mistagent 0.1.0

package/dist/src/tools/code-analyzer/core/ingestion/community-processor.js

@@ -0,0 +1,328 @@
+/**
+ * Community Detection Processor
+ *
+ * Uses the Leiden algorithm (via graphology-communities-leiden) to detect
+ * communities/clusters in the code graph based on CALLS relationships.
+ *
+ * Communities represent groups of code that work together frequently,
+ * helping agents navigate the codebase by functional area rather than file structure.
+ */
+// NOTE: The Leiden algorithm source is vendored from graphology's repo
+// (src/communities-leiden) because it was never published to npm.
+// We use createRequire to load the CommonJS vendored files in ESM context.
+import Graph from 'graphology';
+import { createRequire } from 'node:module';
+import { fileURLToPath } from 'node:url';
+import { dirname, resolve } from 'node:path';
+import { existsSync } from 'node:fs';
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = dirname(__filename);
+// Find project root by walking up until we find vendor/leiden/index.cjs
+// (works from both src/ and dist/ regardless of nesting depth)
+const findLeidenPath = () => {
+    let dir = __dirname;
+    for (let i = 0; i < 10; i++) {
+        const candidate = resolve(dir, 'vendor', 'leiden', 'index.cjs');
+        if (existsSync(candidate))
+            return candidate;
+        const parent = resolve(dir, '..');
+        if (parent === dir)
+            break;
+        dir = parent;
+    }
+    throw new Error('Could not find vendor/leiden/index.cjs');
+};
+const leidenPath = findLeidenPath();
+const _require = createRequire(import.meta.url);
+const leiden = _require(leidenPath);
+// ============================================================================
+// COMMUNITY COLORS (for visualization)
+// ============================================================================
+export const COMMUNITY_COLORS = [
+    '#ef4444', // red
+    '#f97316', // orange
+    '#eab308', // yellow
+    '#22c55e', // green
+    '#06b6d4', // cyan
+    '#3b82f6', // blue
+    '#8b5cf6', // violet
+    '#d946ef', // fuchsia
+    '#ec4899', // pink
+    '#f43f5e', // rose
+    '#14b8a6', // teal
+    '#84cc16', // lime
+];
+export const getCommunityColor = (communityIndex) => {
+    return COMMUNITY_COLORS[communityIndex % COMMUNITY_COLORS.length];
+};
+// ============================================================================
+// MAIN PROCESSOR
+// ============================================================================
+/**
+ * Detect communities in the knowledge graph using Leiden algorithm
+ *
+ * This runs AFTER all relationships (CALLS, IMPORTS, etc.) have been built.
+ * It uses primarily CALLS edges to cluster code that works together.
+ */
+export const processCommunities = async (knowledgeGraph, onProgress) => {
+    onProgress?.('Building graph for community detection...', 0);
+    // Pre-check total symbol count to determine large-graph mode before building
+    let symbolCount = 0;
+    knowledgeGraph.forEachNode(node => {
+        if (node.label === 'Function' || node.label === 'Class' || node.label === 'Method' || node.label === 'Interface') {
+            symbolCount++;
+        }
+    });
+    const isLarge = symbolCount > 10_000;
+    const graph = buildGraphologyGraph(knowledgeGraph, isLarge);
+    if (graph.order === 0) {
+        return {
+            communities: [],
+            memberships: [],
+            stats: { totalCommunities: 0, modularity: 0, nodesProcessed: 0 }
+        };
+    }
+    const nodeCount = graph.order;
+    const edgeCount = graph.size;
+    onProgress?.(`Running Leiden on ${nodeCount} nodes, ${edgeCount} edges${isLarge ? ` (filtered from ${symbolCount} symbols)` : ''}...`, 30);
+    // Large graphs: higher resolution + capped iterations (matching Python leidenalg default of 2).
+    // The first 2 iterations capture ~95%+ of modularity; additional iterations have diminishing returns.
+    // Timeout: abort after 60s for pathological graph structures.
+    const LEIDEN_TIMEOUT_MS = 60_000;
+    let details;
+    try {
+        details = await Promise.race([
+            Promise.resolve(leiden.detailed(graph, {
+                resolution: isLarge ? 2.0 : 1.0,
+                maxIterations: isLarge ? 3 : 0,
+            })),
+            new Promise((_, reject) => setTimeout(() => reject(new Error('Leiden timeout')), LEIDEN_TIMEOUT_MS)),
+        ]);
+    }
+    catch (e) {
+        if (e.message === 'Leiden timeout') {
+            onProgress?.('Community detection timed out, using fallback...', 60);
+            // Fallback: assign all nodes to community 0
+            const communities = {};
+            graph.forEachNode((node) => { communities[node] = 0; });
+            details = { communities, count: 1, modularity: 0 };
+        }
+        else {
+            throw e;
+        }
+    }
+    onProgress?.(`Found ${details.count} communities...`, 60);
+    // Step 3: Create community nodes with heuristic labels
+    const communityNodes = createCommunityNodes(details.communities, details.count, graph, knowledgeGraph);
+    onProgress?.('Creating membership edges...', 80);
+    // Step 4: Create membership mappings
+    const memberships = [];
+    Object.entries(details.communities).forEach(([nodeId, communityNum]) => {
+        memberships.push({
+            nodeId,
+            communityId: `comm_${communityNum}`,
+        });
+    });
+    onProgress?.('Community detection complete!', 100);
+    return {
+        communities: communityNodes,
+        memberships,
+        stats: {
+            totalCommunities: details.count,
+            modularity: details.modularity,
+            nodesProcessed: graph.order,
+        }
+    };
+};
+// ============================================================================
+// HELPER: Build graphology graph from knowledge graph
+// ============================================================================
+/**
+ * Build a graphology graph containing only symbol nodes and clustering edges.
+ * For large graphs (>10K symbols), filter out low-confidence fuzzy-global edges
+ * and degree-1 nodes that add noise and massively increase Leiden runtime.
+ */
+const MIN_CONFIDENCE_LARGE = 0.5;
+const buildGraphologyGraph = (knowledgeGraph, isLarge) => {
+    const graph = new Graph({ type: 'undirected', allowSelfLoops: false });
+    const symbolTypes = new Set(['Function', 'Class', 'Method', 'Interface']);
+    const clusteringRelTypes = new Set(['CALLS', 'EXTENDS', 'IMPLEMENTS']);
+    const connectedNodes = new Set();
+    const nodeDegree = new Map();
+    knowledgeGraph.forEachRelationship(rel => {
+        if (!clusteringRelTypes.has(rel.type) || rel.sourceId === rel.targetId)
+            return;
+        if (isLarge && rel.confidence < MIN_CONFIDENCE_LARGE)
+            return;
+        connectedNodes.add(rel.sourceId);
+        connectedNodes.add(rel.targetId);
+        nodeDegree.set(rel.sourceId, (nodeDegree.get(rel.sourceId) || 0) + 1);
+        nodeDegree.set(rel.targetId, (nodeDegree.get(rel.targetId) || 0) + 1);
+    });
+    knowledgeGraph.forEachNode(node => {
+        if (!symbolTypes.has(node.label) || !connectedNodes.has(node.id))
+            return;
+        // For large graphs, skip degree-1 nodes — they just become singletons or
+        // get absorbed into their single neighbor's community, but cost iteration time.
