@jcode.labs/mimir 0.3.0 → 0.4.0

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.
Files changed (73) hide show
  1. package/CHANGELOG.md +22 -3
  2. package/CONTRIBUTING.md +28 -0
  3. package/README.md +262 -36
  4. package/SECURITY-HARDENING.md +58 -20
  5. package/dist/chunking.d.ts.map +1 -1
  6. package/dist/chunking.js +6 -3
  7. package/dist/chunking.js.map +1 -1
  8. package/dist/cli.js +103 -9
  9. package/dist/cli.js.map +1 -1
  10. package/dist/config.d.ts.map +1 -1
  11. package/dist/config.js +50 -36
  12. package/dist/config.js.map +1 -1
  13. package/dist/defaults.d.ts +11 -0
  14. package/dist/defaults.d.ts.map +1 -0
  15. package/dist/defaults.js +31 -0
  16. package/dist/defaults.js.map +1 -0
  17. package/dist/embeddings.d.ts.map +1 -1
  18. package/dist/embeddings.js +85 -11
  19. package/dist/embeddings.js.map +1 -1
  20. package/dist/files.d.ts +2 -1
  21. package/dist/files.d.ts.map +1 -1
  22. package/dist/files.js +39 -2
  23. package/dist/files.js.map +1 -1
  24. package/dist/gitignore.d.ts +1 -1
  25. package/dist/gitignore.d.ts.map +1 -1
  26. package/dist/gitignore.js +8 -7
  27. package/dist/gitignore.js.map +1 -1
  28. package/dist/ingest.d.ts.map +1 -1
  29. package/dist/ingest.js +2 -1
  30. package/dist/ingest.js.map +1 -1
  31. package/dist/init.d.ts.map +1 -1
  32. package/dist/init.js +4 -24
  33. package/dist/init.js.map +1 -1
  34. package/dist/mcp.d.ts.map +1 -1
  35. package/dist/mcp.js +14 -13
  36. package/dist/mcp.js.map +1 -1
  37. package/dist/parsing.d.ts.map +1 -1
  38. package/dist/parsing.js +138 -0
  39. package/dist/parsing.js.map +1 -1
  40. package/dist/query.d.ts.map +1 -1
  41. package/dist/query.js +14 -22
  42. package/dist/query.js.map +1 -1
  43. package/dist/security.js +16 -18
  44. package/dist/security.js.map +1 -1
  45. package/dist/skill.d.ts +2 -1
  46. package/dist/skill.d.ts.map +1 -1
  47. package/dist/skill.js +24 -9
  48. package/dist/skill.js.map +1 -1
  49. package/dist/store.d.ts.map +1 -1
  50. package/dist/store.js +2 -1
  51. package/dist/store.js.map +1 -1
  52. package/dist/types.d.ts +12 -14
  53. package/dist/types.d.ts.map +1 -1
  54. package/dist/version.d.ts +1 -1
  55. package/dist/version.js +1 -1
  56. package/examples/sovereign-rag-demo/.kb/config.json +22 -0
  57. package/examples/sovereign-rag-demo/.kb/sources.txt +2 -0
  58. package/examples/sovereign-rag-demo/README.md +80 -0
  59. package/examples/sovereign-rag-demo/raw/dataset-inventory.csv +5 -0
  60. package/examples/sovereign-rag-demo/raw/incident-timeline.jsonl +4 -0
  61. package/examples/sovereign-rag-demo/raw/operations-brief.md +16 -0
  62. package/examples/sovereign-rag-demo/raw/review-notes.evidence +11 -0
  63. package/examples/sovereign-rag-demo/raw/security-policy.yaml +14 -0
  64. package/package.json +24 -25
  65. package/skills/mimir/SKILL.md +66 -5
  66. package/skills/mimir-audio-summary/SKILL.md +134 -0
  67. package/skills/mimir-audio-summary/forge-voice.sh +153 -0
  68. package/skills/mimir-audio-summary/split-lines.py +13 -0
  69. package/skills/mimir-audio-summary/xtts-voice.py +46 -0
  70. package/dist/network.d.ts +0 -4
  71. package/dist/network.d.ts.map +0 -1
  72. package/dist/network.js +0 -59
  73. package/dist/network.js.map +0 -1
package/dist/files.js CHANGED
@@ -3,17 +3,51 @@ import { existsSync } from "node:fs";
3
3
  import { readFile, stat } from "node:fs/promises";
4
4
  import path from "node:path";
5
5
  import fg from "fast-glob";
6
- export const SUPPORTED_EXTENSIONS = new Set([
6
+ export const DEFAULT_SUPPORTED_EXTENSIONS = new Set([
7
+ ".atom",
8
+ ".c",
9
+ ".cfg",
10
+ ".conf",
11
+ ".cpp",
12
+ ".cs",
13
+ ".css",
7
14
  ".csv",
15
+ ".docx",
16
+ ".go",
17
+ ".h",
8
18
  ".htm",
9
19
  ".html",
20
+ ".ini",
21
+ ".java",
22
+ ".js",
10
23
  ".json",
24
+ ".jsonl",
25
+ ".jsx",
26
+ ".log",
11
27
  ".md",
12
28
  ".mdx",
29
+ ".ndjson",
30
+ ".odp",
31
+ ".ods",
32
+ ".odt",
13
33
  ".pdf",
34
+ ".php",
35
+ ".pptx",
36
+ ".properties",
37
+ ".py",
38
+ ".rb",
39
+ ".rs",
40
+ ".rss",
41
+ ".rtf",
42
+ ".sql",
14
43
  ".text",
44
+ ".toml",
45
+ ".ts",
15
46
  ".tsv",
47
+ ".tsx",
16
48
  ".txt",
49
+ ".xml",
50
+ ".xlsx",
17
51
  ".yaml",
18
52
  ".yml",
19
53
  ]);
@@ -34,7 +68,7 @@ export async function listSourceFiles(config) {
