sentinelayer-cli 0.1.2 → 0.3.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +996 -996
- package/bin/create-sentinelayer.js +5 -5
- package/bin/sentinelayer-cli.js +4 -4
- package/bin/sl.js +5 -5
- package/package.json +62 -54
- package/src/agents/jules/config/definition.js +209 -209
- package/src/agents/jules/config/system-prompt.js +175 -175
- package/src/agents/jules/error-intake.js +51 -51
- package/src/agents/jules/fix-cycle.js +377 -377
- package/src/agents/jules/loop.js +367 -367
- package/src/agents/jules/pulse.js +319 -319
- package/src/agents/jules/stream.js +186 -186
- package/src/agents/jules/swarm/file-scanner.js +74 -74
- package/src/agents/jules/swarm/index.js +11 -11
- package/src/agents/jules/swarm/orchestrator.js +362 -362
- package/src/agents/jules/swarm/pattern-hunter.js +123 -123
- package/src/agents/jules/swarm/sub-agent.js +308 -308
- package/src/agents/jules/tools/auth-audit.js +226 -222
- package/src/agents/jules/tools/dispatch.js +327 -327
- package/src/agents/jules/tools/file-edit.js +180 -180
- package/src/agents/jules/tools/file-read.js +100 -100
- package/src/agents/jules/tools/frontend-analyze.js +570 -570
- package/src/agents/jules/tools/glob.js +168 -168
- package/src/agents/jules/tools/grep.js +228 -228
- package/src/agents/jules/tools/index.js +29 -29
- package/src/agents/jules/tools/path-guards.js +161 -161
- package/src/agents/jules/tools/runtime-audit.js +493 -493
- package/src/agents/jules/tools/shell.js +383 -383
- package/src/ai/aidenid.js +972 -945
- package/src/ai/client.js +508 -508
- package/src/ai/domain-target-store.js +268 -268
- package/src/ai/identity-store.js +270 -270
- package/src/ai/site-store.js +145 -145
- package/src/audit/agents/architecture.js +180 -180
- package/src/audit/agents/compliance.js +179 -179
- package/src/audit/agents/documentation.js +165 -165
- package/src/audit/agents/performance.js +145 -145
- package/src/audit/agents/security.js +215 -215
- package/src/audit/agents/testing.js +172 -172
- package/src/audit/orchestrator.js +557 -557
- package/src/audit/package.js +204 -204
- package/src/audit/registry.js +284 -284
- package/src/audit/replay.js +103 -103
- package/src/auth/http.js +113 -113
- package/src/auth/service.js +891 -848
- package/src/auth/session-store.js +359 -345
- package/src/cli.js +252 -252
- package/src/commands/ai/identity-lifecycle.js +1338 -1337
- package/src/commands/ai/provision-governance.js +1272 -1246
- package/src/commands/ai/shared.js +147 -147
- package/src/commands/ai.js +11 -11
- package/src/commands/apply.js +12 -12
- package/src/commands/audit.js +1166 -1166
- package/src/commands/auth.js +375 -366
- package/src/commands/chat.js +191 -191
- package/src/commands/config.js +184 -184
- package/src/commands/cost.js +311 -311
- package/src/commands/daemon/core.js +850 -850
- package/src/commands/daemon/extended.js +1048 -1048
- package/src/commands/daemon/shared.js +213 -213
- package/src/commands/daemon.js +11 -11
- package/src/commands/guide.js +174 -174
- package/src/commands/ingest.js +58 -58
- package/src/commands/init.js +55 -55
- package/src/commands/legacy-args.js +10 -10
- package/src/commands/mcp.js +461 -404
- package/src/commands/omargate.js +15 -15
- package/src/commands/persona.js +20 -20
- package/src/commands/plugin.js +260 -260
- package/src/commands/policy.js +132 -132
- package/src/commands/prompt.js +238 -238
- package/src/commands/review.js +704 -704
- package/src/commands/scan.js +866 -788
- package/src/commands/spec.js +716 -716
- package/src/commands/swarm.js +651 -651
- package/src/commands/telemetry.js +202 -202
- package/src/commands/watch.js +510 -510
- package/src/config/agent-dictionary.js +182 -182
- package/src/config/io.js +56 -56
- package/src/config/paths.js +18 -18
- package/src/config/schema.js +55 -55
- package/src/config/service.js +184 -184
- package/src/cost/budget.js +235 -235
- package/src/cost/history.js +188 -188
- package/src/cost/tracker.js +171 -171
- package/src/daemon/artifact-lineage.js +534 -534
- package/src/daemon/assignment-ledger.js +770 -770
- package/src/daemon/ast-parser-layer.js +258 -258
- package/src/daemon/budget-governor.js +633 -633
- package/src/daemon/callgraph-overlay.js +646 -646
- package/src/daemon/error-worker.js +626 -626
- package/src/daemon/hybrid-mapper.js +929 -929
- package/src/daemon/jira-lifecycle.js +632 -632
- package/src/daemon/operator-control.js +657 -657
- package/src/daemon/reliability-lane.js +471 -471
- package/src/daemon/watchdog.js +971 -971
- package/src/guide/generator.js +316 -316
- package/src/ingest/engine.js +918 -918
- package/src/legacy-cli.js +2548 -2435
- package/src/mcp/registry.js +695 -695
- package/src/memory/blackboard.js +301 -301
- package/src/memory/retrieval.js +581 -581
- package/src/plugin/manifest.js +553 -553
- package/src/policy/packs.js +144 -144
- package/src/prompt/generator.js +118 -106
- package/src/review/ai-review.js +669 -669
- package/src/review/local-review.js +1284 -1284
- package/src/review/replay.js +235 -235
- package/src/review/report.js +664 -664
- package/src/review/spec-binding.js +487 -487
- package/src/scaffold/generator.js +67 -0
- package/src/scaffold/templates.js +150 -0
- package/src/scan/generator.js +418 -351
- package/src/scan/gh-secrets.js +107 -0
- package/src/spec/generator.js +519 -519
- package/src/spec/regenerate.js +237 -237
- package/src/spec/templates.js +91 -91
- package/src/swarm/dashboard.js +247 -247
- package/src/swarm/factory.js +363 -363
- package/src/swarm/pentest.js +934 -934
- package/src/swarm/registry.js +419 -419
- package/src/swarm/report.js +158 -158
- package/src/swarm/runtime.js +576 -576
- package/src/swarm/scenario-dsl.js +272 -272
- package/src/telemetry/ledger.js +302 -302
- package/src/telemetry/sync.js +96 -59
- package/src/ui/markdown.js +220 -220
package/src/ingest/engine.js
CHANGED
|
@@ -1,918 +1,918 @@
|
|
|
1
|
-
import { execFile } from "node:child_process";
|
|
2
|
-
import { createHash } from "node:crypto";
|
|
3
|
-
import fs from "node:fs";
|
|
4
|
-
import fsp from "node:fs/promises";
|
|
5
|
-
import path from "node:path";
|
|
6
|
-
import process from "node:process";
|
|
7
|
-
import { promisify } from "node:util";
|
|
8
|
-
|
|
9
|
-
import ignore from "ignore";
|
|
10
|
-
|
|
11
|
-
import { resolveOutputRoot } from "../config/service.js";
|
|
12
|
-
|
|
13
|
-
const DEFAULT_IGNORED_DIRS = new Set([
|
|
14
|
-
".git",
|
|
15
|
-
"node_modules",
|
|
16
|
-
".venv",
|
|
17
|
-
".next",
|
|
18
|
-
"dist",
|
|
19
|
-
"build",
|
|
20
|
-
"coverage",
|
|
21
|
-
".sentinelayer",
|
|
22
|
-
".turbo",
|
|
23
|
-
".idea",
|
|
24
|
-
".vscode",
|
|
25
|
-
]);
|
|
26
|
-
|
|
27
|
-
const MAX_FILE_SIZE_BYTES = 1024 * 1024;
|
|
28
|
-
const FILE_INDEX_LIMIT = 5000;
|
|
29
|
-
const execFileAsync = promisify(execFile);
|
|
30
|
-
const INGEST_CACHE_SCHEMA = "path-size-mtime-sha256-v1";
|
|
31
|
-
|
|
32
|
-
const LANGUAGE_BY_EXTENSION = {
|
|
33
|
-
".js": "JavaScript",
|
|
34
|
-
".cjs": "JavaScript",
|
|
35
|
-
".mjs": "JavaScript",
|
|
36
|
-
".ts": "TypeScript",
|
|
37
|
-
".tsx": "TypeScript",
|
|
38
|
-
".jsx": "JavaScript",
|
|
39
|
-
".py": "Python",
|
|
40
|
-
".go": "Go",
|
|
41
|
-
".rs": "Rust",
|
|
42
|
-
".java": "Java",
|
|
43
|
-
".kt": "Kotlin",
|
|
44
|
-
".swift": "Swift",
|
|
45
|
-
".rb": "Ruby",
|
|
46
|
-
".php": "PHP",
|
|
47
|
-
".cs": "C#",
|
|
48
|
-
".cpp": "C++",
|
|
49
|
-
".cc": "C++",
|
|
50
|
-
".c": "C",
|
|
51
|
-
".h": "C/C++ Header",
|
|
52
|
-
".hpp": "C/C++ Header",
|
|
53
|
-
".sql": "SQL",
|
|
54
|
-
".md": "Markdown",
|
|
55
|
-
".yml": "YAML",
|
|
56
|
-
".yaml": "YAML",
|
|
57
|
-
".json": "JSON",
|
|
58
|
-
".toml": "TOML",
|
|
59
|
-
".tf": "Terraform",
|
|
60
|
-
".sh": "Shell",
|
|
61
|
-
".bash": "Shell",
|
|
62
|
-
".ps1": "PowerShell",
|
|
63
|
-
".dockerfile": "Docker",
|
|
64
|
-
};
|
|
65
|
-
|
|
66
|
-
const MANIFEST_CANDIDATES = new Set([
|
|
67
|
-
"package.json",
|
|
68
|
-
"requirements.txt",
|
|
69
|
-
"pyproject.toml",
|
|
70
|
-
"go.mod",
|
|
71
|
-
"Cargo.toml",
|
|
72
|
-
"Gemfile",
|
|
73
|
-
"pom.xml",
|
|
74
|
-
"build.gradle",
|
|
75
|
-
"build.gradle.kts",
|
|
76
|
-
"composer.json",
|
|
77
|
-
]);
|
|
78
|
-
|
|
79
|
-
const ENTRY_POINT_CANDIDATES = [
|
|
80
|
-
"src/index.ts",
|
|
81
|
-
"src/index.tsx",
|
|
82
|
-
"src/index.js",
|
|
83
|
-
"src/main.ts",
|
|
84
|
-
"src/main.js",
|
|
85
|
-
"src/server.ts",
|
|
86
|
-
"src/server.js",
|
|
87
|
-
"index.ts",
|
|
88
|
-
"index.js",
|
|
89
|
-
"main.ts",
|
|
90
|
-
"main.js",
|
|
91
|
-
"main.py",
|
|
92
|
-
"app.py",
|
|
93
|
-
"server.py",
|
|
94
|
-
"cmd/main.go",
|
|
95
|
-
"src/main.rs",
|
|
96
|
-
];
|
|
97
|
-
|
|
98
|
-
function toPosixPath(value) {
|
|
99
|
-
return String(value || "").replace(/\\/g, "/");
|
|
100
|
-
}
|
|
101
|
-
|
|
102
|
-
function parseIsoToEpoch(value) {
|
|
103
|
-
const normalized = String(value || "").trim();
|
|
104
|
-
if (!normalized) {
|
|
105
|
-
return null;
|
|
106
|
-
}
|
|
107
|
-
const epoch = Date.parse(normalized);
|
|
108
|
-
if (!Number.isFinite(epoch)) {
|
|
109
|
-
return null;
|
|
110
|
-
}
|
|
111
|
-
return epoch;
|
|
112
|
-
}
|
|
113
|
-
|
|
114
|
-
function normalizeMtimeMs(value) {