+        if (isLarge && (nodeDegree.get(node.id) || 0) < 2)
+            return;
+        graph.addNode(node.id, {
+            name: node.properties.name,
+            filePath: node.properties.filePath,
+            type: node.label,
+        });
+    });
+    knowledgeGraph.forEachRelationship(rel => {
+        if (!clusteringRelTypes.has(rel.type))
+            return;
+        if (isLarge && rel.confidence < MIN_CONFIDENCE_LARGE)
+            return;
+        if (graph.hasNode(rel.sourceId) && graph.hasNode(rel.targetId) && rel.sourceId !== rel.targetId) {
+            if (!graph.hasEdge(rel.sourceId, rel.targetId)) {
+                graph.addEdge(rel.sourceId, rel.targetId);
+            }
+        }
+    });
+    return graph;
+};
+// ============================================================================
+// HELPER: Create community nodes with heuristic labels
+// ============================================================================
+/**
+ * Create Community nodes with auto-generated labels based on member file paths
+ */
+const createCommunityNodes = (communities, communityCount, graph, knowledgeGraph) => {
+    // Group node IDs by community
+    const communityMembers = new Map();
+    Object.entries(communities).forEach(([nodeId, commNum]) => {
+        if (!communityMembers.has(commNum)) {
+            communityMembers.set(commNum, []);
+        }
+        communityMembers.get(commNum).push(nodeId);
+    });
+    // Build node lookup for file paths
+    const nodePathMap = new Map();
+    for (const node of knowledgeGraph.iterNodes()) {
+        if (node.properties.filePath) {
+            nodePathMap.set(node.id, node.properties.filePath);
+        }
+    }
+    // Create community nodes - SKIP SINGLETONS (isolated nodes)
+    const communityNodes = [];
+    communityMembers.forEach((memberIds, commNum) => {
+        // Skip singleton communities - they're just isolated nodes
+        if (memberIds.length < 2)
+            return;
+        const heuristicLabel = generateHeuristicLabel(memberIds, nodePathMap, graph, commNum);
+        communityNodes.push({
+            id: `comm_${commNum}`,
+            label: heuristicLabel,
+            heuristicLabel,
+            cohesion: calculateCohesion(memberIds, graph),
+            symbolCount: memberIds.length,
+        });
+    });
+    // Sort by size descending
+    communityNodes.sort((a, b) => b.symbolCount - a.symbolCount);
+    return communityNodes;
+};
+// ============================================================================
+// HELPER: Generate heuristic label from folder patterns
+// ============================================================================
+/**
+ * Generate a human-readable label from the most common folder name in the community
+ */
+const generateHeuristicLabel = (memberIds, nodePathMap, graph, commNum) => {
+    // Collect folder names from file paths
+    const folderCounts = new Map();
+    memberIds.forEach(nodeId => {
+        const filePath = nodePathMap.get(nodeId) || '';
+        const parts = filePath.split('/').filter(Boolean);
+        // Get the most specific folder (parent directory)
+        if (parts.length >= 2) {
+            const folder = parts[parts.length - 2];
+            // Skip generic folder names
+            if (!['src', 'lib', 'core', 'utils', 'common', 'shared', 'helpers'].includes(folder.toLowerCase())) {
+                folderCounts.set(folder, (folderCounts.get(folder) || 0) + 1);
+            }
+        }
+    });
+    // Find most common folder
+    let maxCount = 0;
+    let bestFolder = '';
+    folderCounts.forEach((count, folder) => {
+        if (count > maxCount) {
+            maxCount = count;
+            bestFolder = folder;
+        }
+    });
+    if (bestFolder) {
+        // Capitalize first letter
+        return bestFolder.charAt(0).toUpperCase() + bestFolder.slice(1);
+    }
+    // Fallback: use function names to detect patterns
+    const names = [];
+    memberIds.forEach(nodeId => {
+        const name = graph.getNodeAttribute(nodeId, 'name');
+        if (name)
+            names.push(name);
+    });
+    // Look for common prefixes
+    if (names.length > 2) {
+        const commonPrefix = findCommonPrefix(names);
+        if (commonPrefix.length > 2) {
+            return commonPrefix.charAt(0).toUpperCase() + commonPrefix.slice(1);
+        }
+    }
+    // Last resort: generic name with community ID for uniqueness
+    return `Cluster_${commNum}`;
+};
+/**
+ * Find common prefix among strings
+ */
+const findCommonPrefix = (strings) => {
+    if (strings.length === 0)
+        return '';
+    const sorted = strings.slice().sort();
+    const first = sorted[0];
+    const last = sorted[sorted.length - 1];
+    let i = 0;
+    while (i < first.length && first[i] === last[i]) {
+        i++;
+    }
+    return first.substring(0, i);
+};
+// ============================================================================
+// HELPER: Calculate community cohesion
+// ============================================================================
+/**
+ * Estimate cohesion score (0-1) based on internal edge density.
+ * Uses sampling for large communities to avoid O(N^2) cost.
