Files
rea/src/application/javascript/JavaScriptArtifactGraphFindings.ts
T
morluto b58adafa4f fix(javascript): align module lookup and source evidence
Preserve CommonJS literal paths and confined ESM URL semantics, including package main fallback, directory intent, Unicode identity, and present opaque modules. Parse extraction inputs at their AST/source owner, retain skipped HTML and source-map coverage, and distinguish endpoint candidates from proven receiver facts. Version structural fingerprints with computed literal contents and propagate them through worker contracts.

Reviewed fixes include #1810, #1840, #2642, #2700, #2702, #2714, #2803, #2960, #2968, #2986, #3035, and #3036.

Validation: 158 resolver tests passed; public CLI/MCP module lookup controls, package-main/Unicode workflows, and six builtin workflows passed in the current compiled lane. Worker schema/fingerprint regressions passed. Large public pdf.js completion and real Electron analysis remain unverified.
2026-10-14 08:46:50 +08:00

592 lines
19 KiB
TypeScript

import { createHash } from "node:crypto";
import { compactStringIdentityKey } from "../../domain/compactStringIdentity.js";
import type { ApplicationNode } from "../../domain/javascript/javascriptApplicationGraphSchemas.js";
import type { JavaScriptStaticAnalysis } from "../../domain/javascript/javascriptStaticAnalysisTypes.js";
import type { JavaScriptArtifactFile } from "../../domain/javascript/javascriptArtifactFiles.js";
import {
addStaticInferenceEdge,
artifactLocalIdentity,
chunkLookupKey,
createElectronRoleNode,
javascriptAnalysisCoverage,
linkElectronRoleToAsset,
moduleLookupKey,
selfReferenceOmissions,
sourceNodeFor,
type JavaScriptArtifactGraphContext,
type JavaScriptArtifactGraphCoverage,
} from "./JavaScriptArtifactGraphContext.js";
import {
astObservationEvidence,
staticInferenceEvidence,
} from "./JavaScriptArtifactGraphEvidence.js";
import {
artifactPathResolutionProperties,
resolveArtifactPathByContext,
type ArtifactPathResolution,
} from "./JavaScriptArtifactPathResolution.js";
import { moduleResolutionModeForJavaScriptReference } from "../../domain/javascript/javascriptModuleResolutionMode.js";
type StaticReference = JavaScriptStaticAnalysis["references"][number];
type StaticEndpoint = JavaScriptStaticAnalysis["endpoints"][number];
type StaticStorage = JavaScriptStaticAnalysis["storage"][number];
type StaticRolePath = JavaScriptStaticAnalysis["role_paths"][number];
type StaticSourceMap = JavaScriptStaticAnalysis["source_map_urls"][number];
interface FindingInput<Value> {
readonly file: JavaScriptArtifactFile;
readonly asset: ApplicationNode;
readonly value: Value;
readonly coverage: JavaScriptArtifactGraphCoverage;
}
interface ResolvedReference {
readonly path: string | null;
readonly file: JavaScriptArtifactFile | null;
readonly node: ApplicationNode | null;
readonly limitations: readonly string[];
readonly resolutionStatus?: string;
readonly resolution?: ArtifactPathResolution;
}
/** Project imports, workers, roles, endpoints, storage, and source-map links. */
export function* addJavaScriptStaticFindingsSteps(
context: JavaScriptArtifactGraphContext,
): Generator<void, readonly string[]> {
const omissions = { selfReferences: 0 };
for (const analyzed of context.analysis.files) {
yield;
const { file, javascript } = analyzed;
const asset = context.assetNodes.get(file.path);
if (javascript === null || asset === undefined) continue;
const coverage = javascriptAnalysisCoverage(javascript);
for (const value of javascript.references)
addReference(context, { file, asset, value, coverage }, omissions);
for (const value of javascript.endpoints)
addEndpoint(context, { file, asset, value, coverage });
for (const value of javascript.storage)
addStorage(context, { file, asset, value, coverage });
for (const value of javascript.role_paths)
addDiscoveredRole(context, { file, asset, value, coverage });
for (const value of javascript.source_map_urls)
addSourceMapEdge(context, { file, asset, value, coverage });
}
return selfReferenceOmissions(
omissions.selfReferences,
["static reference", "static references"],
"the referencing module",
);
}
const addReference = (
context: JavaScriptArtifactGraphContext,
input: FindingInput<StaticReference>,
omissions: { selfReferences: number },
): void => {
const source =
sourceNodeFor(
context,
input.file.path,
input.value.module_key,
input.value.module_runtime,
) ?? input.asset;
const resolved =
input.value.specifier === null
? null
: resolveReference(context, input.file.path, input.value, input.coverage);
