mirror/rea
1
0
Fork 0
mirror of https://github.com/morluto/rea synced 2026-10-26 06:16:17 -04:00
Code Issues Packages Projects Releases Wiki Activity

fix(javascript): preserve analysis applicability after interruption (#4207)

Browse Source 
Preserve applicable artifact processing phases and entered/unvisited status after interruption. Centralize JavaScript semantic applicability for full and incremental builders; retain native byte identity, actual JavaScript unknowns and structured-file provenance.

Fixes #3948. Part of #4027.

Validation:78 distinct focused cases across10files, full static/docs, and real Joplin timeout/default public CLI/MCP controls passed. Source pins, every attempt/outlier and compact resource/token projections are retained. No target/native provider execution or speedup claim.

Co-authored-by: Kaoru0822 <208281322+kaoru0822-kitauji@users.noreply.github.com>
This commit is contained in:
Ysopwnable
2026-10-16 03:46:00 +08:00
committed by GitHub
parent 663d07f02c
commit 85eac7b92b
18 changed files with 3717 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/verify/javascript/native-handoffs.mjs
+153 -19
View File
@@ -17,6 +17,10 @@ import {
runOwnedCommand,
} from "../../../dist/process/OwnedCommand.js";
import { closeVerifierResources } from "../../lib/verifier-cleanup.mjs";
import {
matchesStructuredScopeExpectation,
structuredScopeExpectationsSchema,
} from "../../lib/javascript-file-scope-oracle.mjs";
const { values } = parseArgs({
options: {
@@ -24,6 +28,11 @@ const { values } = parseArgs({
"addon-path": { type: "string" },
"native-sha256": { type: "string" },
"source-path": { type: "string", multiple: true },
"structured-path": { type: "string", multiple: true },
"structured-expectations": { type: "string" },
"validate-structured-from-export": { type: "boolean", default: false },
"analysis-timeout-ms": { type: "string" },
"expect-timeout": { type: "boolean", default: false },
rounds: { type: "string", default: "3" },
"cli-parity": { type: "boolean", default: false },
"tokenizer-python": { type: "string" },
@@ -50,6 +59,30 @@ assert.match(
);
const rounds = Number(values.rounds);
assert.ok(Number.isSafeInteger(rounds) && rounds > 0);
const analysisTimeoutMs =
values["analysis-timeout-ms"] === undefined
? undefined
: Number(values["analysis-timeout-ms"]);
assert.ok(
analysisTimeoutMs === undefined ||
(Number.isSafeInteger(analysisTimeoutMs) && analysisTimeoutMs > 0),
);
assert.ok(!values["expect-timeout"] || analysisTimeoutMs !== undefined);
assert.ok(!values["validate-structured-from-export"] || values["cli-parity"]);
const structuredExpectations = structuredScopeExpectationsSchema.parse(
JSON.parse(values["structured-expectations"] ?? "{}"),
);
for (const path of Object.keys(structuredExpectations))
assert.ok(
values["structured-path"]?.includes(path),
`Expected structured path was not selected for verification: ${path}`,
);
if (values["expect-timeout"])
for (const path of values["structured-path"] ?? [])
assert.ok(
Object.hasOwn(structuredExpectations, path),
`--structured-expectations must name the expected operation/status for ${path}`,
);
const environment = {
PATH: process.env.PATH ?? "/usr/bin:/bin",
REA_LOG_LEVEL: "silent",
@@ -139,30 +172,55 @@ try {
),
);
assert.equal(selected.length, 2);
const call = async (name, arguments_) => {
const call = async (name, arguments_, allowTimeout = false) => {
const response = await client.callTool(
{ name, arguments: arguments_ },
{ timeout: 360_000 },
);
assert.notEqual(response.isError, true, serialize(response));
assert.ok(response.structuredContent?.evidence_id);
let projection = response.structuredContent;
if (response.isError === true) {
assert.ok(
allowTimeout,
serialize({ name, arguments: arguments_, response }),
);
const text = response.content.find(({ type }) => type === "text");
assert.ok(text?.type === "text");
projection = JSON.parse(text.text);
assert.equal(projection.error.code, "provider_timeout");
assert.equal(
projection.error.details.partial_observation.kind,
"retained-evidence",
);