|
|
115
|
-
const normalized = Number(value || 0);
|
|
116
|
-
if (!Number.isFinite(normalized) || normalized < 0) {
|
|
117
|
-
return 0;
|
|
118
|
-
}
|
|
119
|
-
return Math.floor(normalized);
|
|
120
|
-
}
|
|
121
|
-
|
|
122
|
-
function appendFingerprintInput(hasher, relativePath, sizeBytes, mtimeMs) {
|
|
123
|
-
hasher.update(
|
|
124
|
-
`${toPosixPath(relativePath)}\u001f${String(Number(sizeBytes || 0))}\u001f${normalizeMtimeMs(
|
|
125
|
-
mtimeMs
|
|
126
|
-
)}\n`,
|
|
127
|
-
"utf-8"
|
|
128
|
-
);
|
|
129
|
-
}
|
|
130
|
-
|
|
131
|
-
function countLoc(text) {
|
|
132
|
-
return String(text || "")
|
|
133
|
-
.split(/\r?\n/)
|
|
134
|
-
.filter((line) => line.trim().length > 0).length;
|
|
135
|
-
}
|
|
136
|
-
|
|
137
|
-
function detectLanguage(relativePath) {
|
|
138
|
-
const normalized = toPosixPath(relativePath);
|
|
139
|
-
const baseName = path.basename(normalized).toLowerCase();
|
|
140
|
-
if (baseName === "dockerfile") {
|
|
141
|
-
return "Docker";
|
|
142
|
-
}
|
|
143
|
-
const extension = path.extname(baseName);
|
|
144
|
-
return LANGUAGE_BY_EXTENSION[extension] || "Other";
|
|
145
|
-
}
|
|
146
|
-
|
|
147
|
-
async function readIgnorePatterns(filePath) {
|
|
148
|
-
try {
|
|
149
|
-
const raw = await fsp.readFile(filePath, "utf-8");
|
|
150
|
-
return String(raw || "")
|
|
151
|
-
.split(/\r?\n/)
|
|
152
|
-
.map((line) => line.trim())
|
|
153
|
-
.filter((line) => line && !line.startsWith("#"));
|
|
154
|
-
} catch (error) {
|
|
155
|
-
if (error && error.code === "ENOENT") {
|
|
156
|
-
return [];
|
|
157
|
-
}
|
|
158
|
-
throw error;
|
|
159
|
-
}
|
|
160
|
-
}
|
|
161
|
-
|
|
162
|
-
async function createIgnoreMatcher(rootPath) {
|
|
163
|
-
const matcher = ignore();
|
|
164
|
-
const gitignorePatterns = await readIgnorePatterns(path.join(rootPath, ".gitignore"));
|
|
165
|
-
const sentinelPatterns = await readIgnorePatterns(path.join(rootPath, ".sentinelayerignore"));
|
|
166
|
-
matcher.add([...gitignorePatterns, ...sentinelPatterns]);
|
|
167
|
-
|
|
168
|
-
return {
|
|
169
|
-
ignores(relativePath, isDirectory) {
|
|
170
|
-
const normalized = toPosixPath(relativePath);
|
|
171
|
-
if (!normalized) {
|
|
172
|
-
return false;
|
|
173
|
-
}
|
|
174
|
-
const candidate = isDirectory ? `${normalized}/` : normalized;
|
|
175
|
-
return matcher.ignores(candidate);
|
|
176
|
-
},
|
|
177
|
-
};
|
|
178
|
-
}
|
|
179
|
-
|
|
180
|
-
async function computeCodebaseContentFingerprint({ rootPath }) {
|
|
181
|
-
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
182
|
-
const ignoreMatcher = await createIgnoreMatcher(resolvedRoot);
|
|
183
|
-
const stack = [resolvedRoot];
|
|
184
|
-
const hasher = createHash("sha256");
|
|
185
|
-
let filesCount = 0;
|
|
186
|
-
let latestFileMtimeMs = 0;
|
|
187
|
-
|
|
188
|
-
while (stack.length > 0) {
|
|
189
|
-
const current = stack.pop();
|
|
190
|
-
if (!current) {
|
|
191
|
-
continue;
|
|
192
|
-
}
|
|
193
|
-
|
|
194
|
-
let entries = [];
|
|
195
|
-
try {
|
|
196
|
-
entries = await fsp.readdir(current, { withFileTypes: true });
|
|
197
|
-
} catch {
|
|
198
|
-
continue;
|
|
199
|
-
}
|
|
200
|
-
|
|
201
|
-
for (const entry of entries) {
|
|
202
|
-
const fullPath = path.join(current, entry.name);
|
|
203
|
-
const relativePath = toPosixPath(path.relative(resolvedRoot, fullPath));
|
|
204
|
-
|
|
205
|
-
if (entry.isDirectory()) {
|
|
206
|
-
if (!relativePath) {
|
|
207
|
-
continue;
|
|
208
|
-
}
|
|
209
|
-
if (DEFAULT_IGNORED_DIRS.has(entry.name)) {
|
|
210
|
-
continue;
|
|
211
|
-
}
|
|
212
|
-
if (ignoreMatcher.ignores(relativePath, true)) {
|
|
213
|
-
continue;
|
|
214
|
-
}
|
|
215
|
-
stack.push(fullPath);
|
|
216
|
-
continue;
|
|
217
|
-
}
|
|
218
|
-
|
|
219
|
-
if (!entry.isFile()) {
|
|
220
|
-
continue;
|
|
221
|
-
}
|
|
222
|
-
if (ignoreMatcher.ignores(relativePath, false)) {
|
|
223
|
-
continue;
|
|
224
|
-
}
|
|
225
|
-
|
|
226
|
-
let stat = null;
|
|
227
|
-
try {
|
|
228
|
-
stat = await fsp.stat(fullPath);
|
|
229
|
-
} catch {
|
|
230
|
-
stat = null;
|
|
231
|
-
}
|
|
232
|
-
if (!stat || stat.size > MAX_FILE_SIZE_BYTES) {
|
|
233
|
-
continue;
|
|
234
|
-
}
|
|
235
|
-
|
|
236
|
-
filesCount += 1;
|
|
237
|
-
latestFileMtimeMs = Math.max(latestFileMtimeMs, normalizeMtimeMs(stat.mtimeMs));
|
|
238
|
-
appendFingerprintInput(hasher, relativePath, stat.size, stat.mtimeMs);
|
|
239
|
-
}
|
|
240
|
-
}
|
|
241
|
-
|
|
242
|
-
return {
|
|
243
|
-
schema: INGEST_CACHE_SCHEMA,
|
|
244
|
-
contentHash: hasher.digest("hex"),
|
|
245
|
-
filesCount,
|
|
246
|
-
latestFileMtimeMs,
|
|
247
|
-
};
|
|
248
|
-
}
|
|
249
|
-
|
|
250
|
-
async function readExistingIngest(outputPath) {
|
|
251
|
-
try {
|
|
252
|
-
const parsed = JSON.parse(await fsp.readFile(outputPath, "utf-8"));
|
|
253
|
-
return parsed && typeof parsed === "object" ? parsed : null;
|
|
254
|
-
} catch (error) {
|
|
255
|
-
if (error && error.code === "ENOENT") {
|
|
256
|
-
return null;
|
|
257
|
-
}
|
|
258
|
-
throw error;
|
|
259
|
-
}
|
|
260
|
-
}
|
|
261
|
-
|
|
262
|
-
async function resolveIngestOutputPath({ rootPath, outputFile = "", outputDir = "" }) {
|
|
263
|
-
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
264
|
-
const explicitOutputFile = String(outputFile || "").trim();
|
|
265
|
-
if (explicitOutputFile) {
|
|
266
|
-
return path.resolve(resolvedRoot, explicitOutputFile);
|
|
267
|
-
}
|
|
268
|
-
const outputRoot = await resolveOutputRoot({
|
|
269
|
-
cwd: resolvedRoot,
|
|
270
|
-
outputDirOverride: outputDir,
|
|
271
|
-
});
|
|
272
|
-
return path.join(outputRoot, "CODEBASE_INGEST.json");
|
|
273
|
-
}
|
|
274
|
-
|
|
275
|
-
async function readGitLastCommitAt(rootPath) {
|
|
276
|
-
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
277
|
-
try {
|
|
278
|
-
const { stdout } = await execFileAsync("git", [
|
|
279
|
-
"-C",
|
|
280
|
-
resolvedRoot,
|
|
281
|
-
"log",
|
|
282
|
-
"-1",
|
|
283
|
-
"--format=%cI",
|
|
284
|
-
]);
|
|
285
|
-
const normalized = String(stdout || "").trim();
|
|
286
|
-
return parseIsoToEpoch(normalized) === null ? "" : normalized;
|
|
287
|
-
} catch {
|
|
288
|
-
return "";
|
|
289
|
-
}
|
|
290
|
-
}
|
|
291
|
-
|
|
292
|
-
function buildIngestStaleness({ existingIngest, fingerprint, lastCommitAt }) {
|
|
293
|
-
if (!existingIngest) {
|
|
294
|
-
return {
|
|
295
|
-
stale: true,
|
|
296
|
-
reasons: ["missing_ingest"],
|
|
297
|
-
};
|
|
298
|
-
}
|
|
299
|
-
|
|
300
|
-
const reasons = [];
|
|
301
|
-
const generatedAtEpoch = parseIsoToEpoch(existingIngest.generatedAt);
|
|
302
|
-
const lastCommitEpoch = parseIsoToEpoch(lastCommitAt);
|
|
303
|
-
if (generatedAtEpoch === null) {
|
|
304
|
-
reasons.push("invalid_generated_at");
|
|
305
|
-
} else if (lastCommitEpoch !== null && generatedAtEpoch < lastCommitEpoch) {
|
|
306
|
-
reasons.push("older_than_last_commit");
|
|
307
|
-
}
|
|
308
|
-
|
|
309
|
-
const existingContentHash = String(existingIngest.cache?.contentHash || "").trim();
|
|
310
|
-
if (existingContentHash && existingContentHash !== fingerprint.contentHash) {
|
|
311
|
-
reasons.push("content_hash_mismatch");
|
|
312
|
-
} else if (!existingContentHash) {
|
|
313
|
-
reasons.push("missing_content_hash");
|
|
314
|
-
}
|
|
315
|
-
|
|
316
|
-
return {
|
|
317
|
-
stale: reasons.length > 0,
|
|
318
|
-
reasons,
|
|
319
|
-
};
|
|
320
|
-
}
|
|
321
|
-
|
|
322
|
-
export function formatIngestResolutionNotice(resolution = {}) {
|
|
323
|
-
const reasons = Array.isArray(resolution.reasons) ? resolution.reasons : [];
|
|
324
|
-
if (resolution.refreshed) {
|
|
325
|
-
return `ingest refreshed (${reasons.join(", ") || "requested"})`;
|
|
326
|
-
}
|
|
327
|
-
if (resolution.stale) {
|
|
328
|
-
return `ingest stale (${reasons.join(", ") || "unknown"}); re-run with --refresh`;
|
|
329
|
-
}
|
|
330
|
-
return "ingest cache hit";
|
|
331
|
-
}
|
|
332
|
-
|
|
333
|
-
function safeJsonParse(raw) {
|
|
334
|
-
try {
|
|
335
|
-
return JSON.parse(raw);
|
|
336
|
-
} catch {
|
|
337
|
-
return null;
|
|
338
|
-
}
|
|
339
|
-
}
|
|
340
|
-
|
|
341
|
-
function normalizeDependencySet(dependencies) {
|
|
342
|
-
if (!dependencies || typeof dependencies !== "object") {
|
|
343
|
-
return new Set();
|
|
344
|
-
}
|
|
345
|
-
return new Set(Object.keys(dependencies).map((value) => String(value || "").toLowerCase()));
|
|
346
|
-
}
|
|
347
|
-
|
|
348
|
-
function detectFrameworks(manifests) {
|
|
349
|
-
const frameworks = new Set();
|
|
350
|
-
|
|
351
|
-
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
352
|
-
if (packageJson) {
|
|
353
|
-
const deps = normalizeDependencySet({
|
|
354
|
-
...(packageJson.dependencies || {}),
|
|
355
|
-
...(packageJson.devDependencies || {}),
|
|
356
|
-
...(packageJson.peerDependencies || {}),
|
|
357
|
-
});
|
|
358
|
-
if (deps.has("next")) frameworks.add("nextjs");
|
|
359
|
-
if (deps.has("react")) frameworks.add("react");
|
|
360
|
-
if (deps.has("vue")) frameworks.add("vue");
|
|
361
|
-
if (deps.has("svelte")) frameworks.add("svelte");
|
|
362
|
-
if (deps.has("express")) frameworks.add("express");
|
|
363
|
-
if (deps.has("fastify")) frameworks.add("fastify");
|
|
364
|
-
if (deps.has("hono")) frameworks.add("hono");
|
|
365
|
-