if (input.value.kind === "worker" || input.value.kind === "service-worker") {
const worker = context.accumulator.addNode({
kind: input.value.kind === "worker" ? "worker" : "service-worker",
identity: artifactLocalIdentity(
input.file.sha256,
input.value.kind,
input.value.specifier ?? input.value.expression,
),
observations: [
{
label: input.value.specifier,
properties: {
declared_specifier: input.value.specifier,
expression: input.value.expression,
resolved_path: resolved?.path ?? null,
resolution_status: resolved?.resolutionStatus ?? null,
...artifactPathResolutionProperties(resolved?.resolution),
},
evidence: staticInferenceEvidence({
sha256: input.file.sha256,
path: input.file.path,
range: input.value.location,
operation: "discover-worker",
coverage: input.coverage,
limitations: resolved?.limitations ?? [],
}),
},
],
});
addStaticInferenceEdge(context, {
source,
target: worker,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: "loads",
properties: {
kind: input.value.kind,
resolution_status: resolved?.resolutionStatus ?? null,
...artifactPathResolutionProperties(resolved?.resolution),
},
limitations: resolved?.limitations ?? [],
});
if (resolved !== null && resolved.path !== null && resolved.node !== null)
addStaticInferenceEdge(context, {
source: worker,
target: resolved.node,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: "maps_to",
properties: { resolved_path: resolved.path },
limitations: resolved.limitations,
});
return;
}
// A reference that resolves to its own source, such as a bundled module
// requiring its own key, cannot become an edge: the application graph
// forbids self-referential edges and one would fail the whole analysis.
if (resolved?.node?.node_id === source.node_id) {
omissions.selfReferences += 1;
return;
}
const target =
resolved?.node ??
unresolvedReferenceNode(context, input.file, input.value, input.coverage);
addStaticInferenceEdge(context, {
source,
target,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: resolved?.file?.kind === "native-addon" ? "loads" : "imports",
properties: {
kind: input.value.kind,
specifier: input.value.specifier,
expression: input.value.expression,
resolution_mode:
input.value.kind === "static-import" ||
input.value.kind === "dynamic-import" ||
input.value.kind === "require"
? moduleResolutionModeForJavaScriptReference(
input.file.path,
input.value.kind,
)
: null,
resolved_path: resolved?.path ?? null,
resolution_status: resolved?.resolutionStatus ?? null,
...artifactPathResolutionProperties(resolved?.resolution),
},
...(resolved === null ? {} : { limitations: resolved.limitations }),
});
};
const addEndpoint = (
context: JavaScriptArtifactGraphContext,
input: FindingInput<StaticEndpoint>,
): void => {
const source =
sourceNodeFor(
context,
input.file.path,
input.value.module_key,
input.value.module_runtime,
) ?? input.asset;
const key = compactStringIdentityKey(input.value.value);
const label = `${input.value.kind} endpoint`;
const node = context.accumulator.addNode({
kind: "endpoint",
identity: artifactLocalIdentity(input.file.sha256, input.value.kind, key),
observations: [
{
label:
input.value.value.length <= label.length ? input.value.value : label,
properties: {
endpoint_kind: input.value.kind,
value: input.value.value,
mechanism: input.value.mechanism,
},
evidence: astObservationEvidence({
sha256: input.file.sha256,
path: input.file.path,
range: input.value.location,
operation: "discover-static-endpoint",
coverage: input.coverage,
}),
},
],
});
addStaticInferenceEdge(context, {
source,
target: node,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: input.value.kind === "route" ? "exposes" : "calls",
properties: { mechanism: input.value.mechanism },
});
};
const addStorage = (
context: JavaScriptArtifactGraphContext,
input: FindingInput<StaticStorage>,
): void => {
const source =
sourceNodeFor(
context,
input.file.path,
input.value.module_key,
input.value.module_runtime,
) ?? input.asset;
const node = context.accumulator.addNode({
kind: "storage",
identity: artifactLocalIdentity(
input.file.sha256,
input.value.kind,
input.value.name ?? input.value.mechanism,
),
observations: [
{
label: input.value.name ?? input.value.kind,
properties: {
storage_kind: input.value.kind,
name: input.value.name,
mechanism: input.value.mechanism,
},
evidence: astObservationEvidence({
sha256: input.file.sha256,
path: input.file.path,
range: input.value.location,
operation: "discover-static-storage",
coverage: input.coverage,
}),
},
],
});
addStaticInferenceEdge(context, {
source,
