/0001", 7]
|
||
CLASS_FIGURES = [1, 2, 14, 15, 12000, 2 ** 53 - 1, 0]
|
||
|
||
|
||
def classes(ins, rng, n):
|
||
"""§2.4.5 rule 2 and its interaction with rule 5: energy claims with every acquisition
|
||
label, uncertainty, filtering state and calibration reference above in fixed
|
||
combinations — each field varied alone from a valid counter, a valid instrument and a
|
||
ModelBased figure — and `n` random combinations of all four with a figure and a grade.
|
||
Each is signed, so its verdict is decided at step 2, 4 or later, never by the signature."""
|
||
base = ins.receipt("energy_accept")
|
||
bases = {
|
||
"OnChipCounter": dict(base["body"], energy_provenance="OnChipCounter",
|
||
energy_uncertainty={"relative_ppm": 20000, "window_us": 1000000},
|
||
energy_counter_filtering="Unknown", grade="measured"),
|
||
"CalibratedInstrument": dict(base["body"], energy_provenance="CalibratedInstrument",
|
||
energy_uncertainty={"relative_ppm": 5000, "window_us": 1000000},
|
||
energy_calibration_ref="cert 7", grade="measured"),
|
||
"ModelBased": dict(base["body"], oer_receipt=None),
|
||
}
|
||
cases = []
|
||
for bname, b in bases.items():
|
||
b = dict(b, oer_receipt=None)
|
||
cases += [(f"{bname}_label_{i}", dict(b, energy_provenance=x)) for i, x in enumerate(LABELS)]
|
||
cases += [(f"{bname}_uncertainty_{i}", dict(b, energy_uncertainty=x)) for i, x in enumerate(UNCERTAINTIES)]
|
||
cases += [(f"{bname}_filtering_{i}", dict(b, energy_counter_filtering=x)) for i, x in enumerate(FILTERINGS)]
|
||
cases += [(f"{bname}_calibration_{i}", dict(b, energy_calibration_ref=x)) for i, x in enumerate(CALIBRATION_REFS)]
|
||
cases += [(f"{bname}_figure_{i}", dict(b, joules_micro=x)) for i, x in enumerate(CLASS_FIGURES)]
|
||
cases += [(f"{bname}_grade_{g}", dict(b, grade=g)) for g in OER.GRADES + ["certified"]]
|
||
for i in range(n):
|
||
cases.append((f"random_{i}", dict(base["body"], oer_receipt=None, energy_provenance=rng.choice(LABELS),
|
||
energy_uncertainty=rng.choice(UNCERTAINTIES),
|
||
energy_counter_filtering=rng.choice(FILTERINGS),
|
||
energy_calibration_ref=rng.choice(CALIBRATION_REFS),
|
||
joules_micro=rng.choice(CLASS_FIGURES), grade=rng.choice(OER.GRADES))))
|
||
for name, body in cases:
|
||
ins.add("class_" + name, CV.jcs(sign(dict(base, body=body), METER)))
|
||
|
||
|
||
# §3.1 step 2 reads a body's members, then each member's type and encoding together in its
|
||
# table's order, a nested object's members where the table reaches it and a lawful-basis
|
||
# body's `record` first. One defect shows where each is caught; two at different positions
|
||
# show the order. Per accepted corpus claim: its signer and, in table order, each position
|
||
# (a member, or a member of a nested object) with the values put there. DELETE removes the
|
||
# member; a position ending in EXTRA adds a member the table does not have. Values stay in
|
||
# what the canonical form can write, so every input is signed — numbers at the edges of
|
||
# what a claim may hold are the `numbers` family's.
|
||
DELETE = object()
|
||
EXTRA = "+"
|
||
BODY_DEFECTS = {
|
||
"lawful_basis_declaration_accept": ("OPERATOR", [
|
||
(("record",), ["consent", "decision", 5, DELETE]),
|
||
(("vocabulary",), [5, None, "w3id.org/dpv", "", "https://w3id.org/ dpv", DELETE]),
|
||
(("fragment",), [5, "zz", "AB" * 32, "ab", DELETE]),
|
||
(("purposes",), ["x", [], [5], {}]),
|
||
(("purposes", 0, "purpose"), [5, "dpv:A B", "", "dpv:\u3000", DELETE]),
|
||
(("purposes", 0, "basis"), [5, None, "", "a b", DELETE]),
|
||
(("purposes", 0, "data"), ["x", [5], [""], ["a b"], [], DELETE]),
|
||
(("purposes", 0, EXTRA), [1]),
|
||
(("purposes", 1, "purpose"), [5, "dpv:A B", "dpv:ServiceProvision", DELETE]),
|
||
(("purposes", 1, "basis"), [5, "", "a\tb", DELETE]),
|
||
(("purposes", 1, "data"), [[5], ["x y"], DELETE]),
|
||
(("purposes", 1, EXTRA), [1]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"lawful_basis_decision_accept": ("VIEWER", [
|
||
(("record",), ["declaration", "x", 5, DELETE]),
|
||
(("vocabulary",), [5, "dpv", DELETE]),
|
||
(("declaration",), [5, "zz", "AB" * 32, DELETE]),
|
||
(("purpose",), [5, "", "a b", DELETE]),
|
||
(("outcome",), [5, "", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"energy_on_chip_counter_accept": ("METER", [
|
||
(("joules_micro",), ["1", 0, True, None, DELETE]),
|
||
(("energy_provenance",), [5, "Measured", "HwShunt", DELETE]),
|
||
(("energy_uncertainty",), ["x", [1, 1], None, {}, DELETE]),
|
||
(("energy_uncertainty", "relative_ppm"), ["1", 0, DELETE]),
|
||
(("energy_uncertainty", "window_us"), [None, 0, DELETE]),
|
||
(("energy_uncertainty", EXTRA), [2]),
|
||
(("energy_counter_filtering",), [5, None, "Maybe", DELETE]),
|
||
(("energy_calibration_ref",), [5, "cert", DELETE]),
|
||
(("coverage",), [5, "Most", DELETE]),
|
||
(("grade",), [5, "certified", DELETE]),
|
||
(("method",), [5, "x", "", DELETE]),
|
||
(("work_units",), ["1", 0, DELETE]),
|
||
(("work_unit",), [5, "", "a b", DELETE]),
|
||
(("oer_receipt",), [5, "md5:1", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"soft_binding_accept": ("PUBLISHER", [
|
||
(("alg",), [5, "", "a b", "bare", DELETE]),
|
||
(("registry",), [5, "x", DELETE]),
|
||
(("binding_type",), [5, "hash", DELETE]),
|
||
(("value",), [5, "zz", "ABCD", "", DELETE]),
|
||
(("resolution_endpoint",), [5, "x", "http://x/", "https://u@x/", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"binding_recovery_accept": ("RESOLVER", [
|
||
(("alg",), [5, "", "bare", DELETE]),
|
||
(("binding_type",), [5, "hash", DELETE]),
|
||
(("value",), [5, "zz", "", DELETE]),
|
||
(("declaration",), [5, "zz", DELETE]),
|
||
(("resolution_endpoint",), [5, "x", "http://x/", DELETE]),
|
||
(("observed",), [5, "zz", DELETE]),
|
||
(("attests",), [5, "object", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"log_inclusion_accept": ("PUBLISHER", [
|
||
(("log",), [5, "x", DELETE]),
|
||
(("profile",), [5, "", DELETE]),
|
||
(("entry",), [5, "", "a b", DELETE]),
|
||
(("log_receipt",), [5, "zz", "", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"stage_accept": ("VIEWER", [
|
||
(("vocabulary",), [5, "staged-delivery", "urn:example:stages:1", DELETE]),
|
||
(("op_requested",), ["x", {}, {"wai-stage:layers": "2"}, {"a b": 1}, {"wai-stage:layers": 256}, None, DELETE]),
|
||
(("op_delivered",), [None, {"wai-stage:layers": 1}, {"wai-stage:layers": 3}, DELETE]),
|
||
(("presented",), [None, 5, {}, {"wai-stage:layers": 1}, DELETE]),
|
||
(("units_by_stage",), ["x", [], [0, 0, 4, 0], [4], [0, "4"], [1, 1, 2], DELETE]),
|
||
(("tier",), [5, "none", "exact", DELETE]),
|
||
(("stage_hash",), [None, "zz", "AB" * 32, DELETE]),
|
||
(("objects",), ["x", [], [5], DELETE]),
|
||
(("objects", 0, "layer"), ["1", 0, 256, DELETE]),
|
||
(("objects", 0, "units"), ["x", {}, DELETE]),
|
||
(("objects", 0, "units", "count"), [0, 4, DELETE]),
|
||
(("objects", 0, "units", EXTRA), [1]),
|
||
(("objects", 0, "status"), [5, "skipped", "late", "partial", "dropped", DELETE]),
|
||
(("objects", 0, "evidence"), ["verified", "none", "inclusion-verified", DELETE]),
|
||
(("objects", 0, "reason"), [5, "too-late", "deadline-missed", DELETE]),
|
||
(("objects", 0, EXTRA), [1]),
|
||
(("objects", 4, "status"), ["dropped", "refused"]),
|
||
(("deadline_misses",), [5, "1", DELETE]),
|
||
(("bytes_envelopes",), [0, DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
"gap_accept": ("RELAY", [
|
||
(("track_alias",), ["8", None, DELETE]),
|
||
(("group_id",), [9, DELETE]),
|
||
(("reason",), [5, "timeout", "discretion", DELETE]),
|
||
(("observed_at",), [5, "2026-02-29T00:00:00Z", "now", DELETE]),
|
||
((EXTRA,), [1]),
|
||
]),
|
||
}
|
||
|
||
|
||
def edit_at(body, path, value):
|
||
"""Put `value` at `path` in `body` (DELETE removes the member there; a path ending in
|
||
EXTRA adds the member `note`). False if an earlier edit left no object on the path."""
|
||
*parents, last = path
|
||
node = body
|
||
for p in parents:
|
||
if isinstance(p, int):
|
||
if not (isinstance(node, list) and p < len(node)):
|
||
return False
|
||
elif not (isinstance(node, dict) and p in node):
|
||
return False
|
||
node = node[p]
|
||
if not isinstance(node, dict):
|
||
return False
|
||
if last == EXTRA:
|
||
node["note"] = value
|
||
elif value is DELETE:
|
||
node.pop(last, None)
|
||
else:
|
||
node[last] = value
|
||
return True
|
||
|
||
|
||
def bodies(ins):
|
||
keys = {"OPERATOR": OPERATOR, "VIEWER": VIEWER, "METER": METER, "PUBLISHER": PUBLISHER, "RESOLVER": RESOLVER,
|
||
"RELAY": RELAY}
|
||
for name, (signer, positions) in BODY_DEFECTS.items():
|
||
base = ins.receipt(name)
|
||
singles = [(path, i, value) for path, values in positions for i, value in enumerate(values)]
|
||
cases = [[s] for s in singles]
|
||
for a, first in enumerate(singles):
|
||
for second in singles[a + 1:]:
|
||
p, q = first[0], second[0]
|
||
if p[:len(q)] != q and q[:len(p)] != p: # different positions, neither inside the other
|
||
cases.append([first, second])
|
||
for case in cases:
|
||
body = copy.deepcopy(base["body"])
|
||
if not all(edit_at(body, path, value) for path, _, value in case):
|
||
continue
|
||
label = "+".join(".".join(map(str, path)) + f"#{i}" for path, i, _ in case)
|
||
ins.add(f"body_{name[:-len('_accept')]}_{label}", CV.jcs(sign(dict(base, body=body), keys[signer])))
|
||
|
||
|
||
def stages(ins, rng, n):
|
||
"""§2.4.10 and §2.4.11 around their rules: stage bodies built at random — a set of
|
||
one to five units, points that agree or not, objects of random layers, spans (some
|
||
past the set, refused before their span was checked or not), statuses, evidence,
|
||
reasons and bytes, in order or not, a tier with or without a hash — signed by the
|
||
sink, so each is decided at step 2, 4 or later; and the corpus's gap
|
||
claim against object receipts at its coordinates and elsewhere, naming its object or
|
||
another."""
|
||
base = ins.receipt("stage_accept")
|
||
hashes = [CV.blake3(b"differential: layer object %d" % i).hex() for i in range(8)]
|
||
for i in range(n):
|
||
units = rng.randint(1, 5)
|
||
top = rng.randint(0, 3)
|
||
ubs = [0] * (top + 1)
|
||
for _ in range(units):
|
||
ubs[rng.randint(0, top)] += 1
|
||
if rng.random() < 0.1:
|
||
ubs.append(rng.choice([0, 1]))
|
||
objects = []
|
||
for _ in range(rng.randint(0, 6)):
|
||
frm = rng.randint(0, units)
|
||
st = rng.choice(["applied", "applied", "partial", "late", "dropped", "refused"])
|
||
objects.append({
|
||
"layer": rng.choice([1, 1, 2, 2, 3, 0, 256]) if rng.random() < 0.2 else rng.randint(1, max(1, top)),
|
||
"object": rng.choice(hashes),
|
||
"units": {"from": frm, "count": rng.choice([1, units - frm, 0, units + 1, units + 7]) if units > frm else rng.choice([0, 1])},
|
||
"bytes": (0 if st == "dropped" else rng.choice([700, 2400])) if rng.random() < 0.8 else rng.choice([0, 700]),
|
||
"status": st,
|
||
"evidence": rng.choice(["delivered-verified", "delivered-verified", "inclusion-verified", "none"]),
|
||
"reason": (None if st in ("applied", "late", "dropped") else "deadline-missed")
|
||
if rng.random() < 0.8 else rng.choice([None, "layer-conflict", "too-late", "refines-malformed", "base-absent"]),
|
||
})
|
||
if rng.random() < 0.8:
|
||
objects.sort(key=lambda o: (o["layer"], o["units"]["from"], o["object"]))
|
||
lowest = next((j for j, c in enumerate(ubs) if c), 0)
|
||
pt = lambda k: {"wai-stage:layers": k}
|
||
delivered = rng.choice([None, pt(rng.randint(0, 3)), pt(lowest)])
|
||
committed = rng.choice([None, pt(2)]) if delivered is not None else rng.choice([None, None, pt(2)])
|
||
tier = rng.choice(["decode-equivalence", "decode-equivalence", "entropy-consistency", "none"])
|
||
has_hash = (tier == "decode-equivalence") != (rng.random() < 0.1)
|
||
body = dict(base["body"], op_requested=rng.choice([None, committed]), op_committed=committed,
|
||
op_delivered=delivered, presented=pt(lowest if rng.random() < 0.9 else rng.randint(0, 3)),
|
||
units_by_stage=ubs, tier=tier, stage_hash=base["body"]["stage_hash"] if has_hash else None,
|
||
objects=objects, conflicts=rng.choice([0, 0, 1, units + 1]),
|
||
deadline_misses=rng.choice([0, 0, 1, units + 1]),
|
||
bytes_envelopes=rng.choice([5000, 5000, 5000, 0]) + max(0, sum(o["bytes"] for o in objects) - rng.choice([0, 1])))
|
||
ins.add(f"stage_random_{i}", CV.jcs(sign(dict(base, body=body), VIEWER)))
|
||
gap = ins.wire("gap_accept")
|
||
g = json.loads(gap)
|
||
for name, ctx in [
|
||
("at", {"track_alias": 8, "group_id": 3, "object_id": 0, "content_hash": g["subject"]}),
|
||
("other_object", {"track_alias": 8, "group_id": 3, "object_id": 0, "content_hash": hashes[0]}),
|
||
("other_track", {"track_alias": 7, "group_id": 3, "object_id": 0, "content_hash": g["subject"]}),
|
||
("other_group", {"track_alias": 8, "group_id": 4, "object_id": 0, "content_hash": g["subject"]}),
|
||
("other_index", {"track_alias": 8, "group_id": 3, "object_id": 1, "content_hash": hashes[1]}),
|
||
]:
|
||
ins.add("gap_context_" + name, gap, {"gap_object": ctx})
|
||
ins.add("gap_context_to_a_stage_claim", ins.wire("stage_accept"),
|
||
{"gap_object": {"track_alias": 8, "group_id": 3, "object_id": 0, "content_hash": g["subject"]}})
|
||
|
||
|
||
SESSION_CAPS = [("wai.video.int_hyper", "primary", True), ("wai.video.av1", "fallback", False),
|
||
("wai.video.int_refine", "primary", False), ("wai.image.png", "fallback", False),
|
||
("wai.neural.int_hyper", "primary", True), ("wai.image.int_refine", "primary", False)]
|
||
SESSION_ENERGY = [
|
||
None,
|
||
{"joules_micro": 250000, "energy_provenance": "OnChipCounter",
|
||
"energy_uncertainty": {"relative_ppm": 20000, "window_us": 1000000}, "energy_counter_filtering": "Unknown",
|
||
"energy_calibration_ref": None},
|
||
{"joules_micro": 1, "energy_provenance": "OnChipCounter", "energy_uncertainty": {"relative_ppm": 288676, "window_us": 1},
|
||
"energy_counter_filtering": "Off", "energy_calibration_ref": None},
|
||
{"joules_micro": 1, "energy_provenance": "OnChipCounter", "energy_uncertainty": {"relative_ppm": 288675, "window_us": 1},
|
||
"energy_counter_filtering": "On", "energy_calibration_ref": None},
|
||
{"joules_micro": 250000, "energy_provenance": "OnChipCounter",
|
||
"energy_uncertainty": {"relative_ppm": 20000, "window_us": 1000000}, "energy_counter_filtering": None,
|
||
"energy_calibration_ref": None},
|
||
{"joules_micro": 9000, "energy_provenance": "CalibratedInstrument",
|
||
"energy_uncertainty": {"relative_ppm": 5000, "window_us": 1000000}, "energy_counter_filtering": None,
|
||
"energy_calibration_ref": "\u3000"},
|
||
{"joules_micro": 9000, "energy_provenance": "CalibratedInstrument",
|
||
"energy_uncertainty": {"relative_ppm": 5000, "window_us": 1000000}, "energy_counter_filtering": None,
|
||
"energy_calibration_ref": "urn:example:certificate"},
|
||
{"joules_micro": 9000, "energy_provenance": "ModelBased", "energy_uncertainty": None,
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
{"joules_micro": 9000, "energy_provenance": "ModelBased", "energy_uncertainty": {"relative_ppm": 1, "window_us": 1},
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
{"joules_micro": 9000, "energy_provenance": "Estimator", "energy_uncertainty": None,
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
{"joules_micro": 9000, "energy_provenance": "HwShunt", "energy_uncertainty": None,
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
{"joules_micro": 0, "energy_provenance": "ModelBased", "energy_uncertainty": None,
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
{"joules_micro": 9000, "energy_provenance": "onchipcounter", "energy_uncertainty": None,
|
||
"energy_counter_filtering": None, "energy_calibration_ref": None},
|
||
]
|
||
|
||
|
||
def session_stage_claim(rng, n_set, at_stage_1, signer, subject, late=0):
|
||
"""A stage claim (§2.4.10) that passes its own rules: `n_set` units, `at_stage_1` of
|
||
them at stage 1 through one applied first layer, the rest at stage 0; and, with
|
||
`late`, a first layer over the next `late` units that arrived late."""