if (deps.has("@nestjs/core")) frameworks.add("nestjs");
|
|
366
|
-
if (deps.has("prisma")) frameworks.add("prisma");
|
|
367
|
-
if (deps.has("typeorm")) frameworks.add("typeorm");
|
|
368
|
-
if (deps.has("drizzle-orm")) frameworks.add("drizzle");
|
|
369
|
-
if (deps.has("playwright")) frameworks.add("playwright");
|
|
370
|
-
if (deps.has("jest")) frameworks.add("jest");
|
|
371
|
-
if (deps.has("vitest")) frameworks.add("vitest");
|
|
372
|
-
if (deps.has("@opentelemetry/api") || deps.has("@sentry/node")) frameworks.add("observability-js");
|
|
373
|
-
}
|
|
374
|
-
|
|
375
|
-
const requirementsText = String(manifests["requirements.txt"] || "").toLowerCase();
|
|
376
|
-
if (/\bfastapi\b/.test(requirementsText)) frameworks.add("fastapi");
|
|
377
|
-
if (/\bdjango\b/.test(requirementsText)) frameworks.add("django");
|
|
378
|
-
if (/\bflask\b/.test(requirementsText)) frameworks.add("flask");
|
|
379
|
-
|
|
380
|
-
const pyprojectText = String(manifests["pyproject.toml"] || "").toLowerCase();
|
|
381
|
-
if (/\bfastapi\b/.test(pyprojectText)) frameworks.add("fastapi");
|
|
382
|
-
if (/\bdjango\b/.test(pyprojectText)) frameworks.add("django");
|
|
383
|
-
if (/\bflask\b/.test(pyprojectText)) frameworks.add("flask");
|
|
384
|
-
|
|
385
|
-
const goModText = String(manifests["go.mod"] || "").toLowerCase();
|
|
386
|
-
if (/gin-gonic\/gin/.test(goModText)) frameworks.add("gin");
|
|
387
|
-
if (/gofiber\/fiber/.test(goModText)) frameworks.add("fiber");
|
|
388
|
-
if (/labstack\/echo/.test(goModText)) frameworks.add("echo");
|
|
389
|
-
|
|
390
|
-
const cargoText = String(manifests["Cargo.toml"] || "").toLowerCase();
|
|
391
|
-
if (/\baxum\b/.test(cargoText)) frameworks.add("axum");
|
|
392
|
-
if (/\bactix-web\b/.test(cargoText)) frameworks.add("actix-web");
|
|
393
|
-
|
|
394
|
-
const gemfileText = String(manifests.Gemfile || "").toLowerCase();
|
|
395
|
-
if (/\brails\b/.test(gemfileText)) frameworks.add("rails");
|
|
396
|
-
|
|
397
|
-
return [...frameworks].sort((left, right) => left.localeCompare(right));
|
|
398
|
-
}
|
|
399
|
-
|
|
400
|
-
function derivePackageMetadata(manifests) {
|
|
401
|
-
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
402
|
-
if (!packageJson || typeof packageJson !== "object") {
|
|
403
|
-
return {
|
|
404
|
-
name: "",
|
|
405
|
-
scripts: [],
|
|
406
|
-
};
|
|
407
|
-
}
|
|
408
|
-
|
|
409
|
-
const scripts =
|
|
410
|
-
packageJson.scripts && typeof packageJson.scripts === "object"
|
|
411
|
-
? Object.keys(packageJson.scripts)
|
|
412
|
-
.map((value) => String(value || "").trim())
|
|
413
|
-
.filter(Boolean)
|
|
414
|
-
.sort((left, right) => left.localeCompare(right))
|
|
415
|
-
: [];
|
|
416
|
-
|
|
417
|
-
return {
|
|
418
|
-
name: String(packageJson.name || "").trim(),
|
|
419
|
-
scripts,
|
|
420
|
-
};
|
|
421
|
-
}
|
|
422
|
-
|
|
423
|
-
function deriveEntryPoints(fileSet, manifests) {
|
|
424
|
-
const entryPoints = new Set();
|
|
425
|
-
for (const candidate of ENTRY_POINT_CANDIDATES) {
|
|
426
|
-
if (fileSet.has(candidate)) {
|
|
427
|
-
entryPoints.add(candidate);
|
|
428
|
-
}
|
|
429
|
-
}
|
|
430
|
-
|
|
431
|
-
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
432
|
-
if (packageJson) {
|
|
433
|
-
if (typeof packageJson.main === "string" && packageJson.main.trim()) {
|
|
434
|
-
entryPoints.add(packageJson.main.trim());
|
|
435
|
-
}
|
|
436
|
-
if (packageJson.bin && typeof packageJson.bin === "object") {
|
|
437
|
-
for (const value of Object.values(packageJson.bin)) {
|
|
438
|
-
const normalized = String(value || "").trim();
|
|
439
|
-
if (normalized) {
|
|
440
|
-
entryPoints.add(normalized);
|
|
441
|
-
}
|
|
442
|
-
}
|
|
443
|
-
}
|
|
444
|
-
}
|
|
445
|
-
|
|
446
|
-
return [...entryPoints].sort((left, right) => left.localeCompare(right));
|
|
447
|
-
}
|
|
448
|
-
|
|
449
|
-
function deriveRiskSurfaces({ fileSet, frameworks, manifests, languageStats }) {
|
|
450
|
-
const surfaces = new Map();
|
|
451
|
-
|
|
452
|
-
const addSurface = (surface, reason) => {
|
|
453
|
-
if (!surfaces.has(surface)) {
|
|
454
|
-
surfaces.set(surface, reason);
|
|
455
|
-
}
|
|
456
|
-
};
|
|
457
|
-
|
|
458
|
-
const hasFile = (predicate) => [...fileSet].some(predicate);
|
|
459
|
-
const hasFramework = (name) => frameworks.includes(name);
|
|
460
|
-
|
|
461
|
-
addSurface("code_quality", "Source files detected.");
|
|
462
|
-
addSurface("security_overlay", "Credential/policy scanning is applicable for any repository ingest.");
|
|
463
|
-
|
|
464
|
-
const hasTests = hasFile((file) => /(^|\/)(test|tests|__tests__)\//.test(file) || /\.(test|spec)\./.test(file));
|
|
465
|
-
if (hasTests || hasFramework("jest") || hasFramework("vitest") || hasFramework("playwright")) {
|
|
466
|
-
addSurface("testing_correctness", "Test assets detected.");
|
|
467
|
-
}
|
|
468
|
-
|
|
469
|
-
const hasFrontend =
|
|
470
|
-
hasFramework("nextjs") ||
|
|
471
|
-
hasFramework("react") ||
|
|
472
|
-
hasFramework("vue") ||
|
|
473
|
-
hasFramework("svelte") ||
|
|
474
|
-
languageStats.JavaScript ||
|
|
475
|
-
languageStats.TypeScript;
|
|
476
|
-
if (hasFrontend) {
|
|
477
|
-
addSurface("frontend_runtime", "Frontend/runtime JavaScript stack detected.");
|
|
478
|
-
}
|
|
479
|
-
|
|
480
|
-
const hasBackend =
|
|
481
|
-
hasFramework("express") ||
|
|
482
|
-
hasFramework("nestjs") ||
|
|
483
|
-
hasFramework("fastify") ||
|
|
484
|
-
hasFramework("hono") ||
|
|
485
|
-
hasFramework("fastapi") ||
|
|
486
|
-
hasFramework("django") ||
|
|
487
|
-
hasFramework("flask") ||
|
|
488
|
-
hasFramework("gin") ||
|
|
489
|
-
hasFramework("fiber") ||
|
|
490
|
-
hasFramework("echo");
|
|
491
|
-
if (hasBackend) {
|
|
492
|
-
addSurface("backend_runtime", "Backend framework/runtime hints detected.");
|
|
493
|
-
}
|
|
494
|
-
|
|
495
|
-
const hasData =
|
|
496
|
-
hasFramework("prisma") ||
|
|
497
|
-
hasFramework("typeorm") ||
|
|
498
|
-
hasFramework("drizzle") ||
|
|
499
|
-
hasFile((file) => /(^|\/)(migrations|db|database|sql)\//.test(file) || file.endsWith(".sql"));
|
|
500
|
-
if (hasData) {
|
|
501
|
-
addSurface("data_layer", "Data-model or migration assets detected.");
|
|
502
|
-
}
|
|
503
|
-
|
|
504
|
-
const hasInfra =
|
|
505
|
-
hasFile(
|
|
506
|
-
(file) =>
|
|
507
|
-
file.endsWith(".tf") ||
|
|
508
|
-
file.includes("docker-compose") ||
|
|
509
|
-
file.endsWith("Dockerfile") ||
|
|
510
|
-
/(^|\/)(k8s|helm|terraform)\//.test(file)
|
|
511
|
-
);
|
|
512
|
-
if (hasInfra) {
|
|
513
|
-
addSurface("infrastructure", "Infrastructure-as-code or container orchestration assets detected.");
|
|
514
|
-
}
|
|
515
|
-
|
|
516
|
-
const hasRelease = hasFile((file) => file.startsWith(".github/workflows/") || file.startsWith(".gitlab-ci"));
|
|
517
|
-
if (hasRelease) {
|
|
518
|
-
addSurface("release_engineering", "CI/CD workflow definitions detected.");
|
|
519
|
-
}
|
|
520
|
-
|
|
521
|
-
const hasSupplyChain =
|
|
522
|
-
Object.keys(manifests).length > 0 ||
|
|
523
|
-
hasFile((file) =>
|
|
524
|
-
[
|
|
525
|
-
"package-lock.json",
|
|
526
|
-
"pnpm-lock.yaml",
|
|
527
|
-
"yarn.lock",
|
|
528
|
-
"poetry.lock",
|
|
529
|
-
"Pipfile.lock",
|
|
530
|
-
"Cargo.lock",
|
|
531
|
-
].some((candidate) => file.endsWith(candidate))
|
|
532
|
-
);
|
|
533
|
-
if (hasSupplyChain) {
|
|
534
|
-
addSurface("supply_chain", "Dependency manifests/lockfiles detected.");
|
|
535
|
-
}
|
|
536
|
-
|
|
537
|
-
const hasObservability =
|
|
538
|
-
hasFramework("observability-js") ||
|
|
539
|
-
hasFile((file) => /sentry|opentelemetry|prometheus|grafana/i.test(file));
|
|
540
|
-
if (hasObservability) {
|
|
541
|
-
addSurface("observability", "Observability tooling indicators detected.");
|
|
542
|
-
}
|
|
543
|
-
|
|
544
|
-
const hasAiPipeline = hasFile((file) => /(^|\/)(prompts|models|llm|agents?)\//i.test(file));
|
|
545
|
-
if (hasAiPipeline) {
|
|
546
|
-
addSurface("ai_pipeline", "AI/agent pipeline assets detected.");
|
|
547
|
-
}
|
|
548
|
-
|
|
549
|
-
const hasDocs = hasFile((file) => file.endsWith(".md") || file.startsWith("docs/"));
|
|
550
|
-
if (hasDocs) {
|
|
551
|
-
addSurface("docs_knowledge", "Documentation assets detected.");
|
|
552
|
-
}
|
|
553
|
-
|
|
554
|
-
if (hasInfra || hasObservability || hasRelease) {
|
|
555
|
-
addSurface("reliability_sre", "Operational and deployment assets detected.");
|
|
556
|
-
}
|
|
557
|
-
|
|
558
|
-
return [...surfaces.entries()]
|
|
559
|
-
.map(([surface, reason]) => ({ surface, reason }))
|
|
560
|
-
.sort((left, right) => left.surface.localeCompare(right.surface));
|
|
561
|
-
}
|
|
562
|
-
|
|
563
|
-
function summarizeLanguageStats(languageStats, totalLoc) {
|
|
564
|
-
return Object.entries(languageStats)
|
|
565
|
-
.map(([language, stats]) => ({
|
|
566
|
-
language,
|
|
567
|
-
files: stats.files,
|
|
568
|
-
loc: stats.loc,
|
|
569
|
-
locShare: totalLoc > 0 ? Number((stats.loc / totalLoc).toFixed(4)) : 0,
|
|
570
|
-
}))
|
|
571
|
-
.sort((left, right) => right.loc - left.loc || left.language.localeCompare(right.language));
|
|
572
|
-
}
|
|
573
|
-
|
|
574
|
-
async function listTopLevel(rootPath, ignoreMatcher) {
|
|
575
|
-
const dirs = [];
|
|
576
|
-
const files = [];
|
|
577
|
-
let entries = [];
|
|
578
|
-
try {
|
|
579