34
68
  });
35
69
  for (const absolutePath of entries) {
36
70
  const extension = path.extname(absolutePath).toLowerCase();
37
- if (!SUPPORTED_EXTENSIONS.has(extension)) {
71
+ if (!supportedExtensions(config).has(extension)) {
38
72
  continue;
39
73
  }
40
74
  const info = await stat(absolutePath);
@@ -52,6 +86,9 @@ export async function listSourceFiles(config) {
52
86
  }
53
87
  return [...files.values()].sort((a, b) => a.relativePath.localeCompare(b.relativePath));
54
88
  }
89
+ export function supportedExtensions(config) {
90
+ return new Set([...DEFAULT_SUPPORTED_EXTENSIONS, ...config.includeExtensions]);
91
+ }
55
92
  async function sourceRoots(config) {
56
93
  const roots = [config.rawDir];
57
94
  if (!existsSync(config.sourcesFile)) {
package/dist/files.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"files.js","sourceRoot":"","sources":["../src/files.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,kBAAkB,CAAA;AACjD,OAAO,IAAI,MAAM,WAAW,CAAA;AAC5B,OAAO,EAAE,MAAM,WAAW,CAAA;AAG1B,MAAM,CAAC,MAAM,oBAAoB,GAAG,IAAI,GAAG,CAAC;IAC1C,MAAM;IACN,MAAM;IACN,OAAO;IACP,OAAO;IACP,KAAK;IACL,MAAM;IACN,MAAM;IACN,OAAO;IACP,MAAM;IACN,MAAM;IACN,OAAO;IACP,MAAM;CACP,CAAC,CAAA;AAEF,MAAM,CAAC,KAAK,UAAU,eAAe,CAAC,MAAc;IAClD,MAAM,KAAK,GAAG,MAAM,WAAW,CAAC,MAAM,CAAC,CAAA;IACvC,MAAM,KAAK,GAAG,IAAI,GAAG,EAAsB,CAAA;IAE3C,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,EAAE,CAAC;YACtB,SAAQ;QACV,CAAC;QAED,MAAM,OAAO,GAAG,MAAM,EAAE,CAAC,MAAM,EAAE;YAC/B,GAAG,EAAE,IAAI;YACT,QAAQ,EAAE,IAAI;YACd,SAAS,EAAE,IAAI;YACf,GAAG,EAAE,KAAK;YACV,mBAAmB,EAAE,KAAK;YAC1B,MAAM,EAAE,CAAC,YAAY,EAAE,oBAAoB,EAAE,WAAW,EAAE,cAAc,CAAC;SAC1E,CAAC,CAAA;QAEF,KAAK,MAAM,YAAY,IAAI,OAAO,EAAE,CAAC;YACnC,MAAM,SAAS,GAAG,IAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC,WAAW,EAAE,CAAA;YAC1D,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,SAAS,CAAC,EAAE,CAAC;gBACzC,SAAQ;YACV,CAAC;YAED,MAAM,IAAI,GAAG,MAAM,IAAI,CAAC,YAAY,CAAC,CAAA;YACrC,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,YAAY,CAAC,CAAA;YAC3C,KAAK,CAAC,GAAG,CAAC,YAAY,EAAE;gBACtB,YAAY;gBACZ,YAAY,EAAE,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,WAAW,EAAE,YAAY,CAAC;gBAC7D,MAAM,EAAE,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,YAAY,CAAC,IAAI,IAAI,CAAC,QAAQ,CAAC,YAAY,CAAC;gBACxE,SAAS;gBACT,KAAK,EAAE,IAAI,CAAC,IAAI;gBAChB,OAAO,EAAE,IAAI,CAAC,OAAO;gBACrB,QAAQ,EAAE,UAAU,CAAC,QAAQ,CAAC,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAC5D,CAAC,CAAA;QACJ,CAAC;IACH,CAAC;IAED,OAAO,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,YAAY,CAAC,aAAa,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAA;AACzF,CAAC;AAED,KAAK,UAAU,WAAW,CAAC,MAAc;IACvC,MAAM,KAAK,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,CAAA;IAC7B,IAAI,CAAC,UAAU,CAAC,MAAM,CAAC,WAAW,CAAC,EAAE,CAAC;QACpC,OAAO,KAAK,CAAA;IACd,CAAC;IAED,MAAM,OAAO,GAAG,MAAM,QAAQ,CAAC,MAAM,CAAC,WAAW,EAAE,MAAM,CAAC,CAAA;IAC1D,KAAK,MAAM,IAAI,IAAI,OAAO,CAAC,KAAK,CAAC,QAAQ,CAAC,EAAE,CAAC;QAC3C,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;QAC3B,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC,EAAE,CAAC;YACxC,SAAQ;QACV,CAAC;QACD,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC,CAAA;IAC5F,CAAC;IAED,OAAO,KAAK,CAAA;AACd,CAAC"}
1
+ {"version":3,"file":"files.js","sourceRoot":"","sources":["../src/files.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,kBAAkB,CAAA;AACjD,OAAO,IAAI,MAAM,WAAW,CAAA;AAC5B,OAAO,EAAE,MAAM,WAAW,CAAA;AAG1B,MAAM,CAAC,MAAM,4BAA4B,GAAG,IAAI,GAAG,CAAC;IAClD,OAAO;IACP,IAAI;IACJ,MAAM;IACN,OAAO;IACP,MAAM;IACN,KAAK;IACL,MAAM;IACN,MAAM;IACN,OAAO;IACP,KAAK;IACL,IAAI;IACJ,MAAM;IACN,OAAO;IACP,MAAM;IACN,OAAO;IACP,KAAK;IACL,OAAO;IACP,QAAQ;IACR,MAAM;IACN,MAAM;IACN,KAAK;IACL,MAAM;IACN,SAAS;IACT,MAAM;IACN,MAAM;IACN,MAAM;IACN,MAAM;IACN,MAAM;IACN,OAAO;IACP,aAAa;IACb,KAAK;IACL,KAAK;IACL,KAAK;IACL,MAAM;IACN,MAAM;IACN,MAAM;IACN,OAAO;IACP,OAAO;IACP,KAAK;IACL,MAAM;IACN,MAAM;IACN,MAAM;IACN,MAAM;IACN,OAAO;IACP,OAAO;IACP,MAAM;CACP,CAAC,CAAA;AAEF,MAAM,CAAC,KAAK,UAAU,eAAe,CAAC,MAAc;IAClD,MAAM,KAAK,GAAG,MAAM,WAAW,CAAC,MAAM,CAAC,CAAA;IACvC,MAAM,KAAK,GAAG,IAAI,GAAG,EAAsB,CAAA;IAE3C,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,EAAE,CAAC;YACtB,SAAQ;QACV,CAAC;QAED,MAAM,OAAO,GAAG,MAAM,EAAE,CAAC,MAAM,EAAE;YAC/B,GAAG,EAAE,IAAI;YACT,QAAQ,EAAE,IAAI;YACd,SAAS,EAAE,IAAI;YACf,GAAG,EAAE,KAAK;YACV,mBAAmB,EAAE,KAAK;YAC1B,MAAM,EAAE,CAAC,YAAY,EAAE,oBAAoB,EAAE,WAAW,EAAE,cAAc,CAAC;SAC1E,CAAC,CAAA;QAEF,KAAK,MAAM,YAAY,IAAI,OAAO,EAAE,CAAC;YACnC,MAAM,SAAS,GAAG,IAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC,WAAW,EAAE,CAAA;YAC1D,IAAI,CAAC,mBAAmB,CAAC,MAAM,CAAC,CAAC,GAAG,CAAC,SAAS,CAAC,EAAE,CAAC;gBAChD,SAAQ;YACV,CAAC;YAED,MAAM,IAAI,GAAG,MAAM,IAAI,CAAC,YAAY,CAAC,CAAA;YACrC,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,YAAY,CAAC,CAAA;YAC3C,KAAK,CAAC,GAAG,CAAC,YAAY,EAAE;gBACtB,YAAY;gBACZ,YAAY,EAAE,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,WAAW,EAAE,YAAY,CAAC;gBAC7D,MAAM,EAAE,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,YAAY,CAAC,IAAI,IAAI,CAAC,QAAQ,CAAC,YAAY,CAAC;gBACxE,SAAS;gBACT,KAAK,EAAE,IAAI,CAAC,IAAI;gBAChB,OAAO,EAAE,IAAI,CAAC,OAAO;gBACrB,QAAQ,EAAE,UAAU,CAAC,QAAQ,CAAC,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAC5D,CAAC,CAAA;QACJ,CAAC;IACH,CAAC;IAED,OAAO,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,YAAY,CAAC,aAAa,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAA;AACzF,CAAC;AAED,MAAM,UAAU,mBAAmB,CAAC,MAAc;IAChD,OAAO,IAAI,GAAG,CAAC,CAAC,GAAG,4BAA4B,EAAE,GAAG,MAAM,CAAC,iBAAiB,CAAC,CAAC,CAAA;AAChF,CAAC;AAED,KAAK,UAAU,WAAW,CAAC,MAAc;IACvC,MAAM,KAAK,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,CAAA;IAC7B,IAAI,CAAC,UAAU,CAAC,MAAM,CAAC,WAAW,CAAC,EAAE,CAAC;QACpC,OAAO,KAAK,CAAA;IACd,CAAC;IAED,MAAM,OAAO,GAAG,MAAM,QAAQ,CAAC,MAAM,CAAC,WAAW,EAAE,MAAM,CAAC,CAAA;IAC1D,KAAK,MAAM,IAAI,IAAI,OAAO,CAAC,KAAK,CAAC,QAAQ,CAAC,EAAE,CAAC;QAC3C,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;QAC3B,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC,EAAE,CAAC;YACxC,SAAQ;QACV,CAAC;QACD,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC,CAAA;IAC5F,CAAC;IAED,OAAO,KAAK,CAAA;AACd,CAAC"}
@@ -1,3 +1,3 @@
1
- export declare const MIMIR_GITIGNORE_ENTRIES: readonly [".kb/", ".mimir/", "private/**", "!private/", "!private/README.md", "!private/**/", "!private/**/.gitkeep"];
1
+ export declare const MIMIR_GITIGNORE_ENTRIES: string[];
2
2
  export declare function ensureMimirGitignore(cwd?: string): Promise<boolean>;
3
3
  //# sourceMappingURL=gitignore.d.ts.map
@@ -1 +1 @@
1
- {"version":3,"file":"gitignore.d.ts","sourceRoot":"","sources":["../src/gitignore.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,uBAAuB,uHAQ1B,CAAA;AAEV,wBAAsB,oBAAoB,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,OAAO,CAAC,CAyBhF"}
1
+ {"version":3,"file":"gitignore.d.ts","sourceRoot":"","sources":["../src/gitignore.ts"],"names":[],"mappings":"AAUA,eAAO,MAAM,uBAAuB,UAQnC,CAAA;AAED,wBAAsB,oBAAoB,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,OAAO,CAAC,CAyBhF"}
package/dist/gitignore.js CHANGED
@@ -1,14 +1,15 @@
1
1
  import { existsSync } from "node:fs";
2
2
  import { readFile, writeFile } from "node:fs/promises";
3
3
  import path from "node:path";
4
+ import { KB_GITIGNORE_ENTRY, MIMIR_GITIGNORE_ENTRY, PRIVATE_DIR, PRIVATE_GITIGNORE_ENTRY, } from "./defaults.js";
4
5
  export const MIMIR_GITIGNORE_ENTRIES = [
5
- ".kb/",
6
- ".mimir/",
7
- "private/**",
8
- "!private/",
9
- "!private/README.md",
10
- "!private/**/",
11
- "!private/**/.gitkeep",
6
+ KB_GITIGNORE_ENTRY,
7
+ MIMIR_GITIGNORE_ENTRY,
8
+ PRIVATE_GITIGNORE_ENTRY,
9
+ `!${PRIVATE_DIR}/`,
10
+ `!${PRIVATE_DIR}/README.md`,
11
+ `!${PRIVATE_DIR}/**/`,
12
+ `!${PRIVATE_DIR}/**/.gitkeep`,
12
13
  ];
13
14
  export async function ensureMimirGitignore(cwd = process.cwd()) {
14
15
  const root = path.resolve(cwd);
@@ -1 +1 @@
1
- {"version":3,"file":"gitignore.js","sourceRoot":"","sources":["../src/gitignore.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACtD,OAAO,IAAI,MAAM,WAAW,CAAA;AAE5B,MAAM,CAAC,MAAM,uBAAuB,GAAG;IACrC,MAAM;IACN,SAAS;IACT,YAAY;IACZ,WAAW;IACX,oBAAoB;IACpB,cAAc;IACd,sBAAsB;CACd,CAAA;AAEV,MAAM,CAAC,KAAK,UAAU,oBAAoB,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAC5D,MAAM,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;IAC9B,MAAM,aAAa,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,YAAY,CAAC,CAAA;IACnD,MAAM,OAAO,GAAG,UAAU,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,MAAM,QAAQ,CAAC,aAAa,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;IACtF,MAAM,YAAY,GAAG,IAAI,GAAG,CAC1B,OAAO;SACJ,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;SAC1B,MAAM,CAAC,OAAO,CAAC,CACnB,CAAA;IACD,MAAM,cAAc,GAAG,uBAAuB,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,YAAY,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,CAAA;IAE1F,IAAI,cAAc,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAChC,OAAO,KAAK,CAAA;IACd,CAAC;IAED,MAAM,cAAc,GAAG,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,YAAY,CAAC,GAAG,CAAC,eAAe,CAAC,CAAA;IACvF,MAAM,KAAK,GAAG,CAAC,cAAc,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,SAAS,EAAE,GAAG,cAAc,CAAC;SACtE,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,KAAK,SAAS,CAAC;SACpC,IAAI,CAAC,IAAI,CAAC,CAAA;IACb,MAAM,MAAM,GAAG,OAAO,CAAC,OAAO,EAAE,CAAA;IAChC,MAAM,IAAI,GAAG,GAAG,MAAM,CAAC,CAAC,CAAC,GAAG,MAAM,MAAM,CAAC,CAAC,CAAC,EAAE,GAAG,KAAK,IAAI,CAAA;IAEzD,MAAM,SAAS,CAAC,aAAa,EAAE,IAAI,EAAE,MAAM,CAAC,CAAA;IAC5C,OAAO,IAAI,CAAA;AACb,CAAC"}
1
+ {"version":3,"file":"gitignore.js","sourceRoot":"","sources":["../src/gitignore.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACtD,OAAO,IAAI,MAAM,WAAW,CAAA;AAC5B,OAAO,EACL,kBAAkB,EAClB,qBAAqB,EACrB,WAAW,EACX,uBAAuB,GACxB,MAAM,eAAe,CAAA;AAEtB,MAAM,CAAC,MAAM,uBAAuB,GAAG;IACrC,kBAAkB;IAClB,qBAAqB;IACrB,uBAAuB;IACvB,IAAI,WAAW,GAAG;IAClB,IAAI,WAAW,YAAY;IAC3B,IAAI,WAAW,MAAM;IACrB,IAAI,WAAW,cAAc;CAC9B,CAAA;AAED,MAAM,CAAC,KAAK,UAAU,oBAAoB,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAC5D,MAAM,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;IAC9B,MAAM,aAAa,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,YAAY,CAAC,CAAA;IACnD,MAAM,OAAO,GAAG,UAAU,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,MAAM,QAAQ,CAAC,aAAa,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;IACtF,MAAM,YAAY,GAAG,IAAI,GAAG,CAC1B,OAAO;SACJ,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;SAC1B,MAAM,CAAC,OAAO,CAAC,CACnB,CAAA;IACD,MAAM,cAAc,GAAG,uBAAuB,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,YAAY,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,CAAA;IAE1F,IAAI,cAAc,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAChC,OAAO,KAAK,CAAA;IACd,CAAC;IAED,MAAM,cAAc,GAAG,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,YAAY,CAAC,GAAG,CAAC,eAAe,CAAC,CAAA;IACvF,MAAM,KAAK,GAAG,CAAC,cAAc,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,SAAS,EAAE,GAAG,cAAc,CAAC;SACtE,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,KAAK,SAAS,CAAC;SACpC,IAAI,CAAC,IAAI,CAAC,CAAA;IACb,MAAM,MAAM,GAAG,OAAO,CAAC,OAAO,EAAE,CAAA;IAChC,MAAM,IAAI,GAAG,GAAG,MAAM,CAAC,CAAC,CAAC,GAAG,MAAM,MAAM,CAAC,CAAC,CAAC,EAAE,GAAG,KAAK,IAAI,CAAA;IAEzD,MAAM,SAAS,CAAC,aAAa,EAAE,IAAI,EAAE,MAAM,CAAC,CAAA;IAC5C,OAAO,IAAI,CAAA;AACb,CAAC"}
@@ -1 +1 @@
1
- {"version":3,"file":"ingest.d.ts","sourceRoot":"","sources":["../src/ingest.ts"],"names":[],"mappings":"AAQA,OAAO,KAAK,EACV,WAAW,EACX,aAAa,EACb,YAAY,EAIb,MAAM,YAAY,CAAA;AAInB,wBAAsB,MAAM,CAAC,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,YAAY,CAAC,CA4D/E;AAED,wBAAsB,KAAK,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,WAAW,CAAC,CAkCrE"}
1
+ {"version":3,"file":"ingest.d.ts","sourceRoot":"","sources":["../src/ingest.ts"],"names":[],"mappings":"AAQA,OAAO,KAAK,EACV,WAAW,EACX,aAAa,EACb,YAAY,EAIb,MAAM,YAAY,CAAA;AAKnB,wBAAsB,MAAM,CAAC,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,YAAY,CAAC,CA4D/E;AAED,wBAAsB,KAAK,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,WAAW,CAAC,CAoCrE"}
package/dist/ingest.js CHANGED
@@ -7,6 +7,7 @@ import { parseFile } from "./parsing.js";
7
7
  import { redactText, totalRedactions } from "./redaction.js";
8
8
  import { openRowsTable, writeRows } from "./store.js";
9
9
  const EMBED_BATCH_SIZE = 32;
10
+ const MAX_AUDIT_ROWS = 100_000;
10
11
  export async function ingest(options = {}) {
11
12
  const config = await loadConfig(String(options.cwd ?? process.cwd()));
12
13
  const files = await listSourceFiles(config);
@@ -72,7 +73,7 @@ export async function audit(cwd = process.cwd()) {
72
73
  totalChunks: 0,
73
74
  };
74
75
  }
75
- const rows = (await table.query().limit(100000).toArray());
76
+ const rows = (await table.query().limit(MAX_AUDIT_ROWS).toArray());
76
77
  const counts = new Map();
77
78
  for (const row of rows) {
78
79
  counts.set(row.relativePath, (counts.get(row.relativePath) ?? 0) + 1);
@@ -1 +1 @@
1