|
||
ubs = [n_set - at_stage_1, at_stage_1] if at_stage_1 else [n_set]
|
||
objects = [] if not at_stage_1 else [{
|
||
"layer": 1, "object": CV.blake3(b"differential: a stage layer").hex(), "units": {"from": 0, "count": at_stage_1},
|
||
"bytes": 700, "status": "applied", "evidence": "none", "reason": None}]
|
||
if late:
|
||
objects.append({"layer": 1, "object": CV.blake3(b"differential: a late stage layer").hex(),
|
||
"units": {"from": at_stage_1, "count": late}, "bytes": 650, "status": "late", "evidence": "none",
|
||
"reason": None})
|
||
body = {"vocabulary": CV.STAGE_VOCABULARY, "op_requested": None, "op_committed": None, "op_delivered": None,
|
||
"presented": {"wai-stage:layers": 0 if ubs[0] else 1}, "units_by_stage": ubs, "tier": "none",
|
||
"stage_hash": None, "objects": objects, "conflicts": 0, "deadline_misses": 0, "bytes_envelopes": 5700}
|
||
c = {"kind": "claim", "class": "stage", "subject_type": "object", "subject": subject, "seq": 0,
|
||
"issued_at": "2026-09-27T12:00:00Z", "body": body, "signer_id": "did:example:sink", "parent_receipt_hash": None}
|
||
return CV.jcs(sign(c, signer))
|
||
|
||
|
||
def restate(body, objects):
|
||
"""The session's counted sums and transfer restated from `objects`, as a sink states
|
||
them from the objects it lists."""
|
||
body["counted_primary"] = sum(o["counted"] for o in objects if o["counted"] is not None and o["path"] == "primary")
|
||
body["counted_fallback"] = sum(o["counted"] for o in objects if o["counted"] is not None and o["path"] == "fallback")
|
||
body["counted_discarded"] = sum(o["counted"] for o in objects if o["counted"] is not None and not o["presented"])
|
||
body["uncounted"] = sum(1 for o in objects if o["counted"] is None)
|
||
body["transferred"] = sum(o["transferred"] for o in objects)
|
||
|
||
|
||
def encrypted_envelope(wai2):
|
||
"""The model session's WAI2 envelope with a `security` member that encrypts it: inert
|
||
(SPEC §6.4)."""
|
||
ml = int.from_bytes(wai2[4:8], "little")
|
||
m = json.loads(wai2[8:8 + ml])
|
||
m["security"] = {"encryption": "mls-sframe", "kid": 1}
|
||
return wai2_with(json.dumps(m, separators=(",", ":")), wai2[8 + ml:])
|
||
|
||
|
||
def sessions(ins, rng, n):
|
||
"""§2.4.12 around its rules: a random presentation list — units with gaps, one to three
|
||
layers each, through one to four decoder entries — and the session it gives, then a
|
||
defect or two in the body or the contexts, signed by the sink."""
|
||
base = ins.receipt("session_accept")
|
||
wai2 = model_envelope(ins, b"WAI2")
|
||
envelope, wai2_hash = wai2.hex(), SV.content_hash(wai2)
|
||
encrypted = encrypted_envelope(wai2)
|
||
reasons = list(CV.SWITCH_REASONS)
|
||
for i in range(n):
|
||
# The entry object: a WAI1 envelope of one capability or another, so that an
|
||
# entry naming it has its capability or not.
|
||
base_cap = rng.choice(["wai.video.int_hyper", "wai.video.int_motion"])
|
||
base_env = SV.wai1(json.dumps({"wai": "1.0", "media": "video", "intent": "replicate",
|
||
"model_requirement": {"capability": base_cap}, "conditioning": {"kind": "none"}},
|
||
separators=(",", ":")).encode(), b"\x33" * 16)
|
||
entry = SV.content_hash(base_env)
|
||
pool = rng.sample(SESSION_CAPS, rng.randint(1, 4))
|
||
ps = []
|
||
for u in range(rng.randint(1, 12)):
|
||
if rng.random() < 0.1:
|
||
continue
|
||
for layers in range(1, rng.randint(1, 3) + 1):
|
||
if layers == 1 or rng.random() < 0.9:
|
||
ps.append({"unit": u, "layers": layers, "decoder": rng.randrange(len(pool)),
|
||
"out_digest": CV.blake3(b"differential: output %d %d %d" % (i, u, layers)).hex()})
|
||
# Entries in first-use order, as an honest sink lists them.
|
||
order = []
|
||
for p in sorted(ps, key=lambda p: (p["layers"], p["unit"])):
|
||
if p["decoder"] not in order:
|
||
order.append(p["decoder"])
|
||
for p in ps:
|
||
p["decoder"] = order.index(p["decoder"])
|
||
decoders = []
|
||
for j, old in enumerate(order):
|
||
cap, path, pinned = pool[old]
|
||
# The envelope's first rendition is wai.video.int_hyper: an entry naming rendition 0
|
||
# with that capability is bound by it, any other is not.
|
||
rendition = rng.choice([None, 0]) if cap == "wai.video.int_hyper" or rng.random() < 0.2 else None
|
||
# Its object: the WAI2 envelope for a rendition, else the entry object or a
|
||
# layer of its own, and now and then the other.
|
||
if rendition is not None:
|
||
source = wai2_hash if rng.random() < 0.9 else entry
|
||
else:
|
||
source = rng.choice([entry, entry, CV.blake3(b"differential: a source %d %d" % (i, j)).hex(), wai2_hash])
|
||
# Its first unit: the first unit decoded through it, as an honest sink states it,
|
||
# or one off; a rendition of one frame covers that unit alone.
|
||
used = [p["unit"] for p in ps if p["decoder"] == j]
|
||
first = min(used) if used else 0
|
||
decoders.append({"capability": cap, "rendition": rendition, "object": source,
|
||
"first_unit": rng.choice([first, first, first, first + 1, max(0, first - 1), 0]), "path": path,
|
||
"decoder": "builtin:[email protected]", "variant": None,
|
||
"pin": CV.blake3(b"differential: a pin").hex() if pinned else None,
|
||
"tier": "decode-equivalence" if pinned else None, "module_status": None,
|
||
"leaves": sum(1 for p in ps if p["decoder"] == j)})
|
||
fin = CV.finals(ps)
|
||
primary = sum(1 for p in fin.values() if decoders[p["decoder"]]["path"] == "primary")
|
||
ordered = sorted(ps, key=lambda p: (p["unit"], p["layers"]))
|
||
root = CV.out_mth([CV.out_leaf(p) for p in ordered]).hex() if ps else None
|
||
deadline = rng.choice([None, 40000])
|
||
late = rng.sample(ordered, min(len(ordered), rng.randint(0, 3))) if deadline else []
|
||
late = sorted(({"unit": p["unit"], "layers": p["layers"], "late_us": rng.randint(1, 5000)} for p in late),
|
||
key=lambda x: (x["unit"], x["layers"]))
|
||
switches = sorted(({"unit": rng.randint(0, 12), "from": "primary", "to": "fallback", "reason": rng.choice(reasons)}
|
||
for _ in range(rng.randint(0, 3))), key=lambda w: w["unit"])
|
||
def obj(h, path="primary", counted=None, presented=True):
|
||
return {"content_hash": h, "path": path, "counted": counted, "transferred": rng.choice([700, 1200, 3000]),
|
||
"presented": presented}
|
||
objects = [obj(CV.blake3(b"differential: object %d %d" % (i, k)).hex(), rng.choice(["primary", "primary", "fallback"]),
|
||
rng.choice([None, 700, 1200, 0]), rng.random() < 0.8) for k in range(rng.randint(0, 5))]
|
||
if rng.random() < 0.85:
|
||
# The entry object's counted bytes: its envelope's, or a figure that is not.
|
||
objects.insert(rng.randint(0, len(objects)), obj(entry, counted=CV.counted_of(base_env) if rng.random() < 0.9 else 5000))
|
||
# The envelopes held, chosen here so the objects received can hold them or not.
|
||
envs = rng.choice([None, None, [], [envelope], [envelope] * 2, [base_env.hex(), envelope], [base_env.hex()]])
|
||
for h in sorted({d["object"] for d in decoders} | ({wai2_hash} if envs and envelope in envs else set())):
|
||
if h != entry and rng.random() < 0.9 and all(o["content_hash"] != h for o in objects):
|
||
held = h == wai2_hash and rng.random() < 0.9
|
||
objects.append(obj(h, counted=CV.counted_of(wai2) if held else rng.choice([None, 700])))
|
||
stage_wires, stage_sets = [], []
|
||
if fin and rng.random() < 0.5:
|
||
# A stage claim over n_set units from a presented unit f: its first `at1` units
|
||
# at stage 1 through one applied first layer, so a unit of the set presented at
|
||
# two layers or more lies within that layer's span or not, and at three layers
|
||
# beyond any layer it lists (§2.4.12 step 6, both bounds).
|
||
units = sorted(fin)
|
||
f = rng.choice(units)
|
||
n_set = rng.randint(1, 3)
|
||
reached = sum(1 for u, p in fin.items() if f <= u < f + n_set and p["layers"] > 1)
|
||
at1 = min(n_set, max(0, reached + rng.choice([0, 0, 1, -1])))
|
||
signer = VIEWER if rng.random() < 0.9 else RELAY
|
||
base_hash = CV.blake3(b"differential: a base %d" % i).hex()
|
||
w = session_stage_claim(rng, n_set, at1, signer, base_hash)
|
||
if rng.random() < 0.05:
|
||
w = w.replace('"tier":"none"', '"tier":"nonе"')
|
||
stage_wires.append(w)
|
||
stage_sets.append({"first_unit": f, "claim": CV.link_hash(json.loads(w)) if rng.random() < 0.95 else "00" * 32})
|
||
# The set's base and its layer object among the objects received, or not.
|
||
if rng.random() < 0.85:
|
||
objects.append(obj(base_hash, counted=4000))
|
||
if at1 and rng.random() < 0.85:
|
||
objects.append(obj(CV.blake3(b"differential: a stage layer").hex(), counted=700))
|
||
# Presented exactly when an entry names it, as an honest sink states it.
|
||
named = {d["object"] for d in decoders}
|
||
for o in objects:
|
||
o["presented"] = o["content_hash"] in named
|
||
counted = lambda path: sum(o["counted"] for o in objects if o["counted"] is not None and o["path"] == path)
|
||
body = dict(base["body"], decoders=decoders, units=len(fin), leaf_count=len(ps), output_root=root,
|
||
primary_units=primary, fallback_units=len(fin) - primary, switches=switches,
|
||
switches_total=len(switches) + rng.choice([0, 0, 0, 2]), deadline_us=deadline,
|
||
missed=len(late) + rng.choice([0, 0, 0, 1]) if deadline else 0, missed_units=late,
|
||
stage_sets=stage_sets, counted_primary=counted("primary"), counted_fallback=counted("fallback"),
|
||
counted_discarded=sum(o["counted"] for o in objects if o["counted"] is not None and not o["presented"]),
|
||
uncounted=sum(1 for o in objects if o["counted"] is None),
|
||
transferred=sum(o["transferred"] for o in objects),
|
||
energy=copy.deepcopy(rng.choice(SESSION_ENERGY)))
|
||
if rng.random() < 0.5:
