scai 0.1.117 → 0.1.119

package/dist/daemon/daemonBatch.js

@@ -1,186 +0,0 @@
-import { indexCodeForFile } from '../db/functionIndex.js';
-import fs from 'fs/promises';
-import fsSync from 'fs';
-import { generateEmbedding } from '../lib/generateEmbedding.js';
-import { log } from '../utils/log.js';
-import lockfile from 'proper-lockfile';
-import { summaryModule } from '../pipeline/modules/summaryModule.js';
-import { classifyFile } from '../fileRules/classifyFile.js';
-import { getDbForRepo, getDbPathForRepo } from '../db/client.js';
-import { markFileAsSkippedByPath, selectUnprocessedFiles, updateFileWithSummaryAndEmbedding, insertGraphTagTemplate, selectGraphTagIdTemplate, insertGraphEntityTagTemplate, } from '../db/sqlTemplates.js';
-import { kgModule } from '../pipeline/modules/kgModule.js';
-const MAX_FILES_PER_BATCH = 5;
-/**
- * Acquire a lock on the DB so only one daemon batch modifies it at a time
- */
-async function lockDb() {
-    try {
-        return await lockfile.lock(getDbPathForRepo());
-    }
-    catch (err) {
-        log('❌ Failed to acquire DB lock: ' + err);
-        throw err;
-    }
-}
-/**
- * Run a daemon batch to process up to MAX_FILES_PER_BATCH files.
- * - Generates summaries & embeddings
- * - Extracts functions
- * - Builds Knowledge Graph (persisting entities, tags, and edges)
- */
-export async function runDaemonBatch() {
-    log('🟡 Starting daemon batch...');
-    const db = getDbForRepo();
-    const rows = db.prepare(selectUnprocessedFiles).all(MAX_FILES_PER_BATCH);
-    if (rows.length === 0) {
-        log('✅ No files left to process.');
-        return false;
-    }
-    const release = await lockDb();
-    for (const row of rows) {
-        log(`📂 Processing file: ${row.path}`);
-        if (!fsSync.existsSync(row.path)) {
-            log(`⚠️ Skipped missing file: ${row.path}`);
-            db.prepare(markFileAsSkippedByPath).run({ path: row.path });
-            continue;
-        }
-        const classification = classifyFile(row.path);
-        if (classification !== 'valid') {
-            log(`⏭️ Skipping (${classification}): ${row.path}`);
-            db.prepare(markFileAsSkippedByPath).run({ path: row.path });
-            continue;
-        }
-        try {
-            const content = await fs.readFile(row.path, 'utf-8');
-            const needsResummary = !row.summary ||
-                !row.indexed_at ||
-                (row.last_modified && new Date(row.last_modified) > new Date(row.indexed_at));
-            if (needsResummary) {
-                log(`📝 Generating summary for ${row.path}...`);
-                const summaryResult = await summaryModule.run({ content, filepath: row.path });
-                const summary = summaryResult?.summary?.trim() || null;
-                let embedding = null;
-                if (summary) {
-                    const vector = await generateEmbedding(summary);
-                    if (vector) {
-                        embedding = JSON.stringify(vector);
-                    }
-                }
-                db.prepare(updateFileWithSummaryAndEmbedding).run({
-                    summary,
-                    embedding,
-                    path: row.path,
-                });
-                log(`✅ Updated summary & embedding for ${row.path}`);
-            }
-            else {
-                log(`⚡ Skipped summary (up-to-date) for ${row.path}`);
-            }
-            const success = await indexCodeForFile(row.path, row.id);
-            if (success) {
-                log(`✅ Indexed code for ${row.path}`);
-                try {
-                    log(`🔗 Building Knowledge Graph for ${row.path}...`);
-                    const kgInput = {
-                        fileId: row.id,
-                        filepath: row.path,
-                        summary: row.summary || undefined,
-                    };
-                    const kgResult = await kgModule.run(kgInput, content);
-                    log(`✅ Knowledge Graph built for ${row.path}`);
-                    log(`Entities: ${kgResult.entities.length}, Edges: ${kgResult.edges.length}`);
-                    // Persist KG entities + tags
-                    if (kgResult.entities.length > 0) {
-                        const insertTag = db.prepare(insertGraphTagTemplate);
-                        const getTagId = db.prepare(selectGraphTagIdTemplate);
-                        const insertEntityTag = db.prepare(insertGraphEntityTagTemplate);
-                        for (const entity of kgResult.entities) {
-                            if (!entity.type || !Array.isArray(entity.tags) || entity.tags.length === 0) {