|
-
entries = await fsp.readdir(rootPath, { withFileTypes: true });
|
|
580
|
-
} catch {
|
|
581
|
-
return { directories: dirs, files };
|
|
582
|
-
}
|
|
583
|
-
|
|
584
|
-
for (const entry of entries) {
|
|
585
|
-
const name = String(entry.name || "");
|
|
586
|
-
if (!name) continue;
|
|
587
|
-
if (DEFAULT_IGNORED_DIRS.has(name)) continue;
|
|
588
|
-
if (ignoreMatcher.ignores(name, entry.isDirectory())) continue;
|
|
589
|
-
if (entry.isDirectory()) {
|
|
590
|
-
dirs.push(name);
|
|
591
|
-
} else if (entry.isFile()) {
|
|
592
|
-
files.push(name);
|
|
593
|
-
}
|
|
594
|
-
}
|
|
595
|
-
|
|
596
|
-
return {
|
|
597
|
-
directories: dirs.sort((left, right) => left.localeCompare(right)).slice(0, 200),
|
|
598
|
-
files: files.sort((left, right) => left.localeCompare(right)).slice(0, 200),
|
|
599
|
-
};
|
|
600
|
-
}
|
|
601
|
-
|
|
602
|
-
export async function collectCodebaseIngest({ rootPath = process.cwd() } = {}) {
|
|
603
|
-
const resolvedRoot = path.resolve(rootPath);
|
|
604
|
-
const ignoreMatcher = await createIgnoreMatcher(resolvedRoot);
|
|
605
|
-
const topLevel = await listTopLevel(resolvedRoot, ignoreMatcher);
|
|
606
|
-
const fingerprintHasher = createHash("sha256");
|
|
607
|
-
let fingerprintFilesCount = 0;
|
|
608
|
-
let latestFileMtimeMs = 0;
|
|
609
|
-
|
|
610
|
-
const stack = [resolvedRoot];
|
|
611
|
-
const fileSet = new Set();
|
|
612
|
-
const languageStats = {};
|
|
613
|
-
const manifests = {};
|
|
614
|
-
|
|
615
|
-
const indexedFiles = [];
|
|
616
|
-
let indexedOmittedCount = 0;
|
|
617
|
-
let filesScanned = 0;
|
|
618
|
-
let directoriesScanned = 0;
|
|
619
|
-
let totalLoc = 0;
|
|
620
|
-
let totalBytes = 0;
|
|
621
|
-
|
|
622
|
-
while (stack.length > 0) {
|
|
623
|
-
const current = stack.pop();
|
|
624
|
-
if (!current) continue;
|
|
625
|
-
|
|
626
|
-
let entries = [];
|
|
627
|
-
try {
|
|
628
|
-
entries = await fsp.readdir(current, { withFileTypes: true });
|
|
629
|
-
} catch {
|
|
630
|
-
continue;
|
|
631
|
-
}
|
|
632
|
-
|
|
633
|
-
directoriesScanned += 1;
|
|
634
|
-
|
|
635
|
-
for (const entry of entries) {
|
|
636
|
-
const fullPath = path.join(current, entry.name);
|
|
637
|
-
const relativePath = toPosixPath(path.relative(resolvedRoot, fullPath));
|
|
638
|
-
|
|
639
|
-
if (entry.isDirectory()) {
|
|
640
|
-
if (!relativePath) {
|
|
641
|
-
continue;
|
|
642
|
-
}
|
|
643
|
-
if (DEFAULT_IGNORED_DIRS.has(entry.name)) {
|
|
644
|
-
continue;
|
|
645
|
-
}
|
|
646
|
-
if (ignoreMatcher.ignores(relativePath, true)) {
|
|
647
|
-
continue;
|
|
648
|
-
}
|
|
649
|
-
stack.push(fullPath);
|
|
650
|
-
continue;
|
|
651
|
-
}
|
|
652
|
-
|
|
653
|
-
if (!entry.isFile()) {
|
|
654
|
-
continue;
|
|
655
|
-
}
|
|
656
|
-
if (ignoreMatcher.ignores(relativePath, false)) {
|
|
657
|
-
continue;
|
|
658
|
-
}
|
|
659
|
-
|
|
660
|
-
let stat;
|
|
661
|
-
try {
|
|
662
|
-
stat = await fsp.stat(fullPath);
|
|
663
|
-
} catch {
|
|
664
|
-
continue;
|
|
665
|
-
}
|
|
666
|
-
if (!stat || stat.size > MAX_FILE_SIZE_BYTES) {
|
|
667
|
-
continue;
|
|
668
|
-
}
|
|
669
|
-
|
|
670
|
-
appendFingerprintInput(fingerprintHasher, relativePath, stat.size, stat.mtimeMs);
|
|
671
|
-
fingerprintFilesCount += 1;
|
|
672
|
-
latestFileMtimeMs = Math.max(latestFileMtimeMs, normalizeMtimeMs(stat.mtimeMs));
|
|
673
|
-
|
|
674
|
-
let text = "";
|
|
675
|
-
try {
|
|
676
|
-
text = await fsp.readFile(fullPath, "utf-8");
|
|
677
|
-
} catch {
|
|
678
|
-
continue;
|
|
679
|
-
}
|
|
680
|
-
|
|
681
|
-
const loc = countLoc(text);
|
|
682
|
-
const language = detectLanguage(relativePath);
|
|
683
|
-
|
|
684
|
-
filesScanned += 1;
|
|
685
|
-
totalLoc += loc;
|
|
686
|
-
totalBytes += stat.size;
|
|
687
|
-
fileSet.add(relativePath);
|
|
688
|
-
|
|
689
|
-
if (!languageStats[language]) {
|
|
690
|
-
languageStats[language] = { files: 0, loc: 0 };
|
|
691
|
-
}
|
|
692
|
-
languageStats[language].files += 1;
|
|
693
|
-
languageStats[language].loc += loc;
|
|
694
|
-
|
|
695
|
-
const baseName = path.basename(relativePath);
|
|
696
|
-
if (MANIFEST_CANDIDATES.has(baseName)) {
|
|
697
|
-
manifests[baseName] = text;
|
|
698
|
-
}
|
|
699
|
-
|
|
700
|
-
if (indexedFiles.length < FILE_INDEX_LIMIT) {
|
|
701
|
-
indexedFiles.push({
|
|
702
|
-
path: relativePath,
|
|
703
|
-
language,
|
|
704
|
-
loc,
|
|
705
|
-
sizeBytes: stat.size,
|
|
706
|
-
});
|
|
707
|
-
} else {
|
|
708
|
-
indexedOmittedCount += 1;
|
|
709
|
-
}
|
|
710
|
-
}
|
|
711
|
-
}
|
|
712
|
-
|
|
713
|
-
const frameworks = detectFrameworks(manifests);
|
|
714
|
-
const packageMetadata = derivePackageMetadata(manifests);
|
|
715
|
-
const entryPoints = deriveEntryPoints(fileSet, manifests);
|
|
716
|
-
const riskSurfaces = deriveRiskSurfaces({
|
|
717
|
-
fileSet,
|
|
718
|
-
frameworks,
|
|
719
|
-
manifests,
|
|
720
|
-
languageStats,
|
|
721
|
-
});
|
|
722
|
-
|
|
723
|
-
return {
|
|
724
|
-
schemaVersion: "1.0.0",
|
|
725
|
-
generatedAt: new Date().toISOString(),
|
|
726
|
-
rootPath: resolvedRoot,
|
|
727
|
-
summary: {
|
|
728
|
-
filesScanned,
|
|
729
|
-
directoriesScanned,
|
|
730
|
-
totalLoc,
|
|
731
|
-
totalBytes,
|
|
732
|
-
},
|
|
733
|
-
topLevel,
|
|
734
|
-
manifests: {
|
|
735
|
-
detected: Object.keys(manifests).sort((left, right) => left.localeCompare(right)),
|
|
736
|
-
},
|
|
737
|
-
languages: summarizeLanguageStats(languageStats, totalLoc),
|
|
738
|
-
frameworks,
|
|
739
|
-
packageMetadata,
|
|
740
|
-
entryPoints,
|
|
741
|
-
riskSurfaces,
|
|
742
|
-
indexedFiles: {
|
|
743
|
-
limit: FILE_INDEX_LIMIT,
|
|
744
|
-
omitted: indexedOmittedCount,
|
|
745
|
-
files: indexedFiles,
|
|
746
|
-
},
|
|
747
|
-
cache: {
|
|
748
|
-
schema: INGEST_CACHE_SCHEMA,
|
|
749
|
-
contentHash: fingerprintHasher.digest("hex"),
|
|
750
|
-
filesCount: fingerprintFilesCount,
|
|
751
|
-
latestFileMtimeMs,
|
|
752
|
-
},
|
|
753
|
-
};
|
|
754
|
-
}
|
|
755
|
-
|
|
756
|
-
export function formatIngestSummary(ingest) {
|
|
757
|
-
const summary = ingest && ingest.summary ? ingest.summary : {};
|
|
758
|
-
const languageHead = Array.isArray(ingest.languages)
|
|
759
|
-
? ingest.languages
|
|
760
|
-
.slice(0, 5)
|
|
761
|
-
.map((item) => `${item.language}(${item.files} files/${item.loc} LOC)`)
|
|
762
|
-
.join(", ")
|
|
763
|
-
: "none";
|
|
764
|
-
const frameworks = Array.isArray(ingest.frameworks) && ingest.frameworks.length
|
|
765
|
-
? ingest.frameworks.join(", ")
|
|
766
|
-
: "none";
|
|
767
|
-
const entryPoints = Array.isArray(ingest.entryPoints) && ingest.entryPoints.length
|
|
768
|
-
? ingest.entryPoints.join(", ")
|
|
769
|
-
: "none";
|
|
770
|
-
const packageName = String(ingest.packageMetadata?.name || "").trim();
|
|
771
|
-
const packageScripts = Array.isArray(ingest.packageMetadata?.scripts)
|
|
772
|
-
? ingest.packageMetadata.scripts
|
|
773
|
-
: [];
|
|
774
|
-
|
|
775
|
-
const lines = [
|
|
776
|
-
`Workspace path: ${ingest.rootPath}`,
|
|
777
|
-
`Top-level directories: ${(ingest.topLevel?.directories || []).slice(0, 20).join(", ") || "none"}`,
|
|
778
|
-
`Top-level files: ${(ingest.topLevel?.files || []).slice(0, 20).join(", ") || "none"}`,
|
|
779
|
-
`Files scanned: ${summary.filesScanned || 0}`,
|
|
780
|
-
`Total LOC: ${summary.totalLoc || 0}`,
|
|
781
|
-
`Languages: ${languageHead}`,
|
|
782
|
-
`Frameworks: ${frameworks}`,
|
|
783
|
-
`Entry points: ${entryPoints}`,
|
|
784
|
-
];
|
|
785
|
-
|
|
786
|
-
if (packageName) {
|
|
787
|
-
lines.push(`package.json name: ${packageName}`);
|
|
788
|
-
}
|
|
789
|
-
if (packageScripts.length > 0) {
|
|
790
|
-
lines.push(`package scripts: ${packageScripts.slice(0, 15).join(", ")}`);
|
|
791
|
-
}
|
|
792
|
-
|
|
793
|
-
return lines.join("\n");
|
|
794
|
-
}
|
|
795
|
-
|
|
796
|
-
export async function writeCodebaseIngest({ ingest, rootPath, outputFile = "", outputDir = "" } = {}) {
|
|
797
|
-
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
798
|
-
const resolvedOutputFile = String(outputFile || "").trim();
|
|
799
|
-
const outputPath = resolvedOutputFile
|
|
800
|
-
? path.resolve(resolvedRoot, resolvedOutputFile)
|
|
801
|
-
: path.join(
|
|
802
|
-
await resolveOutputRoot({
|
|
803
|
-
cwd: resolvedRoot,
|
|
804
|
-
outputDirOverride: outputDir,
|
|
805
|
-
}),
|
|
806
|
-
"CODEBASE_INGEST.json"
|
|
807
|
-
);
|
|
808
|
-
|
|
809
|
-
await fsp.mkdir(path.dirname(outputPath), { recursive: true });
|
|
810
|
-
await fsp.writeFile(outputPath, `${JSON.stringify(ingest, null, 2)}\n`, "utf-8");
|
|
811
|
-
return outputPath;
|
|
812
|
-
}
|
|
813
|
-
|
|
814
|
-
export async function resolveCodebaseIngest({
|
|
815
|
-
rootPath = process.cwd(),
|
|
816
|
-
outputFile = "",
|
|
817
|
-
outputDir = "",
|
|
818
|
-
refresh = false,
|
|
819
|
-
} = {}) {
|
|
820
|
-
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
821
|
-
const outputPath = await resolveIngestOutputPath({
|
|
822
|
-
rootPath: resolvedRoot,
|
|
823
|
-
outputFile,
|
|
824
|
-
outputDir,
|
|
825
|
-
});
|
|
826
|
-
const existingIngest = await readExistingIngest(outputPath);
|
|
827
|
-
const fingerprint = await computeCodebaseContentFingerprint({
|
|
828
|
-
rootPath: resolvedRoot,
|
|
829
|
-
});
|
|
830
|
-
const lastCommitAt = await readGitLastCommitAt(resolvedRoot);