target: node,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: "persists_to",
properties: { mechanism: input.value.mechanism },
});
};
const addDiscoveredRole = (
context: JavaScriptArtifactGraphContext,
input: FindingInput<StaticRolePath>,
): void => {
const resolution = resolveArtifactPathByContext({
declaredPath: input.value.path,
sourcePath: input.file.path,
context: input.value.resolution_context,
files: context.filesByPath,
directoryPaths: context.directoryPaths,
unanalyzedFiles: context.unanalyzedFiles,
...(input.value.role === "preload" ||
(input.value.role === "renderer" &&
input.value.resolution_context === "filesystem-expression")
? { runtimePathKind: "exact-file" as const }
: input.value.role === "renderer" &&
input.value.resolution_context === "module-specifier"
? { runtimePathKind: "url-resource" as const }
: {}),
});
const role = createElectronRoleNode(context, {
kind:
input.value.role === "preload" ? "electron-preload" : "electron-renderer",
anchor: input.file,
resolution,
mechanism: input.value.mechanism,
range: input.value.location,
coverage: input.coverage,
});
const source =
sourceNodeFor(
context,
input.file.path,
input.value.module_key,
input.value.module_runtime,
) ?? input.asset;
addStaticInferenceEdge(context, {
source,
target: role,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: "loads",
properties: { mechanism: input.value.mechanism },
});
linkElectronRoleToAsset(context, {
role,
anchor: input.file,
resolution,
range: input.value.location,
coverage: input.coverage,
});
};
const addSourceMapEdge = (
context: JavaScriptArtifactGraphContext,
input: FindingInput<StaticSourceMap>,
): void => {
const declaredUrlSha256 = createHash("sha256")
.update(input.value.declared_url)
.digest("hex");
const resolution = resolveArtifactPathByContext({
declaredPath: input.value.declared_url,
sourcePath: input.file.path,
context: "url-reference",
files: context.filesByPath,
directoryPaths: context.directoryPaths,
unanalyzedFiles: context.unanalyzedFiles,
});
const resolvedPath = resolution.resolved_path;
const resolvedFile =
resolvedPath === null ? undefined : context.filesByPath.get(resolvedPath);
const resolvedNode =
resolvedPath === null ? undefined : context.fileNodes.get(resolvedPath);
const isAvailable =
resolvedFile?.kind === "source-map" && resolvedNode !== undefined;
const limitations =
isAvailable || resolution.limitations.length > 0
? resolution.limitations
: [
"The declared reference does not identify an available source-map projection.",
];
const target = context.accumulator.addNode({
kind: "source-map",
identity:
isAvailable && resolvedNode !== undefined
? resolvedNode.identity
: artifactLocalIdentity(
input.file.sha256,
"unresolved-source-map",
declaredUrlSha256,
),
observations: [
{
label: "source map reference",
properties: {
declared_url: input.value.declared_url,
declared_url_sha256: declaredUrlSha256,
resolved_path: resolvedPath,
resolution_status: resolution.resolution_status,
...artifactPathResolutionProperties(resolution),
available: isAvailable,
},
evidence: staticInferenceEvidence({
sha256: input.file.sha256,
path: input.file.path,
range: input.value.location,
operation: "discover-source-map",
coverage: input.coverage,
confidence: "medium",
limitations,
}),
},
],
});
addStaticInferenceEdge(context, {
source: input.asset,
target,
file: input.file,
range: input.value.location,
coverage: input.coverage,
relation: "maps_to",
limitations,
properties: {
declared_url_sha256: declaredUrlSha256,
resolved_path: resolvedPath,
resolution_status: resolution.resolution_status,
...artifactPathResolutionProperties(resolution),
},
});
};
const resolveReference = (
context: JavaScriptArtifactGraphContext,
sourcePath: string,
reference: StaticReference,
coverage: JavaScriptArtifactGraphCoverage,
): ResolvedReference | null => {
const specifier = reference.specifier;
if (specifier === null) return null;
const isWorker =
reference.kind === "worker" || reference.kind === "service-worker";
const chunk = isWorker
? null
: resolveBundlerChunkReference(context, sourcePath, reference);
if (chunk !== null) return chunk;
const module =
isWorker || reference.module_runtime === undefined
? undefined
: context.moduleNodes.get(
moduleLookupKey(sourcePath, reference.module_runtime, specifier),
);
if (module !== undefined) {
const file = context.filesByPath.get(sourcePath);
return file === undefined
? null
: {
path: `${sourcePath}#module:${specifier}`,
file,
node: module,
limitations: [],
};
}
// Worker scripts are URLs relative to the script, not module specifiers.
