gitnexus 1.6.6-rc.98 → 1.6.6

package/dist/core/ingestion/parsing-processor.js

@@ -1,19 +1,6 @@
-import Parser from 'tree-sitter';
-import { loadParser, loadLanguage, isLanguageAvailable } from '../tree-sitter/parser-loader.js';
-import { getProvider } from './languages/index.js';
-import { generateId } from '../../lib/utils.js';
-import { getLanguageFromFilename, SupportedLanguages } from '../../_shared/index.js';
-import { extractVueScript, isVueSetupTopLevel } from './vue-sfc-extractor.js';
-import { yieldToEventLoop } from './utils/event-loop.js';
-import { parseSourceSafe } from '../tree-sitter/safe-parse.js';
-import { isVerboseIngestionEnabled } from './utils/verbose.js';
-import { getDefinitionNodeFromCaptures, findEnclosingClassInfo, findObjectLiteralBindingInfo, getLabelFromCaptures, CLASS_CONTAINER_TYPES, } from './utils/ast-helpers.js';
-import { detectFrameworkFromAST } from './framework-detection.js';
-import { buildTypeEnv } from './type-env.js';
-import { buildMethodProps, arityForIdFromInfo, typeTagForId, constTagForId, buildCollisionGroups, parameterShapeIdTag, } from './utils/method-props.js';
-import { extractTemplateArguments, templateArgumentsIdTag, templateConstraintsIdTag, } from './utils/template-arguments.js';
+import { getLanguageFromFilename } from '../../_shared/index.js';
+import { accumulateExportedTypesFromParsedNode } from './call-processor.js';
 import { logger } from '../logger.js';
-import { getTreeSitterBufferSize, getTreeSitterContentByteLength, TREE_SITTER_MAX_BUFFER, } from './constants.js';
 // ============================================================================
 // Worker-based parallel parsing
 // ============================================================================
@@ -21,18 +8,14 @@ import { getTreeSitterBufferSize, getTreeSitterContentByteLength, TREE_SITTER_MA
  * Merge a list of `ParseWorkerResult`s into the running graph + symbol
  * table state and produce the chunk-aggregated `WorkerExtractedData`.
  *
- * Extracted from `processParsingWithWorkers` so the same merge logic can
+ * Split out from the worker-parse path so the same merge logic can
  * be applied to both freshly-parsed worker output AND cached worker
  * output replayed during incremental analyze. Idempotent on the
  * accumulator fields (push-only); idempotent on graph if the caller
  * starts from a clean graph (otherwise duplicate `addNode` calls are
  * silently no-op'd by `KnowledgeGraph`).
  */
-export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
-    const allImports = [];
-    const allCalls = [];
-    const allAssignments = [];
-    const allHeritage = [];
+export const mergeChunkResults = (graph, symbolTable, chunkResults, exportedTypeMap) => {
     const allRoutes = [];
     const allFetchCalls = [];
     const allFetchWrapperDefs = [];
@@ -42,7 +25,6 @@ export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
     const allRouterModuleAliases = [];
     const allToolDefs = [];
     const allORMQueries = [];
-    const allConstructorBindings = [];
     const fileScopeBindingsByFile = [];
     const allParsedFiles = [];
     for (const result of chunkResults) {
@@ -69,14 +51,11 @@ export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
                 qualifiedName: sym.qualifiedName,
             });
         }
-        for (const item of result.imports)
-            allImports.push(item);
-        for (const item of result.calls)
-            allCalls.push(item);
-        for (const item of result.assignments)
-            allAssignments.push(item);
-        for (const item of result.heritage)
-            allHeritage.push(item);
+        if (exportedTypeMap) {
+            for (const node of result.nodes) {
+                accumulateExportedTypesFromParsedNode(exportedTypeMap, node, symbolTable);
+            }
+        }
         for (const item of result.routes)
             allRoutes.push(item);
         for (const item of result.fetchCalls)
@@ -96,8 +75,6 @@ export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
         if (result.ormQueries)
             for (const item of result.ormQueries)
                 allORMQueries.push(item);
-        for (const item of result.constructorBindings)
-            allConstructorBindings.push(item);
         if (result.fileScopeBindings)
             for (const item of result.fileScopeBindings)
                 fileScopeBindingsByFile.push(item);
@@ -106,10 +83,6 @@ export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
                 allParsedFiles.push(item);
     }
     return {
-        imports: allImports,
-        calls: allCalls,
-        assignments: allAssignments,
-        heritage: allHeritage,
         routes: allRoutes,
         fetchCalls: allFetchCalls,
         fetchWrapperDefs: allFetchWrapperDefs,
@@ -119,21 +92,28 @@ export const mergeChunkResults = (graph, symbolTable, chunkResults) => {
         routerModuleAliases: allRouterModuleAliases,
         toolDefs: allToolDefs,
         ormQueries: allORMQueries,
-        constructorBindings: allConstructorBindings,
         fileScopeBindings: fileScopeBindingsByFile,
         parsedFiles: allParsedFiles,
     };
 };
-const processParsingWithWorkers = async (graph, files, symbolTable, astCache, workerPool, onFileProgress, 
 /**
- * When provided, populated with the raw worker results before merging.