- {"version":3,"file":"ingest.js","sourceRoot":"","sources":["../src/ingest.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAA;AAC9C,OAAO,EAAE,aAAa,EAAE,MAAM,eAAe,CAAA;AAC7C,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAA;AAC5C,OAAO,EAAE,eAAe,EAAE,MAAM,YAAY,CAAA;AAC5C,OAAO,EAAE,SAAS,EAAE,MAAM,cAAc,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,eAAe,EAAE,MAAM,gBAAgB,CAAA;AAC5D,OAAO,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,YAAY,CAAA;AAUrD,MAAM,gBAAgB,GAAG,EAAE,CAAA;AAE3B,MAAM,CAAC,KAAK,UAAU,MAAM,CAAC,UAAyB,EAAE;IACtD,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,KAAK,GAAG,MAAM,eAAe,CAAC,MAAM,CAAC,CAAA;IAC3C,MAAM,SAAS,GAAgB,EAAE,CAAA;IACjC,MAAM,MAAM,GAA2B,EAAE,CAAA;IACzC,MAAM,eAAe,GAAqB,EAAE,CAAA;IAC5C,IAAI,YAAY,GAAG,CAAC,CAAA;IAEpB,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,CAAC;YACH,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,IAAI,CAAC,CAAA;YACpC,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;YAChD,eAAe,CAAC,IAAI,CAAC,GAAG,QAAQ,CAAC,MAAM,CAAC,CAAA;YACxC,MAAM,MAAM,GAAG,aAAa,CAC1B,EAAE,GAAG,MAAM,EAAE,IAAI,EAAE,QAAQ,CAAC,IAAI,EAAE,EAClC,MAAM,CAAC,SAAS,EAChB,MAAM,CAAC,YAAY,CACpB,CAAA;YACD,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACxB,YAAY,IAAI,CAAC,CAAA;YACnB,CAAC;YACD,SAAS,CAAC,IAAI,CAAC,GAAG,MAAM,CAAC,CAAA;QAC3B,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,CAAC,IAAI,CAAC;gBACV,IAAI,EAAE,IAAI,CAAC,YAAY;gBACvB,OAAO,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAChE,CAAC,CAAA;QACJ,CAAC;IACH,CAAC;IAED,MAAM,IAAI,GAAgB,EAAE,CAAA;IAC5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC,IAAI,gBAAgB,EAAE,CAAC;QAC5D,MAAM,KAAK,GAAG,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,GAAG,gBAAgB,CAAC,CAAA;QACtD,MAAM,UAAU,GAAG,MAAM,UAAU,CACjC,KAAK,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,EAChC,MAAM,CACP,CAAA;QACD,KAAK,MAAM,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,EAAE,CAAC;YAC7C,MAAM,MAAM,GAAG,UAAU,CAAC,KAAK,CAAC,CAAA;YAChC,IAAI,CAAC,MAAM,EAAE,CAAC;gBACZ,MAAM,IAAI,KAAK,CAAC,+BAA+B,KAAK,CAAC,YAAY,IAAI,KAAK,CAAC,UAAU,GAAG,CAAC,CAAA;YAC3F,CAAC;YACD,IAAI,CAAC,IAAI,CAAC,EAAE,GAAG,KAAK,EAAE,MAAM,EAAE,CAAC,CAAA;QACjC,CAAC;IACH,CAAC;IAED,MAAM,SAAS,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IAC7B,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,QAAQ;QAChB,WAAW,EAAE,IAAI,CAAC,MAAM;QACxB,UAAU,EAAE,eAAe,CAAC,eAAe,CAAC;KAC7C,CAAC,CAAA;IAEF,OAAO;QACL,YAAY,EAAE,IAAI,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI;QAC/D,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,YAAY;QACZ,UAAU,EAAE,eAAe,CAAC,eAAe,CAAC;QAC5C,MAAM;KACP,CAAA;AACH,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,KAAK,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAC7C,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;IACpC,MAAM,KAAK,GAAG,MAAM,eAAe,CAAC,MAAM,CAAC,CAAA;IAC3C,MAAM,cAAc,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;IAC7D,MAAM,KAAK,GAAG,MAAM,aAAa,CAAC,MAAM,CAAC,CAAA;IAEzC,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO;YACL,YAAY,EAAE,EAAE;YAChB,cAAc;YACd,gBAAgB,EAAE,cAAc;YAChC,YAAY,EAAE,EAAE;YAChB,WAAW,EAAE,CAAC;SACf,CAAA;IACH,CAAC;IAED,MAAM,IAAI,GAAG,CAAC,MAAM,KAAK,CAAC,KAAK,EAAE,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,OAAO,EAAE,CAAoC,CAAA;IAC7F,MAAM,MAAM,GAAG,IAAI,GAAG,EAAkB,CAAA;IACxC,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QACvB,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,YAAY,EAAE,CAAC,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,YAAY,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;IACvE,CAAC;IAED,MAAM,YAAY,GAAG,IAAI,GAAG,CAAC,cAAc,CAAC,CAAA;IAC5C,MAAM,UAAU,GAAG,IAAI,GAAG,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC,CAAA;IAEzC,OAAO;QACL,YAAY,EAAE,CAAC,GAAG,MAAM,CAAC,OAAO,EAAE,CAAC;aAChC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC;aACtC,GAAG,CAAC,CAAC,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,MAAM,EAAE,CAAC,CAAC;QAClD,cAAc;QACd,gBAAgB,EAAE,cAAc,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,UAAU,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;QACxE,YAAY,EAAE,CAAC,GAAG,UAAU,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,EAAE;QAC9E,WAAW,EAAE,IAAI,CAAC,MAAM;KACzB,CAAA;AACH,CAAC"}
1