|
||
body["classical_declared"], body["classical_source"] = None, None
|
||
ctx = {"outputs": copy.deepcopy(ps), "objects": objects}
|
||
# One or two defects, in the body or in a context.
|
||
inert = None
|
||
for _ in range(rng.choice([0, 1, 1, 2])):
|
||
k = rng.randrange(31)
|
||
ds = body["decoders"]
|
||
if k == 0:
|
||
body["primary_units"] += rng.choice([1, -1]) if body["primary_units"] else 1
|
||
elif k == 1:
|
||
body["units"] += 1
|
||
elif k == 2 and len(ds) > 1 and ds[0]["leaves"] > 1:
|
||
ds[0]["leaves"] -= 1
|
||
ds[1]["leaves"] += 1
|
||
elif k == 3:
|
||
body["leaf_count"] += rng.choice([1, -1]) if body["leaf_count"] else 1
|
||
elif k == 4:
|
||
body["output_root"] = rng.choice([None, CV.blake3(b"differential: another root").hex()])
|
||
elif k == 5 and late:
|
||
late[0]["layers"] += 3
|
||
elif k == 6:
|
||
body["switches_total"] = max(0, len(switches) - 1)
|
||
elif k == 7:
|
||
body["switches"] = [{"unit": u, "from": "primary", "to": "fallback", "reason": "recovered"} for u in range(1025)]
|
||
body["switches_total"] = 1025
|
||
elif k == 8 and ps:
|
||
body["deadline_us"], body["missed"] = 40000, 257
|
||
body["missed_units"] = [{"unit": u, "layers": 1, "late_us": 1} for u in range(257)]
|
||
elif k == 9:
|
||
body["deadline_us"], body["missed"] = None, max(1, body["missed"])
|
||
elif k == 10:
|
||
body["classical_declared"], body["classical_source"] = rng.choice([(700, None), (None, "wai2-table"), (700, "referenced")])
|
||
elif k == 11 and ctx["outputs"]:
|
||
ctx["outputs"][rng.randrange(len(ctx["outputs"]))]["decoder"] = len(ds) + rng.choice([0, 1])
|
||
elif k == 12 and len(ds) > 1:
|
||
ds[0], ds[1] = ds[1], ds[0]
|
||
elif k == 13:
|
||
body["counted_discarded"] += 1
|
||
elif k == 14 and ctx["objects"]:
|
||
o = ctx["objects"][0]
|
||
o["counted"] = None if o["counted"] is not None else 5
|
||
elif k == 15:
|
||
body["host"] = dict(body["host"], threads=rng.choice([0, 3, 6, 2 ** 52]))
|
||
elif k == 16 and ds:
|
||
ds[-1]["decoder"] = rng.choice(["builtin:wai", "builtin:wai@x", "module:sha256:" + "a" * 64, "builtin:Wai@1"])
|
||
elif k == 17 and ds:
|
||
ds[-1]["pin"] = rng.choice([ds[-1]["pin"], CV.blake3(b"differential: a pin").hex()])
|
||
ds[-1]["tier"] = rng.choice(["none", "decode-equivalence", "entropy-consistency", "bit-exact", None])
|
||
elif k == 18 and ds:
|
||
ds.append(dict(ds[0]))
|
||
elif k == 19 and ctx["outputs"]:
|
||
ctx["outputs"].pop(rng.randrange(len(ctx["outputs"])))
|
||
elif k == 20 and ctx["outputs"]:
|
||
ctx["outputs"][0]["out_digest"] = CV.blake3(b"differential: not this output").hex()
|
||
elif k == 21 and stage_sets:
|
||
stage_sets.append(dict(stage_sets[0]))
|
||
elif k == 22:
|
||
body["framing"] = rng.choice(["registered", "wai-measure/0"])
|
||
elif k == 23 and ds:
|
||
ds[0]["module_status"] = rng.choice(["pinned", "fetched", "cached"])
|
||
elif k == 24:
|
||
body["counted_primary"] += rng.choice([1, -1]) if body["counted_primary"] else 1
|
||
elif k == 25 and switches:
|
||
body["switches"] = [dict(switches[0], to="primary")] + switches[1:]
|
||
elif k == 26 and objects:
|
||
# An object listed twice, or a fetch listed apart from the one before it, on
|
||
# the other path; the sums restated to agree.
|
||
o = dict(rng.choice(objects))
|
||
if rng.random() < 0.3:
|
||
o["path"] = "fallback" if o["path"] == "primary" else "primary"
|
||
objects.append(o)
|
||
restate(body, objects)
|
||
elif k == 27 and objects:
|
||
o = rng.choice(objects)
|
||
o["presented"] = not o["presented"]
|
||
restate(body, objects)
|
||
elif k == 28:
|
||
inert = rng.choice(["encrypted", "bytes"])
|
||
elif k == 29 and ds:
|
||
d = rng.choice(ds)
|
||
d["first_unit"] = d["first_unit"] + 1 if d["first_unit"] == 0 or rng.random() < 0.5 else d["first_unit"] - 1
|
||
elif k == 30 and objects:
|
||
# A held envelope's object told a figure other than its counted bytes.
|
||
o = rng.choice(objects)
|
||
o["counted"] = rng.choice([None, 0, (o["counted"] or 0) + 1, max(0, (o["counted"] or 0) - 1)])
|
||
restate(body, objects)
|
||
if rng.random() < 0.5:
|
||
ctx["redecoded"] = [{"index": j, "out_digest": rng.choice([p["out_digest"], CV.blake3(b"differential: redecoded").hex()])}
|
||
for j, p in enumerate(ps) if rng.random() < 0.3]
|
||
if envs is not None:
|
||
ctx["envelopes"] = [{"envelope": e} for e in envs]
|
||
n2 = sum(1 for e in envs if e == envelope)
|
||
if n2:
|
||
body["classical_declared"], body["classical_source"] = rng.choice(
|
||
[(700, "wai2-table"), (700 * n2, "wai2-table"), (701, "wai2-table"), (None, None)])
|
||
if inert:
|
||
# Inert bytes received — an encrypted envelope, or bytes of neither magic — listed
|
||
# as an object neither counted nor presented, held or not, and named by an entry
|
||
# or not.
|
||
data = encrypted if inert == "encrypted" else b"differential: inert bytes %d" % i
|
||
h = SV.content_hash(data)
|
||
if rng.random() < 0.8:
|
||
ctx.setdefault("envelopes", []).append({"envelope": data.hex()})
|
||
if body["decoders"] and rng.random() < 0.3:
|
||
d = rng.choice(body["decoders"])
|
||
d["object"], d["rendition"] = h, None
|
||
if "objects" in ctx:
|
||
named = {d["object"] for d in body["decoders"]}
|
||
ctx["objects"].append(obj(h, counted=None, presented=h in named))
|
||
restate(body, ctx["objects"])
|
||
if stage_wires:
|
||
ctx["stage_claims"] = stage_wires
|
||
if rng.random() < 0.1:
|
||
del ctx["outputs"]
|
||
if rng.random() < 0.1:
|
||
del ctx["objects"]
|
||
ins.add(f"session_random_{i}", CV.jcs(sign(dict(base, subject=entry, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def session_stage_bounds(ins, rng, n):
|
||
"""§2.4.12 step 6 against a stage claim, both ways, on sessions that pass steps 1-5: a
|
||
set of one to six units whose first layer applied to some and arrived late for others,
|
||
and presentations of its units at one to three layers. A unit presented at two layers
|
||
lies within an arrived first layer or not, one at three layers never has an arrived
|
||
second layer, and the units presented at two layers or more reach the units the claim
|
||
counts at stage 1 or fall short. The objects received hold the set's base and its
|
||
layers, or lack either."""
|
||
base = ins.receipt("session_accept")
|
||
dec = [dict(base["body"]["decoders"][0], leaves=0), dict(base["body"]["decoders"][3], leaves=0)]
|
||
for i in range(n):
|
||
n_set = rng.randint(1, 6)
|
||
at1 = rng.randint(0, n_set)
|
||
late = rng.randint(0, n_set - at1)
|
||
ps = []
|
||
for u in range(n_set):
|
||
top = 1
|
||
if u < at1 + late and rng.random() < 0.8:
|
||
top = rng.choice([2, 2, 3])
|
||
elif rng.random() < 0.15:
|
||
top = 2
|
||
for layers in range(1, top + 1):
|
||
ps.append({"unit": u, "layers": layers, "decoder": 0 if layers == 1 else 1,
|
||
"out_digest": CV.blake3(b"differential: bounds %d %d %d" % (i, u, layers)).hex()})
|
||
base_hash = CV.blake3(b"differential: a bounded base %d" % i).hex()
|
||
decoders = [dict(dec[0], object=base_hash, leaves=sum(1 for p in ps if p["decoder"] == 0))]
|
||
if any(p["decoder"] == 1 for p in ps):
|
||
decoders.append(dict(dec[1], object=CV.blake3(b"differential: a stage layer").hex(), leaves=sum(1 for p in ps if p["decoder"] == 1)))
|
||
if rng.random() < 0.15:
|
||
# The base's entry, or its layer's, at another first unit than the set's.
|
||
rng.choice(decoders)["first_unit"] += 1
|
||
w = session_stage_claim(rng, n_set, rng.choice([at1, at1, max(0, at1 - 1), min(n_set, at1 + 1)]), VIEWER, base_hash,
|
||
late=late if late and at1 + late <= n_set else 0)
|
||
objects = []
|
||
for h, cnt in [(base_hash, 4000), (CV.blake3(b"differential: a stage layer").hex(), 700),
|
||
(CV.blake3(b"differential: a late stage layer").hex(), 650)]:
|
||
if rng.random() < 0.9:
|
||
objects.append({"content_hash": h, "path": "primary", "counted": cnt, "transferred": cnt,
|
||
"presented": h in {d["object"] for d in decoders}})
|
||
entry = objects[0]["content_hash"] if objects else base_hash
|
||
ordered = sorted(ps, key=lambda p: (p["unit"], p["layers"]))
|
||
body = dict(base["body"], decoders=decoders, units=n_set, leaf_count=len(ps), primary_units=n_set, fallback_units=0,
|
||
output_root=CV.out_mth([CV.out_leaf(p) for p in ordered]).hex(), switches=[], switches_total=0,
|
||
deadline_us=None, missed=0, missed_units=[], stage_sets=[{"first_unit": 0, "claim": CV.link_hash(json.loads(w))}],
|
||
counted_primary=sum(o["counted"] for o in objects), counted_fallback=0,
|
||
counted_discarded=sum(o["counted"] for o in objects if not o["presented"]), uncounted=0,
|
||
transferred=sum(o["transferred"] for o in objects), classical_declared=None, classical_source=None,
|
||
energy=None)
|
||
ctx = {"outputs": ps, "objects": objects, "stage_claims": [w]}
|
||
if rng.random() < 0.1:
|
||
del ctx["objects"]
|
||
ins.add(f"session_bounds_{i}", CV.jcs(sign(dict(base, subject=entry, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def epoch_stage_claim(rng, i, k, n_set, subject):
|
||
"""A stage claim (§2.4.10) that passes its own rules, as the claim of one epoch of a
|
||
set of `n_set` units: a first layer of any status over a span, perhaps a second over
|
||
other units, and perhaps a second layer applied or not within an applied first; its
|
||
units at each stage as many as its applied layers can have reached, or fewer."""
|
||
def obj(layer, j, frm, cnt, status):
|
||
h = CV.blake3(b"differential: epoch layer %d %d %d %d" % (i, k, layer, j)).hex()
|
||
return {"layer": layer, "object": h, "units": {"from": frm, "count": cnt},
|
||
"bytes": 0 if status == "dropped" else 600 + 10 * j + layer, "status": status, "evidence": "none",
|
||
"reason": {"partial": "layer-gap", "refused": "geometry-mismatch"}.get(status)}
|
||
objects = []
|
||
cut = rng.randint(1, n_set)
|
||
st1 = rng.choice(["applied", "applied", "partial", "late", "refused", "dropped"])
|
||
if st1 == "partial" and cut < 2:
|
||
st1 = "applied"
|
||
objects.append(obj(1, 0, 0, cut, st1))
|
||
if cut < n_set and rng.random() < 0.6:
|
||
st = rng.choice(["applied", "late", "refused", "dropped"])
|
||
objects.append(obj(1, 1, cut, n_set - cut, st))
|
||
reach1 = sum(o["units"]["count"] if o["status"] == "applied" else o["units"]["count"] - 1 if o["status"] == "partial" else 0
|
||
for o in objects)
|
||
reach2 = 0
|
||
if st1 == "applied" and rng.random() < 0.6:
|
||
st = rng.choice(["applied", "applied", "late", "refused"])
|
||
objects.append(obj(2, 0, 0, 1, st))
|
||
reach2 = 1 if st == "applied" else 0
|
||
s1 = rng.randint(0, reach1)
|
||
s2 = rng.randint(0, min(reach2, s1))
|
||
ubs = [n_set - s1, s1 - s2, s2] if s2 else ([n_set - s1, s1] if s1 else [n_set])
|
||
objects.sort(key=lambda o: (o["layer"], o["units"]["from"], o["object"]))
|
||
body = {"vocabulary": CV.STAGE_VOCABULARY, "op_requested": None, "op_committed": None, "op_delivered": None,
|
||
"presented": {"wai-stage:layers": next(j for j, c in enumerate(ubs) if c)}, "units_by_stage": ubs, "tier": "none",
|
||
"stage_hash": None, "objects": objects, "conflicts": 0, "deadline_misses": 0,
|
||
"bytes_envelopes": 5000 + sum(o["bytes"] for o in objects)}
|
||
c = {"kind": "claim", "class": "stage", "subject_type": "object", "subject": subject, "seq": k,
|
||
"issued_at": "2026-09-27T12:00:00Z", "body": body, "signer_id": "did:example:sink", "parent_receipt_hash": None}
|
||
return CV.jcs(sign(c, VIEWER)), objects
|
||
|
||
|
||
def session_stage_epochs(ins, rng, n):
|
||
"""§2.4.12 step 6 from above across epochs: one to three stage claims of one set, as a
|
||
sink that forgot its base signs one per epoch, each listing first and second layers of
|
||
every status; sometimes one named one unit further on. The session presents each unit
|
||
at one layer and some at two or three, within what one of the claims holding the unit
|
||
applied or past it, and receives the arrived layers or misses one."""
|
||
base = ins.receipt("session_accept")
|
||
dec = [dict(base["body"]["decoders"][0], leaves=0), dict(base["body"]["decoders"][3], leaves=0)]
|
||
for i in range(n):
|
||
n_set = rng.randint(2, 6)
|
||
base_hash = CV.blake3(b"differential: an epoch base %d" % i).hex()
|
||
claims = [epoch_stage_claim(rng, i, k, n_set, base_hash) for k in range(rng.randint(1, 3))]
|
||
named = [(1 if k and rng.random() < 0.15 else 0, w) for k, (w, _) in enumerate(claims)]
|
||
ps = []
|
||
for u in range(n_set + 1):
|
||
top = rng.choice([1, 1, 2, 2, 2, 3])
|
||
for layers in range(1, top + 1):
|
||
ps.append({"unit": u, "layers": layers, "decoder": 0 if layers == 1 else 1,
|
||
"out_digest": CV.blake3(b"differential: epochs %d %d %d" % (i, u, layers)).hex()})
|
||
decoders = [dict(dec[0], object=base_hash, leaves=sum(1 for p in ps if p["decoder"] == 0))]
|
||
objects = [{"content_hash": base_hash, "path": "primary", "counted": 4000, "transferred": 4000, "presented": True}]
|
||
seen = set()
|
||
for _, objs in claims:
|
||
for o in objs:
|
||
if o["status"] != "dropped" and o["object"] not in seen and rng.random() < 0.95:
|
||
seen.add(o["object"])
|
||
objects.append({"content_hash": o["object"], "path": "primary", "counted": o["bytes"],
|
||
"transferred": o["bytes"], "presented": o["status"] in ("applied", "partial")})
|
||
if any(p["decoder"] == 1 for p in ps):
|
||
# The refinement's entry names a layer the session received, or its base.
|
||
layer = next((o["content_hash"] for o in objects[1:] if o["presented"]), base_hash)
|
||
decoders.append(dict(dec[1], object=layer, leaves=sum(1 for p in ps if p["decoder"] == 1)))
|
||
# Presented exactly when an entry names it, as an honest sink states it.
|
||
for o in objects:
|
||
o["presented"] = o["content_hash"] in {d["object"] for d in decoders}
|
||
if rng.random() < 0.15:
|
||
# The base's entry, or a layer's, at another first unit than the claims'.
|
||
rng.choice(decoders)["first_unit"] += rng.choice([1, 2])
|
||
units = n_set + 1
|
||
ordered = sorted(ps, key=lambda p: (p["unit"], p["layers"]))
|
||
sets = sorted({(f, CV.link_hash(json.loads(w))) for f, w in named})
|
||
body = dict(base["body"], decoders=decoders, units=units, leaf_count=len(ps), primary_units=units, fallback_units=0,
|
||
output_root=CV.out_mth([CV.out_leaf(p) for p in ordered]).hex(), switches=[], switches_total=0,
|
||
deadline_us=None, missed=0, missed_units=[], stage_sets=[{"first_unit": f, "claim": h} for f, h in sets],
|
||
counted_primary=sum(o["counted"] for o in objects), counted_fallback=0,
|
||
counted_discarded=sum(o["counted"] for o in objects if not o["presented"]), uncounted=0,
|
||
transferred=sum(o["transferred"] for o in objects), classical_declared=None, classical_source=None,
|
||
energy=None)
|
||
ctx = {"outputs": ps, "objects": objects, "stage_claims": [w for w, _ in claims]}
|
||
ins.add(f"session_epochs_{i}", CV.jcs(sign(dict(base, subject=base_hash, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def wiv1(frames, keyframe):
|
||
"""A WIV1 clip of `frames` frames, 16 by 16, built as the corpus generator builds its
|
||
own: `keyframe` its WIH1 keyframe, its residual stream filler that nothing decodes."""