- * Used by the incremental-indexing parse cache to capture the per-chunk
- * worker output for caching across runs. The mutation happens in-place
- * so the caller (parse-impl) can keep a reference. See
- * `gitnexus/src/storage/parse-cache.ts`.
+ * Dispatch a chunk's files to the worker pool and return the RAW per-worker
+ * results, WITHOUT merging them into the graph. Split out from
+ * {@link processParsing} so the parse loop can overlap one chunk's
+ * merge (main-thread, via {@link mergeChunkResults}) with the NEXT chunk's
+ * worker parse — the merge is the only remaining serial main-thread step once
+ * ParsedFile serialization moved into the workers (#worker-idle pipelining).
+ * Returns `[]` for an all-unparseable chunk (the caller merges `[]` → empty).
  */
-outRawResults) => {
-    // Filter to parseable files only
+export const dispatchChunkParse = async (files, workerPool, onFileProgress, 
+/** Populated in-place with the raw results (parse-cache capture). */
+outRawResults, 
+/**
+ * Content hash of this parse chunk. When set, the workers tag their durable
+ * ParsedFile shards with it so a future warm cache hit can restore them
+ * (#2038). `undefined` ⇒ no durable write (tests / no-cache path).
+ */
+chunkHash) => {
     const parseableFiles = [];
     for (const file of files) {
         const lang = getLanguageFromFilename(file.path);
@@ -141,39 +121,17 @@ outRawResults) => {
             parseableFiles.push({ path: file.path, content: file.content });
     }
     if (parseableFiles.length === 0)
-        return {
-            imports: [],
-            calls: [],
-            assignments: [],
-            heritage: [],
-            routes: [],
-            fetchCalls: [],
-            fetchWrapperDefs: [],
-            decoratorRoutes: [],
-            routerIncludes: [],
-            routerImports: [],
-            routerModuleAliases: [],
-            toolDefs: [],
-            ormQueries: [],
-            constructorBindings: [],
-            fileScopeBindings: [],
-            parsedFiles: [],
-        };
+        return [];
     const total = files.length;
-    // Dispatch to worker pool — pool handles splitting into chunks and sub-batching
     const chunkResults = await workerPool.dispatch(parseableFiles, (filesProcessed) => {
         onFileProgress?.(Math.min(filesProcessed, total), total, 'Parsing...');
-    });
-    // Capture the raw chunk results for the incremental parse cache before
-    // merging — the cache stores the unmerged worker output so a future run
-    // can re-merge them into a fresh graph state.
+    }, chunkHash);
+    // Capture raw results for the incremental parse cache before merging.
     if (outRawResults) {
         for (const r of chunkResults)
             outRawResults.push(r);
     }
-    // Merge results from all workers into graph and symbol table.
-    const merged = mergeChunkResults(graph, symbolTable, chunkResults);
-    // Merge and log skipped languages from workers
+    // Skipped-language telemetry (worker output, independent of the merge).
     const skippedLanguages = new Map();
     for (const result of chunkResults) {
         for (const [lang, count] of Object.entries(result.skippedLanguages)) {
@@ -186,497 +144,8 @@ outRawResults) => {
             .join(', ');
         logger.warn(`  Skipped unsupported languages: ${summary}`);
     }
-    // Final progress
     onFileProgress?.(total, total, 'done');
-    return merged;
-};
-// ============================================================================
-// Sequential fallback (original implementation)
-// ============================================================================
-// Inline caches to avoid repeated parent-walks per node (same pattern as parse-worker.ts).
-// Keyed by tree-sitter node reference — cleared at the start of each file.
-const classInfoCache = new Map();
-const exportCache = new Map();
-const cachedFindEnclosingClassInfo = (node, filePath, resolveEnclosingOwner) => {
-    const cached = classInfoCache.get(node);
-    if (cached !== undefined)
-        return cached;
-    const result = findEnclosingClassInfo(node, filePath, resolveEnclosingOwner);
-    classInfoCache.set(node, result);
-    return result;
-};
-const cachedExportCheck = (checker, node, name) => {
-    const cached = exportCache.get(node);
-    if (cached !== undefined)
-        return cached;
-    const result = checker(node, name);
-    exportCache.set(node, result);
-    return result;
-};
-// FieldExtractor cache for sequential path — same pattern as parse-worker.ts
-const seqFieldInfoCache = new Map();
-// MethodExtractor cache for sequential path — avoids re-traversing the same class
-// body once per method. Keyed on classNode.id (tree-sitter node identity number).
-const seqMethodExtractCache = new Map();
-// Derived method map + collision groups cache — avoids rebuilding per method.
-const seqMethodMapCache = new Map();
-/** Provider-aware enclosing container lookup.
- *  Walks up from `node` until a CLASS_CONTAINER_TYPES node is found.