+ {"version":3,"file":"ingest.js","sourceRoot":"","sources":["../src/ingest.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAA;AAC9C,OAAO,EAAE,aAAa,EAAE,MAAM,eAAe,CAAA;AAC7C,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAA;AAC5C,OAAO,EAAE,eAAe,EAAE,MAAM,YAAY,CAAA;AAC5C,OAAO,EAAE,SAAS,EAAE,MAAM,cAAc,CAAA;AACxC,OAAO,EAAE,UAAU,EAAE,eAAe,EAAE,MAAM,gBAAgB,CAAA;AAC5D,OAAO,EAAE,aAAa,EAAE,SAAS,EAAE,MAAM,YAAY,CAAA;AAUrD,MAAM,gBAAgB,GAAG,EAAE,CAAA;AAC3B,MAAM,cAAc,GAAG,OAAO,CAAA;AAE9B,MAAM,CAAC,KAAK,UAAU,MAAM,CAAC,UAAyB,EAAE;IACtD,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,KAAK,GAAG,MAAM,eAAe,CAAC,MAAM,CAAC,CAAA;IAC3C,MAAM,SAAS,GAAgB,EAAE,CAAA;IACjC,MAAM,MAAM,GAA2B,EAAE,CAAA;IACzC,MAAM,eAAe,GAAqB,EAAE,CAAA;IAC5C,IAAI,YAAY,GAAG,CAAC,CAAA;IAEpB,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,CAAC;YACH,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,IAAI,CAAC,CAAA;YACpC,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;YAChD,eAAe,CAAC,IAAI,CAAC,GAAG,QAAQ,CAAC,MAAM,CAAC,CAAA;YACxC,MAAM,MAAM,GAAG,aAAa,CAC1B,EAAE,GAAG,MAAM,EAAE,IAAI,EAAE,QAAQ,CAAC,IAAI,EAAE,EAClC,MAAM,CAAC,SAAS,EAChB,MAAM,CAAC,YAAY,CACpB,CAAA;YACD,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACxB,YAAY,IAAI,CAAC,CAAA;YACnB,CAAC;YACD,SAAS,CAAC,IAAI,CAAC,GAAG,MAAM,CAAC,CAAA;QAC3B,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,CAAC,IAAI,CAAC;gBACV,IAAI,EAAE,IAAI,CAAC,YAAY;gBACvB,OAAO,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAChE,CAAC,CAAA;QACJ,CAAC;IACH,CAAC;IAED,MAAM,IAAI,GAAgB,EAAE,CAAA;IAC5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC,IAAI,gBAAgB,EAAE,CAAC;QAC5D,MAAM,KAAK,GAAG,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,GAAG,gBAAgB,CAAC,CAAA;QACtD,MAAM,UAAU,GAAG,MAAM,UAAU,CACjC,KAAK,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,EAChC,MAAM,CACP,CAAA;QACD,KAAK,MAAM,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,EAAE,CAAC;YAC7C,MAAM,MAAM,GAAG,UAAU,CAAC,KAAK,CAAC,CAAA;YAChC,IAAI,CAAC,MAAM,EAAE,CAAC;gBACZ,MAAM,IAAI,KAAK,CAAC,+BAA+B,KAAK,CAAC,YAAY,IAAI,KAAK,CAAC,UAAU,GAAG,CAAC,CAAA;YAC3F,CAAC;YACD,IAAI,CAAC,IAAI,CAAC,EAAE,GAAG,KAAK,EAAE,MAAM,EAAE,CAAC,CAAA;QACjC,CAAC;IACH,CAAC;IAED,MAAM,SAAS,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IAC7B,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,QAAQ;QAChB,WAAW,EAAE,IAAI,CAAC,MAAM;QACxB,UAAU,EAAE,eAAe,CAAC,eAAe,CAAC;KAC7C,CAAC,CAAA;IAEF,OAAO;QACL,YAAY,EAAE,IAAI,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI;QAC/D,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,YAAY;QACZ,UAAU,EAAE,eAAe,CAAC,eAAe,CAAC;QAC5C,MAAM;KACP,CAAA;AACH,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,KAAK,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAC7C,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;IACpC,MAAM,KAAK,GAAG,MAAM,eAAe,CAAC,MAAM,CAAC,CAAA;IAC3C,MAAM,cAAc,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;IAC7D,MAAM,KAAK,GAAG,MAAM,aAAa,CAAC,MAAM,CAAC,CAAA;IAEzC,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO;YACL,YAAY,EAAE,EAAE;YAChB,cAAc;YACd,gBAAgB,EAAE,cAAc;YAChC,YAAY,EAAE,EAAE;YAChB,WAAW,EAAE,CAAC;SACf,CAAA;IACH,CAAC;IAED,MAAM,IAAI,GAAG,CAAC,MAAM,KAAK,CAAC,KAAK,EAAE,CAAC,KAAK,CAAC,cAAc,CAAC,CAAC,OAAO,EAAE,CAE/D,CAAA;IACF,MAAM,MAAM,GAAG,IAAI,GAAG,EAAkB,CAAA;IACxC,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QACvB,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,YAAY,EAAE,CAAC,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,YAAY,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;IACvE,CAAC;IAED,MAAM,YAAY,GAAG,IAAI,GAAG,CAAC,cAAc,CAAC,CAAA;IAC5C,MAAM,UAAU,GAAG,IAAI,GAAG,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC,CAAA;IAEzC,OAAO;QACL,YAAY,EAAE,CAAC,GAAG,MAAM,CAAC,OAAO,EAAE,CAAC;aAChC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC;aACtC,GAAG,CAAC,CAAC,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,MAAM,EAAE,CAAC,CAAC;QAClD,cAAc;QACd,gBAAgB,EAAE,cAAc,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,UAAU,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;QACxE,YAAY,EAAE,CAAC,GAAG,UAAU,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,EAAE;QAC9E,WAAW,EAAE,IAAI,CAAC,MAAM;KACzB,CAAA;AACH,CAAC"}
@@ -1 +1 @@
1
- {"version":3,"file":"init.d.ts","sourceRoot":"","sources":["../src/init.ts"],"names":[],"mappings":"AA2BA,wBAAsB,WAAW,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,CAwCxE"}
1
+ {"version":3,"file":"init.d.ts","sourceRoot":"","sources":["../src/init.ts"],"names":[],"mappings":"AAMA,wBAAsB,WAAW,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,CAwCxE"}