|
|
831
|
-
const staleness = buildIngestStaleness({
|
|
832
|
-
existingIngest,
|
|
833
|
-
fingerprint,
|
|
834
|
-
lastCommitAt,
|
|
835
|
-
});
|
|
836
|
-
const staleBeforeRefresh = staleness.stale;
|
|
837
|
-
|
|
838
|
-
let ingest = existingIngest;
|
|
839
|
-
let refreshed = false;
|
|
840
|
-
let refreshedBecause = "";
|
|
841
|
-
if (!existingIngest) {
|
|
842
|
-
refreshed = true;
|
|
843
|
-
refreshedBecause = "missing_ingest";
|
|
844
|
-
} else if (refresh) {
|
|
845
|
-
refreshed = true;
|
|
846
|
-
refreshedBecause = "refresh_requested";
|
|
847
|
-
}
|
|
848
|
-
|
|
849
|
-
if (refreshed) {
|
|
850
|
-
ingest = await collectCodebaseIngest({
|
|
851
|
-
rootPath: resolvedRoot,
|
|
852
|
-
});
|
|
853
|
-
ingest.generatedAt = new Date().toISOString();
|
|
854
|
-
if (!ingest.cache || typeof ingest.cache !== "object") {
|
|
855
|
-
ingest.cache = {};
|
|
856
|
-
}
|
|
857
|
-
ingest.cache.schema = INGEST_CACHE_SCHEMA;
|
|
858
|
-
ingest.cache.contentHash = fingerprint.contentHash;
|
|
859
|
-
ingest.cache.filesCount = fingerprint.filesCount;
|
|
860
|
-
ingest.cache.latestFileMtimeMs = fingerprint.latestFileMtimeMs;
|
|
861
|
-
await writeCodebaseIngest({
|
|
862
|
-
ingest,
|
|
863
|
-
rootPath: resolvedRoot,
|
|
864
|
-
outputFile,
|
|
865
|
-
outputDir,
|
|
866
|
-
});
|
|
867
|
-
}
|
|
868
|
-
|
|
869
|
-
const resolutionReasons = refreshed
|
|
870
|
-
? [refreshedBecause, ...staleness.reasons].filter(Boolean)
|
|
871
|
-
: staleness.reasons;
|
|
872
|
-
|
|
873
|
-
return {
|
|
874
|
-
ingest,
|
|
875
|
-
outputPath,
|
|
876
|
-
refreshed,
|
|
877
|
-
stale: refreshed ? false : staleness.stale,
|
|
878
|
-
staleBeforeRefresh,
|
|
879
|
-
reasons: resolutionReasons,
|
|
880
|
-
refreshedBecause,
|
|
881
|
-
refreshRequested: Boolean(refresh),
|
|
882
|
-
lastCommitAt,
|
|
883
|
-
fingerprint,
|
|
884
|
-
event:
|
|
885
|
-
refreshed || staleBeforeRefresh
|
|
886
|
-
? {
|
|
887
|
-
event: "ingest_refresh",
|
|
888
|
-
payload: {
|
|
889
|
-
refreshed,
|
|
890
|
-
stale: refreshed ? false : staleness.stale,
|
|
891
|
-
reason:
|
|
892
|
-
refreshedBecause || (staleness.reasons.length > 0 ? staleness.reasons.join(",") : "cache_hit"),
|
|
893
|
-
contentHash: fingerprint.contentHash,
|
|
894
|
-
filesCount: fingerprint.filesCount,
|
|
895
|
-
lastCommitAt,
|
|
896
|
-
},
|
|
897
|
-
}
|
|
898
|
-
: null,
|
|
899
|
-
};
|
|
900
|
-
}
|
|
901
|
-
|
|
902
|
-
export async function generateCodebaseIngest({
|
|
903
|
-
rootPath = process.cwd(),
|
|
904
|
-
outputFile = "",
|
|
905
|
-
outputDir = "",
|
|
906
|
-
} = {}) {
|
|
907
|
-
const ingest = await collectCodebaseIngest({ rootPath });
|
|
908
|
-
const outputPath = await writeCodebaseIngest({
|
|
909
|
-
ingest,
|
|
910
|
-
rootPath,
|
|
911
|
-
outputFile,
|
|
912
|
-
outputDir,
|
|
913
|
-
});
|
|
914
|
-
return {
|
|
915
|
-
ingest,
|
|
916
|
-
outputPath,
|
|
917
|
-
};
|
|
918
|
-
}
|
|
1
|
+
import { execFile } from "node:child_process";
|
|
2
|
+
import { createHash } from "node:crypto";
|
|
3
|
+
import fs from "node:fs";
|
|
4
|
+
import fsp from "node:fs/promises";
|
|
5
|
+
import path from "node:path";
|
|
6
|
+
import process from "node:process";
|
|
7
|
+
import { promisify } from "node:util";
|
|
8
|
+
|
|
9
|
+
import ignore from "ignore";
|
|
10
|
+
|
|
11
|
+
import { resolveOutputRoot } from "../config/service.js";
|
|
12
|
+
|
|
13
|
+
const DEFAULT_IGNORED_DIRS = new Set([
|
|
14
|
+
".git",
|
|
15
|
+
"node_modules",
|
|
16
|
+
".venv",
|
|
17
|
+
".next",
|
|
18
|
+
"dist",
|
|
19
|
+
"build",
|
|
20
|
+
"coverage",
|
|
21
|
+
".sentinelayer",
|
|
22
|
+
".turbo",
|
|
23
|
+
".idea",
|
|
24
|
+
".vscode",
|
|
25
|
+
]);
|
|
26
|
+
|
|
27
|
+
const MAX_FILE_SIZE_BYTES = 1024 * 1024;
|
|
28
|
+
const FILE_INDEX_LIMIT = 5000;
|
|
29
|
+
const execFileAsync = promisify(execFile);
|
|
30
|
+
const INGEST_CACHE_SCHEMA = "path-size-mtime-sha256-v1";
|
|
31
|
+
|
|
32
|
+
const LANGUAGE_BY_EXTENSION = {
|
|
33
|
+
".js": "JavaScript",
|
|
34
|
+
".cjs": "JavaScript",
|
|
35
|
+
".mjs": "JavaScript",
|
|
36
|
+
".ts": "TypeScript",
|
|
37
|
+
".tsx": "TypeScript",
|
|
38
|
+
".jsx": "JavaScript",
|
|
39
|
+
".py": "Python",
|
|
40
|
+
".go": "Go",
|
|
41
|
+
".rs": "Rust",
|
|
42
|
+
".java": "Java",
|
|
43
|
+
".kt": "Kotlin",
|
|
44
|
+
".swift": "Swift",
|
|
45
|
+
".rb": "Ruby",
|
|
46
|
+
".php": "PHP",
|
|
47
|
+
".cs": "C#",
|
|
48
|
+
".cpp": "C++",
|
|
49
|
+
".cc": "C++",
|
|
50
|
+
".c": "C",
|
|
51
|
+
".h": "C/C++ Header",
|
|
52
|
+
".hpp": "C/C++ Header",
|
|
53
|
+
".sql": "SQL",
|
|
54
|
+
".md": "Markdown",
|
|
55
|
+
".yml": "YAML",
|
|
56
|
+
".yaml": "YAML",
|
|
57
|
+
".json": "JSON",
|
|
58
|
+
".toml": "TOML",
|
|
59
|
+
".tf": "Terraform",
|
|
60
|
+
".sh": "Shell",
|
|
61
|
+
".bash": "Shell",
|
|
62
|
+
".ps1": "PowerShell",
|
|
63
|
+
".dockerfile": "Docker",
|
|
64
|
+
};
|
|
65
|
+
|
|
66
|
+
const MANIFEST_CANDIDATES = new Set([
|
|
67
|
+
"package.json",
|
|
68
|
+
"requirements.txt",
|
|
69
|
+
"pyproject.toml",
|
|
70
|
+
"go.mod",
|
|
71
|
+
"Cargo.toml",
|
|
72
|
+
"Gemfile",
|
|
73
|
+
"pom.xml",
|
|
74
|
+
"build.gradle",
|
|
75
|
+
"build.gradle.kts",
|
|
76
|
+
"composer.json",
|
|
77
|
+
]);
|
|
78
|
+
|
|
79
|
+
const ENTRY_POINT_CANDIDATES = [
|
|
80
|
+
"src/index.ts",
|
|
81
|
+
"src/index.tsx",
|
|
82
|
+
"src/index.js",
|
|
83
|
+
"src/main.ts",
|
|
84
|
+
"src/main.js",
|
|
85
|
+
"src/server.ts",
|
|
86
|
+
"src/server.js",
|
|
87
|
+
"index.ts",
|
|
88
|
+
"index.js",
|
|
89
|
+
"main.ts",
|
|
90
|
+
"main.js",
|
|
91
|
+
"main.py",
|
|
92
|
+
"app.py",
|
|
93
|
+
"server.py",
|
|
94
|
+
"cmd/main.go",
|
|
95
|
+
"src/main.rs",
|
|
96
|
+
];
|
|
97
|
+
|
|
98
|
+
function toPosixPath(value) {
|
|
99
|
+
return String(value || "").replace(/\\/g, "/");
|
|
100
|
+
}
|
|
101
|
+
|
|
102
|
+
function parseIsoToEpoch(value) {
|
|
103
|
+
const normalized = String(value || "").trim();
|
|
104
|
+
if (!normalized) {
|
|
105
|
+
return null;
|
|
106
|
+
}
|
|
107
|
+
const epoch = Date.parse(normalized);
|
|
108
|
+
if (!Number.isFinite(epoch)) {
|
|
109
|
+
return null;
|
|
110
|
+
}
|
|
111
|
+
return epoch;
|
|
112
|
+
}
|
|
113
|
+
|
|
114
|
+
function normalizeMtimeMs(value) {
|
|
115
|
+
const normalized = Number(value || 0);
|
|
116
|
+
if (!Number.isFinite(normalized) || normalized < 0) {
|
|
117
|
+
return 0;
|
|
118
|
+
}
|
|
119
|
+
return Math.floor(normalized);
|
|
120
|
+
}
|
|
121
|
+
|
|
122
|
+
function appendFingerprintInput(hasher, relativePath, sizeBytes, mtimeMs) {
|
|
123
|
+
hasher.update(
|
|
124
|
+
`${toPosixPath(relativePath)}\u001f${String(Number(sizeBytes || 0))}\u001f${normalizeMtimeMs(
|
|
125
|
+
mtimeMs
|
|
126
|
+
)}\n`,
|
|
127
|
+
"utf-8"
|
|
128
|
+
);
|
|
129
|
+
}
|
|
130
|
+
|
|
131
|
+
function countLoc(text) {
|
|
132
|
+
return String(text || "")
|
|
133
|
+
.split(/\r?\n/)
|
|
134
|
+
.filter((line) => line.trim().length > 0).length;
|
|
135
|
+
}
|
|
136
|
+
|
|
137
|
+
function detectLanguage(relativePath) {
|
|
138
|
+
const normalized = toPosixPath(relativePath);
|
|
139
|
+
const baseName = path.basename(normalized).toLowerCase();
|
|
140
|
+
if (baseName === "dockerfile") {
|
|
141
|
+
return "Docker";
|
|
142
|
+
}
|
|
143
|
+
const extension = path.extname(baseName);
|
|
144
|
+
return LANGUAGE_BY_EXTENSION[extension] || "Other";
|
|
145
|
+
}
|
|
146
|
+
|
|
147
|
+
async function readIgnorePatterns(filePath) {
|
|
148
|
+
try {
|
|
149
|
+
const raw = await fsp.readFile(filePath, "utf-8");
|
|
150
|
+
return String(raw || "")
|
|
151
|
+
.split(/\r?\n/)
|
|
152
|
+
.map((line) => line.trim())
|
|
153
|
+
.filter((line) => line && !line.startsWith("#"));
|
|
154
|
+
} catch (error) {
|
|
155
|
+
if (error && error.code === "ENOENT") {
|
|
156
|
+
return [];
|
|
157
|
+
}
|
|
158
|
+
throw error;
|
|
159
|
+
}
|
|
160
|
+
}
|
|
161
|
+
|
|
162
|
+
async function createIgnoreMatcher(rootPath) {
|
|
163
|
+
const matcher = ignore();
|
|
164
|
+
const gitignorePatterns = await readIgnorePatterns(path.join(rootPath, ".gitignore"));
|
|
165
|
+
const sentinelPatterns = await readIgnorePatterns(path.join(rootPath, ".sentinelayerignore"));
|
|
166
|
+
matcher.add([...gitignorePatterns, ...sentinelPatterns]);
|
|
167
|
+
|
|
168
|
+
return {
|
|
169
|
+
ignores(relativePath, isDirectory) {
|
|
170
|
+
const normalized = toPosixPath(relativePath);
|
|
171
|
+
if (!normalized) {
|
|
172
|
+
return false;
|
|
173
|
+
}
|
|
174
|
+
const candidate = isDirectory ? `${normalized}/` : normalized;
|
|
175
|
+
return matcher.ignores(candidate);
|
|
176
|
+
},
|
|
177
|
+
};
|
|
178
|
+
}
|
|
179
|
+
|
|
180
|
+
async function computeCodebaseContentFingerprint({ rootPath }) {
|
|
181
|
+
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
182
|
+
const ignoreMatcher = await createIgnoreMatcher(resolvedRoot);
|
|
183
|
+
const stack = [resolvedRoot];
|
|
184
|
+
const hasher = createHash("sha256");
|
|
185
|
+
let filesCount = 0;
|
|
186
|
+
let latestFileMtimeMs = 0;
|
|
187
|
+
|
|
188
|
+
while (stack.length > 0) {
|
|
189
|
+
const current = stack.pop();
|
|
190
|
+
if (!current) {
|
|
191
|
+
continue;
|
|
192
|
+
}
|
|
193
|
+
|
|
194
|
+
let entries = [];