|
||
block = 1 if frames == 1 else 16
|
||
nb = (16 // block) ** 2
|
||
return CV.IPV.wiv1_from_json({"n_frames": frames, "h": 16, "w": 16, "block": block, "q": 1, "offset": 0,
|
||
"cdf": [0, 32768, 65536], "mvs": [[]] + [[[0, 0]] * nb for _ in range(frames - 1)],
|
||
"rans_b64": base64.b64encode(b"\x33" * 8).decode()}, keyframe=keyframe)
|
||
|
||
|
||
def wai2_of(entries, payloads):
|
||
"""A WAI2 envelope of the manifest entries `entries`, each with its payload's SHA-256
|
||
unless it states its own, over `payloads`."""
|
||
m = {"wai": "1.1", "media": "video", "intent": "replicate",
|
||
"renditions": [dict({"sha256": hashlib.sha256(p).hexdigest()}, **e) for e, p in zip(entries, payloads)], "target": None}
|
||
t = json.dumps(m, separators=(",", ":")).encode()
|
||
table, o = b"", 0
|
||
for p in payloads:
|
||
table += o.to_bytes(4, "little") + len(p).to_bytes(4, "little")
|
||
o += len(p)
|
||
return b"WAI2" + len(t).to_bytes(4, "little") + t + len(payloads).to_bytes(2, "little") + table + b"".join(payloads)
|
||
|
||
|
||
def wir1_units(n, _cache={}):
|
||
"""A WIR1 payload of `n` units that parses (staged-delivery Appendix A): the corpus's
|
||
four-unit layer's header and table, its first unit's stream for every unit. Nothing
|
||
decodes it."""
|
||
if "four" not in _cache:
|
||
raise RuntimeError("wir1_units before session_held read the corpus's layer")
|
||
p = _cache["four"]
|
||
w = SV.parse_wir1(p, "rgb8")
|
||
cdf_len = struct.unpack_from("<H", p, 40)[0]
|
||
head = p[:8] + struct.pack("<I", n) + p[12:44]
|
||
tables = p[44:44 + 4 * cdf_len * (n if w["per_unit"] else 1)] if not w["per_unit"] else p[44:44 + 4 * cdf_len] * n
|
||
s0 = w["streams"][0]
|
||
return head + tables + struct.pack("<I", len(s0)) * n + s0 * n
|
||
|
||
|
||
def session_held(ins, rng, n):
|
||
"""§2.4.12 step 6 against the envelopes with the outputs and the objects, around the
|
||
rules that read what a held envelope gives: a base of one to six frames — a WAI1
|
||
envelope whose manifest may name a fallback the base's entry decodes through, or the
|
||
base component of a WAI2 envelope — and layers refining it at layer 1 or 2 over a span
|
||
or over all of its units, as WAI1 envelopes or as refinement components of that WAI2
|
||
envelope, now and then naming another base or one that is a layer, at layer 0, with an
|
||
empty span or one past the base, or with no `refines`; presentations through each at
|
||
one to three layers, of units within what the envelope covers and past it, through
|
||
entries whose `first_unit` is the base's first unit or one off, the base's and the
|
||
layers' each; the objects each envelope's figures — counted bytes, length, fetched once
|
||
or more, presented exactly when an entry names it — or one told otherwise, listed
|
||
twice, or inert bytes received beside them; the base held or not."""
|
||
model = ins.by_name["session_held_layer_accept"]
|
||
base = json.loads(model["receipt"])
|
||
envs = [bytes.fromhex(e["envelope"]) for e in model["context"]["envelopes"]]
|
||
whole = [bytes.fromhex(e["envelope"]) for e in ins.by_name["session_unit_whole_layer_accept"]["context"]["envelopes"]][1]
|
||
base_m, _, base_p = SV.read_wai1(envs[0])
|
||
keyframe = CV.IPV.parse_wiv1(base_p)["wih1"]
|
||
one_m, _, one_p = SV.read_wai1(envs[1]) # a layer coding one unit
|
||
_, _, four_p = SV.read_wai1(whole) # and one coding four
|
||
wir1_units.__defaults__[0]["four"] = four_p
|
||
jtext = lambda m: json.dumps(m, separators=(",", ":")).encode()
|
||
dec = {d["capability"]: d for d in base["body"]["decoders"]}
|
||
encrypted = encrypted_envelope(model_envelope(ins, b"WAI2"))
|
||
HYPER, REFINE = "wai.video.int_hyper", "wai.video.int_refine"
|
||
for i in range(n):
|
||
frames = rng.choice([1, 2, 3, 4, 4, 6])
|
||
first = rng.choice([0, 0, 1, 2, 5])
|
||
form = rng.choice(["wai1", "wai1", "wai2", "alias"])
|
||
listed_only = [] # objects received and listed whose bytes the verifier does not hold
|
||
# The layers: at layer 1 or 2, over a span or the whole base, and now and then one a
|
||
# sink cannot apply.
|
||
specs = []
|
||
for k in range(rng.choice([0, 1, 1, 2, 3])):
|
||
level = 1 if k == 0 or rng.random() < 0.5 else 2
|
||
frm = rng.randint(0, frames)
|
||
count = rng.randint(1, max(1, frames - min(frm, frames - 1)))
|
||
span = None if rng.random() < 0.3 else {"from": frm, "count": count}
|
||
bad = rng.choice(["layer0", "empty", "absent", "past", "on_layer"]) if rng.random() < 0.12 else None
|
||
specs.append((level, span, bad))
|
||
|
||
def refines(base_ref, prev, level, span, bad, on_layer):
|
||
r = {"base": base_ref, "prev": prev, "layer": 0 if bad == "layer0" else level}
|
||
if bad == "empty":
|
||
r["span"] = {"from": 0, "count": 0}
|
||
elif bad == "past":
|
||
r["span"] = {"from": frames, "count": 2}
|
||
elif span is not None:
|
||
r["span"] = span
|
||
if bad == "on_layer" and on_layer is not None:
|
||
r["base"] = r["prev"] = on_layer
|
||
r["layer"] = 1
|
||
return r
|
||
|
||
# A layer's WIR1 payload: a unit for each unit it covers, or now and then one fewer.
|
||
payload_of = lambda span: wir1_units(max(1, (span["count"] if span is not None and span["count"] else frames) # noqa: E731
|
||
- (1 if rng.random() < 0.08 else 0)))
|
||
refine_cap = lambda: "wai.image.int_refine" if rng.random() < 0.05 else REFINE # noqa: E731
|
||
if form in ("wai1", "alias"):
|
||
if form == "wai1":
|
||
# The base's capability and fallback: one format, or a fallback of another
|
||
# format than the capability's, or a companion-only capability.
|
||
cap, fallback = rng.choice([(HYPER, None), (HYPER, None), ("wai.video.int_motion", None), (HYPER, "wai.video.av1"),
|
||
("wai.neural.int_hyper", HYPER), ("wai.audio.int_codec", "wai.video.int_motion"),
|
||
("wai.meta.hdr10plus", HYPER)])
|
||
mr = {"capability": cap, "fallback": fallback} if fallback else {"capability": cap}
|
||
base_env = SV.wai1(jtext(dict(base_m, model_requirement=mr)), wiv1(frames, keyframe) if rng.random() < 0.95 else b"\x33" * 16)
|
||
subject = layer_base = SV.content_hash(base_env)
|
||
held_envs = [base_env]
|
||
base_entry = dict(dec[HYPER], capability=cap, rendition=None, object=subject, first_unit=first)
|
||
if fallback and rng.random() < 0.5:
|
||
base_entry["capability"] = fallback # decoded through the fallback
|
||
else:
|
||
# The base a WAI2 envelope's component, which the layers refine by its
|
||
# exploded envelope, and which its entry names in either form.
|
||
w = wai2_of([{"capability": HYPER, "kind": "int_hyper", "id": "b"}], [wiv1(frames, keyframe)])
|
||
x = SV.explode_each(w)[0]
|
||
layer_base = SV.content_hash(x)
|
||
if rng.random() < 0.7:
|
||
subject, held_envs = SV.content_hash(w), [w] + ([x] if rng.random() < 0.3 else [])
|
||
base_entry = dict(dec[HYPER], rendition=0, object=subject, first_unit=first)
|
||
else:
|
||
# The exploded envelope's own bytes held, or only the WAI2 the entry
|
||
# resolves to, or neither.
|
||
subject = layer_base
|
||
held_envs = ([x] if rng.random() < 0.5 else []) + ([w] if rng.random() < 0.8 else [])
|
||
listed_only = [] if x in held_envs else [x]
|
||
base_entry = dict(dec[HYPER], rendition=None, object=subject, first_unit=first)
|
||
if base_entry["capability"] != HYPER:
|
||
base_entry.update(pin=None, tier=None) # no pin above its capability's class
|
||
layer_objects, layer_caps = [], []
|
||
for k, (level, span, bad) in enumerate(specs):
|
||
ref_base = layer_base if rng.random() < 0.9 else CV.blake3(b"differential: another base %d" % i).hex()
|
||
prev = ref_base if level == 1 or not layer_objects else layer_objects[-1]
|
||
lm = copy.deepcopy(one_m)
|
||
lc = refine_cap()
|
||
lm["model_requirement"] = dict(lm["model_requirement"], capability=lc)
|
||
if bad == "absent":
|
||
lm.pop("refines", None)
|
||
else:
|
||
lm["refines"] = refines(ref_base, prev, level, span, bad, layer_objects[0] if layer_objects else None)
|
||
e = SV.wai1(jtext(lm), payload_of(span))
|
||
held_envs.append(e)
|
||
layer_objects.append(SV.content_hash(e))
|
||
layer_caps.append(lc)
|
||
layer_entries = [dict(dec[REFINE], capability=lc, rendition=None, object=h) for h, lc in zip(layer_objects, layer_caps)]
|
||
else:
|
||
comps, payloads = [{"capability": HYPER, "kind": "int_hyper", "id": "b"}], [wiv1(frames, keyframe)]
|
||
if rng.random() < 0.05:
|
||
comps[0]["role"] = "companion"
|
||
for k, (level, span, bad) in enumerate(specs):
|
||
base_id = "b" if rng.random() < 0.92 else rng.choice(["c", "B", "r0"])
|
||
prev = base_id if level == 1 or k == 0 else f"r{k - 1}"
|
||
e = {"capability": refine_cap(), "kind": "refine", "id": f"r{k}", "role": "refinement"}
|
||
if bad != "absent":
|
||
e["refines"] = refines(base_id, prev, level, span, bad, "r0" if k else None)
|
||
# Now and then a refinement that does not conform, or does not match its digest.
|
||
flaw = rng.choice(["no_sha", "no_role", "wrong_sha"]) if rng.random() < 0.08 else None
|
||
if flaw == "no_sha":
|
||
e["sha256"] = None
|
||
elif flaw == "no_role":
|
||
del e["role"]
|
||
elif flaw == "wrong_sha":
|
||
e["sha256"] = hashlib.sha256(b"differential: another payload").hexdigest()
|
||
comps.append(e)
|
||
payloads.append(payload_of(span))
|
||
env = wai2_of(comps, payloads)
|
||
subject = SV.content_hash(env)
|
||
held_envs = [env]
|
||
base_entry = dict(dec[HYPER], rendition=0, object=subject, first_unit=first)
|
||
each = SV.explode_each(env)
|
||
if rng.random() < 0.25 and each[0] is not None:
|
||
# The base component named by its exploded envelope, its own bytes held or
|
||
# not (then the entry resolves to the WAI2's rendition).
|
||
x = each[0]
|
||
if rng.random() < 0.5:
|
||
held_envs.append(x)
|
||
else:
|
||
listed_only.append(x)
|
||
subject = SV.content_hash(x)
|
||
base_entry = dict(dec[HYPER], rendition=None, object=subject, first_unit=first)
|
||
layer_entries = []
|
||
for k in range(len(specs)):
|
||
xk = each[k + 1] if k + 1 < len(each) else None
|
||
if xk is not None and rng.random() < 0.25:
|
||
# A refinement named by its exploded envelope's content hash.
|
||
layer_entries.append(dict(dec[REFINE], capability=comps[k + 1]["capability"], rendition=None, object=SV.content_hash(xk)))
|
||
if rng.random() < 0.3:
|
||
held_envs.append(xk)
|
||
else:
|
||
listed_only.append(xk)
|
||
else:
|
||
layer_entries.append(dict(dec[REFINE], capability=comps[k + 1]["capability"], rendition=k + 1, object=SV.content_hash(env)))
|
||
for d in layer_entries:
|
||
d["first_unit"] = first if rng.random() < 0.9 else rng.choice([0, first + 1])
|
||
# Each unit at one layer through the base, and some at two or three through a layer
|
||
# of that level — one covering the unit, or any.
|
||
units = list(range(first, first + frames))
|
||
if rng.random() < 0.2:
|
||
units.append(rng.choice([first + frames, max(0, first - 1)]))
|
||
ps = []
|
||
for u in sorted(set(units)):
|
||
ps.append({"unit": u, "layers": 1, "decoder": 0})
|
||
for level in (1, 2):
|
||
cands = [j for j, (lv, _, _) in enumerate(specs) if lv == level]
|
||
cover = [j for j in cands if specs[j][1] is None or specs[j][1]["from"] <= u - first < specs[j][1]["from"] + specs[j][1]["count"]]
|
||
pick = cover if cover and rng.random() < 0.85 else cands
|
||
if not pick or rng.random() < 0.4:
|
||
break
|
||
ps.append({"unit": u, "layers": level + 1, "decoder": 1 + rng.choice(pick)})
|
||
if rng.random() < 0.15 and specs:
|
||
# A presentation through an entry whose envelope gives another layer count.
|
||
p = rng.choice(ps)
|
||
p["decoder"] = 1 + rng.randrange(len(specs)) if p["layers"] == 1 or rng.random() < 0.6 else 0
|
||
for p in ps:
|
||
p["out_digest"] = CV.blake3(b"differential: held %d %d %d" % (i, p["unit"], p["layers"])).hex()
|
||
# Entries in first-use order, the unused layers named by none.
|
||
order = []
|
||
for p in sorted(ps, key=lambda p: (p["layers"], p["unit"])):
|
||
if p["decoder"] not in order:
|
||
order.append(p["decoder"])
|
||
for p in ps:
|
||
p["decoder"] = order.index(p["decoder"])
|
||
all_entries = [base_entry] + layer_entries
|
||
decoders = []
|
||
for j, old in enumerate(order):
|
||
d = dict(all_entries[old])
|
||
d["leaves"] = sum(1 for p in ps if p["decoder"] == j)
|
||
decoders.append(d)
|
||
if rng.random() < 0.1:
|
||
decoders[0]["first_unit"] = rng.choice([first + 1, max(0, first - 1)])
|
||
held = list(held_envs)
|
||
named = {d["object"] for d in decoders}
|
||
listed = held + [e for e in listed_only if e not in held]
|
||
objects = [{"content_hash": SV.content_hash(e), "path": "primary", "counted": CV.counted_of(e),
|
||
"transferred": len(e) * rng.choice([1, 1, 1, 2, 3]), "presented": SV.content_hash(e) in named}
|
||
for e in listed]