- *  When `resolveEnclosingOwner` is provided, delegates language-specific
- *  container remapping (e.g., Ruby singleton_class → enclosing class).
- *  Without the hook, returns the first matching container directly (raw lookup). */
-function seqFindEnclosingOwnerNode(node, resolveEnclosingOwner) {
-    let current = node.parent;
-    while (current) {
-        if (CLASS_CONTAINER_TYPES.has(current.type)) {
-            if (resolveEnclosingOwner) {
-                const resolved = resolveEnclosingOwner(current);
-                if (resolved === null) {
-                    // Provider says skip this container — keep walking up.
-                    current = current.parent;
-                    continue;
-                }
-                return resolved;
-            }
-            return current;
-        }
-        current = current.parent;
-    }
-    return null;
-}
-/** Minimal no-op SymbolTable stub for sequential extractor contexts. The real
- *  SymbolTable is not fully populated yet at this stage, so use the stub for safety.
- *  Implements the full {@link SymbolTableReader} surface so future extractor additions
- *  don't silently fall off an `as unknown as` cast. */
-const NOOP_SYMBOL_TABLE_SEQ = {
-    lookupExact: () => undefined,
-    lookupExactFull: () => undefined,
-    lookupExactAll: () => [],
-    lookupCallableByName: () => [],
-    getFiles: () => [][Symbol.iterator](),
-    getStats: () => ({ fileCount: 0 }),
-};
-function seqGetFieldInfo(classNode, provider, context) {
-    if (!provider.fieldExtractor)
-        return undefined;
-    const cacheKey = classNode.startIndex;
-    let cached = seqFieldInfoCache.get(cacheKey);
-    if (cached)
-        return cached;
-    const extracted = provider.fieldExtractor.extract(classNode, context);
-    if (!extracted?.fields?.length)
-        return undefined;
-    cached = new Map();
-    for (const field of extracted.fields)
-        cached.set(field.name, field);
-    seqFieldInfoCache.set(cacheKey, cached);
-    return cached;
-}
-const processParsingSequential = async (graph, files, symbolTable, astCache, scopeTreeCache, onFileProgress) => {
-    const parser = await loadParser();
-    const total = files.length;
-    const logSkipped = isVerboseIngestionEnabled();
-    const skippedByLang = logSkipped ? new Map() : null;
-    for (let i = 0; i < files.length; i++) {
-        const file = files[i];
-        // Reset memoization before each new file (node refs are per-tree)
-        classInfoCache.clear();
-        exportCache.clear();
-        seqFieldInfoCache.clear();
-        seqMethodExtractCache.clear();
-        seqMethodMapCache.clear();
-        onFileProgress?.(i + 1, total, file.path);
-        if (i % 20 === 0)
-            await yieldToEventLoop();
-        const language = getLanguageFromFilename(file.path);
-        if (!language)
-            continue;
-        if (!isLanguageAvailable(language)) {
-            if (skippedByLang) {
-                skippedByLang.set(language, (skippedByLang.get(language) ?? 0) + 1);
-            }
-            continue;
-        }
-        // Skip files larger than the max tree-sitter buffer (32 MB)
-        if (getTreeSitterContentByteLength(file.content) > TREE_SITTER_MAX_BUFFER)
-            continue;
-        // Vue SFC preprocessing: extract <script> block content
-        let parseContent = file.content;
-        let lineOffset = 0;
-        let isVueSetup = false;
-        if (language === SupportedLanguages.Vue) {
-            const extracted = extractVueScript(file.content);
-            if (!extracted)
-                continue; // skip .vue files with no script block
-            parseContent = extracted.scriptContent;
-            lineOffset = extracted.lineOffset;
-            isVueSetup = extracted.isSetup;
-        }
-        // Per-language source-text transform (e.g., UE macro stripping for C++).
-        // Length-preserving — see LanguageProvider.preprocessSource contract.
-        parseContent =
-            getProvider(language).preprocessSource?.(parseContent, file.path) ?? parseContent;
-        try {
-            await loadLanguage(language, file.path);
-        }
-        catch {
-            continue; // parser unavailable — safety net
-        }
-        let tree;
-        try {
-            tree = parseSourceSafe(parser, parseContent, undefined, {
-                bufferSize: getTreeSitterBufferSize(parseContent),
-            });
-        }
-        catch (parseError) {
-            logger.warn(`Skipping unparseable file: ${file.path}`);
-            continue;
-        }
-        astCache.set(file.path, tree);
-        const provider = getProvider(language);
-        // Mirror into the cross-phase cache only when the language has a
-        // scope-resolution consumer — otherwise we retain Trees no one
-        // reads. parse-impl clears `astCache` between chunks;
-        // `scopeTreeCache` survives until scope-resolution disposes it.