|
|
195
|
+
try {
|
|
196
|
+
entries = await fsp.readdir(current, { withFileTypes: true });
|
|
197
|
+
} catch {
|
|
198
|
+
continue;
|
|
199
|
+
}
|
|
200
|
+
|
|
201
|
+
for (const entry of entries) {
|
|
202
|
+
const fullPath = path.join(current, entry.name);
|
|
203
|
+
const relativePath = toPosixPath(path.relative(resolvedRoot, fullPath));
|
|
204
|
+
|
|
205
|
+
if (entry.isDirectory()) {
|
|
206
|
+
if (!relativePath) {
|
|
207
|
+
continue;
|
|
208
|
+
}
|
|
209
|
+
if (DEFAULT_IGNORED_DIRS.has(entry.name)) {
|
|
210
|
+
continue;
|
|
211
|
+
}
|
|
212
|
+
if (ignoreMatcher.ignores(relativePath, true)) {
|
|
213
|
+
continue;
|
|
214
|
+
}
|
|
215
|
+
stack.push(fullPath);
|
|
216
|
+
continue;
|
|
217
|
+
}
|
|
218
|
+
|
|
219
|
+
if (!entry.isFile()) {
|
|
220
|
+
continue;
|
|
221
|
+
}
|
|
222
|
+
if (ignoreMatcher.ignores(relativePath, false)) {
|
|
223
|
+
continue;
|
|
224
|
+
}
|
|
225
|
+
|
|
226
|
+
let stat = null;
|
|
227
|
+
try {
|
|
228
|
+
stat = await fsp.stat(fullPath);
|
|
229
|
+
} catch {
|
|
230
|
+
stat = null;
|
|
231
|
+
}
|
|
232
|
+
if (!stat || stat.size > MAX_FILE_SIZE_BYTES) {
|
|
233
|
+
continue;
|
|
234
|
+
}
|
|
235
|
+
|
|
236
|
+
filesCount += 1;
|
|
237
|
+
latestFileMtimeMs = Math.max(latestFileMtimeMs, normalizeMtimeMs(stat.mtimeMs));
|
|
238
|
+
appendFingerprintInput(hasher, relativePath, stat.size, stat.mtimeMs);
|
|
239
|
+
}
|
|
240
|
+
}
|
|
241
|
+
|
|
242
|
+
return {
|
|
243
|
+
schema: INGEST_CACHE_SCHEMA,
|
|
244
|
+
contentHash: hasher.digest("hex"),
|
|
245
|
+
filesCount,
|
|
246
|
+
latestFileMtimeMs,
|
|
247
|
+
};
|
|
248
|
+
}
|
|
249
|
+
|
|
250
|
+
async function readExistingIngest(outputPath) {
|
|
251
|
+
try {
|
|
252
|
+
const parsed = JSON.parse(await fsp.readFile(outputPath, "utf-8"));
|
|
253
|
+
return parsed && typeof parsed === "object" ? parsed : null;
|
|
254
|
+
} catch (error) {
|
|
255
|
+
if (error && error.code === "ENOENT") {
|
|
256
|
+
return null;
|
|
257
|
+
}
|
|
258
|
+
throw error;
|
|
259
|
+
}
|
|
260
|
+
}
|
|
261
|
+
|
|
262
|
+
async function resolveIngestOutputPath({ rootPath, outputFile = "", outputDir = "" }) {
|
|
263
|
+
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
264
|
+
const explicitOutputFile = String(outputFile || "").trim();
|
|
265
|
+
if (explicitOutputFile) {
|
|
266
|
+
return path.resolve(resolvedRoot, explicitOutputFile);
|
|
267
|
+
}
|
|
268
|
+
const outputRoot = await resolveOutputRoot({
|
|
269
|
+
cwd: resolvedRoot,
|
|
270
|
+
outputDirOverride: outputDir,
|
|
271
|
+
});
|
|
272
|
+
return path.join(outputRoot, "CODEBASE_INGEST.json");
|
|
273
|
+
}
|
|
274
|
+
|
|
275
|
+
async function readGitLastCommitAt(rootPath) {
|
|
276
|
+
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
277
|
+
try {
|
|
278
|
+
const { stdout } = await execFileAsync("git", [
|
|
279
|
+
"-C",
|
|
280
|
+
resolvedRoot,
|
|
281
|
+
"log",
|
|
282
|
+
"-1",
|
|
283
|
+
"--format=%cI",
|
|
284
|
+
]);
|
|
285
|
+
const normalized = String(stdout || "").trim();
|
|
286
|
+
return parseIsoToEpoch(normalized) === null ? "" : normalized;
|
|
287
|
+
} catch {
|
|
288
|
+
return "";
|
|
289
|
+
}
|
|
290
|
+
}
|
|
291
|
+
|
|
292
|
+
function buildIngestStaleness({ existingIngest, fingerprint, lastCommitAt }) {
|
|
293
|
+
if (!existingIngest) {
|
|
294
|
+
return {
|
|
295
|
+
stale: true,
|
|
296
|
+
reasons: ["missing_ingest"],
|
|
297
|
+
};
|
|
298
|
+
}
|
|
299
|
+
|
|
300
|
+
const reasons = [];
|
|
301
|
+
const generatedAtEpoch = parseIsoToEpoch(existingIngest.generatedAt);
|
|
302
|
+
const lastCommitEpoch = parseIsoToEpoch(lastCommitAt);
|
|
303
|
+
if (generatedAtEpoch === null) {
|
|
304
|
+
reasons.push("invalid_generated_at");
|
|
305
|
+
} else if (lastCommitEpoch !== null && generatedAtEpoch < lastCommitEpoch) {
|
|
306
|
+
reasons.push("older_than_last_commit");
|
|
307
|
+
}
|
|
308
|
+
|
|
309
|
+
const existingContentHash = String(existingIngest.cache?.contentHash || "").trim();
|
|
310
|
+
if (existingContentHash && existingContentHash !== fingerprint.contentHash) {
|
|
311
|
+
reasons.push("content_hash_mismatch");
|
|
312
|
+
} else if (!existingContentHash) {
|
|
313
|
+
reasons.push("missing_content_hash");
|
|
314
|
+
}
|
|
315
|
+
|
|
316
|
+
return {
|
|
317
|
+
stale: reasons.length > 0,
|
|
318
|
+
reasons,
|
|
319
|
+
};
|
|
320
|
+
}
|
|
321
|
+
|
|
322
|
+
export function formatIngestResolutionNotice(resolution = {}) {
|
|
323
|
+
const reasons = Array.isArray(resolution.reasons) ? resolution.reasons : [];
|
|
324
|
+
if (resolution.refreshed) {
|
|
325
|
+
return `ingest refreshed (${reasons.join(", ") || "requested"})`;
|
|
326
|
+
}
|
|
327
|
+
if (resolution.stale) {
|
|
328
|
+
return `ingest stale (${reasons.join(", ") || "unknown"}); re-run with --refresh`;
|
|
329
|
+
}
|
|
330
|
+
return "ingest cache hit";
|
|
331
|
+
}
|
|
332
|
+
|
|
333
|
+
function safeJsonParse(raw) {
|
|
334
|
+
try {
|
|
335
|
+
return JSON.parse(raw);
|
|
336
|
+
} catch {
|
|
337
|
+
return null;
|
|
338
|
+
}
|
|
339
|
+
}
|
|
340
|
+
|
|
341
|
+
function normalizeDependencySet(dependencies) {
|
|
342
|
+
if (!dependencies || typeof dependencies !== "object") {
|
|
343
|
+
return new Set();
|
|
344
|
+
}
|
|
345
|
+
return new Set(Object.keys(dependencies).map((value) => String(value || "").toLowerCase()));
|
|
346
|
+
}
|
|
347
|
+
|
|
348
|
+
function detectFrameworks(manifests) {
|
|
349
|
+
const frameworks = new Set();
|
|
350
|
+
|
|
351
|
+
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
352
|
+
if (packageJson) {
|
|
353
|
+
const deps = normalizeDependencySet({
|
|
354
|
+
...(packageJson.dependencies || {}),
|
|
355
|
+
...(packageJson.devDependencies || {}),
|
|
356
|
+
...(packageJson.peerDependencies || {}),
|
|
357
|
+
});
|
|
358
|
+
if (deps.has("next")) frameworks.add("nextjs");
|
|
359
|
+
if (deps.has("react")) frameworks.add("react");
|
|
360
|
+
if (deps.has("vue")) frameworks.add("vue");
|
|
361
|
+
if (deps.has("svelte")) frameworks.add("svelte");
|
|
362
|
+
if (deps.has("express")) frameworks.add("express");
|
|
363
|
+
if (deps.has("fastify")) frameworks.add("fastify");
|
|
364
|
+
if (deps.has("hono")) frameworks.add("hono");
|
|
365
|
+
if (deps.has("@nestjs/core")) frameworks.add("nestjs");
|
|
366
|
+
if (deps.has("prisma")) frameworks.add("prisma");
|
|
367
|
+
if (deps.has("typeorm")) frameworks.add("typeorm");
|
|
368
|
+
if (deps.has("drizzle-orm")) frameworks.add("drizzle");
|
|
369
|
+
if (deps.has("playwright")) frameworks.add("playwright");
|
|
370
|
+
if (deps.has("jest")) frameworks.add("jest");
|
|
371
|
+
if (deps.has("vitest")) frameworks.add("vitest");
|
|
372
|
+
if (deps.has("@opentelemetry/api") || deps.has("@sentry/node")) frameworks.add("observability-js");
|
|
373
|
+
}
|
|
374
|
+
|
|
375
|
+
const requirementsText = String(manifests["requirements.txt"] || "").toLowerCase();
|
|
376
|
+
if (/\bfastapi\b/.test(requirementsText)) frameworks.add("fastapi");
|
|
377
|
+
if (/\bdjango\b/.test(requirementsText)) frameworks.add("django");
|
|
378
|
+
if (/\bflask\b/.test(requirementsText)) frameworks.add("flask");
|
|
379
|
+
|
|
380
|
+
const pyprojectText = String(manifests["pyproject.toml"] || "").toLowerCase();
|
|
381
|
+
if (/\bfastapi\b/.test(pyprojectText)) frameworks.add("fastapi");
|
|
382
|
+
if (/\bdjango\b/.test(pyprojectText)) frameworks.add("django");
|
|
383
|
+
if (/\bflask\b/.test(pyprojectText)) frameworks.add("flask");
|
|
384
|
+
|
|
385
|
+
const goModText = String(manifests["go.mod"] || "").toLowerCase();
|
|
386
|
+
if (/gin-gonic\/gin/.test(goModText)) frameworks.add("gin");
|
|
387
|
+
if (/gofiber\/fiber/.test(goModText)) frameworks.add("fiber");
|
|
388
|
+
if (/labstack\/echo/.test(goModText)) frameworks.add("echo");
|
|
389
|
+
|
|
390
|
+
const cargoText = String(manifests["Cargo.toml"] || "").toLowerCase();
|
|
391
|
+
if (/\baxum\b/.test(cargoText)) frameworks.add("axum");
|
|
392
|
+
if (/\bactix-web\b/.test(cargoText)) frameworks.add("actix-web");
|
|
393
|
+
|
|
394
|
+
const gemfileText = String(manifests.Gemfile || "").toLowerCase();
|
|
395
|
+
if (/\brails\b/.test(gemfileText)) frameworks.add("rails");
|
|
396
|
+
|
|
397
|
+
return [...frameworks].sort((left, right) => left.localeCompare(right));
|
|
398
|
+
}
|
|
399
|
+
|
|
400
|
+
function derivePackageMetadata(manifests) {
|
|
401
|
+
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
402
|
+
if (!packageJson || typeof packageJson !== "object") {
|
|
403
|
+
return {
|
|
404
|
+
name: "",
|
|
405
|
+
scripts: [],
|
|
406
|
+
};
|
|
407
|
+
}
|
|
408
|
+
|
|
409
|
+
const scripts =
|
|
410
|
+
packageJson.scripts && typeof packageJson.scripts === "object"
|
|
411
|
+
? Object.keys(packageJson.scripts)
|
|
412
|
+
.map((value) => String(value || "").trim())
|
|
413
|
+
.filter(Boolean)
|
|
414
|
+
.sort((left, right) => left.localeCompare(right))
|
|
415
|
+
: [];
|
|
416
|
+
|
|
417
|
+
return {
|
|
418
|
+
name: String(packageJson.name || "").trim(),
|
|
419
|
+
scripts,
|
|
420
|
+
};
|
|
421
|
+
}
|
|
422
|
+
|
|
423
|
+
function deriveEntryPoints(fileSet, manifests) {
|
|
424
|
+
const entryPoints = new Set();
|
|
425
|
+
for (const candidate of ENTRY_POINT_CANDIDATES) {