} else assert.ok(projection?.evidence_id);
return {
request: { name, arguments: arguments_ },
response: response.structuredContent,
response: projection,
wire_response: response,
};
};
const prepStart = performance.now();
const preparation = await call("analyze_javascript_application", {
input_path: values.artifact,
detail: "summary",
...(values["source-path"] === undefined
? {}
: { source_paths: values["source-path"] }),
});
const preparation = await call(
"analyze_javascript_application",
{
input_path: values.artifact,
detail: "summary",
...(analysisTimeoutMs === undefined
? {}
: { analysis_timeout_ms: analysisTimeoutMs }),
...(values["source-path"] === undefined
? {}
: { source_paths: values["source-path"] }),
},
values["expect-timeout"],
);
const preparationMs = performance.now() - prepStart;
assert.equal(preparation.response.subject.digest.sha256, archiveSha256);
const sourceId = preparation.response.normalized_result.parent_evidence_id;
const timedOut = preparation.wire_response.isError === true;
assert.equal(timedOut, values["expect-timeout"]);
if (!timedOut)
assert.equal(preparation.response.subject.digest.sha256, archiveSha256);
const sourceId = timedOut
? preparation.response.error.details.partial_observation.evidence_id
: preparation.response.normalized_result.parent_evidence_id;
assert.match(sourceId, /^ev_[a-f0-9]{64}$/);
const query = {
application: { kind: "retained-evidence", evidence_id: sourceId },
@@ -221,6 +279,38 @@ try {
trace.native_handoffs[0].status === "evidence-linked" &&
serialize(trace.native_handoffs[0].evidence_ids) ===
serialize([native.evidence_id]));
const analysisScopes = (result, path) =>
result.graph.nodes.flatMap((node) =>
node.kind !== "unknown"
? []
: node.observations.flatMap((observation) =>
observation.properties.path === path &&
observation.properties.analysis_failure !== undefined
? [{ node_id: node.node_id, ...observation }]
: [],
),
);
const nativeScopes = analysisScopes(trace, values["addon-path"]);
const structuredQueries = [];
const structuredScopes = [];
for (const path of values["validate-structured-from-export"]
? []
: (values["structured-path"] ?? [])) {
const result = await call("trace_application_feature", {
...query,
native_observations: [],
seed: { ...query.seed, value: path },
});
structuredQueries.push(result);
structuredScopes.push({
source: "retained-feature-trace",
path,
scopes: analysisScopes(result.response.normalized_result, path),
evidence_id: result.response.evidence_id,
source_coverage: result.response.normalized_result.coverage,
graph_coverage: result.response.normalized_result.graph.coverage,
});
}
let cliParity = null;
if (values["cli-parity"]) {
temporary = await PrivateRuntimeRoot.create({
@@ -237,6 +327,25 @@ try {
({ evidence_id }) => evidence_id === sourceId,
);
assert.ok(parent, "Export omitted the retained parent");
assert.equal(parent.normalized_result.root_artifact_sha256, archiveSha256);
if (values["validate-structured-from-export"])
for (const path of values["structured-path"] ?? []) {
const retained = parent.normalized_result;
const matchingNodes = retained.graph.nodes.filter((node) =>
node.observations.some(({ properties }) => properties.path === path),
);
assert.ok(matchingNodes.length > 0, `Missing retained path: ${path}`);
// Diagnostic validation of the complete backend record is distinct
// from the inline native analyst task used by this control's tokens.
structuredScopes.push({
source: "exported-parent-validation",
path,
scopes: analysisScopes(retained, path),
source_evidence_id: parent.evidence_id,
retained_matching_nodes: matchingNodes.length,
graph_coverage: retained.graph.coverage,
});
}
const request = join(temporary.path, "trace.json");
// The CLI's portable file contract takes complete Evidence; MCP resolves a
// same-session reference into the same shared application workflow.
@@ -259,6 +368,16 @@ try {
evidence_id: evidence.evidence_id,
};
}
const safeApplicability =
nativeScopes.length === 0 &&
structuredScopes.every(({ path, scopes }) =>
timedOut
? matchesStructuredScopeExpectation(
scopes,
