@duytransipher/gitnexus 1.0.0

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

@@ -0,0 +1,312 @@
+/**
+ * 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';
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = dirname(__filename);
+// Navigate to package root (works from both src/ and dist/)
+const leidenPath = resolve(__dirname, '..', '..', '..', 'vendor', 'leiden', 'index.cjs');
+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);
+};

package/dist/core/ingestion/constants.d.ts

@@ -0,0 +1,16 @@
+/**
+ * Default minimum buffer size for tree-sitter parsing (512 KB).
+ * tree-sitter requires bufferSize >= file size in bytes.
+ */
+export declare const TREE_SITTER_BUFFER_SIZE: number;
+/**
+ * Maximum buffer size cap (32 MB) to prevent OOM on huge files.
+ * Also used as the file-size skip threshold — files larger than this are not parsed.
+ */
+export declare const TREE_SITTER_MAX_BUFFER: number;
+/**
+ * Compute adaptive buffer size for tree-sitter parsing.
+ * Uses 2× file size, clamped between 512 KB and 32 MB.
+ * Previous 256 KB fixed limit silently skipped files > ~200 KB (e.g., imgui.h at 411 KB).
+ */
+export declare const getTreeSitterBufferSize: (contentLength: number) => number;

package/dist/core/ingestion/constants.js

@@ -0,0 +1,16 @@
+/**
+ * Default minimum buffer size for tree-sitter parsing (512 KB).
+ * tree-sitter requires bufferSize >= file size in bytes.
+ */
+export const TREE_SITTER_BUFFER_SIZE = 512 * 1024;
+/**
+ * Maximum buffer size cap (32 MB) to prevent OOM on huge files.
+ * Also used as the file-size skip threshold — files larger than this are not parsed.
+ */
+export const TREE_SITTER_MAX_BUFFER = 32 * 1024 * 1024;
+/**
+ * Compute adaptive buffer size for tree-sitter parsing.
+ * Uses 2× file size, clamped between 512 KB and 32 MB.
+ * Previous 256 KB fixed limit silently skipped files > ~200 KB (e.g., imgui.h at 411 KB).
+ */
+export const getTreeSitterBufferSize = (contentLength) => Math.min(Math.max(contentLength * 2, TREE_SITTER_BUFFER_SIZE), TREE_SITTER_MAX_BUFFER);

package/dist/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.
+ */
+import { SupportedLanguages } from '../../config/supported-languages.js';
+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: SupportedLanguages, 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;