|
|
426
|
+
if (fileSet.has(candidate)) {
|
|
427
|
+
entryPoints.add(candidate);
|
|
428
|
+
}
|
|
429
|
+
}
|
|
430
|
+
|
|
431
|
+
const packageJson = manifests["package.json"] ? safeJsonParse(manifests["package.json"]) : null;
|
|
432
|
+
if (packageJson) {
|
|
433
|
+
if (typeof packageJson.main === "string" && packageJson.main.trim()) {
|
|
434
|
+
entryPoints.add(packageJson.main.trim());
|
|
435
|
+
}
|
|
436
|
+
if (packageJson.bin && typeof packageJson.bin === "object") {
|
|
437
|
+
for (const value of Object.values(packageJson.bin)) {
|
|
438
|
+
const normalized = String(value || "").trim();
|
|
439
|
+
if (normalized) {
|
|
440
|
+
entryPoints.add(normalized);
|
|
441
|
+
}
|
|
442
|
+
}
|
|
443
|
+
}
|
|
444
|
+
}
|
|
445
|
+
|
|
446
|
+
return [...entryPoints].sort((left, right) => left.localeCompare(right));
|
|
447
|
+
}
|
|
448
|
+
|
|
449
|
+
function deriveRiskSurfaces({ fileSet, frameworks, manifests, languageStats }) {
|
|
450
|
+
const surfaces = new Map();
|
|
451
|
+
|
|
452
|
+
const addSurface = (surface, reason) => {
|
|
453
|
+
if (!surfaces.has(surface)) {
|
|
454
|
+
surfaces.set(surface, reason);
|
|
455
|
+
}
|
|
456
|
+
};
|
|
457
|
+
|
|
458
|
+
const hasFile = (predicate) => [...fileSet].some(predicate);
|
|
459
|
+
const hasFramework = (name) => frameworks.includes(name);
|
|
460
|
+
|
|
461
|
+
addSurface("code_quality", "Source files detected.");
|
|
462
|
+
addSurface("security_overlay", "Credential/policy scanning is applicable for any repository ingest.");
|
|
463
|
+
|
|
464
|
+
const hasTests = hasFile((file) => /(^|\/)(test|tests|__tests__)\//.test(file) || /\.(test|spec)\./.test(file));
|
|
465
|
+
if (hasTests || hasFramework("jest") || hasFramework("vitest") || hasFramework("playwright")) {
|
|
466
|
+
addSurface("testing_correctness", "Test assets detected.");
|
|
467
|
+
}
|
|
468
|
+
|
|
469
|
+
const hasFrontend =
|
|
470
|
+
hasFramework("nextjs") ||
|
|
471
|
+
hasFramework("react") ||
|
|
472
|
+
hasFramework("vue") ||
|
|
473
|
+
hasFramework("svelte") ||
|
|
474
|
+
languageStats.JavaScript ||
|
|
475
|
+
languageStats.TypeScript;
|
|
476
|
+
if (hasFrontend) {
|
|
477
|
+
addSurface("frontend_runtime", "Frontend/runtime JavaScript stack detected.");
|
|
478
|
+
}
|
|
479
|
+
|
|
480
|
+
const hasBackend =
|
|
481
|
+
hasFramework("express") ||
|
|
482
|
+
hasFramework("nestjs") ||
|
|
483
|
+
hasFramework("fastify") ||
|
|
484
|
+
hasFramework("hono") ||
|
|
485
|
+
hasFramework("fastapi") ||
|
|
486
|
+
hasFramework("django") ||
|
|
487
|
+
hasFramework("flask") ||
|
|
488
|
+
hasFramework("gin") ||
|
|
489
|
+
hasFramework("fiber") ||
|
|
490
|
+
hasFramework("echo");
|
|
491
|
+
if (hasBackend) {
|
|
492
|
+
addSurface("backend_runtime", "Backend framework/runtime hints detected.");
|
|
493
|
+
}
|
|
494
|
+
|
|
495
|
+
const hasData =
|
|
496
|
+
hasFramework("prisma") ||
|
|
497
|
+
hasFramework("typeorm") ||
|
|
498
|
+
hasFramework("drizzle") ||
|
|
499
|
+
hasFile((file) => /(^|\/)(migrations|db|database|sql)\//.test(file) || file.endsWith(".sql"));
|
|
500
|
+
if (hasData) {
|
|
501
|
+
addSurface("data_layer", "Data-model or migration assets detected.");
|
|
502
|
+
}
|
|
503
|
+
|
|
504
|
+
const hasInfra =
|
|
505
|
+
hasFile(
|
|
506
|
+
(file) =>
|
|
507
|
+
file.endsWith(".tf") ||
|
|
508
|
+
file.includes("docker-compose") ||
|
|
509
|
+
file.endsWith("Dockerfile") ||
|
|
510
|
+
/(^|\/)(k8s|helm|terraform)\//.test(file)
|
|
511
|
+
);
|
|
512
|
+
if (hasInfra) {
|
|
513
|
+
addSurface("infrastructure", "Infrastructure-as-code or container orchestration assets detected.");
|
|
514
|
+
}
|
|
515
|
+
|
|
516
|
+
const hasRelease = hasFile((file) => file.startsWith(".github/workflows/") || file.startsWith(".gitlab-ci"));
|
|
517
|
+
if (hasRelease) {
|
|
518
|
+
addSurface("release_engineering", "CI/CD workflow definitions detected.");
|
|
519
|
+
}
|
|
520
|
+
|
|
521
|
+
const hasSupplyChain =
|
|
522
|
+
Object.keys(manifests).length > 0 ||
|
|
523
|
+
hasFile((file) =>
|
|
524
|
+
[
|
|
525
|
+
"package-lock.json",
|
|
526
|
+
"pnpm-lock.yaml",
|
|
527
|
+
"yarn.lock",
|
|
528
|
+
"poetry.lock",
|
|
529
|
+
"Pipfile.lock",
|
|
530
|
+
"Cargo.lock",
|
|
531
|
+
].some((candidate) => file.endsWith(candidate))
|
|
532
|
+
);
|
|
533
|
+
if (hasSupplyChain) {
|
|
534
|
+
addSurface("supply_chain", "Dependency manifests/lockfiles detected.");
|
|
535
|
+
}
|
|
536
|
+
|
|
537
|
+
const hasObservability =
|
|
538
|
+
hasFramework("observability-js") ||
|
|
539
|
+
hasFile((file) => /sentry|opentelemetry|prometheus|grafana/i.test(file));
|
|
540
|
+
if (hasObservability) {
|
|
541
|
+
addSurface("observability", "Observability tooling indicators detected.");
|
|
542
|
+
}
|
|
543
|
+
|
|
544
|
+
const hasAiPipeline = hasFile((file) => /(^|\/)(prompts|models|llm|agents?)\//i.test(file));
|
|
545
|
+
if (hasAiPipeline) {
|
|
546
|
+
addSurface("ai_pipeline", "AI/agent pipeline assets detected.");
|
|
547
|
+
}
|
|
548
|
+
|
|
549
|
+
const hasDocs = hasFile((file) => file.endsWith(".md") || file.startsWith("docs/"));
|
|
550
|
+
if (hasDocs) {
|
|
551
|
+
addSurface("docs_knowledge", "Documentation assets detected.");
|
|
552
|
+
}
|
|
553
|
+
|
|
554
|
+
if (hasInfra || hasObservability || hasRelease) {
|
|
555
|
+
addSurface("reliability_sre", "Operational and deployment assets detected.");
|
|
556
|
+
}
|
|
557
|
+
|
|
558
|
+
return [...surfaces.entries()]
|
|
559
|
+
.map(([surface, reason]) => ({ surface, reason }))
|
|
560
|
+
.sort((left, right) => left.surface.localeCompare(right.surface));
|
|
561
|
+
}
|
|
562
|
+
|
|
563
|
+
function summarizeLanguageStats(languageStats, totalLoc) {
|
|
564
|
+
return Object.entries(languageStats)
|
|
565
|
+
.map(([language, stats]) => ({
|
|
566
|
+
language,
|
|
567
|
+
files: stats.files,
|
|
568
|
+
loc: stats.loc,
|
|
569
|
+
locShare: totalLoc > 0 ? Number((stats.loc / totalLoc).toFixed(4)) : 0,
|
|
570
|
+
}))
|
|
571
|
+
.sort((left, right) => right.loc - left.loc || left.language.localeCompare(right.language));
|
|
572
|
+
}
|
|
573
|
+
|
|
574
|
+
async function listTopLevel(rootPath, ignoreMatcher) {
|
|
575
|
+
const dirs = [];
|
|
576
|
+
const files = [];
|
|
577
|
+
let entries = [];
|
|
578
|
+
try {
|
|
579
|
+
entries = await fsp.readdir(rootPath, { withFileTypes: true });
|
|
580
|
+
} catch {
|
|
581
|
+
return { directories: dirs, files };
|
|
582
|
+
}
|
|
583
|
+
|
|
584
|
+
for (const entry of entries) {
|
|
585
|
+
const name = String(entry.name || "");
|
|
586
|
+
if (!name) continue;
|
|
587
|
+
if (DEFAULT_IGNORED_DIRS.has(name)) continue;
|
|
588
|
+
if (ignoreMatcher.ignores(name, entry.isDirectory())) continue;
|
|
589
|
+
if (entry.isDirectory()) {
|
|
590
|
+
dirs.push(name);
|
|
591
|
+
} else if (entry.isFile()) {
|
|
592
|
+
files.push(name);
|
|
593
|
+
}
|
|
594
|
+
}
|
|
595
|
+
|
|
596
|
+
return {
|
|
597
|
+
directories: dirs.sort((left, right) => left.localeCompare(right)).slice(0, 200),
|
|
598
|
+
files: files.sort((left, right) => left.localeCompare(right)).slice(0, 200),
|
|
599
|
+
};
|
|
600
|
+
}
|
|
601
|
+
|
|
602
|
+
export async function collectCodebaseIngest({ rootPath = process.cwd() } = {}) {
|
|
603
|
+
const resolvedRoot = path.resolve(rootPath);
|
|
604
|
+
const ignoreMatcher = await createIgnoreMatcher(resolvedRoot);
|
|
605
|
+
const topLevel = await listTopLevel(resolvedRoot, ignoreMatcher);
|
|
606
|
+
const fingerprintHasher = createHash("sha256");
|
|
607
|
+
let fingerprintFilesCount = 0;
|
|
608
|
+
let latestFileMtimeMs = 0;
|
|
609
|
+
|
|
610
|
+
const stack = [resolvedRoot];
|
|
611
|
+
const fileSet = new Set();
|
|
612
|
+
const languageStats = {};
|
|
613
|
+
const manifests = {};
|
|
614
|
+
|
|
615
|
+
const indexedFiles = [];
|
|
616
|
+
let indexedOmittedCount = 0;
|
|
617
|
+
let filesScanned = 0;
|
|
618
|
+
let directoriesScanned = 0;
|
|
619
|
+
let totalLoc = 0;
|
|
620
|
+
let totalBytes = 0;
|
|
621
|
+
|
|
622
|
+
while (stack.length > 0) {
|
|
623
|
+
const current = stack.pop();
|
|
624
|
+
if (!current) continue;
|
|
625
|
+
|
|
626
|
+
let entries = [];
|
|
627
|
+
try {
|
|
628
|
+
entries = await fsp.readdir(current, { withFileTypes: true });
|
|
629
|
+
} catch {
|
|
630
|
+
continue;
|
|
631
|
+
}
|
|
632
|
+
|
|
633
|
+
directoriesScanned += 1;
|
|
634
|
+
|
|
635
|
+
for (const entry of entries) {
|
|
636
|
+
const fullPath = path.join(current, entry.name);
|
|
637
|
+
const relativePath = toPosixPath(path.relative(resolvedRoot, fullPath));
|
|
638
|
+
|
|
639
|
+
if (entry.isDirectory()) {
|
|
640
|
+
if (!relativePath) {
|
|
641
|
+
continue;
|
|
642
|
+
}
|
|
643
|
+
if (DEFAULT_IGNORED_DIRS.has(entry.name)) {
|
|
644
|
+
continue;
|
|
645
|
+
}
|
|
646
|
+
if (ignoreMatcher.ignores(relativePath, true)) {
|
|
647
|
+
continue;
|
|
648
|
+
}
|
|
649
|
+
stack.push(fullPath);
|
|
650
|
+
continue;
|
|
651
|
+
}
|
|
652
|
+
|
|
653
|
+
if (!entry.isFile()) {
|
|
654
|
+
continue;
|
|
655
|
+
}
|
|
656
|
+
if (ignoreMatcher.ignores(relativePath, false)) {
|
|
657
|
+
continue;
|
|
658
|
+
}
|
|
659
|
+
|
|
660
|
+
let stat;
|
|
661
|
+
try {
|
|
662
|
+
stat = await fsp.stat(fullPath);
|
|
663
|
+