+ */
+const calculateCohesion = (memberIds, graph) => {
+    if (memberIds.length <= 1)
+        return 1.0;
+    const memberSet = new Set(memberIds);
+    // Sample up to 50 members for large communities
+    const SAMPLE_SIZE = 50;
+    const sample = memberIds.length <= SAMPLE_SIZE
+        ? memberIds
+        : memberIds.slice(0, SAMPLE_SIZE);
+    let internalEdges = 0;
+    let totalEdges = 0;
+    for (const nodeId of sample) {
+        if (!graph.hasNode(nodeId))
+            continue;
+        graph.forEachNeighbor(nodeId, (neighbor) => {
+            totalEdges++;
+            if (memberSet.has(neighbor)) {
+                internalEdges++;
+            }
+        });
+    }
+    // Cohesion = fraction of edges that stay internal
+    if (totalEdges === 0)
+        return 1.0;
+    return Math.min(1.0, internalEdges / totalEdges);
+};
+//# sourceMappingURL=community-processor.js.map

package/dist/src/tools/code-analyzer/core/ingestion/community-processor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"community-processor.js","sourceRoot":"","sources":["../../../../../../src/tools/code-analyzer/core/ingestion/community-processor.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AAEH,uEAAuE;AACvE,kEAAkE;AAClE,2EAA2E;AAC3E,OAAO,KAAK,MAAM,YAAY,CAAC;AAC/B,OAAO,EAAE,aAAa,EAAE,MAAM,aAAa,CAAC;AAC5C,OAAO,EAAE,aAAa,EAAE,MAAM,UAAU,CAAC;AACzC,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,MAAM,WAAW,CAAC;AAC7C,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAC;AAGrC,MAAM,UAAU,GAAG,aAAa,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;AAClD,MAAM,SAAS,GAAG,OAAO,CAAC,UAAU,CAAC,CAAC;AACtC,wEAAwE;AACxE,+DAA+D;AAC/D,MAAM,cAAc,GAAG,GAAW,EAAE;IAClC,IAAI,GAAG,GAAG,SAAS,CAAC;IACpB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,EAAE,EAAE,CAAC,EAAE,EAAE,CAAC;QAC5B,MAAM,SAAS,GAAG,OAAO,CAAC,GAAG,EAAE,QAAQ,EAAE,QAAQ,EAAE,WAAW,CAAC,CAAC;QAChE,IAAI,UAAU,CAAC,SAAS,CAAC;YAAE,OAAO,SAAS,CAAC;QAC5C,MAAM,MAAM,GAAG,OAAO,CAAC,GAAG,EAAE,IAAI,CAAC,CAAC;QAClC,IAAI,MAAM,KAAK,GAAG;YAAE,MAAM;QAC1B,GAAG,GAAG,MAAM,CAAC;IACf,CAAC;IACD,MAAM,IAAI,KAAK,CAAC,wCAAwC,CAAC,CAAC;AAC5D,CAAC,CAAC;AACF,MAAM,UAAU,GAAG,cAAc,EAAE,CAAC;AACpC,MAAM,QAAQ,GAAG,aAAa,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;AAChD,MAAM,MAAM,GAAG,QAAQ,CAAC,UAAU,CAAC,CAAC;AA6BpC,+EAA+E;AAC/E,uCAAuC;AACvC,+EAA+E;AAE/E,MAAM,CAAC,MAAM,gBAAgB,GAAG;IAC9B,SAAS,EAAE,MAAM;IACjB,SAAS,EAAE,SAAS;IACpB,SAAS,EAAE,SAAS;IACpB,SAAS,EAAE,QAAQ;IACnB,SAAS,EAAE,OAAO;IAClB,SAAS,EAAE,OAAO;IAClB,SAAS,EAAE,SAAS;IACpB,SAAS,EAAE,UAAU;IACrB,SAAS,EAAE,OAAO;IAClB,SAAS,EAAE,OAAO;IAClB,SAAS,EAAE,OAAO;IAClB,SAAS,EAAE,OAAO;CACnB,CAAC;AAEF,MAAM,CAAC,MAAM,iBAAiB,GAAG,CAAC,cAAsB,EAAU,EAAE;IAClE,OAAO,gBAAgB,CAAC,cAAc,GAAG,gBAAgB,CAAC,MAAM,CAAC,CAAC;AACpE,CAAC,CAAC;AAEF,+EAA+E;AAC/E,iBAAiB;AACjB,+EAA+E;AAE/E;;;;;GAKG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAG,KAAK,EACrC,cAA8B,EAC9B,UAAwD,EACrB,EAAE;IACrC,UAAU,EAAE,CAAC,2CAA2C,EAAE,CAAC,CAAC,CAAC;IAE7D,6EAA6E;IAC7E,IAAI,WAAW,GAAG,CAAC,CAAC;IACpB,cAAc,CAAC,WAAW,CAAC,IAAI,CAAC,EAAE;QAChC,IAAI,IAAI,CAAC,KAAK,KAAK,UAAU,IAAI,IAAI,CAAC,KAAK,KAAK,OAAO,IAAI,IAAI,CAAC,KAAK,KAAK,QAAQ,IAAI,IAAI,CAAC,KAAK,KAAK,WAAW,EAAE,CAAC;YACjH,WAAW,EAAE,CAAC;QAChB,CAAC;IACH,CAAC,CAAC,CAAC;IACH,MAAM,OAAO,GAAG,WAAW,GAAG,MAAM,CAAC;IAErC,MAAM,KAAK,GAAG,oBAAoB,CAAC,cAAc,EAAE,OAAO,CAAC,CAAC;IAE5D,IAAI,KAAK,CAAC,KAAK,KAAK,CAAC,EAAE,CAAC;QACtB,OAAO;YACL,WAAW,EAAE,EAAE;YACf,WAAW,EAAE,EAAE;YACf,KAAK,EAAE,EAAE,gBAAgB,EAAE,CAAC,EAAE,UAAU,EAAE,CAAC,EAAE,cAAc,EAAE,CAAC,EAAE;SACjE,CAAC;IACJ,CAAC;IAED,MAAM,SAAS,GAAG,KAAK,CAAC,KAAK,CAAC;IAC9B,MAAM,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC;IAE7B,UAAU,EAAE,CAAC,qBAAqB,SAAS,WAAW,SAAS,SAAS,OAAO,CAAC,CAAC,CAAC,mBAAmB,WAAW,WAAW,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,EAAE,CAAC,CAAC;IAE3I,gGAAgG;IAChG,sGAAsG;IACtG,8DAA8D;IAC9D,MAAM,iBAAiB,GAAG,MAAM,CAAC;IACjC,IAAI,OAAY,CAAC;IACjB,IAAI,CAAC;QACH,OAAO,GAAG,MAAM,OAAO,CAAC,IAAI,CAAC;YAC3B,OAAO,CAAC,OAAO,CAAE,MAAc,CAAC,QAAQ,CAAC,KAAK,EAAE;gBAC9C,UAAU,EAAE,OAAO,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG;gBAC/B,aAAa,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;aAC/B,CAAC,CAAC;YACH,IAAI,OAAO,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CACxB,UAAU,CAAC,GAAG,EAAE,CAAC,MAAM,CAAC,IAAI,KAAK,CAAC,gBAAgB,CAAC,CAAC,EAAE,iBAAiB,CAAC,CACzE;SACF,CAAC,CAAC;IACL,CAAC;IAAC,OAAO,CAAM,EAAE,CAAC;QAChB,IAAI,CAAC,CAAC,OAAO,KAAK,gBAAgB,EAAE,CAAC;YACnC,UAAU,EAAE,CAAC,kDAAkD,EAAE,EAAE,CAAC,CAAC;YACrE,4CAA4C;YAC5C,MAAM,WAAW,GAA2B,EAAE,CAAC;YAC/C,KAAK,CAAC,WAAW,CAAC,CAAC,IAAY,EAAE,EAAE,GAAG,WAAW,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAChE,OAAO,GAAG,EAAE,WAAW,EAAE,KAAK,EAAE,CAAC,EAAE,UAAU,EAAE,CAAC,EAAE,CAAC;QACrD,CAAC;aAAM,CAAC;YACN,MAAM,CAAC,CAAC;QACV,CAAC;IACH,CAAC;IAED,UAAU,EAAE,CAAC,SAAS,OAAO,CAAC,KAAK,iBAAiB,EAAE,EAAE,CAAC,CAAC;IAE1D,uDAAuD;IACvD,MAAM,cAAc,GAAG,oBAAoB,CACzC,OAAO,CAAC,WAAqC,EAC7C,OAAO,CAAC,KAAK,EACb,KAAK,EACL,cAAc,CACf,CAAC;IAEF,UAAU,EAAE,CAAC,8BAA8B,EAAE,EAAE,CAAC,CAAC;IAEjD,qCAAqC;IACrC,MAAM,WAAW,GAA0B,EAAE,CAAC;IAC9C,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,WAAW,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,EAAE,YAAY,CAAC,EAAE,EAAE;QACrE,WAAW,CAAC,IAAI,CAAC;YACf,MAAM;YACN,WAAW,EAAE,QAAQ,YAAY,