-                                console.warn(`⚠ Skipping entity due to missing type or tags:`, entity);
-                                continue;
-                            }
-                            for (const tag of entity.tags) {
-                                if (!tag || typeof tag !== 'string') {
-                                    console.warn(`⚠ Skipping invalid tag for entity ${entity.type}:`, tag);
-                                    continue;
-                                }
-                                try {
-                                    insertTag.run({ name: tag });
-                                    const tagRow = getTagId.get({ name: tag });
-                                    if (!tagRow) {
-                                        console.warn(`⚠ Could not find tag ID for: ${tag}`);
-                                        continue;
-                                    }
-                                    insertEntityTag.run({
-                                        entity_type: entity.type,
-                                        entity_unique_id: `${entity.name}@${row.path}`,
-                                        tag_id: tagRow.id,
-                                    });
-                                }
-                                catch (err) {
-                                    console.error(`❌ Failed to persist entity/tag:`, { entity, tag, error: err });
-                                }
-                            }
-                        }
-                        log(`✅ Persisted entities + tags for ${row.path}`);
-                    }
-                    else {
-                        log(`⚠️ No entities found for ${row.path}, skipping DB inserts`);
-                    }
-                    // Persist KG entities + tags (from LLM)
-                    if (kgResult.entities.length > 0) {
-                        const insertTag = db.prepare(insertGraphTagTemplate);
-                        const getTagId = db.prepare(selectGraphTagIdTemplate);
-                        const insertEntityTag = db.prepare(insertGraphEntityTagTemplate);
-                        for (const entity of kgResult.entities) {
-                            if (!entity.type || !Array.isArray(entity.tags) || entity.tags.length === 0) {
-                                console.warn(`⚠ Skipping entity due to missing type or tags:`, entity);
-                                continue;
-                            }
-                            for (const tag of entity.tags) {
-                                if (!tag || typeof tag !== 'string') {
-                                    console.warn(`⚠ Skipping invalid tag for entity ${entity.type}:`, tag);
-                                    continue;
-                                }
-                                try {
-                                    insertTag.run({ name: tag });
-                                    const tagRow = getTagId.get({ name: tag });
-                                    if (!tagRow) {
-                                        console.warn(`⚠ Could not find tag ID for: ${tag}`);
-                                        continue;
-                                    }
-                                    insertEntityTag.run({
-                                        entity_type: entity.type,
-                                        entity_unique_id: `${entity.name}@${row.path}`,
-                                        tag_id: tagRow.id,
-                                    });
-                                }
-                                catch (err) {
-                                    console.error(`❌ Failed to persist entity/tag:`, { entity, tag, error: err });
-                                }
-                            }
-                        }
-                        log(`✅ Persisted entities + tags for ${row.path}`);
-                    }
-                    else {
-                        log(`⚠️ No entities found for ${row.path}, skipping DB inserts`);
-                    }
-                }
-                catch (kgErr) {
-                    log(`❌ KG build failed for ${row.path}: ${kgErr instanceof Error ? kgErr.message : String(kgErr)}`);
-                }
-            }
-            else {
-                log(`ℹ️ No code elements extracted for ${row.path}`);
-            }
-        }
-        catch (err) {
-            log(`❌ Failed: ${row.path}: ${err instanceof Error ? err.message : String(err)}\n`);
-        }
-        // Short delay between files
-        await new Promise((resolve) => setTimeout(resolve, 200));
-    }
-    await release();
-    log('✅ Finished daemon batch.\n\n');
-    return true;
-}