|
||
# One or two defects among the objects and envelopes.
|
||
inert = None
|
||
for _ in range(rng.choice([0, 0, 1, 1, 2])):
|
||
k = rng.randrange(6)
|
||
o = rng.choice(objects)
|
||
if k == 0:
|
||
objects.append(dict(o))
|
||
elif k == 1:
|
||
o["presented"] = not o["presented"]
|
||
elif k == 2:
|
||
o["counted"] = rng.choice([None, 0, (o["counted"] or 0) + 1, max(0, (o["counted"] or 0) - 1)])
|
||
elif k == 3 and len(held) > 1:
|
||
held = held[1:] # the base not held
|
||
elif k == 4:
|
||
inert = rng.choice([encrypted, b"differential: inert bytes %d" % i])
|
||
elif k == 5:
|
||
objects.append(dict(o, path="fallback")) # fetched again on the other path
|
||
envelopes = list(held)
|
||
if inert is not None:
|
||
ih = SV.content_hash(inert)
|
||
envelopes.append(inert)
|
||
if rng.random() < 0.3 and len(decoders) > 1:
|
||
decoders[-1]["object"] = ih
|
||
objects.append({"content_hash": ih, "path": "primary", "counted": None, "transferred": len(inert),
|
||
"presented": ih in {d["object"] for d in decoders}})
|
||
rng.shuffle(envelopes)
|
||
fin = CV.finals(ps)
|
||
ordered = sorted(ps, key=lambda p: (p["unit"], p["layers"]))
|
||
body = dict(base["body"], decoders=decoders, units=len(fin), leaf_count=len(ps), primary_units=len(fin),
|
||
fallback_units=0, output_root=CV.out_mth([CV.out_leaf(p) for p in ordered]).hex())
|
||
restate(body, objects)
|
||
ctx = {"outputs": ps, "objects": objects, "envelopes": [{"envelope": e.hex()} for e in envelopes]}
|
||
if rng.random() < 0.05:
|
||
del ctx["outputs"]
|
||
ins.add(f"session_held_{form}_{i}", CV.jcs(sign(dict(base, subject=subject, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def session_encrypted(ins, rng, n):
|
||
"""§2.4.12 against an encrypted envelope, whose manifest and rendition table are in the
|
||
clear (SPEC §6.4): the model session with its WAI2 envelope encrypted — its security
|
||
well formed or not — and the entry decoding it naming a rendition in the table, out
|
||
of it or none, its capability or another; the named rendition's sha256 kept, absent,
|
||
well formed and unmatched, or malformed (uppercase, 63 digits, not hex), which matches
|
||
no payload whether or not the envelope is encrypted; the classical length stated as
|
||
the table gives it, or not; the plaintext envelope held beside it or not."""
|
||
model = ins.by_name["session_envelope_encrypted"]
|
||
base = json.loads(model["receipt"])
|
||
enc = bytes.fromhex(model["context"]["envelopes"][0]["envelope"])
|
||
ml = int.from_bytes(enc[4:8], "little")
|
||
m0, rest = json.loads(enc[8:8 + ml]), enc[8 + ml:]
|
||
plain = model_envelope(ins, b"WAI2")
|
||
for i in range(n):
|
||
m = copy.deepcopy(m0)
|
||
m["security"] = rng.choice([m0["security"], m0["security"], {"encryption": "mls-sframe", "kid": -1}, {"encryption": 5}, {"encryption": "none"}])
|
||
r0 = m["renditions"][0]
|
||
good = r0.get("sha256") or hashlib.sha256(b"a payload").hexdigest()
|
||
form = rng.choice(["keep", "keep", "keep", "absent", "unmatched", "upper", "short", "not_hex"])
|
||
if form == "absent":
|
||
r0.pop("sha256", None)
|
||
elif form != "keep":
|
||
r0["sha256"] = {"unmatched": hashlib.sha256(b"another payload").hexdigest(), "upper": good.upper(),
|
||
"short": good[:63], "not_hex": "z" * 64}[form]
|
||
e = wai2_with(json.dumps(m, separators=(",", ":")), rest)
|
||
body = copy.deepcopy(base["body"])
|
||
d = body["decoders"][1]
|
||
d["object"] = SV.content_hash(e)
|
||
d["rendition"] = rng.choice([0, 0, 1, 2, 9, None])
|
||
d["capability"] = rng.choice(["wai.video.int_hyper", "wai.video.int_hyper", "wai.video.avc", "wai.audio.opus"])
|
||
if d["capability"] != "wai.video.int_hyper":
|
||
d["pin"], d["tier"] = None, None
|
||
envs = [e] + ([plain] if rng.random() < 0.3 else [])
|
||
body["classical_declared"] = rng.choice([700, 700, None, 700 * len(envs), 701])
|
||
body["classical_source"] = None if body["classical_declared"] is None else "wai2-table"
|
||
ctx = dict(model["context"], envelopes=[{"envelope": x.hex()} for x in envs])
|
||
ins.add(f"session_encrypted_{i}", CV.jcs(sign(dict(base, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def session_roles_sweep(ins):
|
||
"""SPEC §7.1's component rules, which decide whether a WAI2 offers its refinements, over
|
||
every registered capability and an unregistered string: a refinement beside a companion
|
||
of each capability bound to the base, and beside an HDR10+ descriptor and an
|
||
enhancement layer bound to a rendition of each capability."""
|
||
model = ins.by_name["session_roles_companion_and_refinement_accept"]
|
||
base = json.loads(model["receipt"])
|
||
w = bytes.fromhex(model["context"]["envelopes"][0]["envelope"])
|
||
_, m, payloads = CV.read_envelope(w)
|
||
b, comp, ref = m["renditions"]
|
||
for cap in sorted(CV.REGISTRY) + ["org.example.meta"]:
|
||
cases = [
|
||
("companion", [b, dict(comp, capability=cap), ref], [payloads[0], payloads[1], payloads[2]], 2),
|
||
("hdr_over", [b, {"capability": cap, "kind": "k", "id": "x"}, dict(comp, companion_of=cap, companion_of_id=None), ref],
|
||
[payloads[0], payloads[0], payloads[1], payloads[2]], 3),
|
||
("lcevc_over", [b, {"capability": cap, "kind": "k", "id": "x"}, dict(comp, capability="wai.video.lcevc", companion_of=cap, companion_of_id=None), ref],
|
||
[payloads[0], payloads[0], payloads[1], payloads[2]], 3),
|
||
]
|
||
for tag, entries, pls, ri in cases:
|
||
entries = [{k: v for k, v in e.items() if v is not None} for e in entries]
|
||
e2 = wai2_of(entries, pls)
|
||
h = SV.content_hash(e2)
|
||
body = copy.deepcopy(base["body"])
|
||
body["decoders"][0]["object"] = h
|
||
body["decoders"][1].update(object=h, rendition=ri)
|
||
objects = [{"content_hash": h, "path": "primary", "counted": CV.counted_of(e2), "transferred": len(e2), "presented": True}]
|
||
restate(body, objects)
|
||
ctx = dict(model["context"], objects=objects, envelopes=[{"envelope": e2.hex()}])
|
||
ins.add(f"session_roles_{tag}_{cap}", CV.jcs(sign(dict(base, subject=h, body=body), VIEWER)), ctx)
|
||
|
||
|
||
def wai2_with(manifest_text, rest):
|
||
"""A WAI2 envelope of the manifest text (or bytes) given, with the rest of the session
|
||
corpus's envelope: its rendition count, table and payloads."""
|
||
t = manifest_text.encode() if isinstance(manifest_text, str) else manifest_text
|
||
return b"WAI2" + len(t).to_bytes(4, "little") + t + rest
|
||
|
||
|
||
def model_envelope(ins, magic):
|
||
"""The model session's envelope of this magic: its base's WAI1, or its WAI2."""
|
||
for e in ins.by_name["session_accept"]["context"]["envelopes"]:
|
||
b = bytes.fromhex(e["envelope"])
|
||
if b[:4] == magic:
|
||
return b
|
||
raise ValueError(f"the model session holds no {magic!r} envelope")
|
||
|
||
|
||
def sourced(ins, entry, env):
|
||
"""The model session with decoder entry `entry`'s object the envelope given, signed
|
||
again: the session that decoded from it."""
|
||
c = copy.deepcopy(ins.receipt("session_accept"))
|
||
c["body"]["decoders"][entry]["object"] = SV.content_hash(env)
|
||
return CV.jcs(sign(c, VIEWER))
|
||
|
||
|
||
def session_envelopes(ins):
|
||
"""§2.4.12 step 6, the envelopes: the model session against its WAI2 envelope with
|
||
manifests every reader may read otherwise — members of the wrong type, pins that are
|
||
not pins, a `security` that encrypts or is no security object, duplicate members,
|
||
numbers past every range, nesting at and past the depth a reader allows, byte-order
|
||
marks, lone surrogates — with the claim unchanged, so only the envelope decides."""
|
||
env = model_envelope(ins, b"WAI2")
|
||
ml = int.from_bytes(env[4:8], "little")
|
||
text, rest = env[8:8 + ml].decode(), env[8 + ml:]
|
||
idx = 1 # the classical rendition
|
||
|
||
def mutate(f):
|
||
m = json.loads(text)
|
||
f(m)
|
||
return json.dumps(m, separators=(",", ":"))
|
||
|
||
def setr(k, val, i=idx):
|
||
def f(m):
|
||
m["renditions"][i][k] = val
|
||
return f
|
||
|
||
deep = lambda n: "[" * n + "]" * n
|
||
cases = {
|
||
"control": text,
|
||
"id_int": mutate(setr("id", 5)),
|
||
"sha256_int": mutate(setr("sha256", 5)),
|
||
"companion_of_array": mutate(setr("companion_of", [])),
|
||
"companion_of_id_object": mutate(setr("companion_of_id", {})),
|
||
"prior_string": mutate(setr("prior", "x")),
|
||
"prior_empty_object": mutate(setr("prior", {})),
|
||
"prior_no_determinism": mutate(setr("prior", {"id": "a", "sha256": "b"})),
|
||
"prior_id_int": mutate(setr("prior", {"id": 1, "sha256": "b", "determinism": "c"})),
|
||
"prior_tier_int": mutate(setr("prior", {"id": "a", "sha256": "b", "determinism": "c", "tier": 5})),
|
||
"prior_supersedes_string": mutate(setr("prior", {"id": "a", "sha256": "b", "determinism": "c", "supersedes": "x"})),
|
||
"other_entry_id_int": mutate(setr("id", 7, 0)),
|
||
"other_entry_prior_string": mutate(setr("prior", "x", 0)),
|
||
"role_int": mutate(setr("role", 5)),
|
||
"security_string": mutate(lambda m: m.__setitem__("security", "none")),
|
||
"security_int": mutate(lambda m: m.__setitem__("security", 5)),
|
||
"security_encrypted": mutate(lambda m: m.__setitem__("security", {"encryption": "aes-128-gcm", "kid": 1})),
|
||
"security_none": mutate(lambda m: m.__setitem__("security", {"encryption": "none"})),
|
||
"security_kid_negative": mutate(lambda m: m.__setitem__("security", {"encryption": "none", "kid": -1})),
|
||
"duplicate_ids": mutate(lambda m: [r.__setitem__("id", "a") for r in m["renditions"]]),
|
||
"missing_intent": mutate(lambda m: m.pop("intent")),
|
||
"wai_number": mutate(lambda m: m.__setitem__("wai", 1.1)),
|
||
"target_absent": mutate(lambda m: m.pop("target")),
|
||
"target_1e400": text.replace('"target":null', '"target":1e400'),
|
||
"target_neg_zero": text.replace('"target":null', '"target":-0'),
|
||
"target_2pow64": text.replace('"target":null', '"target":18446744073709551616'),
|
||
"target_depth_70": text.replace('"target":null', '"target":' + deep(70)),
|
||
"target_depth_126": text.replace('"target":null', '"target":' + deep(126)),
|
||
"target_depth_130": text.replace('"target":null', '"target":' + deep(130)),
|
||
"duplicate_member_media": text.replace('"media":"video"', '"media":"video","media":"video"'),
|
||
"duplicate_member_in_entry": text.replace('"kind":"second"', '"kind":"second","kind":"second"'),
|
||
"trailing_garbage": text + " x",
|
||
"trailing_space": text + " ",
|
||
"bom": "\ufeff" + text,
|
||
"lone_surrogate_target": text.replace('"target":null', '"target":"\\ud800"'),
|
||
"manifest_array": "[]",
|
||
"kind_null": mutate(setr("kind", None)),
|
||
"capability_int": mutate(setr("capability", 5)),
|
||
"entry_not_object": mutate(lambda m: m["renditions"].__setitem__(0, "x")),
|
||
"renditions_object": mutate(lambda m: m.__setitem__("renditions", {})),
|
||
"envelope_manifest_int": mutate(setr("envelope_manifest", 5)),
|
||
"refines_int": mutate(setr("refines", 5)),
|
||
"id_null": mutate(setr("id", None)),
|
||
"prior_null": mutate(setr("prior", None)),
|
||
"role_companion": mutate(setr("role", "companion")),
|
||
"role_unknown": mutate(setr("role", "rendition")),
|
||
"media_audio": mutate(lambda m: m.__setitem__("media", "audio")),
|
||
}
|
||
for name, t in cases.items():
|
||
e = wai2_with(t, rest)
|
||
ins.add("session_envelope_" + name, sourced(ins, 1, e), {"envelopes": [{"envelope": e.hex()}]})
|
||
# The table and payload framing around a sound manifest.
|
||
t = text.encode()
|
||
framing = {
|
||
"count_short": b"WAI2" + len(t).to_bytes(4, "little") + t + (1).to_bytes(2, "little") + rest[2:],
|
||
"truncated": wai2_with(t, rest)[:-1],
|
||
"trailing_bytes": wai2_with(t, rest) + b"\x00",
|
||
"manifest_length_past_end": b"WAI2" + (len(t) + len(rest) + 1).to_bytes(4, "little") + t + rest,
|
||
}
|
||
for name, e in framing.items():
|
||
ins.add("session_envelope_" + name, sourced(ins, 1, e), {"envelopes": [{"envelope": e.hex()}]})
|
||
|
||
|
||
def session_wai1_envelopes(ins):
|
||
"""§2.4.12 step 6, the envelopes: the model session against its base's WAI1 envelope
|
||
with manifests every reader may read otherwise, a capability or fallback the base's
|
||
entry does or does not name, and framing short or long; the session signed again
|
||
with the base's entry naming that envelope, so only the envelope decides."""