EAAE;SACpC,CAAC,CAAC;IACL,CAAC,CAAC,CAAC;IAEH,UAAU,EAAE,CAAC,+BAA+B,EAAE,GAAG,CAAC,CAAC;IAEnD,OAAO;QACL,WAAW,EAAE,cAAc;QAC3B,WAAW;QACX,KAAK,EAAE;YACL,gBAAgB,EAAE,OAAO,CAAC,KAAK;YAC/B,UAAU,EAAE,OAAO,CAAC,UAAU;YAC9B,cAAc,EAAE,KAAK,CAAC,KAAK;SAC5B;KACF,CAAC;AACJ,CAAC,CAAC;AAEF,+EAA+E;AAC/E,sDAAsD;AACtD,+EAA+E;AAE/E;;;;GAIG;AACH,MAAM,oBAAoB,GAAG,GAAG,CAAC;AAEjC,MAAM,oBAAoB,GAAG,CAAC,cAA8B,EAAE,OAAgB,EAAO,EAAE;IACrF,MAAM,KAAK,GAAG,IAAK,KAAa,CAAC,EAAE,IAAI,EAAE,YAAY,EAAE,cAAc,EAAE,KAAK,EAAE,CAAC,CAAC;IAEhF,MAAM,WAAW,GAAG,IAAI,GAAG,CAAY,CAAC,UAAU,EAAE,OAAO,EAAE,QAAQ,EAAE,WAAW,CAAC,CAAC,CAAC;IACrF,MAAM,kBAAkB,GAAG,IAAI,GAAG,CAAC,CAAC,OAAO,EAAE,SAAS,EAAE,YAAY,CAAC,CAAC,CAAC;IACvE,MAAM,cAAc,GAAG,IAAI,GAAG,EAAU,CAAC;IACzC,MAAM,UAAU,GAAG,IAAI,GAAG,EAAkB,CAAC;IAE7C,cAAc,CAAC,mBAAmB,CAAC,GAAG,CAAC,EAAE;QACvC,IAAI,CAAC,kBAAkB,CAAC,GAAG,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,GAAG,CAAC,QAAQ,KAAK,GAAG,CAAC,QAAQ;YAAE,OAAO;QAC/E,IAAI,OAAO,IAAI,GAAG,CAAC,UAAU,GAAG,oBAAoB;YAAE,OAAO;QAE7D,cAAc,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;QACjC,cAAc,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;QACjC,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,EAAE,CAAC,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;QACtE,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,EAAE,CAAC,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;IACxE,CAAC,CAAC,CAAC;IAEH,cAAc,CAAC,WAAW,CAAC,IAAI,CAAC,EAAE;QAChC,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,cAAc,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC;YAAE,OAAO;QACzE,yEAAyE;QACzE,gFAAgF;QAChF,IAAI,OAAO,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC;YAAE,OAAO;QAE1D,KAAK,CAAC,OAAO,CAAC,IAAI,CAAC,EAAE,EAAE;YACrB,IAAI,EAAE,IAAI,CAAC,UAAU,CAAC,IAAI;YAC1B,QAAQ,EAAE,IAAI,CAAC,UAAU,CAAC,QAAQ;YAClC,IAAI,EAAE,IAAI,CAAC,KAAK;SACjB,CAAC,CAAC;IACL,CAAC,CAAC,CAAC;IAEH,cAAc,CAAC,mBAAmB,CAAC,GAAG,CAAC,EAAE;QACvC,IAAI,CAAC,kBAAkB,CAAC,GAAG,CAAC,GAAG,CAAC,IAAI,CAAC;YAAE,OAAO;QAC9C,IAAI,OAAO,IAAI,GAAG,CAAC,UAAU,GAAG,oBAAoB;YAAE,OAAO;QAC7D,IAAI,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,GAAG,CAAC,QAAQ,KAAK,GAAG,CAAC,QAAQ,EAAE,CAAC;YAChG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,QAAQ,EAAE,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;gBAC/C,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,QAAQ,EAAE,GAAG,CAAC,QAAQ,CAAC,CAAC;YAC5C,CAAC;QACH,CAAC;IACH,CAAC,CAAC,CAAC;IAEH,OAAO,KAAK,CAAC;AACf,CAAC,CAAC;AAEF,+EAA+E;AAC/E,uDAAuD;AACvD,+EAA+E;AAE/E;;GAEG;AACH,MAAM,oBAAoB,GAAG,CAC3B,WAAmC,EACnC,cAAsB,EACtB,KAAU,EACV,cAA8B,EACb,EAAE;IACnB,8BAA8B;IAC9B,MAAM,gBAAgB,GAAG,IAAI,GAAG,EAAoB,CAAC;IAErD,MAAM,CAAC,OAAO,CAAC,WAAW,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,EAAE,OAAO,CAAC,EAAE,EAAE;QACxD,IAAI,CAAC,gBAAgB,CAAC,GAAG,CAAC,OAAO,CAAC,EAAE,CAAC;YACnC,gBAAgB,CAAC,GAAG,CAAC,OAAO,EAAE,EAAE,CAAC,CAAC;QACpC,CAAC;QACD,gBAAgB,CAAC,GAAG,CAAC,OAAO,CAAE,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;IAC9C,CAAC,CAAC,CAAC;IAEH,mCAAmC;IACnC,MAAM,WAAW,GAAG,IAAI,GAAG,EAAkB,CAAC;IAC9C,KAAK,MAAM,IAAI,IAAI,cAAc,CAAC,SAAS,EAAE,EAAE,CAAC;QAC9C,IAAI,IAAI,CAAC,UAAU,CAAC,QAAQ,EAAE,CAAC;YAC7B,WAAW,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,EAAE,IAAI,CAAC,UAAU,CAAC,QAAQ,CAAC,CAAC;QACrD,CAAC;IACH,CAAC;IAED,4DAA4D;IAC5D,MAAM,cAAc,GAAoB,EAAE,CAAC;IAE3C,gBAAgB,CAAC,OAAO,CAAC,CAAC,SAAS,EAAE,OAAO,EAAE,EAAE;QAC9C,2DAA2D;QAC3D,IAAI,SAAS,CAAC,MAAM,GAAG,CAAC;YAAE,OAAO;QAEjC,MAAM,cAAc,GAAG,sBAAsB,CAAC,SAAS,EAAE,WAAW,EAAE,KAAK,EAAE,OAAO,CAAC,CAAC;QAEtF,cAAc,CAAC,IAAI,CAAC;YAClB,EAAE,EAAE,QAAQ,OAAO,EAAE;YACrB,KAAK,EAAE,cAAc;YACrB,cAAc;YACd,QAAQ,EAAE,iBAAiB,CAAC,SAAS,EAAE,KAAK,CAAC;YAC7C,WAAW,EAAE,SAAS,CAAC,MAAM;SAC9B,CAAC,CAAC;IACL,CAAC,CAAC,CAAC;IAEH,0BAA0B;IAC1B,cAAc,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,WAAW,GAAG,CAAC,CAAC,WAAW,CAAC,CAAC;IAE7D,OAAO,cAAc,CAAC;AACxB,CAAC,CAAC;AAEF,+EAA+E;AAC/E,wDAAwD;AACxD,+EAA+E;AAE/E;;GAEG;AACH,MAAM,sBAAsB,GAAG,CAC7B,SAAmB,EACnB,WAAgC,EAChC,KAAU,EACV,OAAe,EACP,EAAE;IACV,uCAAuC;IACvC,MAAM,YAAY,GAAG,IAAI,GAAG,EAAkB,CAAC;IAE/C,SAAS,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE;QACzB,MAAM,QAAQ,GAAG,WAAW,CAAC,GAAG,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC;QAC/C,MAAM,KAAK,GAAG,QAAQ,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;QAElD,kDAAkD;QAClD,IAAI,KAAK,CAAC,MAAM,IAAI,CAAC,EAAE,CAAC;YACtB,MAAM,MAAM,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;YACvC,4BAA4B;YAC5B,IAAI,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,MAAM,EAAE,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC,QAAQ,CAAC,MAAM,CAAC,WAAW,EAAE,CAAC,EAAE,CAAC;gBACnG,YAAY,CAAC,GAAG,CAAC,MAAM,EAAE,CAAC,YAAY,CAAC,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;YAChE,CAAC;QACH,CAAC;IACH,CAAC,CAAC,CAAC;IAEH,0BAA0B;IAC1B,IAAI,QAAQ,GAAG,CAAC,CAAC;IACjB,IAAI,UAAU,GAAG,EAAE,CAAC;IAEpB,YAAY,CAAC,OAAO,CAAC,CAAC,KAAK,EAAE,MAAM,EAAE,EAAE;QACrC,IAAI,KAAK,GAAG,QAAQ,EAAE,CAAC;YACrB,QAAQ,GAAG,KAAK,CAAC;YACjB,UAAU,GAAG,MAAM,CAAC;QACtB,CAAC;IACH,CAAC,CAAC,CAAC;IAEH,IAAI,UAAU,EAAE,CAAC;QACf,0BAA0B;QAC1B,OAAO,UAAU,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;IAClE,CAAC;IAED,kDAAkD;IAClD,MAAM,KAAK,GAAa,EAAE,CAAC;IAC3B,SAAS,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE;QACzB,MAAM,IAAI,GAAG,KAAK,CAAC