-        if (provider.emitScopeCaptures !== undefined) {
-            scopeTreeCache?.set(file.path, tree);
-        }
-        const queryString = provider.treeSitterQueries;
-        if (!queryString) {
-            continue;
-        }
-        let query;
-        let matches;
-        try {
-            const language = parser.getLanguage();
-            query = new Parser.Query(language, queryString);
-            matches = query.matches(tree.rootNode);
-        }
-        catch (queryError) {
-            logger.warn({ queryError }, `Query error for ${file.path}:`);
-            continue;
-        }
-        // Build per-file type environment for FieldExtractor context (lightweight — skipped if no fieldExtractor).
-        //
-        // Note: this TypeEnv is intentionally NOT flushed into the BindingAccumulator.
-        // The accumulator feed happens later in `call-processor.ts` via its own
-        // `typeEnv.flush(accumulator)` call. Flushing here would double-count
-        // file-scope bindings and break the single-use invariant of `flush()`.
-        // See the BindingAccumulator class JSDoc for the full accumulator
-        // lifecycle and flush-site ownership rules.
-        const typeEnv = provider.fieldExtractor
-            ? buildTypeEnv(tree, language, {
-                enclosingFunctionFinder: provider.enclosingFunctionFinder,
-                extractFunctionName: provider.methodExtractor?.extractFunctionName,
-            })
-            : null;
-        matches.forEach((match) => {
-            const captureMap = {};
-            match.captures.forEach((c) => {
-                captureMap[c.name] = c.node;
-            });
-            const definitionNodeForRange = getDefinitionNodeFromCaptures(captureMap);
-            const definitionNode = getDefinitionNodeFromCaptures(captureMap);
-            const defaultNodeLabel = getLabelFromCaptures(captureMap, provider);
-            if (!defaultNodeLabel)
-                return;
-            const nameNode = captureMap['name'];
-            const extractedClassSymbol = definitionNode && provider.classExtractor?.isTypeDeclaration(definitionNode)
-                ? provider.classExtractor.extract(definitionNode, {
-                    name: nameNode?.text,
-                    type: defaultNodeLabel,
-                })
-                : null;
-            const nodeLabel = extractedClassSymbol?.type ?? defaultNodeLabel;
-            const isClassLikeLabel = nodeLabel === 'Class' ||
-                nodeLabel === 'Struct' ||
-                nodeLabel === 'Interface' ||
-                nodeLabel === 'Enum' ||
-                nodeLabel === 'Record';
-            if (isClassLikeLabel &&
-                provider.classExtractor?.shouldSkipClassCapture?.({
-                    captureMap,
-                    definitionNode,
-                    nameNode,
-                    nodeLabel,
-                }) === true) {
-                return;
-            }
-            // Synthesize name for constructors without explicit @name capture (e.g. Swift init)
-            if (!nameNode && nodeLabel !== 'Constructor' && !extractedClassSymbol)
-                return;
-            const nodeName = extractedClassSymbol?.name ?? (nameNode ? nameNode.text : 'init');
-            const startLine = definitionNodeForRange
-                ? definitionNodeForRange.startPosition.row + lineOffset
-                : nameNode
-                    ? nameNode.startPosition.row + lineOffset
-                    : lineOffset;
-            // Compute enclosing class BEFORE node ID — needed to qualify method IDs
-            const needsOwner = nodeLabel === 'Method' ||
-                nodeLabel === 'Constructor' ||
-                nodeLabel === 'Property' ||
-                nodeLabel === 'Function';
-            const enclosingClassInfo = needsOwner
-                ? cachedFindEnclosingClassInfo(nameNode || definitionNodeForRange, file.path, provider.resolveEnclosingOwner)
-                : null;
-            const enclosingClassId = enclosingClassInfo?.classId ?? null;
-            const objectLiteralOwnerInfo = !enclosingClassId && nodeLabel === 'Method' && definitionNode
-                ? findObjectLiteralBindingInfo(definitionNode, file.path)
-                : null;
-            // Qualify method/property IDs with enclosing class name to avoid collisions
-            // e.g. "Method:animal.dart:Animal.speak" vs "Method:animal.dart:Dog.speak"
-            const qualifiedName = enclosingClassInfo
-                ? `${enclosingClassInfo.className}.${nodeName}`
-                : nodeName;
-            // Extract method metadata for Function/Method/Constructor nodes BEFORE generating
-            // the node ID — parameterCount is needed to disambiguate overloaded methods.
-            // Use the per-language MethodExtractor for method metadata (isAbstract, isStatic,
-            // visibility, annotations, parameterCount, parameterTypes, returnType, etc.).
-            const isMethodLike = nodeLabel === 'Function' || nodeLabel === 'Method' || nodeLabel === 'Constructor';
-            let methodProps = {};
-            let arityForId; // raw param count for ID, even for variadic
-            let seqDefMethodInfo;
-            let seqDefMethods;
-            let seqClassNodeId;
-            if (isMethodLike && definitionNode) {
-                let enriched = false;
-                if (provider.methodExtractor) {
-                    // Try class-based extraction (method inside a class/struct/trait body).
-                    // Raw lookup (no resolveEnclosingOwner) so the method extractor sees
-                    // the actual container node (e.g. singleton_class) for static detection.