structuredExpectations[path],
)
: scopes.length === 0,
);
let tokenCosts = null;
if (values["tokenizer-python"]) {
temporary ??= await PrivateRuntimeRoot.create({
@@ -267,7 +386,12 @@ try {
const input = join(temporary.path, "token-input.json");
await writeFile(
input,
serialize({ catalog, selected, preparation, query: first }),
serialize({
catalog,
selected,
preparation,
queries: [first, ...structuredQueries],
}),
);
const result = await ownedCommand(values["tokenizer-python"], [
"-c",
@@ -275,9 +399,9 @@ try {
with open(sys.argv[1],encoding='utf-8') as f:d=json.load(f)
enc=tiktoken.get_encoding('cl100k_base')
def n(v):return len(enc.encode(json.dumps(v,ensure_ascii=False,separators=(',',':')),disallowed_special=()))
requests=n(d['preparation']['request'])+n(d['query']['request'])
structured=n(d['preparation']['response'])+n(d['query']['response'])
full_reply=n(d['preparation']['wire_response'])+n(d['query']['wire_response'])
requests=n(d['preparation']['request'])+sum(n(q['request']) for q in d['queries'])
structured=n(d['preparation']['response'])+sum(n(q['response']) for q in d['queries'])
full_reply=n(d['preparation']['wire_response'])+sum(n(q['wire_response']) for q in d['queries'])
s=n(d['selected']);a=n(d['catalog'])
print(json.dumps(dict(tokenizer='cl100k_base',tiktoken_version=tiktoken.__version__,selected_catalog=s,full_catalog=a,requests=requests,structured_replies=structured,full_tool_replies=full_reply,selected_catalog_structured_task=s+requests+structured,full_catalog_structured_task=a+requests+structured,selected_catalog_full_reply_task=s+requests+full_reply,full_catalog_full_reply_task=a+requests+full_reply)))`,
input,
@@ -292,11 +416,19 @@ print(json.dumps(dict(tokenizer='cl100k_base',tiktoken_version=tiktoken.__versio
companion_sha256_observed: companionSha256,
prior_native_sha256_supplied: values["native-sha256"],
source_paths: values["source-path"] ?? null,
preparation_outcome: timedOut ? "provider_timeout" : "completed",
analysis_timeout_ms: analysisTimeoutMs ?? null,
parent_evidence_id: sourceId,
trace_evidence_id: first.response.evidence_id,
trace_subject: first.response.subject,
preparation_ms: preparationMs,
samples,
safe_identity: safeIdentity,
safe_applicability: safeApplicability,
native_analysis_scopes: nativeScopes,
structured_scopes: structuredScopes,
structured_expectations: structuredExpectations,
modeled_structured_trace_count: structuredQueries.length,
node: {
node_id: addon.node_id,
identity: addon.identity,
@@ -322,10 +454,12 @@ print(json.dumps(dict(tokenizer='cl100k_base',tiktoken_version=tiktoken.__versio
cli_parity: cliParity,
token_costs: tokenCosts,
conditions:
"Fresh MCP session, one selected analysis followed by repeated identical retained traces. Filesystem/build caches uncontrolled. Task token counts include one analysis summary and one trace with name/description/inputSchema catalog projections; structured and complete-tool-reply projections are alternatives, not billing. Repeated rounds, export, CLI parity and tokenizer are validation overhead outside that one-task token count but inside total resource receipts. A wrong-identity baseline is not a successful equivalent task or efficiency credit.",
"Fresh MCP session, selected analysis followed by retained traces. Task token counts include one analysis summary or parsed timeout error, one native trace and each requested structured-path trace, with name/description/inputSchema catalog projections. Structured Evidence/parsed error and complete-tool-reply projections are alternatives, not billing. Repeated native rounds, export, CLI parity and tokenizer are validation overhead outside task tokens but inside total resource receipts. Filesystem/build caches uncontrolled. Incorrect identity/applicability is not a successful equivalent task or efficiency credit.",
};
if (values["require-safe"])
if (values["require-safe"]) {
assert.ok(safeIdentity, "Native identity/linkage oracle failed");
assert.ok(safeApplicability, "File analysis applicability oracle failed");
}
} catch (cause) {
primaryError = cause;
throw cause;