|
||
env = model_envelope(ins, b"WAI1")
|
||
multi = {"envelope": model_envelope(ins, b"WAI2").hex()}
|
||
ml = int.from_bytes(env[4:8], "little")
|
||
text, payload = env[8:8 + ml].decode(), env[12 + ml:]
|
||
|
||
def mutate(f):
|
||
m = json.loads(text)
|
||
f(m)
|
||
return json.dumps(m, separators=(",", ":"))
|
||
|
||
def mr(k, val):
|
||
return mutate(lambda m: m["model_requirement"].__setitem__(k, val))
|
||
|
||
deep = lambda n: "[" * n + "]" * n
|
||
cases = {
|
||
"control": text,
|
||
"capability_other": mr("capability", "wai.video.av1"),
|
||
"fallback_is_entry": mutate(lambda m: m["model_requirement"].update(capability="wai.video.av1", fallback="wai.video.int_hyper")),
|
||
"fallback_other": mr("fallback", "wai.video.hevc"),
|
||
"capability_int": mr("capability", 5),
|
||
"fallback_int": mr("fallback", 5),
|
||
"min_sim_float": mr("min_sim", 0.5),
|
||
"min_sim_string": mr("min_sim", "0.5"),
|
||
"min_sim_huge": text.replace('"capability":"wai.video.int_hyper"', '"capability":"wai.video.int_hyper","min_sim":1e400'),
|
||
"sim_metric_int": mr("sim_metric", 5),
|
||
"model_requirement_missing": mutate(lambda m: m.pop("model_requirement")),
|
||
"model_requirement_array": mutate(lambda m: m.__setitem__("model_requirement", [])),
|
||
"conditioning_missing": mutate(lambda m: m.pop("conditioning")),
|
||
"conditioning_kind_int": mutate(lambda m: m["conditioning"].__setitem__("kind", 5)),
|
||
"prior_string": mutate(lambda m: m["conditioning"].__setitem__("prior", "x")),
|
||
"security_none": mutate(lambda m: m.__setitem__("security", {"encryption": "none"})),
|
||
"security_null": mutate(lambda m: m.__setitem__("security", None)),
|
||
"security_encrypted": mutate(lambda m: m.__setitem__("security", {"encryption": "aes-128-gcm", "kid": 1})),
|
||
"security_string": mutate(lambda m: m.__setitem__("security", "none")),
|
||
"security_kid_negative": mutate(lambda m: m.__setitem__("security", {"encryption": "none", "kid": -1})),
|
||
"security_kid_2pow64": mutate(lambda m: m.__setitem__("security", {"kid": 2 ** 64})),
|
||
"missing_intent": mutate(lambda m: m.pop("intent")),
|
||
"media_int": mutate(lambda m: m.__setitem__("media", 5)),
|
||
"wai_number": mutate(lambda m: m.__setitem__("wai", 1.0)),
|
||
"unknown_member": mutate(lambda m: m.__setitem__("x_unknown", 1)),
|
||
"target_object": mutate(lambda m: m.__setitem__("target", {"a": 1})),
|
||
"target_depth_130": mutate(lambda m: m.__setitem__("target", None)).replace('"target":null', '"target":' + deep(130)),
|
||
"refines_int": mutate(lambda m: m.__setitem__("refines", 5)),
|
||
"derived_from_string": mutate(lambda m: m.__setitem__("derived_from", "x")),
|
||
"duplicate_member": text.replace('"media":"video"', '"media":"video","media":"video"'),
|
||
"duplicate_in_requirement": text.replace('"capability":"wai.video.int_hyper"', '"capability":"wai.video.int_hyper","capability":"wai.video.int_hyper"'),
|
||
"bom": "\ufeff" + text,
|
||
"trailing_garbage": text + " x",
|
||
"lone_surrogate": mutate(lambda m: m.__setitem__("target", None)).replace('"target":null', '"target":"\\ud800"'),
|
||
"manifest_array": "[]",
|
||
}
|
||
for name, t in cases.items():
|
||
tb = t.encode()
|
||
e = SV.wai1(tb, payload)
|
||
ins.add("session_wai1_" + name, sourced(ins, 0, e), {"envelopes": [{"envelope": e.hex()}, multi]})
|
||
tb = text.encode()
|
||
framing = {
|
||
"payload_short": SV.wai1(tb, payload)[:-1],
|
||
"trailing_bytes": SV.wai1(tb, payload) + b"\x00",
|
||
"manifest_length_past_end": b"WAI1" + (len(tb) + 100).to_bytes(4, "little") + tb,
|
||
"no_payload_length": b"WAI1" + len(tb).to_bytes(4, "little") + tb,
|
||
"magic_other": b"WAI3" + SV.wai1(tb, payload)[4:],
|
||
"short": b"WAI",
|
||
}
|
||
for name, e in framing.items():
|
||
ins.add("session_wai1_" + name, sourced(ins, 0, e), {"envelopes": [{"envelope": e.hex()}, multi]})
|
||
|
||
|
||
def session_classical_sweep(ins):
|
||
"""§2.4.12's classical rendition: every registered capability, an unregistered string
|
||
and the model's own, as the second rendition of the model session's WAI2 envelope —
|
||
once in a manifest of the session's media and once in a manifest of the capability's
|
||
own media — so that every rule that admits or excludes one is reached."""
|
||
env = model_envelope(ins, b"WAI2")
|
||
ml = int.from_bytes(env[4:8], "little")
|
||
m0, rest = json.loads(env[8:8 + ml]), env[8 + ml:]
|
||
for cap in sorted(CV.REGISTRY) + ["org.example.codec", "wai.video.int_hyper"]:
|
||
medias = ["video"] + ([CV.REGISTRY[cap]["media"]] if cap in CV.REGISTRY and CV.REGISTRY[cap]["media"] not in (None, "video") else [])
|
||
for media in medias:
|
||
m = copy.deepcopy(m0)
|
||
m["media"] = media
|
||
m["renditions"][1]["capability"] = cap
|
||
e = wai2_with(json.dumps(m, separators=(",", ":")), rest)
|
||
ins.add(f"session_classical_{cap}_{media}", sourced(ins, 1, e), {"envelopes": [{"envelope": e.hex()}]})
|
||
|
||
|
||
def session_caps(ins):
|
||
"""§2.4.12 rules 7 and 8 at their caps, where nothing else decides: 300 units presented
|
||
once and late, with the missed presentations and the switches written as many as their
|
||
totals, as the cap, and one either side of each."""
|
||
base = ins.receipt("session_accept")
|
||
ps = [{"unit": u, "layers": 1, "decoder": 0, "out_digest": CV.blake3(b"differential: capped %d" % u).hex()}
|
||
for u in range(300)]
|
||
root = CV.out_mth([CV.out_leaf(p) for p in ps]).hex()
|
||
decoder = dict(base["body"]["decoders"][0], leaves=300)
|
||
for missed, written in [(300, 255), (300, 256), (300, 257), (255, 255), (256, 256), (257, 257), (256, 255), (255, 256)]:
|
||
for total, switched in [(0, 0), (1100, 1023), (1100, 1024), (1100, 1025), (1023, 1023), (1024, 1024), (1025, 1024),
|
||
(1023, 1024)]:
|
||
sw = [{"unit": u // 4, "from": "primary" if u % 2 == 0 else "fallback",
|
||
"to": "fallback" if u % 2 == 0 else "primary", "reason": "recovered"} for u in range(switched)]
|
||
body = dict(base["body"], decoders=[decoder], units=300, leaf_count=300, output_root=root,
|
||
primary_units=300, fallback_units=0, switches=sw, switches_total=total,
|
||
deadline_us=40000, missed=missed, stage_sets=[],
|
||
missed_units=[{"unit": u, "layers": 1, "late_us": 1 + u} for u in range(written)])
|
||
ins.add(f"session_caps_{missed}_{written}_{total}_{switched}", CV.jcs(sign(dict(base, body=body), VIEWER)),
|
||
{"outputs": copy.deepcopy(ps)})
|
||
|
||
|
||
def figures(ins, rng, n):
|
||
oer_t = ins.by_name["energy_matches_oer_accept"]["context"]["oer"]["receipt"]
|
||
claim_t = ins.receipt("energy_matches_oer_accept")
|
||
keys = {OER_SIGNER: pk(METER)}
|
||
cases = [(f, uj) for f in FIGURES + NUMERIC_FIGURES + RANGE_FIGURES for uj in (1, 12000, 12001, 11999, 2, 3)]
|
||
for _ in range(n):
|
||
uj = rng.randrange(1, 10 ** rng.randrange(1, 16)) # a claim's integers stop at 2^53 - 1
|
||
frac = "".join(rng.choice("0123456789") for _ in range(rng.randrange(0, 50)))
|
||
pick = rng.randrange(4)
|
||
if pick == 0: # exactly half-way, with trailing zeros
|
||
f = f"{uj // 10**6}.{uj % 10**6:06d}5" + "0" * rng.randrange(0, 40)
|
||
elif pick == 1: # just either side of half-way
|
||
f = f"{uj // 10**6}.{uj % 10**6:06d}{rng.choice(['49', '50', '51'])}{frac}"
|
||
elif pick == 2: # exponent form of the same digits
|
||
s = f"{uj}5{frac}"
|
||
f = f"{s}e-{len(s) - len(str(uj)) + 6}"
|
||
else: # a double near the figure
|
||
f = (uj + rng.choice([0.5, 0.4999999, 0.5000001, 0.25])) / 1e6
|
||
cases.append((f, rng.choice([uj, uj + 1, uj - 1 if uj > 1 else uj])))
|
||
for f, uj in cases:
|
||
r = seal_oer(oer_t, f)
|
||
body = dict(claim_t["body"], joules_micro=uj, oer_receipt=r["receipt_id"])
|
||
c = sign(dict(claim_t, body=body), METER)
|
||
ins.add(f"figure_{str(f)[:40]}_uj{uj}", CV.jcs(c), {"oer": {"receipt": r, "keys": keys}})
|
||
|
||
|
||
def sums(ins, rng, n):
|
||
"""OER receipts whose breakdown or floor exercises §6's exact arithmetic: large terms
|
||
that cancel, differences at and one unit past the 1e-9 tolerance, floors exactly on
|
||
the 1% bound, and quantities at §4's limits. The claim names each receipt with the
|
||
figure's microjoules, so the verdict turns on whether each OER verifier accepts the
|
||
receipt."""
|
||
oer_t = ins.by_name["energy_matches_oer_accept"]["context"]["oer"]["receipt"]
|
||
claim_t = ins.receipt("energy_matches_oer_accept")
|
||
keys = {OER_SIGNER: pk(METER)}
|
||
e37 = "1" + "0" * 37
|
||
cases = [
|
||
("cancel_1e24", "0.012", ["1000000000000000000000000.012", "-1000000000000000000000000", "0", "0", "0"], None),
|
||
("cancel_1e30", "0.012", ["1000000000000000000000000000000.012", "-1000000000000000000000000000000", "0", "0", "0"], None),
|
||
("cancel_1e37", "0.012", ["10000000000000000000000000000000000000.012", "-" + e37, "0", "0", "0"], None),
|
||
("wide_1e37_1e-38", e37, [e37, "0." + "0" * 37 + "1", "0", "0", "0"], None),
|
||
("at_tolerance", "9999999990000000000000000000000000000", [e37, "0", "0", "0", "0"], None),
|
||
("past_tolerance", "9999999989999999999999999999999999999", [e37, "0", "0", "0", "0"], None),
|
||
("at_tolerance_unit", "1.000000001", ["2", "-1", "0", "0", "0"], None),
|
||
("past_tolerance_unit", "1.000000001000000000000000000000000001", ["2", "-1", "0", "0", "0"], None),
|
||
("negative_total", "-0.012", ["-0.012", "0", "0", "0", "0"], None),
|
||
("floor_tiny_temperature", "0.012", None, (256, "1e-38", "0")),
|
||
("floor_huge_temperature", "0.012", None, (256, "9e37", "1.6e+17")),
|
||
("floor_zero_bits", "0.012", None, (0, "300", "0")),
|
||
("floor_zero_bits_tiny_floor_j", "0.012", None, (0, "300", "1e-38")),
|
||
("floor_bits_beyond_64", "0.012", None, (2 ** 64, "300", "1e-18")),
|
||
("floor_bits_float", "0.012", None, (256.0, "300", "1.838e-18")),
|
||
]
|
||
# floor_j exactly on each 1% bound, and one unit of its last digit either side: §6's
|
||
# check is exact, so a difference exactly at the tolerance passes and one past it does
|
||
# not, in both verifiers or neither.
|
||
with localcontext(OER.EXACT):
|
||
for bits in (1, 256, 510, 4096, 2 ** 40):
|
||
for temp in ("300", "3.000000e+02", "0.001", "77.5"):
|
||
exact = Decimal(bits) * Decimal("1.380649e-23") * Decimal(temp) * Decimal("0.6931471805599453")
|
||
for side in ("0.99", "1.01"):
|
||
on = exact * Decimal(side)
|
||
unit = Decimal((0, (1,), on.as_tuple().exponent))
|
||
for name, fj in (("on", on), ("below", on - unit), ("above", on + unit)):
|
||
cases.append((f"floor_{name}_bound_{bits}_{temp}_{side}", "0.012", None, (bits, temp, format(fj, "e"))))
|
||
def term():
|
||
whole = "".join(rng.choice("0123456789") for _ in range(rng.randrange(1, 20))).lstrip("0") or "0"
|
||
frac = "".join(rng.choice("0123456789") for _ in range(rng.randrange(1, 25)))
|
||
return ("-" if rng.random() < 0.3 else "") + whole + "." + frac
|
||
|
||
for _ in range(n):
|
||
# Five random terms, some negative, up to 43 digits (so some exceed §4's 34 and make
|
||
# the receipt malformed), and a total exactly at, inside or outside the tolerance of
|
||
# their sum.
|
||
terms = [term() for _ in range(5)]
|
||
with localcontext(OER.EXACT):
|
||
total = sum(Decimal(t) for t in terms)
|
||
tol = Decimal("1e-9") * max(Decimal(1), abs(total))
|
||
jt = total + rng.choice([Decimal(0), tol, -tol, tol * 2, Decimal("1e-30")])
|
||
cases.append((f"random_sum_{len(cases)}", format(jt, "f"), terms, None))
|
||
for name, jt, terms, floor in cases:
|
||
r = seal_oer(oer_t, jt, terms, floor)
|
||
uj = Decimal(jt) * 1000000 if OER.DECIMAL.fullmatch(jt) else 1
|
||
uj = max(1, min(int(abs(uj).to_integral_value()), 2 ** 53 - 1))
|
||
body = dict(claim_t["body"], joules_micro=uj, oer_receipt=r["receipt_id"])
|
||
ins.add("sum_" + name, CV.jcs(sign(dict(claim_t, body=body), METER)), {"oer": {"receipt": r, "keys": keys}})
|
||
|
||
|
||
def readings(ins):
|
||
"""oer-receipts §4 "Numbers": numbers a receipt body must not carry — `-0` however
|
||
spelt, fractions, exponents, integers beyond I-JSON's range — beside the range's own
|
||
bounds, in figures and in members no check reads. A body holding a forbidden number is
|
||
`malformed` before its content address is computed, so the verdict is the same however
|
||
a reader would have rounded it. And a delivered object's microjoules spelt as numbers
|
||
the corpus does not give them as."""