package/dist/init.js CHANGED
@@ -1,36 +1,16 @@
1
1
  import { existsSync } from "node:fs";
2
2
  import { mkdir, writeFile } from "node:fs/promises";
3
3
  import path from "node:path";
4
+ import { CONFIG_PATH, DEFAULT_CONFIG, KB_DIR, PRIVATE_DIR } from "./defaults.js";
4
5
  import { ensureMimirGitignore } from "./gitignore.js";
5
- const DEFAULT_CONFIG = {
6
- rawDir: "private",
7
- storageDir: ".kb/storage",
8
- sourcesFile: ".kb/sources.txt",
9
- accessLogPath: ".kb/access.log",
10
- tableName: "chunks",
11
- ollamaHost: "http://localhost:11434",
12
- networkPolicy: "local-only",
13
- embedModel: "nomic-embed-text",
14
- llmModel: "gemma4:latest",
15
- redaction: {
16
- enabled: true,
17
- builtIn: true,
18
- patterns: [],
19
- },
20
- accessLog: true,
21
- mcpMaxTopK: 10,
22
- topK: 5,
23
- chunkSize: 1200,
24
- chunkOverlap: 150,
25
- };
26
6
  export async function initProject(cwd = process.cwd()) {
27
7
  const root = path.resolve(cwd);
28
- const kbDir = path.join(root, ".kb");
29
- const privateDir = path.join(root, "private");
8
+ const kbDir = path.join(root, KB_DIR);
9
+ const privateDir = path.join(root, PRIVATE_DIR);
30
10
  const created = [];
31
11
  await mkdir(kbDir, { recursive: true });
32
12
  await mkdir(privateDir, { recursive: true });
33
- const configPath = path.join(kbDir, "config.json");
13
+ const configPath = path.join(root, CONFIG_PATH);
34
14
  if (!existsSync(configPath)) {
35
15
  await writeFile(configPath, `${JSON.stringify(DEFAULT_CONFIG, null, 2)}\n`, "utf8");
36
16
  created.push(path.relative(root, configPath));
package/dist/init.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"init.js","sourceRoot":"","sources":["../src/init.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACnD,OAAO,IAAI,MAAM,WAAW,CAAA;AAC5B,OAAO,EAAE,oBAAoB,EAAE,MAAM,gBAAgB,CAAA;AAErD,MAAM,cAAc,GAAG;IACrB,MAAM,EAAE,SAAS;IACjB,UAAU,EAAE,aAAa;IACzB,WAAW,EAAE,iBAAiB;IAC9B,aAAa,EAAE,gBAAgB;IAC/B,SAAS,EAAE,QAAQ;IACnB,UAAU,EAAE,wBAAwB;IACpC,aAAa,EAAE,YAAY;IAC3B,UAAU,EAAE,kBAAkB;IAC9B,QAAQ,EAAE,eAAe;IACzB,SAAS,EAAE;QACT,OAAO,EAAE,IAAI;QACb,OAAO,EAAE,IAAI;QACb,QAAQ,EAAE,EAAE;KACb;IACD,SAAS,EAAE,IAAI;IACf,UAAU,EAAE,EAAE;IACd,IAAI,EAAE,CAAC;IACP,SAAS,EAAE,IAAI;IACf,YAAY,EAAE,GAAG;CAClB,CAAA;AAED,MAAM,CAAC,KAAK,UAAU,WAAW,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IACnD,MAAM,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;IAC9B,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,KAAK,CAAC,CAAA;IACpC,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,SAAS,CAAC,CAAA;IAC7C,MAAM,OAAO,GAAa,EAAE,CAAA;IAE5B,MAAM,KAAK,CAAC,KAAK,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IACvC,MAAM,KAAK,CAAC,UAAU,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAE5C,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,aAAa,CAAC,CAAA;IAClD,IAAI,CAAC,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5B,MAAM,SAAS,CAAC,UAAU,EAAE,GAAG,IAAI,CAAC,SAAS,CAAC,cAAc,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;QACnF,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC,CAAA;IAC/C,CAAC;IAED,MAAM,WAAW,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,aAAa,CAAC,CAAA;IACnD,IAAI,CAAC,UAAU,CAAC,WAAW,CAAC,EAAE,CAAC;QAC7B,MAAM,SAAS,CACb,WAAW,EACX,8FAA8F,EAC9F,MAAM,CACP,CAAA;QACD,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,WAAW,CAAC,CAAC,CAAA;IAChD,CAAC;IAED,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,UAAU,EAAE,WAAW,CAAC,CAAA;IACrD,IAAI,CAAC,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5B,MAAM,SAAS,CACb,UAAU,EACV,6FAA6F,EAC7F,MAAM,CACP,CAAA;QACD,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC,CAAA;IAC/C,CAAC;IAED,IAAI,MAAM,oBAAoB,CAAC,IAAI,CAAC,EAAE,CAAC;QACrC,OAAO,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;IAC5B,CAAC;IAED,OAAO,OAAO,CAAA;AAChB,CAAC"}
1