,gBAAgB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;QACpD,IAAI,IAAI;YAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAC7B,CAAC,CAAC,CAAC;IAEH,2BAA2B;IAC3B,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACrB,MAAM,YAAY,GAAG,gBAAgB,CAAC,KAAK,CAAC,CAAC;QAC7C,IAAI,YAAY,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YAC5B,OAAO,YAAY,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,GAAG,YAAY,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;QACtE,CAAC;IACH,CAAC;IAED,6DAA6D;IAC7D,OAAO,WAAW,OAAO,EAAE,CAAC;AAC9B,CAAC,CAAC;AAEF;;GAEG;AACH,MAAM,gBAAgB,GAAG,CAAC,OAAiB,EAAU,EAAE;IACrD,IAAI,OAAO,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,EAAE,CAAC;IAEpC,MAAM,MAAM,GAAG,OAAO,CAAC,KAAK,EAAE,CAAC,IAAI,EAAE,CAAC;IACtC,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,CAAC,CAAC;IACxB,MAAM,IAAI,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;IAEvC,IAAI,CAAC,GAAG,CAAC,CAAC;IACV,OAAO,CAAC,GAAG,KAAK,CAAC,MAAM,IAAI,KAAK,CAAC,CAAC,CAAC,KAAK,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC;QAChD,CAAC,EAAE,CAAC;IACN,CAAC;IAED,OAAO,KAAK,CAAC,SAAS,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;AAC/B,CAAC,CAAC;AAEF,+EAA+E;AAC/E,uCAAuC;AACvC,+EAA+E;AAE/E;;;GAGG;AACH,MAAM,iBAAiB,GAAG,CAAC,SAAmB,EAAE,KAAU,EAAU,EAAE;IACpE,IAAI,SAAS,CAAC,MAAM,IAAI,CAAC;QAAE,OAAO,GAAG,CAAC;IAEtC,MAAM,SAAS,GAAG,IAAI,GAAG,CAAC,SAAS,CAAC,CAAC;IAErC,gDAAgD;IAChD,MAAM,WAAW,GAAG,EAAE,CAAC;IACvB,MAAM,MAAM,GAAG,SAAS,CAAC,MAAM,IAAI,WAAW;QAC5C,CAAC,CAAC,SAAS;QACX,CAAC,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC;IAEpC,IAAI,aAAa,GAAG,CAAC,CAAC;IACtB,IAAI,UAAU,GAAG,CAAC,CAAC;IAEnB,KAAK,MAAM,MAAM,IAAI,MAAM,EAAE,CAAC;QAC5B,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC;YAAE,SAAS;QACrC,KAAK,CAAC,eAAe,CAAC,MAAM,EAAE,CAAC,QAAgB,EAAE,EAAE;YACjD,UAAU,EAAE,CAAC;YACb,IAAI,SAAS,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;gBAC5B,aAAa,EAAE,CAAC;YAClB,CAAC;QACH,CAAC,CAAC,CAAC;IACL,CAAC;IAED,kDAAkD;IAClD,IAAI,UAAU,KAAK,CAAC;QAAE,OAAO,GAAG,CAAC;IACjC,OAAO,IAAI,CAAC,GAAG,CAAC,GAAG,EAAE,aAAa,GAAG,UAAU,CAAC,CAAC;AACnD,CAAC,CAAC"}

package/dist/src/tools/code-analyzer/core/ingestion/entry-point-scoring.d.ts

@@ -0,0 +1,40 @@
+/**
+ * Entry Point Scoring
+ *
+ * Calculates entry point scores for process detection based on:
+ * 1. Call ratio (existing algorithm - callees / (callers + 1))
+ * 2. Export status (exported functions get higher priority)
+ * 3. Name patterns (functions matching entry point patterns like handle*, on*, *Controller)
+ * 4. Framework detection (path-based detection for Next.js, Express, Django, etc.)
+ *
+ * This module is language-agnostic - language-specific patterns are defined per language.
+ */
+export interface EntryPointScoreResult {
+    score: number;
+    reasons: string[];
+}
+/**
+ * Calculate an entry point score for a function/method
+ *
+ * Higher scores indicate better entry point candidates.
+ * Score = baseScore × exportMultiplier × nameMultiplier
+ *
+ * @param name - Function/method name
+ * @param language - Programming language
+ * @param isExported - Whether the function is exported/public
+ * @param callerCount - Number of functions that call this function
+ * @param calleeCount - Number of functions this function calls
+ * @returns Score and array of reasons explaining the score
+ */
+export declare function calculateEntryPointScore(name: string, language: string, isExported: boolean, callerCount: number, calleeCount: number, filePath?: string): EntryPointScoreResult;
+/**
+ * Check if a file path is a test file (should be excluded from entry points)
+ * Covers common test file patterns across all supported languages
+ */
+export declare function isTestFile(filePath: string): boolean;
+/**
+ * Check if a file path is likely a utility/helper file
+ * These might still have entry points but should be lower priority
+ */
+export declare function isUtilityFile(filePath: string): boolean;
+//# sourceMappingURL=entry-point-scoring.d.ts.map

package/dist/src/tools/code-analyzer/core/ingestion/entry-point-scoring.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"entry-point-scoring.d.ts","sourceRoot":"","sources":["../../../../../../src/tools/code-analyzer/core/ingestion/entry-point-scoring.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AA8HH,MAAM,WAAW,qBAAqB;IACpC,KAAK,EAAE,MAAM,CAAC;IACd,OAAO,EAAE,MAAM,EAAE,CAAC;CACnB;AAMD;;;;;;;;;;;;GAYG;AACH,wBAAgB,wBAAwB,CACtC,IAAI,EAAE,MAAM,EACZ,QAAQ,EAAE,MAAM,EAChB,UAAU,EAAE,OAAO,EACnB,WAAW,EAAE,MAAM,EACnB,WAAW,EAAE,MAAM,EACnB,QAAQ,GAAE,MAAW,GACpB,qBAAqB,CAuDvB;AAMD;;;GAGG;AACH,wBAAgB,UAAU,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CA0BpD;AAED;;;GAGG;AACH,wBAAgB,aAAa,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAkBvD"}

package/dist/src/tools/code-analyzer/core/ingestion/entry-point-scoring.js

@@ -0,0 +1,236 @@
+/**
+ * Entry Point Scoring
+ *
+ * Calculates entry point scores for process detection based on:
+ * 1. Call ratio (existing algorithm - callees / (callers + 1))
+ * 2. Export status (exported functions get higher priority)
+ * 3. Name patterns (functions matching entry point patterns like handle*, on*, *Controller)
+ * 4. Framework detection (path-based detection for Next.js, Express, Django, etc.)