|
||
oer_t = ins.by_name["energy_matches_oer_accept"]["context"]["oer"]["receipt"]
|
||
claim_t = ins.receipt("energy_matches_oer_accept")
|
||
keys = {OER_SIGNER: pk(METER)}
|
||
def k_floor(bits): # temperature_k "300", computed as §6 fixes it
|
||
with localcontext(OER.EXACT):
|
||
return format(Decimal(bits) * OER.KBOLTZ * Decimal(300) * OER.LN2, "e")
|
||
|
||
def at(path, spelling):
|
||
def edit(r):
|
||
*outer, last = path
|
||
for p in outer:
|
||
r = r[p]
|
||
r[last] = raw(spelling)
|
||
return edit
|
||
|
||
def both(*edits):
|
||
return lambda r: [e(r) for e in edits]
|
||
|
||
zero_terms = ["0.012000", "0", "0", "0", "0"]
|
||
cases = [ # (name, joules_total, breakdown, floor, edit, microjoules)
|
||
("term_neg_zero", "0.012000", zero_terms, None, at(("energy", "breakdown", "movement_j"), "-0"), 12000),
|
||
("term_neg_zero_float", "0.012000", zero_terms, None, at(("energy", "breakdown", "movement_j"), "-0.0"), 12000),
|
||
("term_neg_zero_exp", "0.012000", zero_terms, None, at(("energy", "breakdown", "movement_j"), "-0e7"), 12000),
|
||
("total_neg_zero", "0", ["0", "0", "0", "0", "0"], None,
|
||
both(at(("energy", "joules_total"), "-0"), at(("energy", "breakdown", "compute_j"), "-0")), 1),
|
||
("latency_neg_zero", "0.012000", None, None, at(("work", "latency_s"), "-0"), 12000),
|
||
("status_neg_zero", "0.012000", None, None, at(("work", "status"), "-0"), 12000),
|
||
("status_i64_min", "0.012000", None, None, at(("work", "status"), "-9223372036854775808"), 12000),
|
||
("status_below_i64", "0.012000", None, None, at(("work", "status"), "-9223372036854775809"), 12000),
|
||
("status_u64_max", "0.012000", None, None, at(("work", "status"), "18446744073709551615"), 12000),
|
||
("status_i_json_max", "0.012000", None, None, at(("work", "status"), "9007199254740991"), 12000),
|
||
("status_i_json_min", "0.012000", None, None, at(("work", "status"), "-9007199254740991"), 12000),
|
||
("status_beyond_i_json", "0.012000", None, None, at(("work", "status"), "9007199254740992"), 12000),
|
||
("status_below_i_json", "0.012000", None, None, at(("work", "status"), "-9007199254740992"), 12000),
|
||
("status_one_point_zero", "0.012000", None, None, at(("work", "status"), "1.0"), 12000),
|
||
("status_exponent", "0.012000", None, None, at(("work", "status"), "1e0"), 12000),
|
||
("status_beyond_u64", "0.012000", None, None, at(("work", "status"), "18446744073709551616"), 12000),
|
||
("status_1e30_integer", "0.012000", None, None, at(("work", "status"), "1" + "0" * 30), 12000),
|
||
("status_308_digits", "0.012000", None, None, at(("work", "status"), "1" + "0" * 308), 12000),
|
||
("status_exponent_zeros", "0.012000", None, None, at(("work", "status"), "1e" + "0" * 400 + "5"), 12000),
|
||
("status_underflow", "0.012000", None, None, at(("work", "status"), "-1e-400"), 12000),
|
||
("bits_beyond_u64", "0.012000", None, (2 ** 64, "300", k_floor(float(2 ** 64))),
|
||
at(("floor", "bits_committed"), "18446744073709551616"), 12000),
|
||
("bits_1e30_integer", "0.012000", None, (10 ** 30, "300", k_floor(1e30)),
|
||
at(("floor", "bits_committed"), "1" + "0" * 30), 12000),
|
||
("bits_neg_zero", "0.012000", None, (0, "300", "0"), at(("floor", "bits_committed"), "-0"), 12000),
|
||
("total_beyond_u64_micro", "18446744073709.551616", ["18446744073709.551616", "0", "0", "0", "0"], None,
|
||
at(("energy", "joules_total"), "18446744073709551616e-6"), 2 ** 53 - 1),
|
||
("total_number_spelt_long", "0.012000", zero_terms, None,
|
||
at(("energy", "joules_total"), "0.01200000000000000000000000000000000000000000000001"), 12000),
|
||
]
|
||
for name, jt, terms, floor, edit, uj in cases:
|
||
r = seal_oer(oer_t, jt, terms, floor, edit)
|
||
body = dict(claim_t["body"], joules_micro=uj, oer_receipt=r["receipt_id"])
|
||
c = CV.jcs(sign(dict(claim_t, body=body), METER))
|
||
ins.add(f"reading_{name}", c, {"oer": {"receipt": r, "keys": keys}})
|
||
delivery = ins.by_name["energy_matches_delivery_accept"]
|
||
for spelling in ["12000", "-0", "12000.0", "1.2e4", "18446744073709551616", "-12000"]:
|
||
ctx = dict(delivery["context"], object=dict(delivery["context"]["object"], joules_micro=raw(spelling)))
|
||
ins.add("delivered_joules_" + spelling[:24], delivery["receipt"], ctx)
|
||
|
||
|
||
# (name, spelling) of a value placed in an OER receipt's `work.outcome`, each in a file of
|
||
# its own: whether the file is read at all is the first thing compared.
|
||
DOCUMENTS = [
|
||
("control", '"encoded"'), ("max_double", "1.7976931348623157e308"), ("below_max_half_ulp", "1.7976931348623158e308"),
|
||
("past_max_half_ulp", "1.7976931348623159e308"), ("e400", "1e400"), ("neg_e400", "-1e400"), ("e308_8", "1.8e308"),
|
||
("digits_400", "1" + "0" * 400), ("neg_digits_400", "-1" + "0" * 400), ("zero_e_huge", "0e99999999999999999999"),
|
||
("one_e_huge", "1e99999999999999999999"), ("one_e_neg_huge", "1e-99999999999999999999"),
|
||
("nan", "NaN"), ("infinity", "Infinity"), ("neg_infinity", "-Infinity"),
|
||
("lone_high_surrogate", '"\\ud800"'), ("lone_low_surrogate", '"\\udc00"'), ("paired_surrogates", '"\\ud83d\\ude00"'),
|
||
("high_surrogate_then_letter", '"\\ud800\\u0041"'),
|
||
# A repeated member name, in an object no check reads, and a repeated figure.
|
||
("repeated_member", '{"a":1,"a":2}'), ("repeated_member_same_value", '{"a":1,"a":1}'),
|
||
("distinct_members", '{"a":1,"b":2}'),
|
||
]
|
||
# work.outcome sits at level 7 of a corpus file ({vectors:[{context:{oer:{receipt:{work:
|
||
# ...}}}}]}), so arrays nested d deep inside it reach level 7 + d; §4 reads 64 levels.
|
||
DOCUMENTS += [(f"nesting_to_level_{7 + d}", "[" * d + "]" * d) for d in (55, 56, 57, 58, 121, 200, 1100, 5000)]
|
||
|
||
|
||
def documents(corpus):
|
||
"""One corpus-format file per DOCUMENTS entry: an energy claim naming an OER receipt that
|
||
holds the value, sealed over §4's reading where the text is one §4 reads."""
|
||
ins = Inputs(corpus)
|
||
oer_t = ins.by_name["energy_matches_oer_accept"]["context"]["oer"]["receipt"]
|
||
claim_t = ins.receipt("energy_matches_oer_accept")
|
||
keys = {OER_SIGNER: pk(METER)}
|
||
out = []
|
||
for name, spelling in DOCUMENTS:
|
||
def edit(r, spelling=spelling):
|
||
r["work"]["outcome"] = raw(spelling)
|
||
try:
|
||
r = seal_oer(oer_t, "0.012000", None, None, edit)
|
||
except (ValueError, RecursionError):
|
||
# §4 does not read the receipt's own text. Sealed with a placeholder, which only
|
||
# matters if a verifier reads the file anyway — and then the verdicts differ.
|
||
r = seal_oer(oer_t, "0.012000")
|
||
edit(r)
|
||
body = dict(claim_t["body"], joules_micro=12000, oer_receipt=r["receipt_id"])
|
||
one = Inputs(corpus)
|
||
one.add("document_" + name, CV.jcs(sign(dict(claim_t, body=body), METER)), {"oer": {"receipt": r, "keys": keys}})
|
||
out.append((name, text_of({"about": "one differential input", "vectors": one.out})))
|
||
return out
|
||
|
||
|
||
def reports(ins, rng, n):
|
||
"""measurement and operating-point claims (§2.4.8-§2.4.9) around the edges of their
|
||
rules: brackets at and one ppm below the least declaration, figures one off, ties at a
|
||
half microjoule, the ratio thresholds, flat brackets and a held value, products past 128 bits,
|
||
every pattern of liveness facts; operating points in every presence combination; and
|
||
reports and energy claims against figure views and labelled objects."""
|
||
m0 = ins.receipt("measurement_accept")
|
||
fig0 = ins.by_name["measurement_accept"]["context"]["figure"]
|
||
op0 = ins.receipt("operating_point_accept")
|
||
hw = ["OnChipCounter", "CalibratedInstrument"]
|
||
|
||
def summary(n_, mean, sd, live=None, run=None):
|
||
live = n_ if live is None else live
|
||
return {"n": n_, "mean_uw": mean, "sd_uw": sd, "live_intervals": live, "longest_live_run": n_ if run is None else run}
|
||
|
||
def class_fields(label, ppm, window, figure):
|
||
b = {"energy_provenance": label, "energy_uncertainty": None, "energy_counter_filtering": None,
|
||
"energy_calibration_ref": None}
|
||
if label == "OnChipCounter":
|
||
b.update(energy_uncertainty={"relative_ppm": ppm, "window_us": window}, energy_counter_filtering="Unknown")
|
||
elif label == "CalibratedInstrument":
|
||
b.update(energy_uncertainty={"relative_ppm": ppm, "window_us": window},
|
||
energy_calibration_ref="https://calibration.example.org/certificates/0001")
|
||
return dict(b, joules_micro=figure)
|
||
|
||
def report(br, label, ppm, window, figure, name, ctx=None):
|
||
body = dict(m0["body"], bracket=br, **class_fields(label, ppm, window, figure))
|
||
ins.add(name, CV.jcs(sign(dict(m0, body=body), METER)), ctx)
|
||
|
||
def no_reader(br, label, ppm, window, figure, name, ctx=None):
|
||
body = dict(m0["body"], bracket=br, power_reader=None, **class_fields(label, ppm, window, figure))
|
||
ins.add(name, CV.jcs(sign(dict(m0, body=body), METER)), ctx)
|
||
|
||
def around(name, br, labels=("OnChipCounter", "CalibratedInstrument", "ModelBased")):
|
||
f = CV.bracket_figure(br)
|
||
for label in labels:
|
||
if f is None:
|
||
report(br, label, 300_000, br["window_us"], 1, f"{name}_nofig_{label}")
|
||
continue
|
||
p = CV.least_ppm(br, f) or 2 ** 32 - 1
|
||
for d_ppm, d_fig, d_win in [(0, 0, 0), (-1, 0, 0), (1, 0, 0), (0, 1, 0), (0, -1, 0), (0, 0, 1)]:
|
||
ppm = max(1, min(2 ** 32 - 1, p + d_ppm))
|
||
fig = f + d_fig
|
||
if fig <= 0 or fig > CV.MAX_INT:
|
||
continue
|
||
report(br, label, ppm, br["window_us"] + d_win, fig, f"{name}_{label}_{d_ppm}_{d_fig}_{d_win}")
|
||
|
||
def bracket(idle, busy, t, c, nn, opening):
|
||
return {"idle": idle, "busy": busy, "window_us": t, "completions": c, "in_flight": nn, "opening": opening}
|
||
|
||
# Fixed edges.
|
||
around("half_tie", bracket(summary(10, 20_000_000, 1_000), summary(14, 21_000_000, 1_000), 2_500_000, 1_000_000, 1, "running"))
|
||
around("flat_idle", bracket(summary(10, 20_000_000, 0), summary(14, 30_000_000, 115_471), 3_000_000, 60, 10, "from-rest"))
|
||
around("flat_busy", bracket(summary(10, 20_000_000, 115_471), summary(14, 30_000_000, 0), 3_000_000, 60, 10, "from-rest"))
|
||
around("held_value", bracket(summary(10, 5_000_000, 0, 1, 1), summary(14, 10_500_000, 2_980_708, 2, 1), 3_000_000, 60, 10, "from-rest"))
|
||
# A report that names no power reader, with and without a bracket, under each class.
|
||
for label in ("OnChipCounter", "CalibratedInstrument", "ModelBased", "Estimator"):
|
||
unit_br = m0["body"]["bracket"]
|
||
f = CV.bracket_figure(unit_br)
|
||
p = CV.least_ppm(unit_br, f)
|
||
for br in (unit_br, None):
|
||
no_reader(br, label, p, unit_br["window_us"], f, f"no_reader_{label}_{br is None}", fig0)
|
||
around("wide", bracket(summary(10, 20_000_000, 115_471), summary(14, 30_000_000, 115_471), 10 ** 11, 10 ** 13, 64, "from-rest"))
|
||
for k in (10, 3, 1):
|
||
for d in (0, -1):
|
||
around(f"ratio_{k}_{d}", bracket(summary(10, 20_000_000, 100_000), summary(14, 20_000_000 + k * 100_000 + d, 100_000),
|
||
3_000_000, 60, 10, "from-rest"), labels=("OnChipCounter", "ModelBased"))
|
||
# Liveness facts: every pattern around the thresholds, in each bracket.
|
||
for n_ in (4, 5, 6, 10, 14):
|
||
for live in sorted({0, 1, 2, 4, 5, n_ * 4 // 5, n_ * 4 // 5 + 1, n_ - 1, n_, n_ + 1}):
|
||
for run in sorted({0, 1, 2, 3, live, live + 1}):
|
||
idle = summary(10, 20_000_000, 115_471)
|
||
busy = summary(n_, 30_000_000, 115_471, live, run)
|
||
br = bracket(idle, busy, 3_000_000, 60, 10, "from-rest")
|
||
f = CV.bracket_figure(br)
|
||
p = CV.least_ppm(br, f)
|
||
report(br, "OnChipCounter", p, 3_000_000, f, f"live_busy_{n_}_{live}_{run}")
|
||
br2 = bracket(summary(n_, 20_000_000, 115_471, live, run), summary(14, 30_000_000, 115_471), 3_000_000, 60, 10, "from-rest")
|
||
report(br2, "OnChipCounter", CV.least_ppm(br2, f), 3_000_000, f, f"live_idle_{n_}_{live}_{run}")
|
||
# Random brackets.
|
||
for i in range(n):
|
||
ni, nb = rng.choice([1, 2, 3, 5, 10, 14, 50]), rng.choice([1, 2, 3, 5, 10, 14, 50])
|
||
idle_mean = rng.randrange(0, 50_000_000)
|
||
busy_mean = idle_mean + rng.choice([-1, 0, 1, rng.randrange(1, 40_000_000)])
|
||
sd = lambda: rng.choice([0, 1, rng.randrange(1, 5_000_000)]) # noqa: E731
|
||
live = lambda m: rng.choice([m, m - 1, 0, 2, 5]) # noqa: E731
|
||
idle = summary(ni, idle_mean, sd(), live(ni))
|
||
busy = summary(nb, max(0, busy_mean), sd(), live(nb))
|
||
t = rng.choice([0, 1, rng.randrange(1, 10 ** 7), rng.randrange(1, 10 ** 12)])
|
||
c = rng.choice([0, 1, rng.randrange(1, 10 ** 6), rng.randrange(1, 10 ** 13)])
|
||
nn = rng.choice([0, 1, rng.randrange(1, 65)])
|
||
br = bracket(idle, busy, t, c, nn, rng.choice(["from-rest", "running"]))
|
||
f = CV.bracket_figure(br)
|
||
label = rng.choice(hw + ["ModelBased", "Estimator"])
|
||
if f is None:
|
||
report(br, label, 300_000, max(1, t), rng.randrange(1, 10 ** 6), f"random_{i}_nofig")
|
||
continue
|
||
p = CV.least_ppm(br, f)
|
||
ppm = max(1, min(2 ** 32 - 1, (p or 2 ** 32 - 1) + rng.choice([0, 0, -1, 1, 1000])))
|
||
if f * ppm < 288_676:
|
||
ppm = 288_676 // f + 1
|
||
report(br, label, ppm, rng.choice([max(1, t), t + 1]), f + rng.choice([0, 0, 0, 1, -1]) or 1, f"random_{i}")
|
||
|
||
# Operating points: every presence combination, and points that are empty, not terms,
|
||
# or not integers.
|
||
pts = [None, {}, {"height": 1080}, {"height": 720}, {"frame rate": 1}, {"height": "1080"}, {"height": -1}]
|
||
pred = op0["body"]["prediction"]
|
||
for ri, r in enumerate(pts):
|
||
for ci, c in enumerate(pts):
|
||
for di, d in enumerate(pts):
|
||
for pi, pr in enumerate([None, pred]):
|
||
body = dict(op0["body"], op_requested=r, op_committed=c, op_delivered=d, prediction=pr)
|
||
ins.add(f"op_{ri}{ci}{di}{pi}", CV.jcs(sign(dict(op0, body=body), METER)) if all(
|
||
not isinstance(x, dict) or all(isinstance(v, int) and v >= 0 for v in x.values()) for x in (r, c, d))
|
||
else text_of(sign_raw(dict(op0, body=body), METER)), {"figure": fig0})