} catch {
|
|
664
|
+
continue;
|
|
665
|
+
}
|
|
666
|
+
if (!stat || stat.size > MAX_FILE_SIZE_BYTES) {
|
|
667
|
+
continue;
|
|
668
|
+
}
|
|
669
|
+
|
|
670
|
+
appendFingerprintInput(fingerprintHasher, relativePath, stat.size, stat.mtimeMs);
|
|
671
|
+
fingerprintFilesCount += 1;
|
|
672
|
+
latestFileMtimeMs = Math.max(latestFileMtimeMs, normalizeMtimeMs(stat.mtimeMs));
|
|
673
|
+
|
|
674
|
+
let text = "";
|
|
675
|
+
try {
|
|
676
|
+
text = await fsp.readFile(fullPath, "utf-8");
|
|
677
|
+
} catch {
|
|
678
|
+
continue;
|
|
679
|
+
}
|
|
680
|
+
|
|
681
|
+
const loc = countLoc(text);
|
|
682
|
+
const language = detectLanguage(relativePath);
|
|
683
|
+
|
|
684
|
+
filesScanned += 1;
|
|
685
|
+
totalLoc += loc;
|
|
686
|
+
totalBytes += stat.size;
|
|
687
|
+
fileSet.add(relativePath);
|
|
688
|
+
|
|
689
|
+
if (!languageStats[language]) {
|
|
690
|
+
languageStats[language] = { files: 0, loc: 0 };
|
|
691
|
+
}
|
|
692
|
+
languageStats[language].files += 1;
|
|
693
|
+
languageStats[language].loc += loc;
|
|
694
|
+
|
|
695
|
+
const baseName = path.basename(relativePath);
|
|
696
|
+
if (MANIFEST_CANDIDATES.has(baseName)) {
|
|
697
|
+
manifests[baseName] = text;
|
|
698
|
+
}
|
|
699
|
+
|
|
700
|
+
if (indexedFiles.length < FILE_INDEX_LIMIT) {
|
|
701
|
+
indexedFiles.push({
|
|
702
|
+
path: relativePath,
|
|
703
|
+
language,
|
|
704
|
+
loc,
|
|
705
|
+
sizeBytes: stat.size,
|
|
706
|
+
});
|
|
707
|
+
} else {
|
|
708
|
+
indexedOmittedCount += 1;
|
|
709
|
+
}
|
|
710
|
+
}
|
|
711
|
+
}
|
|
712
|
+
|
|
713
|
+
const frameworks = detectFrameworks(manifests);
|
|
714
|
+
const packageMetadata = derivePackageMetadata(manifests);
|
|
715
|
+
const entryPoints = deriveEntryPoints(fileSet, manifests);
|
|
716
|
+
const riskSurfaces = deriveRiskSurfaces({
|
|
717
|
+
fileSet,
|
|
718
|
+
frameworks,
|
|
719
|
+
manifests,
|
|
720
|
+
languageStats,
|
|
721
|
+
});
|
|
722
|
+
|
|
723
|
+
return {
|
|
724
|
+
schemaVersion: "1.0.0",
|
|
725
|
+
generatedAt: new Date().toISOString(),
|
|
726
|
+
rootPath: resolvedRoot,
|
|
727
|
+
summary: {
|
|
728
|
+
filesScanned,
|
|
729
|
+
directoriesScanned,
|
|
730
|
+
totalLoc,
|
|
731
|
+
totalBytes,
|
|
732
|
+
},
|
|
733
|
+
topLevel,
|
|
734
|
+
manifests: {
|
|
735
|
+
detected: Object.keys(manifests).sort((left, right) => left.localeCompare(right)),
|
|
736
|
+
},
|
|
737
|
+
languages: summarizeLanguageStats(languageStats, totalLoc),
|
|
738
|
+
frameworks,
|
|
739
|
+
packageMetadata,
|
|
740
|
+
entryPoints,
|
|
741
|
+
riskSurfaces,
|
|
742
|
+
indexedFiles: {
|
|
743
|
+
limit: FILE_INDEX_LIMIT,
|
|
744
|
+
omitted: indexedOmittedCount,
|
|
745
|
+
files: indexedFiles,
|
|
746
|
+
},
|
|
747
|
+
cache: {
|
|
748
|
+
schema: INGEST_CACHE_SCHEMA,
|
|
749
|
+
contentHash: fingerprintHasher.digest("hex"),
|
|
750
|
+
filesCount: fingerprintFilesCount,
|
|
751
|
+
latestFileMtimeMs,
|
|
752
|
+
},
|
|
753
|
+
};
|
|
754
|
+
}
|
|
755
|
+
|
|
756
|
+
export function formatIngestSummary(ingest) {
|
|
757
|
+
const summary = ingest && ingest.summary ? ingest.summary : {};
|
|
758
|
+
const languageHead = Array.isArray(ingest.languages)
|
|
759
|
+
? ingest.languages
|
|
760
|
+
.slice(0, 5)
|
|
761
|
+
.map((item) => `${item.language}(${item.files} files/${item.loc} LOC)`)
|
|
762
|
+
.join(", ")
|
|
763
|
+
: "none";
|
|
764
|
+
const frameworks = Array.isArray(ingest.frameworks) && ingest.frameworks.length
|
|
765
|
+
? ingest.frameworks.join(", ")
|
|
766
|
+
: "none";
|
|
767
|
+
const entryPoints = Array.isArray(ingest.entryPoints) && ingest.entryPoints.length
|
|
768
|
+
? ingest.entryPoints.join(", ")
|
|
769
|
+
: "none";
|
|
770
|
+
const packageName = String(ingest.packageMetadata?.name || "").trim();
|
|
771
|
+
const packageScripts = Array.isArray(ingest.packageMetadata?.scripts)
|
|
772
|
+
? ingest.packageMetadata.scripts
|
|
773
|
+
: [];
|
|
774
|
+
|
|
775
|
+
const lines = [
|
|
776
|
+
`Workspace path: ${ingest.rootPath}`,
|
|
777
|
+
`Top-level directories: ${(ingest.topLevel?.directories || []).slice(0, 20).join(", ") || "none"}`,
|
|
778
|
+
`Top-level files: ${(ingest.topLevel?.files || []).slice(0, 20).join(", ") || "none"}`,
|
|
779
|
+
`Files scanned: ${summary.filesScanned || 0}`,
|
|
780
|
+
`Total LOC: ${summary.totalLoc || 0}`,
|
|
781
|
+
`Languages: ${languageHead}`,
|
|
782
|
+
`Frameworks: ${frameworks}`,
|
|
783
|
+
`Entry points: ${entryPoints}`,
|
|
784
|
+
];
|
|
785
|
+
|
|
786
|
+
if (packageName) {
|
|
787
|
+
lines.push(`package.json name: ${packageName}`);
|
|
788
|
+
}
|
|
789
|
+
if (packageScripts.length > 0) {
|
|
790
|
+
lines.push(`package scripts: ${packageScripts.slice(0, 15).join(", ")}`);
|
|
791
|
+
}
|
|
792
|
+
|
|
793
|
+
return lines.join("\n");
|
|
794
|
+
}
|
|
795
|
+
|
|
796
|
+
export async function writeCodebaseIngest({ ingest, rootPath, outputFile = "", outputDir = "" } = {}) {
|
|
797
|
+
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
798
|
+
const resolvedOutputFile = String(outputFile || "").trim();
|
|
799
|
+
const outputPath = resolvedOutputFile
|
|
800
|
+
? path.resolve(resolvedRoot, resolvedOutputFile)
|
|
801
|
+
: path.join(
|
|
802
|
+
await resolveOutputRoot({
|
|
803
|
+
cwd: resolvedRoot,
|
|
804
|
+
outputDirOverride: outputDir,
|
|
805
|
+
}),
|
|
806
|
+
"CODEBASE_INGEST.json"
|
|
807
|
+
);
|
|
808
|
+
|
|
809
|
+
await fsp.mkdir(path.dirname(outputPath), { recursive: true });
|
|
810
|
+
await fsp.writeFile(outputPath, `${JSON.stringify(ingest, null, 2)}\n`, "utf-8");
|
|
811
|
+
return outputPath;
|
|
812
|
+
}
|
|
813
|
+
|
|
814
|
+
export async function resolveCodebaseIngest({
|
|
815
|
+
rootPath = process.cwd(),
|
|
816
|
+
outputFile = "",
|
|
817
|
+
outputDir = "",
|
|
818
|
+
refresh = false,
|
|
819
|
+
} = {}) {
|
|
820
|
+
const resolvedRoot = path.resolve(rootPath || process.cwd());
|
|
821
|
+
const outputPath = await resolveIngestOutputPath({
|
|
822
|
+
rootPath: resolvedRoot,
|
|
823
|
+
outputFile,
|
|
824
|
+
outputDir,
|
|
825
|
+
});
|
|
826
|
+
const existingIngest = await readExistingIngest(outputPath);
|
|
827
|
+
const fingerprint = await computeCodebaseContentFingerprint({
|
|
828
|
+
rootPath: resolvedRoot,
|
|
829
|
+
});
|
|
830
|
+
const lastCommitAt = await readGitLastCommitAt(resolvedRoot);
|
|
831
|
+
const staleness = buildIngestStaleness({
|
|
832
|
+
existingIngest,
|
|
833
|
+
fingerprint,
|
|
834
|
+
lastCommitAt,
|
|
835
|
+
});
|
|
836
|
+
const staleBeforeRefresh = staleness.stale;
|
|
837
|
+
|
|
838
|
+
let ingest = existingIngest;
|
|
839
|
+
let refreshed = false;
|
|
840
|
+
let refreshedBecause = "";
|
|
841
|
+
if (!existingIngest) {
|
|
842
|
+
refreshed = true;
|
|
843
|
+
refreshedBecause = "missing_ingest";
|
|
844
|
+
} else if (refresh) {
|
|
845
|
+
refreshed = true;
|
|
846
|
+
refreshedBecause = "refresh_requested";
|
|
847
|
+
}
|
|
848
|
+
|
|
849
|
+
if (refreshed) {
|
|
850
|
+
ingest = await collectCodebaseIngest({
|
|
851
|
+
rootPath: resolvedRoot,
|
|
852
|
+
});
|
|
853
|
+
ingest.generatedAt = new Date().toISOString();
|
|
854
|
+
if (!ingest.cache || typeof ingest.cache !== "object") {
|
|
855
|
+
ingest.cache = {};
|
|
856
|
+
}
|
|
857
|
+
ingest.cache.schema = INGEST_CACHE_SCHEMA;
|
|
858
|
+
ingest.cache.contentHash = fingerprint.contentHash;
|
|
859
|
+
ingest.cache.filesCount = fingerprint.filesCount;
|
|
860
|
+
ingest.cache.latestFileMtimeMs = fingerprint.latestFileMtimeMs;
|
|
861
|
+
await writeCodebaseIngest({
|
|
862
|
+
ingest,
|
|
863
|
+
rootPath: resolvedRoot,
|
|
864
|
+
outputFile,
|
|
865
|
+
outputDir,
|
|
866
|
+
});
|
|
867
|
+
}
|
|
868
|
+
|
|
869
|
+
const resolutionReasons = refreshed
|
|
870
|
+
? [refreshedBecause, ...staleness.reasons].filter(Boolean)
|
|
871
|
+
: staleness.reasons;
|
|
872
|
+
|
|
873
|
+
return {
|
|
874
|
+
ingest,
|
|
875
|
+
outputPath,
|
|
876
|
+
refreshed,
|
|
877
|
+
stale: refreshed ? false : staleness.stale,
|
|
878
|
+
staleBeforeRefresh,
|
|
879
|
+
reasons: resolutionReasons,
|
|
880
|
+
refreshedBecause,
|
|
881
|
+
refreshRequested: Boolean(refresh),
|
|
882
|
+
lastCommitAt,
|
|
883
|
+
fingerprint,
|
|
884
|
+
event:
|
|
885
|
+
refreshed || staleBeforeRefresh
|
|
886
|
+
? {
|
|
887
|
+
event: "ingest_refresh",
|
|
888
|
+
payload: {
|
|
889
|
+
refreshed,
|
|
890
|
+
stale: refreshed ? false : staleness.stale,
|
|
891
|
+
reason:
|
|
892
|
+
refreshedBecause || (staleness.reasons.length > 0 ? staleness.reasons.join(",") : "cache_hit"),
|
|
893
|
+
contentHash: fingerprint.contentHash,
|
|
894
|
+
filesCount: fingerprint.filesCount,
|
|
895
|
+
lastCommitAt,
|
|
896
|
+
},
|
|
897
|
+
}
|
|
898
|
+
: null,
|
|
899
|
+
};
|
|
900
|
+
}
|
|
901
|
+
|
|
902
|
+
export async function generateCodebaseIngest({
|
|
903
|
+
rootPath = process.cwd(),
|
|
904
|
+
outputFile = "",
|
|
905
|
+
outputDir = "",
|
|
906
|
+
} = {}) {
|
|
907
|
+
const ingest = await collectCodebaseIngest({ rootPath });
|
|
908
|
+
const outputPath = await writeCodebaseIngest({
|
|
909
|
+
ingest,
|
|
910
|
+
rootPath,
|
|
911
|
+
outputFile,
|
|
912
|
+
outputDir,
|
|
913
|
+
});
|
|
914
|
+
return {
|
|
915
|
+
ingest,
|
|
916
|
+
outputPath,
|
|
917
|
+
};
|
|
918
|
+
}
|