+ *
+ * This module is language-agnostic - language-specific patterns are defined per language.
+ */
+import { detectFrameworkFromPath } from './framework-detection.js';
+// ============================================================================
+// NAME PATTERNS - All 9 supported languages
+// ============================================================================
+/**
+ * Common entry point naming patterns by language
+ * These patterns indicate functions that are likely feature entry points
+ */
+const ENTRY_POINT_PATTERNS = {
+    // Universal patterns (apply to all languages)
+    '*': [
+        /^(main|init|bootstrap|start|run|setup|configure)$/i,
+        /^handle[A-Z]/, // handleLogin, handleSubmit
+        /^on[A-Z]/, // onClick, onSubmit
+        /Handler$/, // RequestHandler
+        /Controller$/, // UserController
+        /^process[A-Z]/, // processPayment
+        /^execute[A-Z]/, // executeQuery
+        /^perform[A-Z]/, // performAction
+        /^dispatch[A-Z]/, // dispatchEvent
+        /^trigger[A-Z]/, // triggerAction
+        /^fire[A-Z]/, // fireEvent
+        /^emit[A-Z]/, // emitEvent
+    ],
+    // JavaScript/TypeScript
+    'javascript': [
+        /^use[A-Z]/, // React hooks (useEffect, etc.)
+    ],
+    'typescript': [
+        /^use[A-Z]/, // React hooks
+    ],
+    // Python
+    'python': [
+        /^app$/, // Flask/FastAPI app
+        /^(get|post|put|delete|patch)_/i, // REST conventions
+        /^api_/, // API functions
+        /^view_/, // Django views
+    ],
+    // Java
+    'java': [
+        /^do[A-Z]/, // doGet, doPost (Servlets)
+        /^create[A-Z]/, // Factory patterns
+        /^build[A-Z]/, // Builder patterns
+        /Service$/, // UserService
+    ],
+    // C#
+    'csharp': [
+        /^(Get|Post|Put|Delete)/, // ASP.NET conventions
+        /Action$/, // MVC actions
+        /^On[A-Z]/, // Event handlers
+        /Async$/, // Async entry points
+    ],
+    // Go
+    'go': [
+        /Handler$/, // http.Handler pattern
+        /^Serve/, // ServeHTTP
+        /^New[A-Z]/, // Constructor pattern (returns new instance)
+        /^Make[A-Z]/, // Make functions
+    ],
+    // Rust
+    'rust': [
+        /^(get|post|put|delete)_handler$/i,
+        /^handle_/, // handle_request
+        /^new$/, // Constructor pattern
+        /^run$/, // run entry point
+        /^spawn/, // Async spawn
+    ],
+    // C - explicit main() boost (critical for C programs)
+    'c': [
+        /^main$/, // THE entry point
+        /^init_/, // Initialization functions
+        /^start_/, // Start functions
+        /^run_/, // Run functions
+    ],
+    // C++ - same as C plus class patterns
+    'cpp': [
+        /^main$/, // THE entry point
+        /^init_/,
+        /^Create[A-Z]/, // Factory patterns
+        /^Run$/, // Run methods
+        /^Start$/, // Start methods
+    ],
+};
+// ============================================================================
+// UTILITY PATTERNS - Functions that should be penalized
+// ============================================================================
+/**
+ * Patterns that indicate utility/helper functions (NOT entry points)
+ * These get penalized in scoring
+ */
+const UTILITY_PATTERNS = [
+    /^(get|set|is|has|can|should|will|did)[A-Z]/, // Accessors/predicates
+    /^_/, // Private by convention
+    /^(format|parse|validate|convert|transform)/i, // Transformation utilities
+    /^(log|debug|error|warn|info)$/i, // Logging
+    /^(to|from)[A-Z]/, // Conversions
+    /^(encode|decode)/i, // Encoding utilities
+    /^(serialize|deserialize)/i, // Serialization
+    /^(clone|copy|deep)/i, // Cloning utilities
+    /^(merge|extend|assign)/i, // Object utilities
+    /^(filter|map|reduce|sort|find)/i, // Collection utilities (standalone)
+    /Helper$/,
+    /Util$/,
+    /Utils$/,
+    /^utils?$/i,
+    /^helpers?$/i,
+];
+// ============================================================================
+// MAIN SCORING FUNCTION
+// ============================================================================
+/**
+ * Calculate an entry point score for a function/method
+ *
+ * Higher scores indicate better entry point candidates.