-                    const methodOwnerNode = seqFindEnclosingOwnerNode(definitionNode);
-                    if (methodOwnerNode) {
-                        // Cache extract() results per class node to avoid re-traversing the
-                        // same class body for every method it contains (O(N) -> O(1) per hit).
-                        let result = seqMethodExtractCache.get(methodOwnerNode.id);
-                        if (result === undefined) {
-                            result =
-                                provider.methodExtractor.extract(methodOwnerNode, {
-                                    filePath: file.path,
-                                    language,
-                                }) ?? null;
-                            seqMethodExtractCache.set(methodOwnerNode.id, result);
-                        }
-                        if (result?.methods?.length) {
-                            const defLine = definitionNode.startPosition.row + 1;
-                            const info = result.methods.find((m) => m.name === nodeName && m.line === defLine);
-                            if (info) {
-                                enriched = true;
-                                arityForId = arityForIdFromInfo(info);
-                                methodProps = buildMethodProps(info);
-                                seqDefMethodInfo = info;
-                                seqDefMethods = result.methods;
-                                seqClassNodeId = methodOwnerNode.id;
-                            }
-                        }
-                    }
-                    // For top-level methods (e.g. Go method_declaration), try extractFromNode
-                    if (!enriched && provider.methodExtractor.extractFromNode) {
-                        const info = provider.methodExtractor.extractFromNode(definitionNode, {
-                            filePath: file.path,
-                            language,
-                        });
-                        if (info) {
-                            enriched = true;
-                            arityForId = arityForIdFromInfo(info);
-                            methodProps = buildMethodProps(info);
-                        }
-                    }
-                }
-            }
-            // Append #<paramCount> to owned callable IDs to disambiguate overloads.
-            // Top-level Function IDs stay stable; functions inside an owner may overload.
-            // When same-arity collisions exist, append ~type1,type2 for further disambiguation.
-            const needsAritySuffix = nodeLabel === 'Method' ||
-                nodeLabel === 'Constructor' ||
-                (nodeLabel === 'Function' && enclosingClassId !== null);
-            let arityTag = needsAritySuffix && arityForId !== undefined ? `#${arityForId}` : '';
-            if (arityTag && seqDefMethods && seqDefMethodInfo && seqClassNodeId !== undefined) {
-                // Use cached method map + collision groups (built once per class, not per method)
-                let cached = seqMethodMapCache.get(seqClassNodeId);
-                if (!cached) {
-                    const tempMap = new Map();
-                    for (const m of seqDefMethods)
-                        tempMap.set(`${m.name}:${m.line}`, m);
-                    cached = { map: tempMap, groups: buildCollisionGroups(tempMap) };
-                    seqMethodMapCache.set(seqClassNodeId, cached);
-                }
-                arityTag += typeTagForId(cached.map, nodeName, arityForId, seqDefMethodInfo, language, cached.groups);
-                arityTag += constTagForId(cached.map, nodeName, arityForId, seqDefMethodInfo, cached.groups);
-            }
-            const parameterShapeTag = nodeLabel === 'Function' || nodeLabel === 'Method'
-                ? parameterShapeIdTag(methodProps.parameterTypes, methodProps.parameterTypeClasses)
-                : '';
-            const classTemplateArguments = extractedClassSymbol?.templateArguments ??
-                provider.classExtractor?.extractTemplateArgumentsFromCapture?.({
-                    captureMap,
-                    definitionNode,
-                    nameNode,
-                }) ??
-                (captureMap['template-arguments']
-                    ? extractTemplateArguments(captureMap['template-arguments'].text)
-                    : undefined) ??
-                (nameNode && nameNode.text ? extractTemplateArguments(nameNode.text) : undefined);
-            const classTemplateTag = (nodeLabel === 'Class' ||
-                nodeLabel === 'Struct' ||
-                nodeLabel === 'Interface' ||
-                nodeLabel === 'Enum' ||
-                nodeLabel === 'Record') &&
-                classTemplateArguments !== undefined &&
-                classTemplateArguments.length > 0
-                ? templateArgumentsIdTag(classTemplateArguments)
-                : '';
-            // SFINAE / `requires`-clause aware ID disambiguation (issue #1579).
-            // Function-template overloads with identical parameterTypes but
-            // mutually-exclusive constraints (e.g. `enable_if_t<is_integral_v<T>>`
-            // vs `enable_if_t<is_floating_point_v<T>>`) need distinct graph
-            // nodes so the constraint-filter step in `narrowOverloadCandidates`
-            // has two candidates to narrow between. Without this tag they
-            // collapse to a single Function node and the SFINAE call resolves
-            // to only one edge regardless of which overload's constraint holds.
-            // The provider hook is the right invocation point — parsing-processor
-            // sees raw tree-sitter matches without the `@`-prefixed synthetic
-            // captures `scope-extractor` consumes, so we delegate extraction to
-            // the language adapter (C++ implements this; other languages opt out).