const resolution = isWorker
? resolveArtifactPathByContext({
declaredPath: specifier,
sourcePath,
context: "url-reference",
files: context.filesByPath,
directoryPaths: context.directoryPaths,
unanalyzedFiles: context.unanalyzedFiles,
})
: resolveArtifactPathByContext({
declaredPath: specifier,
sourcePath,
context: "module-specifier",
files: context.filesByPath,
directoryPaths: context.directoryPaths,
unanalyzedFiles: context.unanalyzedFiles,
...("require" === reference.kind ||
"static-import" === reference.kind ||
"dynamic-import" === reference.kind
? {
moduleKind: moduleResolutionModeForJavaScriptReference(
sourcePath,
reference.kind,
),
}
: {}),
});
const path = resolution.resolved_path;
if (path === null) {
const source = context.filesByPath.get(sourcePath);
return source === undefined
? null
: {
path: null,
file: null,
node: isWorker
? null
: unresolvedReferenceNode(
context,
source,
reference,
coverage,
resolution,
),
limitations: resolution.limitations,
resolutionStatus: resolution.resolution_status,
resolution,
};
}
const file = context.filesByPath.get(path);
const node = context.fileNodes.get(path);
return file === undefined || node === undefined
? null
: {
path,
file,
node,
limitations: resolution.limitations,
resolutionStatus: resolution.resolution_status,
resolution,
};
};
const resolveBundlerChunkReference = (
context: JavaScriptArtifactGraphContext,
sourcePath: string,
reference: StaticReference,
): ResolvedReference | null => {
const key = reference.specifier?.startsWith("chunk:")
? reference.specifier.slice("chunk:".length)
: null;
if (key === null) return null;
const file = context.filesByPath.get(sourcePath);
if (file === undefined) return null;
const registrations = context.analysis.files
.find(({ file: analyzedFile }) => analyzedFile.path === sourcePath)
?.javascript?.bundler_registrations.filter(({ runtime, modules }) =>
reference.module_runtime === undefined
? reference.module_key === null
: runtime === reference.module_runtime &&
modules.some(
({ module_key: moduleKey }) => moduleKey === reference.module_key,
),
);
if (registrations === undefined) return null;
const candidates = registrations.flatMap(({ runtime }) => {
const node = context.chunkNodes.get(
chunkLookupKey(sourcePath, runtime, key),
);
return node === undefined ? [] : [node];
});
const unique = [
...new Map(candidates.map((node) => [node.node_id, node])).values(),
];
if (unique.length !== 1) return null;
const node = unique[0];
return node === undefined
? null
: { path: `${sourcePath}#chunk:${key}`, file, node, limitations: [] };
};
const unresolvedReferenceNode = (
context: JavaScriptArtifactGraphContext,
file: JavaScriptArtifactFile,
reference: StaticReference,
coverage: JavaScriptArtifactGraphCoverage,
resolution?: ArtifactPathResolution,
): ApplicationNode =>
context.accumulator.addNode({
kind: "javascript-module",
identity: artifactLocalIdentity(
file.sha256,
"unresolved-reference",
`${reference.kind}:${reference.specifier ?? reference.expression}`,
),
observations: [
{
label: reference.specifier,
properties: {
reference_kind: reference.kind,
specifier: reference.specifier,
expression: reference.expression,
resolution: "unresolved",
...(resolution === undefined
? {}
: { resolution_status: resolution.resolution_status }),
...artifactPathResolutionProperties(resolution),
},
evidence: staticInferenceEvidence({
sha256: file.sha256,
path: file.path,
range: reference.location,
operation: "resolve-static-reference",
coverage,
confidence: "low",
limitations: [
"The static reference could not be resolved within the inventoried artifact.",
...(resolution?.limitations ?? []),
],
}),
},
],
});