+ * Score = baseScore × exportMultiplier × nameMultiplier
+ *
+ * @param name - Function/method name
+ * @param language - Programming language
+ * @param isExported - Whether the function is exported/public
+ * @param callerCount - Number of functions that call this function
+ * @param calleeCount - Number of functions this function calls
+ * @returns Score and array of reasons explaining the score
+ */
+export function calculateEntryPointScore(name, language, isExported, callerCount, calleeCount, filePath = '' // Optional for backwards compatibility
+) {
+    const reasons = [];
+    // Must have outgoing calls to be an entry point (we need to trace forward)
+    if (calleeCount === 0) {
+        return { score: 0, reasons: ['no-outgoing-calls'] };
+    }
+    // Base score: call ratio (existing algorithm)
+    // High ratio = calls many, called by few = likely entry point
+    const baseScore = calleeCount / (callerCount + 1);
+    reasons.push(`base:${baseScore.toFixed(2)}`);
+    // Export bonus: exported/public functions are more likely entry points
+    const exportMultiplier = isExported ? 2.0 : 1.0;
+    if (isExported) {
+        reasons.push('exported');
+    }
+    // Name pattern scoring
+    let nameMultiplier = 1.0;
+    // Check negative patterns first (utilities get penalized)
+    if (UTILITY_PATTERNS.some(p => p.test(name))) {
+        nameMultiplier = 0.3; // Significant penalty
+        reasons.push('utility-pattern');
+    }
+    else {
+        // Check positive patterns
+        const universalPatterns = ENTRY_POINT_PATTERNS['*'] || [];
+        const langPatterns = ENTRY_POINT_PATTERNS[language] || [];
+        const allPatterns = [...universalPatterns, ...langPatterns];
+        if (allPatterns.some(p => p.test(name))) {
+            nameMultiplier = 1.5; // Bonus for matching entry point pattern
+            reasons.push('entry-pattern');
+        }
+    }
+    // Framework detection bonus (Phase 2)
+    let frameworkMultiplier = 1.0;
+    if (filePath) {
+        const frameworkHint = detectFrameworkFromPath(filePath);
+        if (frameworkHint) {
+            frameworkMultiplier = frameworkHint.entryPointMultiplier;
+            reasons.push(`framework:${frameworkHint.reason}`);
+        }
+    }
+    // Calculate final score
+    const finalScore = baseScore * exportMultiplier * nameMultiplier * frameworkMultiplier;
+    return {
+        score: finalScore,
+        reasons,
+    };
+}
+// ============================================================================
+// HELPER FUNCTIONS
+// ============================================================================
+/**
+ * Check if a file path is a test file (should be excluded from entry points)
+ * Covers common test file patterns across all supported languages
+ */
+export function isTestFile(filePath) {
+    const p = filePath.toLowerCase().replace(/\\/g, '/');
+    return (
+    // JavaScript/TypeScript test patterns
+    p.includes('.test.') ||
+        p.includes('.spec.') ||
+        p.includes('__tests__/') ||
+        p.includes('__mocks__/') ||
+        // Generic test folders
+        p.includes('/test/') ||
+        p.includes('/tests/') ||
+        p.includes('/testing/') ||
+        // Python test patterns
+        p.endsWith('_test.py') ||
+        p.includes('/test_') ||
+        // Go test patterns
+        p.endsWith('_test.go') ||
+        // Java test patterns
+        p.includes('/src/test/') ||
+        // Rust test patterns (inline tests are different, but test files)
+        p.includes('/tests/') ||
+        // C# test patterns
+        p.includes('.tests/') ||
+        p.includes('tests.cs'));
+}
+/**
+ * Check if a file path is likely a utility/helper file
+ * These might still have entry points but should be lower priority
+ */
+export function isUtilityFile(filePath) {
+    const p = filePath.toLowerCase().replace(/\\/g, '/');
+    return (p.includes('/utils/') ||
+        p.includes('/util/') ||
+        p.includes('/helpers/') ||
+        p.includes('/helper/') ||
+        p.includes('/common/') ||
+        p.includes('/shared/') ||
+        p.includes('/lib/') ||
+        p.endsWith('/utils.ts') ||
+        p.endsWith('/utils.js') ||
+        p.endsWith('/helpers.ts') ||
+        p.endsWith('/helpers.js') ||
+        p.endsWith('_utils.py') ||
+        p.endsWith('_helpers.py'));
+}
+//# sourceMappingURL=entry-point-scoring.js.map

package/dist/src/tools/code-analyzer/core/ingestion/entry-point-scoring.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"entry-point-scoring.js","sourceRoot":"","sources":["../../../../../../src/tools/code-analyzer/core/ingestion/entry-point-scoring.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,EAAE,uBAAuB,EAAE,MAAM,0BAA0B,CAAC;AAEnE,+EAA+E;AAC/E,4CAA4C;AAC5C,+EAA+E;AAE/E;;;GAGG;AACH,MAAM,oBAAoB,GAA6B;IACrD,8CAA8C;IAC9C,GAAG,EAAE;QACH,oDAAoD;QACpD,cAAc,EAAY,4BAA4B;QACtD,UAAU,EAAgB,oBAAoB;QAC9C,UAAU,EAAgB,iBAAiB;QAC3C,aAAa,EAAa,iBAAiB;QAC3C,eAAe,EAAW,iBAAiB;QAC3C,eAAe,EAAW,eAAe;QACzC,eAAe,EAAW,gBAAgB;QAC1C,gBAAgB,EAAU,gBAAgB;QAC1C,eAAe,EAAW,gBAAgB;QAC1C,YAAY,EAAc,YAAY;QACtC,YAAY,EAAc,YAAY;KACvC;IAED,wBAAwB;IACxB,YAAY,EAAE;QACZ,WAAW,EAAe,gCAAgC;KAC3D;IACD,YAAY,EAAE;QACZ,WAAW,EAAe,cAAc;KACzC;IAED,SAAS;IACT,QAAQ,EAAE;QACR,OAAO,EAAmB,oBAAoB;QAC9C,gCAAgC,EAAG,mBAAmB;QACtD,OAAO,EAAmB,gBAAgB;QAC1C,QAAQ,EAAkB,eAAe;KAC1C;IAED,OAAO;IACP,MAAM,EAAE;QACN,UAAU,EAAgB,2BAA2B;QACrD,cAAc,EAAY,mBAAmB;QAC7C,aAAa,EAAa,mBAAmB;QAC7C,UAAU,EAAgB,cAAc;KACzC;IAED,KAAK;IACL,QAAQ,EAAE;QACR,wBAAwB,EAAG,sBAAsB;QACjD,SAAS,EAAkB,cAAc;QACzC,UAAU,EAAgB,iBAAiB;QAC3C,QAAQ,EAAkB,qBAAqB;KAChD;IAED,KAAK;IACL,IAAI,EAAE;QACJ,UAAU,EAAgB,uBAAuB