package/dist/fileRules/scoreFiles.js

@@ -1,71 +0,0 @@
-import { stringSimilarity } from 'string-similarity-js';
-export function scoreFiles(query, embedding, candidates) {
-    const terms = query.toLowerCase().split(/\s+/);
-    const bm25Min = Math.min(...candidates.map(r => r.bm25Score));
-    const bm25Max = Math.max(...candidates.map(r => r.bm25Score));
-    return candidates.map(result => {
-        let finalScore = 0;
-        let sim = 0;
-        const path = result.path.toLowerCase();
-        const filename = result.filename.toLowerCase();
-        const summary = (result.summary || '').toLowerCase();
-        // 🎯 Normalize BM25
-        const normalizedBm25 = 1 - ((result.bm25Score - bm25Min) / (bm25Max - bm25Min + 1e-5));
-        // 🧠 Embedding similarity
-        if (result.embedding) {
-            try {
-                const vector = JSON.parse(result.embedding);
-                sim = cosineSimilarity(embedding, vector);
-            }
-            catch {
-                sim = 0;
-            }
-        }
-        // 🧩 Match ratio
-        let matchRatio = 0;
-        let matchedTerms = 0;
-        for (const term of terms) {
-            if (path.includes(term) || summary.includes(term))
-                matchedTerms++;
-        }
-        matchRatio = matchedTerms / terms.length;
-        const termBoost = matchRatio >= 1 ? 1.0 : matchRatio >= 0.5 ? 0.5 : 0;
-        // 🪜 Path heuristics
-        const isHtml = path.endsWith('.html');
-        const isSrc = path.includes('/src/') || path.includes('/controls/');
-        const isDoc = path.includes('/docs/') || path.includes('/mvndist/');
-        const isExactMatch = filename === `${terms[0]}.js`;
-        let weight = 1;
-        if (isHtml)
-            weight *= 0.85;
-        if (isDoc)
-            weight *= 0.8;
-        if (isSrc)
-            weight *= 1.2;
-        if (isExactMatch)
-            weight *= 1.5;
-        // 🧪 Fuzzy score
-        const fuzzyScore = stringSimilarity(query, `${path} ${summary}`);
-        // 🧮 Final composite score
-        finalScore =
-            (0.4 * normalizedBm25) +
-                (0.4 * sim) +
-                (0.2 * termBoost) +
-                (fuzzyScore * 0.5); // scale fuzzy match moderately
-        finalScore *= weight;
-        return {
-            id: result.id,
-            path: result.path,
-            summary: result.summary,
-            score: finalScore,
-            sim,
-            bm25: result.bm25Score,
-        };
-    }).sort((a, b) => b.score - a.score);
-}
-function cosineSimilarity(a, b) {
-    const dot = a.reduce((sum, ai, i) => sum + ai * b[i], 0);
-    const magA = Math.sqrt(a.reduce((sum, ai) => sum + ai * ai, 0));
-    const magB = Math.sqrt(b.reduce((sum, bi) => sum + bi * bi, 0));
-    return dot / (magA * magB);
-}

package/dist/lib/generateEmbedding.js

@@ -1,22 +0,0 @@
-export async function generateEmbedding(text) {
-    try {
-        const res = await fetch('http://localhost:11434/api/embeddings', {
-            method: 'POST',
-            headers: { 'Content-Type': 'application/json' },
-            body: JSON.stringify({
-                model: 'mistral', // or 'llama3' — whatever works best
-                prompt: text,
-            }),
-        });
-        if (!res.ok) {
-            console.error('❌ Failed to generate embedding:', await res.text());
-            return null;
-        }
-        const data = await res.json();
-        return data.embedding;
-    }
-    catch (err) {
-        console.error('❌ Embedding error:', err instanceof Error ? err.message : err);
-        return null;
-    }
-}