-            let parsedTemplateConstraints = undefined;
-            let constraintsTag = '';
-            if ((nodeLabel === 'Function' || nodeLabel === 'Method') &&
-                provider.extractTemplateConstraints !== undefined &&
-                definitionNode !== null) {
-                try {
-                    parsedTemplateConstraints = provider.extractTemplateConstraints(definitionNode);
-                    if (parsedTemplateConstraints !== undefined) {
-                        constraintsTag = templateConstraintsIdTag(parsedTemplateConstraints);
-                    }
-                }
-                catch {
-                    parsedTemplateConstraints = undefined;
-                    constraintsTag = '';
-                }
-            }
-            const nodeId = generateId(nodeLabel, `${file.path}:${qualifiedName}${classTemplateTag}${arityTag}${constraintsTag}${parameterShapeTag}`);
-            const classNodeForSymbol = definitionNodeForRange || definitionNode || nameNode;
-            const qualifiedTypeName = extractedClassSymbol?.qualifiedName ??
-                (classNodeForSymbol && provider.classExtractor?.isTypeDeclaration(classNodeForSymbol)
-                    ? (provider.classExtractor.extractQualifiedName(classNodeForSymbol, nodeName) ?? nodeName)
-                    : undefined);
-            const frameworkHint = definitionNode
-                ? detectFrameworkFromAST(language, (definitionNode.text || '').slice(0, 300))
-                : null;
-            const node = {
-                id: nodeId,
-                label: nodeLabel,
-                properties: {
-                    name: nodeName,
-                    filePath: file.path,
-                    startLine: definitionNodeForRange
-                        ? definitionNodeForRange.startPosition.row + lineOffset
-                        : startLine,
-                    endLine: definitionNodeForRange
-                        ? definitionNodeForRange.endPosition.row + lineOffset
-                        : startLine,
-                    language: language,
-                    isExported: language === SupportedLanguages.Vue && isVueSetup
-                        ? isVueSetupTopLevel(nameNode || definitionNodeForRange)
-                        : cachedExportCheck(provider.exportChecker, nameNode || definitionNodeForRange, nodeName),
-                    ...(qualifiedTypeName !== undefined ? { qualifiedName: qualifiedTypeName } : {}),
-                    ...(classTemplateArguments !== undefined && classTemplateArguments.length > 0
-                        ? { templateArguments: classTemplateArguments }
-                        : {}),
-                    ...(parsedTemplateConstraints !== undefined
-                        ? { templateConstraints: parsedTemplateConstraints }
-                        : {}),
-                    ...(frameworkHint
-                        ? {
-                            astFrameworkMultiplier: frameworkHint.entryPointMultiplier,
-                            astFrameworkReason: frameworkHint.reason,
-                        }
-                        : {}),
-                    ...methodProps,
-                },
-            };
-            graph.addNode(node);
-            // enclosingClassId already computed above (before nodeId generation)
-            // Extract declared type and field metadata for Property nodes
-            let declaredType;
-            let seqVisibility;
-            let seqIsStatic;
-            let seqIsReadonly;
-            if (nodeLabel === 'Property' && definitionNode) {
-                // FieldExtractor is the single source of truth when available
-                if (provider.fieldExtractor && typeEnv) {
-                    const classNode = seqFindEnclosingOwnerNode(definitionNode, provider.resolveEnclosingOwner);
-                    if (classNode) {
-                        const fieldMap = seqGetFieldInfo(classNode, provider, {
-                            typeEnv,
-                            symbolTable: NOOP_SYMBOL_TABLE_SEQ,
-                            filePath: file.path,
-                            language,
-                        });
-                        const info = fieldMap?.get(nodeName);
-                        if (info) {
-                            declaredType = info.type ?? undefined;
-                            seqVisibility = info.visibility;
-                            seqIsStatic = info.isStatic;
-                            seqIsReadonly = info.isReadonly;
-                        }
-                    }
-                }
-                // All 15 tree-sitter languages register a FieldExtractor — no fallback needed.