;QACjD,QAAQ,EAAkB,YAAY;QACtC,WAAW,EAAe,6CAA6C;QACvE,YAAY,EAAc,iBAAiB;KAC5C;IAED,OAAO;IACP,MAAM,EAAE;QACN,kCAAkC;QAClC,UAAU,EAAgB,iBAAiB;QAC3C,OAAO,EAAmB,sBAAsB;QAChD,OAAO,EAAmB,kBAAkB;QAC5C,QAAQ,EAAkB,cAAc;KACzC;IAED,sDAAsD;IACtD,GAAG,EAAE;QACH,QAAQ,EAAkB,kBAAkB;QAC5C,QAAQ,EAAkB,2BAA2B;QACrD,SAAS,EAAiB,kBAAkB;QAC5C,OAAO,EAAmB,gBAAgB;KAC3C;IAED,sCAAsC;IACtC,KAAK,EAAE;QACL,QAAQ,EAAkB,kBAAkB;QAC5C,QAAQ;QACR,cAAc,EAAY,mBAAmB;QAC7C,OAAO,EAAmB,cAAc;QACxC,SAAS,EAAiB,gBAAgB;KAC3C;CACF,CAAC;AAEF,+EAA+E;AAC/E,wDAAwD;AACxD,+EAA+E;AAE/E;;;GAGG;AACH,MAAM,gBAAgB,GAAa;IACjC,4CAA4C,EAAG,uBAAuB;IACtE,IAAI,EAA6C,wBAAwB;IACzE,6CAA6C,EAAG,2BAA2B;IAC3E,gCAAgC,EAAgB,UAAU;IAC1D,iBAAiB,EAA+B,cAAc;IAC9D,mBAAmB,EAA6B,qBAAqB;IACrE,2BAA2B,EAAqB,gBAAgB;IAChE,qBAAqB,EAA2B,oBAAoB;IACpE,yBAAyB,EAAuB,mBAAmB;IACnE,iCAAiC,EAAc,oCAAoC;IACnF,SAAS;IACT,OAAO;IACP,QAAQ;IACR,WAAW;IACX,aAAa;CACd,CAAC;AAWF,+EAA+E;AAC/E,wBAAwB;AACxB,+EAA+E;AAE/E;;;;;;;;;;;;GAYG;AACH,MAAM,UAAU,wBAAwB,CACtC,IAAY,EACZ,QAAgB,EAChB,UAAmB,EACnB,WAAmB,EACnB,WAAmB,EACnB,WAAmB,EAAE,CAAE,uCAAuC;;IAE9D,MAAM,OAAO,GAAa,EAAE,CAAC;IAE7B,2EAA2E;IAC3E,IAAI,WAAW,KAAK,CAAC,EAAE,CAAC;QACtB,OAAO,EAAE,KAAK,EAAE,CAAC,EAAE,OAAO,EAAE,CAAC,mBAAmB,CAAC,EAAE,CAAC;IACtD,CAAC;IAED,8CAA8C;IAC9C,8DAA8D;IAC9D,MAAM,SAAS,GAAG,WAAW,GAAG,CAAC,WAAW,GAAG,CAAC,CAAC,CAAC;IAClD,OAAO,CAAC,IAAI,CAAC,QAAQ,SAAS,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC;IAE7C,uEAAuE;IACvE,MAAM,gBAAgB,GAAG,UAAU,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC;IAChD,IAAI,UAAU,EAAE,CAAC;QACf,OAAO,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;IAC3B,CAAC;IAED,uBAAuB;IACvB,IAAI,cAAc,GAAG,GAAG,CAAC;IAEzB,0DAA0D;IAC1D,IAAI,gBAAgB,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC;QAC7C,cAAc,GAAG,GAAG,CAAC,CAAE,sBAAsB;QAC7C,OAAO,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAC;IAClC,CAAC;SAAM,CAAC;QACN,0BAA0B;QAC1B,MAAM,iBAAiB,GAAG,oBAAoB,CAAC,GAAG,CAAC,IAAI,EAAE,CAAC;QAC1D,MAAM,YAAY,GAAG,oBAAoB,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC;QAC1D,MAAM,WAAW,GAAG,CAAC,GAAG,iBAAiB,EAAE,GAAG,YAAY,CAAC,CAAC;QAE5D,IAAI,WAAW,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC;YACxC,cAAc,GAAG,GAAG,CAAC,CAAE,yCAAyC;YAChE,OAAO,CAAC,IAAI,CAAC,eAAe,CAAC,CAAC;QAChC,CAAC;IACH,CAAC;IAED,sCAAsC;IACtC,IAAI,mBAAmB,GAAG,GAAG,CAAC;IAC9B,IAAI,QAAQ,EAAE,CAAC;QACb,MAAM,aAAa,GAAG,uBAAuB,CAAC,QAAQ,CAAC,CAAC;QACxD,IAAI,aAAa,EAAE,CAAC;YAClB,mBAAmB,GAAG,aAAa,CAAC,oBAAoB,CAAC;YACzD,OAAO,CAAC,IAAI,CAAC,aAAa,aAAa,CAAC,MAAM,EAAE,CAAC,CAAC;QACpD,CAAC;IACH,CAAC;IAED,wBAAwB;IACxB,MAAM,UAAU,GAAG,SAAS,GAAG,gBAAgB,GAAG,cAAc,GAAG,mBAAmB,CAAC;IAEvF,OAAO;QACL,KAAK,EAAE,UAAU;QACjB,OAAO;KACR,CAAC;AACJ,CAAC;AAED,+EAA+E;AAC/E,mBAAmB;AACnB,+EAA+E;AAE/E;;;GAGG;AACH,MAAM,UAAU,UAAU,CAAC,QAAgB;IACzC,MAAM,CAAC,GAAG,QAAQ,CAAC,WAAW,EAAE,CAAC,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC;IAErD,OAAO;IACL,sCAAsC;IACtC,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC;QACpB,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC;QACpB,CAAC,CAAC,QAAQ,CAAC,YAAY,CAAC;QACxB,CAAC,CAAC,QAAQ,CAAC,YAAY,CAAC;QACxB,uBAAuB;QACvB,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC;QACpB,CAAC,CAAC,QAAQ,CAAC,SAAS,CAAC;QACrB,CAAC,CAAC,QAAQ,CAAC,WAAW,CAAC;QACvB,uBAAuB;QACvB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC;QACtB,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC;QACpB,mBAAmB;QACnB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC;QACtB,qBAAqB;QACrB,CAAC,CAAC,QAAQ,CAAC,YAAY,CAAC;QACxB,kEAAkE;QAClE,CAAC,CAAC,QAAQ,CAAC,SAAS,CAAC;QACrB,mBAAmB;QACnB,CAAC,CAAC,QAAQ,CAAC,SAAS,CAAC;QACrB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC,CACvB,CAAC;AACJ,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,aAAa,CAAC,QAAgB;IAC5C,MAAM,CAAC,GAAG,QAAQ,CAAC,WAAW,EAAE,CAAC,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC;IAErD,OAAO,CACL,CAAC,CAAC,QAAQ,CAAC,SAAS,CAAC;QACrB,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC;QACpB,CAAC,CAAC,QAAQ,CAAC,WAAW,CAAC;QACvB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC;QACtB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC;QACtB,CAAC,CAAC,QAAQ,CAAC,UAAU,CAAC;QACtB,CAAC,CAAC,QAAQ,CAAC,OAAO,CAAC;QACnB,CAAC,CAAC,QAAQ,CAAC,WAAW,CAAC;QACvB,CAAC,CAAC,QAAQ,CAAC,WAAW,CAAC;QACvB,CAAC,CAAC,QAAQ,CAAC,aAAa,CAAC;QACzB,CAAC,CAAC,QAAQ,CAAC,aAAa,CAAC;QACzB,CAAC,CAAC,QAAQ,CAAC,WAAW,CAAC;QACvB,CAAC,CAAC,QAAQ,CAAC,aAAa,CAAC,CAC1B,CAAC;AACJ,CAAC"}

package/dist/src/tools/code-analyzer/core/ingestion/filesystem-walker.d.ts

@@ -0,0 +1,29 @@
+export interface FileEntry {
+    path: string;
+    content: string;
+}
+/** Lightweight entry — path + size from stat, no content in memory */
+export interface ScannedFile {
+    path: string;
+    size: number;
+}
+/** Path-only reference (for type signatures) */
+export interface FilePath {
+    path: string;
+}
+/**
+ * Phase 1: Scan repository — stat files to get paths + sizes, no content loaded.
+ * Memory: ~10MB for 100K files vs ~1GB+ with content.
+ */
+export declare const walkRepositoryPaths: (repoPath: string, onProgress?: (current: number, total: number, filePath: string) => void) => Promise<ScannedFile[]>;
+/**
+ * Phase 2: Read file contents for a specific set of relative paths.
+ * Returns a Map for O(1) lookup. Silently skips files that fail to read.
+ */
+export declare const readFileContents: (repoPath: string, relativePaths: string[]) => Promise<Map<string, string>>;
+/**
+ * Legacy API — scans and reads everything into memory.
+ * Used by sequential fallback path only.
+ */
+export declare const walkRepository: (repoPath: string, onProgress?: (current: number, total: number, filePath: string) => void) => Promise<FileEntry[]>;
+//# sourceMappingURL=filesystem-walker.d.ts.map