|
||
|
||
# Figure views that agree and disagree in each member.
|
||
e = json.loads(fig0["claim"])
|
||
eb = e["body"]
|
||
base = {"profile": "moq-jwp:claim-link", "identity": CV.link_hash(e), "group": None,
|
||
"figures": [{"index": None, "joules_micro": eb["joules_micro"], "energy_provenance": eb["energy_provenance"],
|
||
"energy_uncertainty": eb["energy_uncertainty"],
|
||
"energy_counter_filtering": eb["energy_counter_filtering"]}],
|
||
"signer": pk(METER)}
|
||
variants = [("same", {}), ("profile", {"profile": "wai:asset-receipt-id"}), ("identity", {"identity": "ab" * 32}),
|
||
("group", {"group": "cd" * 32}), ("signer", {"signer": pk(OPERATOR)}), ("no_signer", {"signer": None})]
|
||
fig_variants = [("figure_plus", {"joules_micro": eb["joules_micro"] + 1}), ("unlabelled", {"energy_provenance": None,
|
||
"energy_uncertainty": None, "energy_counter_filtering": None}),
|
||
("filtering_absent", {"energy_counter_filtering": None}), ("filtering_off", {"energy_counter_filtering": "Off"}),
|
||
("ppm", {"energy_uncertainty": dict(eb["energy_uncertainty"], relative_ppm=eb["energy_uncertainty"]["relative_ppm"] + 1)}),
|
||
("index", {"index": 0})]
|
||
for name, edit in variants:
|
||
ins.add("view_" + name, CV.jcs(m0), {"figure": {"view": dict(base, **edit)}})
|
||
ins.add("view_op_" + name, CV.jcs(op0), {"figure": {"view": dict(base, **edit)}})
|
||
for name, edit in fig_variants:
|
||
fv = [{k: v for k, v in dict(base["figures"][0], **edit).items() if v is not None or k == "index"}]
|
||
ins.add("view_fig_" + name, CV.jcs(m0), {"figure": {"view": dict(base, figures=fv)}})
|
||
ins.add("view_none_given_claim_not_json", CV.jcs(m0), {"figure": {"claim": "not json"}})
|
||
|
||
# Energy claims against labelled objects.
|
||
ec = ins.receipt("energy_on_chip_counter_accept")
|
||
b = ec["body"]
|
||
lab = {"energy_provenance": "OnChipCounter", "energy_uncertainty": b["energy_uncertainty"], "energy_counter_filtering": "Unknown"}
|
||
for name, j, extra in [("same", b["joules_micro"], lab), ("no_label", b["joules_micro"], {}),
|
||
("filtering_absent", b["joules_micro"], {k: v for k, v in lab.items() if k != "energy_counter_filtering"}),
|
||
("model", b["joules_micro"], {"energy_provenance": "ModelBased"}),
|
||
("ppm", b["joules_micro"], dict(lab, energy_uncertainty={"relative_ppm": 20_001, "window_us": 1_000_000})),
|
||
("figure", b["joules_micro"] + 1, lab)]:
|
||
ins.add("labelled_object_" + name, CV.jcs(ec), {"object": dict({"content_hash": ec["subject"], "joules_micro": j}, **extra)})
|
||
|
||
|
||
def sign_raw(c, k):
|
||
"""`c` signed over its JCS where JCS can serialise it, with a member JCS refuses (a
|
||
string or negative point value) left for the verifier to refuse at step 2. The
|
||
signature covers a canonical form computed with that member as written."""
|
||
c = copy.deepcopy(c)
|
||
c.pop("sig", None)
|
||
c["signer_pubkey"] = pk(k)
|
||
def canon(v):
|
||
if isinstance(v, int) and not isinstance(v, bool) and v < 0:
|
||
return str(v)
|
||
if isinstance(v, dict):
|
||
keys = sorted(v, key=lambda x: x.encode("utf-16-be"))
|
||
return "{" + ",".join(json.dumps(x, ensure_ascii=False) + ":" + canon(v[x]) for x in keys) + "}"
|
||
if isinstance(v, list):
|
||
return "[" + ",".join(canon(x) for x in v) + "]"
|
||
return CV.jcs(v)
|
||
c["sig"] = k.sign(b"moq-jwp:claim\x01" + canon(c).encode()).hex()
|
||
return c
|
||
|
||
|
||
def contexts(ins):
|
||
decl = ins.receipt("soft_binding_accept")
|
||
record = ins.receipt("binding_recovery_accept")
|
||
lawful = ins.receipt("lawful_basis_declaration_accept")
|
||
decision = ins.receipt("lawful_basis_decision_accept")
|
||
e0 = ins.receipt("energy_accept")
|
||
e1 = ins.receipt("energy_chain_accept")
|
||
ppk = pk(PUBLISHER)
|
||
|
||
def recovery_case(name, declaration_wire, rec, algs=(WM_ALG,), publisher=ppk):
|
||
ctx = {"declaration": declaration_wire, "publisher": publisher, "exact_match_algs": list(algs)}
|
||
ins.add(name, CV.jcs(rec), ctx)
|
||
|
||
def naming(rec_template, declaration, k, **body):
|
||
b = dict(rec_template["body"], declaration=CV.link_hash(declaration), **body)
|
||
return sign(dict(rec_template, body=b), k)
|
||
|
||
# A resolver-signed declaration with its signer_pubkey rewritten to the publisher's
|
||
# key, and a record naming that forgery: a declaration parsed but not verified.
|
||
forged = dict(sign(decl, RESOLVER), signer_pubkey=ppk)
|
||
recovery_case("recovery_forged_declaration", CV.jcs(forged), naming(record, forged, RESOLVER))
|
||
forged_pub = dict(decl, seq=decl["seq"] + 1) # the publisher's declaration, altered after signing
|
||
recovery_case("recovery_altered_declaration", CV.jcs(forged_pub), naming(record, forged_pub, RESOLVER))
|
||
recovery_case("recovery_declaration_not_json", "not json", record)
|
||
recovery_case("recovery_declaration_not_a_claim", CV.jcs(dict(decl, kind="group")), record)
|
||
recovery_case("recovery_declaration_lawful", CV.jcs(lawful), naming(record, lawful, RESOLVER))
|
||
recovery_case("recovery_declaration_energy_forged", CV.jcs(dict(e0, seq=5)), naming(record, dict(e0, seq=5), RESOLVER))
|
||
recovery_case("recovery_record_alg_differs", CV.jcs(decl), naming(record, decl, RESOLVER, alg="org.example.watermark.2"))
|
||
recovery_case("recovery_record_type_differs", CV.jcs(decl), naming(record, decl, RESOLVER, binding_type="fingerprint"))
|
||
recovery_case("recovery_record_no_matcher", CV.jcs(decl), record, algs=())
|
||
recovery_case("recovery_record_ok", CV.jcs(decl), record)
|
||
recovery_case("recovery_record_rogue_publisher", CV.jcs(decl), record, publisher=pk(RESOLVER))
|
||
# A decision naming a declaration altered after the operator signed it (a purpose added).
|
||
altered = copy.deepcopy(lawful)
|
||
altered["body"]["purposes"].append({"purpose": "dpv:PersonalisedAdvertising", "basis": "dpv:Consent", "data": []})
|
||
dec_b = dict(decision["body"], declaration=CV.link_hash(altered), purpose="dpv:PersonalisedAdvertising")
|
||
ins.add("decision_altered_declaration", CV.jcs(sign(dict(decision, body=dec_b), VIEWER)), {"declaration": CV.jcs(altered)})
|
||
ins.add("decision_declaration_not_json", CV.jcs(decision), {"declaration": "{"})
|
||
ins.add("decision_declaration_is_decision", CV.jcs(sign(dict(decision, body=dict(decision["body"], declaration=CV.link_hash(decision))), VIEWER)),
|
||
{"declaration": CV.jcs(decision)})
|
||
ins.add("decision_declaration_is_soft_binding", CV.jcs(sign(dict(decision, body=dict(decision["body"], declaration=CV.link_hash(decl))), VIEWER)),
|
||
{"declaration": CV.jcs(decl)})
|
||
st = sign(dict(decision, subject_type="receipt"), VIEWER)
|
||
ins.add("decision_subject_type_differs", CV.jcs(st), {"declaration": CV.jcs(lawful)})
|
||
ins.add("declaration_given_to_a_declaration", CV.jcs(lawful), {"declaration": CV.jcs(lawful)})
|
||
ins.add("declaration_given_to_energy", CV.jcs(e0), {"declaration": "not json"})
|
||
# Step 5 against a previous claim that does not itself verify.
|
||
bad_prev = dict(e0, seq=0, issued_at="2026-09-27T12:00:01Z") # altered after signing
|
||
e1_bad = sign(dict(e1, parent_receipt_hash=CV.link_hash(bad_prev)), METER)
|
||
ins.add("previous_altered_named", CV.jcs(e1_bad), {"previous": CV.jcs(bad_prev)})
|
||
ins.add("previous_altered_head", CV.jcs(sign(dict(e1, parent_receipt_hash=None), METER)), {"previous": CV.jcs(bad_prev)})
|
||
ins.add("previous_not_json", CV.jcs(e1), {"previous": "[]"})
|
||
ins.add("previous_ok", CV.jcs(e1), {"previous": CV.jcs(e0)})
|
||
# §3.2 with the same declarations.
|
||
for name, d in [("forged", forged), ("altered", forged_pub), ("ok", decl)]:
|
||
ins.add("recover_" + name, None, {"declaration": CV.jcs(d), "publisher": ppk, "observed_value": decl["body"]["value"],
|
||
"observed_hash": None, "exact_match_algs": [WM_ALG]}, procedure="recover")
|
||
|
||
|
||
# ------------------------------------------------------------------ run and compare
|
||
|
||
def run_raw(cmd):
|
||
env = dict(os.environ, PYTHONIOENCODING="utf-8", PYTHONDONTWRITEBYTECODE="1")
|
||
return subprocess.run(cmd, capture_output=True, text=True, encoding="utf-8", env=env)
|
||
|
||
|
||
def run(cmd):
|
||
p = run_raw(cmd)
|
||
if p.returncode != 0:
|
||
sys.exit(f"{' '.join(cmd)} exited {p.returncode}: {p.stderr.strip()[:400]}")
|
||
return [line.split("\t", 1) for line in p.stdout.splitlines()]
|
||
|
||
|
||
def document_outcome(p, python):
|
||
"""`refused`, the verdict lines, or `crash: ...`. The reference refuses a file with exit
|
||
status 1, no output and `not read:` on stderr; anything else that exits non-zero is a
|
||
crash. wai-rs exits 1 with its reader's error on stderr."""
|
||
if p.returncode == 0:
|
||
return "\n".join(p.stdout.splitlines())
|
||
if p.stdout or p.returncode != 1 or (python and "not read:" not in p.stderr):
|
||
return "crash: exit %d: %s" % (p.returncode, p.stderr.strip().splitlines()[-1:] or "")
|
||
return "refused"
|
||
|
||
|
||
def main(argv):
|
||
ap = argparse.ArgumentParser()
|
||
ap.add_argument("rust_bin")
|
||
ap.add_argument("--random", type=int, default=200)
|
||
ap.add_argument("--seed", type=int, default=1)
|
||
ap.add_argument("--keep")
|
||
a = ap.parse_args(argv[1:])
|
||
rng = random.Random(a.seed)
|
||
corpus = OER.load(open(os.path.join(HERE, "vectors.json"), encoding="utf-8"))
|
||
ins = Inputs(corpus)
|
||
numbers(ins)
|
||
strings(ins)
|
||
nesting(ins)
|
||
dates(ins, rng, a.random)
|
||
endpoints(ins, rng, a.random)
|
||
figures(ins, rng, a.random)
|
||
sums(ins, rng, a.random)
|
||
classes(ins, random.Random(f"classes {a.seed}"), a.random) # its own stream: other families' draws are unchanged
|
||
readings(ins)
|
||
contexts(ins)
|
||
bodies(ins)
|
||
reports(ins, random.Random(f"reports {a.seed}"), a.random) # its own stream, as classes()
|
||
stages(ins, random.Random(f"stages {a.seed}"), a.random) # its own stream, as classes()
|
||
sessions(ins, random.Random(f"sessions {a.seed}"), a.random) # its own stream, as classes()
|
||
session_caps(ins)
|
||
session_stage_bounds(ins, random.Random(f"session bounds {a.seed}"), a.random) # its own stream, as classes()
|
||
session_stage_epochs(ins, random.Random(f"session epochs {a.seed}"), a.random) # its own stream, as classes()
|
||
session_held(ins, random.Random(f"session held {a.seed}"), a.random) # its own stream, as classes()
|
||
session_envelopes(ins)
|
||
session_wai1_envelopes(ins)
|
||
session_classical_sweep(ins)
|
||
session_roles_sweep(ins)
|
||
session_encrypted(ins, random.Random(f"session encrypted {a.seed}"), a.random) # its own stream, as classes()
|
||
doc = text_of({"about": "differential inputs; no outcome is declared", "vectors": ins.out})
|
||
with tempfile.NamedTemporaryFile("w", suffix=".json", delete=False, encoding="utf-8") as f:
|
||
f.write(doc)
|
||
path = f.name
|
||
try:
|
||
rust = run([a.rust_bin, "verdicts", path])
|
||
py = run([sys.executable, os.path.join(HERE, "claim_verify.py"), "verdicts", path])
|
||
finally:
|
||
if a.keep:
|
||
os.replace(path, a.keep)
|
||
else:
|
||
os.unlink(path)
|
||
if len(rust) != len(ins.out) or len(py) != len(ins.out):
|
||
sys.exit(f"verdict counts differ: {len(ins.out)} inputs, wai-rs {len(rust)}, python {len(py)}")
|
||
differ = [(n, r, p) for (n, r), (_, p) in zip(rust, py) if r != p]
|
||
crashes = [(n, v) for n, v in rust + py if v.startswith("crash")]
|
||
tally = {}
|
||
for _, v in rust:
|
||
k = v.split(" link_hash=")[0].split(" ancestor=")[0]
|
||
tally[k] = tally.get(k, 0) + 1
|
||
oer = "reject step=6 reason=energy_oer_invalid"
|
||
for n, r, p in differ:
|
||
# Marked, never set apart: the verdicts differ either way.
|
||
mark = " (one side's OER verifier refused the receipt)" if oer in (r, p) else ""
|
||
print(f"DIFFER {n}{mark}\n wai-rs: {r}\n python: {p}")
|
||
for n, v in crashes:
|
||
print(f"CRASH {n}: {v}")
|
||
print(f"{len(ins.out)} inputs; verdicts: " + ", ".join(f"{k} x{c}" for k, c in sorted(tally.items())))
|
||
# Files of their own: read by both or refused by both, and the same verdicts if read.
|
||
doc_differ, doc_crash, doc_tally = 0, 0, {}
|
||
with tempfile.TemporaryDirectory() as d:
|
||
for name, text in documents(corpus):
|
||
path = os.path.join(d, name + ".json")
|
||
with open(path, "w", encoding="utf-8") as f:
|
||
f.write(text)
|
||
r = document_outcome(run_raw([a.rust_bin, "verdicts", path]), False)
|
||
p = document_outcome(run_raw([sys.executable, os.path.join(HERE, "claim_verify.py"), "verdicts", path]), True)
|
||
key = r.split("\t", 1)[-1].split(" link_hash=")[0]
|
||
doc_tally[key] = doc_tally.get(key, 0) + 1
|
||
if r.startswith("crash") or p.startswith("crash"):
|
||
doc_crash += 1
|
||
print(f"CRASH document {name}\n wai-rs: {r}\n python: {p}")
|
||
elif r != p:
|
||
doc_differ += 1
|
||
print(f"DIFFER document {name}\n wai-rs: {r}\n python: {p}")
|
||
print(f"{len(DOCUMENTS)} documents; wai-rs: " + ", ".join(f"{k} x{c}" for k, c in sorted(doc_tally.items())))
|
||
print(f"{len(differ)} differ, {len(crashes)} crash; documents: {doc_differ} differ, {doc_crash} crash")
|
||
return 1 if differ or crashes or doc_differ or doc_crash else 0
|
||
|
||
|
||
if __name__ == "__main__":
|
||
sys.exit(main(sys.argv))
|