-            }
-            // Apply field metadata to the graph node retroactively
-            if (seqVisibility !== undefined)
-                node.properties.visibility = seqVisibility;
-            if (seqIsStatic !== undefined)
-                node.properties.isStatic = seqIsStatic;
-            if (seqIsReadonly !== undefined)
-                node.properties.isReadonly = seqIsReadonly;
-            if (declaredType !== undefined)
-                node.properties.declaredType = declaredType;
-            symbolTable.add(file.path, nodeName, nodeId, nodeLabel, {
-                parameterCount: methodProps.parameterCount,
-                requiredParameterCount: methodProps.requiredParameterCount,
-                parameterTypes: methodProps.parameterTypes,
-                parameterTypeClasses: methodProps.parameterTypeClasses,
-                returnType: methodProps.returnType,
-                declaredType,
-                templateArguments: classTemplateArguments,
-                ownerId: enclosingClassId ?? objectLiteralOwnerInfo?.ownerId ?? undefined,
-                qualifiedName: qualifiedTypeName,
-            });
-            const fileId = generateId('File', file.path);
-            const relId = generateId('DEFINES', `${fileId}->${nodeId}`);
-            const relationship = {
-                id: relId,
-                sourceId: fileId,
-                targetId: nodeId,
-                type: 'DEFINES',
-                confidence: 1.0,
-                reason: '',
-            };
-            graph.addRelationship(relationship);
-            // ── HAS_METHOD / HAS_PROPERTY: link member to enclosing class ──
-            const ownerIdForMemberEdge = enclosingClassId ?? objectLiteralOwnerInfo?.ownerId ?? null;
-            if (ownerIdForMemberEdge) {
-                const memberEdgeType = nodeLabel === 'Property' ? 'HAS_PROPERTY' : 'HAS_METHOD';
-                graph.addRelationship({
-                    id: generateId(memberEdgeType, `${ownerIdForMemberEdge}->${nodeId}`),
-                    sourceId: ownerIdForMemberEdge,
-                    targetId: nodeId,
-                    type: memberEdgeType,
-                    confidence: 1.0,
-                    reason: objectLiteralOwnerInfo
-                        ? 'object literal method belongs to exported object binding'
-                        : '',
-                });
-            }
-        });
-    }
-    if (skippedByLang && skippedByLang.size > 0) {
-        for (const [lang, count] of skippedByLang.entries()) {
-            logger.warn(`[ingestion] Skipped ${count} ${lang} file(s) in parsing processing — ${lang} parser not available.`);
-        }
-    }
+    return chunkResults;
 };
 // ============================================================================
 // Public API
@@ -688,23 +157,13 @@ const processParsingSequential = async (graph, files, symbolTable, astCache, sco
  * when the pool is garbage-collected.
  */
 const loggedQuarantineByPool = new WeakMap();
-export const processParsing = async (graph, files, symbolTable, astCache, 
+export const processParsing = async (graph, files, symbolTable, workerPool, onFileProgress, 
 /**
- * Persistent tree cache (separate from `astCache`, which the caller
- * clears between chunks). Sequential parses additionally write the
- * Tree here so cross-phase consumers (scope-resolution) can read it.
- * Worker-mode parses skip — Trees can't cross MessageChannels.
- * Pass `undefined` if no consumer needs cross-phase access.
+ * Optional out-parameter for the incremental parse cache. When provided,
+ * populated with the raw `ParseWorkerResult[]` from the workers (pre-merge).
+ * See `gitnexus/src/storage/parse-cache.ts`.
  */
-scopeTreeCache, onFileProgress, workerPool, 
-/**
- * Optional out-parameter for the incremental parse cache. When
- * provided AND the worker-pool path runs successfully, populated
- * with the raw `ParseWorkerResult[]` from the workers (pre-merge).
- * Stays empty for the sequential fallback path (no per-chunk
- * artifact to cache there). See `gitnexus/src/storage/parse-cache.ts`.
- */
-outRawResults) => {
+outRawResults, exportedTypeMap) => {
     let lastProgress = 0;
     const reportProgress = onFileProgress
         ? (current, total, detail) => {
@@ -712,71 +171,48 @@ outRawResults) => {
             onFileProgress(lastProgress, total, detail);
         }
         : undefined;
-    if (workerPool) {
-        if (scopeTreeCache !== undefined && process.env.PROF_SCOPE_RESOLUTION === '1') {
-            // Trees can't cross MessageChannels, so worker-parsed files land
-            // in scope-resolution with an empty cache and get re-parsed.
-            // Surfacing this in PROF mode prevents silent perf cliffs when
-            // a repo crosses the worker-pool threshold.
-            logger.warn(`[scope-resolution prof] worker pool engaged for ${files.length} files — cross-phase tree cache will be empty; scope-resolution re-parses.`);
-        }
-        // U20 design pivot: the worker pool's resilience layers
-        // (respawn budget, circuit breaker, quarantine, slot-attribution,
-        // cumulative timeout) are the SOLE contract for handling worker
-        // failures. There is no sequential-parser fallback for either
-        // partial quarantine or full pool failure — the operator must see
-        // a clear hard signal when workers can't recover, instead of a
-        // silently-degraded graph from a possibly-crashing main-thread
-        // sequential parser. A failing tree-sitter native binding that
-        // quarantined a worker would, under the previous design, re-trigger
-        // the same SIGSEGV on the main thread; we avoid that risk entirely.
-        //
-        // - Partial quarantine: the file is missing from this run's graph;
-        //   the per-chunk warn log below surfaces it; U2's chunk-cache
-        //   write-guard in parse-impl.ts keeps the chunk uncached so the
-        //   next analyze gets a cache miss and a fresh pool retries.
-        // - Full pool failure: `WorkerPoolDispatchError` propagates from
-        //   `processParsingWithWorkers` up through this function. The
-        //   analyze run errors out instead of falling back to sequential.
-        const data = await processParsingWithWorkers(graph, files, symbolTable, astCache, workerPool, reportProgress, outRawResults);
-        // Session-scoped quarantine (worker-pool resilience Layer 3): surface
-        // any files this pool has decided are unsafe for workers so the
-        // operator can see what was skipped. The pool already filtered them
-        // out of dispatch; we only need to log + progress-report. Quarantine
-        // is session-scoped per pool instance — a fresh `createWorkerPool`
-        // call clears it.
-        //
-        // Dedup: log full path list only for entries newly quarantined since
-        // the previous dispatch on the same pool. The per-chunk progress
-        // message still surfaces the count for UX continuity, but the
-        // structured `quarantinedFiles` payload is only emitted when there
-        // is new signal — prevents O(quarantine × chunks) log spam.
-        const quarantineSnapshot = workerPool.getQuarantinedPaths?.() ?? [];
-        const quarantineSet = new Set(quarantineSnapshot);
-        if (quarantineSet.size > 0) {
-            const quarantinedInChunk = files.filter((file) => quarantineSet.has(file.path));
-            if (quarantinedInChunk.length > 0) {
-                const seenForPool = loggedQuarantineByPool.get(workerPool) ?? new Set();
-                const newlyQuarantined = quarantinedInChunk
-                    .map((file) => file.path)
-                    .filter((p) => !seenForPool.has(p));
-                for (const p of newlyQuarantined)
-                    seenForPool.add(p);
-                loggedQuarantineByPool.set(workerPool, seenForPool);
-                if (newlyQuarantined.length > 0) {
-                    logger.warn({
-                        newlyQuarantined,
-                        cumulativeQuarantine: quarantineSet.size,
-                        chunkSkipped: quarantinedInChunk.length,
-                    }, `Worker quarantine: ${newlyQuarantined.length} new file(s) skipped this chunk ` +
-                        `(${quarantinedInChunk.length} skipped total, ${quarantineSet.size} cumulative).`);
-                }
-                reportProgress?.(lastProgress, files.length, `${quarantinedInChunk.length} worker-quarantined file(s) skipped`);
+    // U20 design pivot: the worker pool's resilience layers (respawn budget,
+    // circuit breaker, quarantine, slot-attribution, cumulative timeout) are the
+    // SOLE contract for handling worker failures. There is no sequential parser:
+    // a partial quarantine drops the file from this run's graph (surfaced by the
+    // per-chunk warn below; the chunk-cache write-guard in parse-impl.ts keeps the
+    // chunk uncached so the next analyze retries with a fresh pool), and a full
+    // pool failure propagates `WorkerPoolDispatchError` so the run errors out.
+    const chunkResults = await dispatchChunkParse(files, workerPool, reportProgress, outRawResults);
+    const data = mergeChunkResults(graph, symbolTable, chunkResults, exportedTypeMap);
+    // Session-scoped quarantine (worker-pool resilience Layer 3): surface any
+    // files this pool has decided are unsafe for workers so the operator can see
+    // what was skipped. The pool already filtered them out of dispatch; we only
+    // need to log + progress-report. Quarantine is session-scoped per pool
+    // instance — a fresh `createWorkerPool` call clears it.
+    //
+    // Dedup: log the full path list only for entries newly quarantined since the
+    // previous dispatch on the same pool. The per-chunk progress message still
+    // surfaces the count for UX continuity, but the structured `quarantinedFiles`
+    // payload is only emitted when there is new signal — prevents
+    // O(quarantine × chunks) log spam.
+    const quarantineSnapshot = workerPool.getQuarantinedPaths?.() ?? [];
+    const quarantineSet = new Set(quarantineSnapshot);
+    if (quarantineSet.size > 0) {
+        const quarantinedInChunk = files.filter((file) => quarantineSet.has(file.path));
+        if (quarantinedInChunk.length > 0) {
+            const seenForPool = loggedQuarantineByPool.get(workerPool) ?? new Set();
+            const newlyQuarantined = quarantinedInChunk
+                .map((file) => file.path)
+                .filter((p) => !seenForPool.has(p));
+            for (const p of newlyQuarantined)
+                seenForPool.add(p);
+            loggedQuarantineByPool.set(workerPool, seenForPool);
+            if (newlyQuarantined.length > 0) {
+                logger.warn({
+                    newlyQuarantined,
+                    cumulativeQuarantine: quarantineSet.size,
+                    chunkSkipped: quarantinedInChunk.length,
+                }, `Worker quarantine: ${newlyQuarantined.length} new file(s) skipped this chunk ` +
+                    `(${quarantinedInChunk.length} skipped total, ${quarantineSet.size} cumulative).`);
             }
+            reportProgress?.(lastProgress, files.length, `${quarantinedInChunk.length} worker-quarantined file(s) skipped`);
         }
-        return data;
     }
-    // Fallback: sequential parsing (no pre-extracted data)
-    await processParsingSequential(graph, files, symbolTable, astCache, scopeTreeCache, reportProgress);
-    return null;
+    return data;
 };