+ {"version":3,"file":"init.js","sourceRoot":"","sources":["../src/init.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAA;AACpC,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACnD,OAAO,IAAI,MAAM,WAAW,CAAA;AAC5B,OAAO,EAAE,WAAW,EAAE,cAAc,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,eAAe,CAAA;AAChF,OAAO,EAAE,oBAAoB,EAAE,MAAM,gBAAgB,CAAA;AAErD,MAAM,CAAC,KAAK,UAAU,WAAW,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IACnD,MAAM,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;IAC9B,MAAM,KAAK,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IACrC,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,WAAW,CAAC,CAAA;IAC/C,MAAM,OAAO,GAAa,EAAE,CAAA;IAE5B,MAAM,KAAK,CAAC,KAAK,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IACvC,MAAM,KAAK,CAAC,UAAU,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAE5C,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,WAAW,CAAC,CAAA;IAC/C,IAAI,CAAC,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5B,MAAM,SAAS,CAAC,UAAU,EAAE,GAAG,IAAI,CAAC,SAAS,CAAC,cAAc,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;QACnF,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC,CAAA;IAC/C,CAAC;IAED,MAAM,WAAW,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,aAAa,CAAC,CAAA;IACnD,IAAI,CAAC,UAAU,CAAC,WAAW,CAAC,EAAE,CAAC;QAC7B,MAAM,SAAS,CACb,WAAW,EACX,8FAA8F,EAC9F,MAAM,CACP,CAAA;QACD,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,WAAW,CAAC,CAAC,CAAA;IAChD,CAAC;IAED,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,CAAC,UAAU,EAAE,WAAW,CAAC,CAAA;IACrD,IAAI,CAAC,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5B,MAAM,SAAS,CACb,UAAU,EACV,6FAA6F,EAC7F,MAAM,CACP,CAAA;QACD,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC,CAAA;IAC/C,CAAC;IAED,IAAI,MAAM,oBAAoB,CAAC,IAAI,CAAC,EAAE,CAAC;QACrC,OAAO,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;IAC5B,CAAC;IAED,OAAO,OAAO,CAAA;AAChB,CAAC"}
package/dist/mcp.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"mcp.d.ts","sourceRoot":"","sources":["../src/mcp.ts"],"names":[],"mappings":"AAUA,wBAAsB,QAAQ,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,IAAI,CAAC,CAiFjE"}
1
+ {"version":3,"file":"mcp.d.ts","sourceRoot":"","sources":["../src/mcp.ts"],"names":[],"mappings":"AAeA,wBAAsB,QAAQ,CAAC,GAAG,SAAgB,GAAG,OAAO,CAAC,IAAI,CAAC,CA6EjE"}
package/dist/mcp.js CHANGED
@@ -7,6 +7,10 @@ import { ask, search } from "./query.js";
7
7
  import { securityAudit } from "./security.js";
8
8
  import { countRows } from "./store.js";
9
9
  import { VERSION } from "./version.js";
10
+ const queryToolInputSchema = z.object({
11
+ query: z.string().min(1),
12
+ topK: z.number().int().positive().optional(),
13
+ });
10
14
  export async function serveMcp(cwd = process.cwd()) {
11
15
  const server = new McpServer({
12
16
  name: "mimir",
@@ -24,11 +28,14 @@ export async function serveMcp(cwd = process.cwd()) {
24
28
  rawDir: config.rawDir,
25
29
  storageDir: config.storageDir,
26
30
  sourcesFile: config.sourcesFile,
27
- embedModel: config.embedModel,
28
- llmModel: config.llmModel,
29
- networkPolicy: config.networkPolicy,
31
+ embeddingProvider: config.embeddingProvider,
32
+ embeddingModel: config.embeddingModel,
33
+ embeddingModelPath: config.embeddingModelPath,
34
+ transformersAllowRemoteModels: config.transformersAllowRemoteModels,
35
+ llmGeneration: false,
30
36
  redactionEnabled: config.redaction.enabled,
31
37
  mcpMaxTopK: config.mcpMaxTopK,
38
+ includeExtensions: config.includeExtensions,
32
39
  chunksIndexed,
33
40
  };
34
41
  return textResult(output);
@@ -36,18 +43,12 @@ export async function serveMcp(cwd = process.cwd()) {
36
43
  server.registerTool("mimir_search", {
37
44
  title: "Mimir Search",
38
45
  description: "Retrieve relevant passages from the local Mimir knowledge base.",
39
- inputSchema: z.object({
40
- query: z.string().min(1),
41
- topK: z.number().int().positive().optional(),
42
- }),
46
+ inputSchema: queryToolInputSchema,
43
47
  }, async ({ query, topK }) => textResult(await search(query, await searchOptions(cwd, topK))));
44
48
  server.registerTool("mimir_ask", {
45
49
  title: "Mimir Ask",
46
- description: "Answer a question using local retrieved passages and the configured Ollama model.",
47
- inputSchema: z.object({
48
- query: z.string().min(1),
49
- topK: z.number().int().positive().optional(),
50
- }),
50
+ description: "Return cited retrieval context for a question without calling an LLM.",
51
+ inputSchema: queryToolInputSchema,
51
52
  }, async ({ query, topK }) => textResult(await ask(query, await searchOptions(cwd, topK))));
52
53
  server.registerTool("mimir_audit", {
53
54
  title: "Mimir Audit",
@@ -56,7 +57,7 @@ export async function serveMcp(cwd = process.cwd()) {
56
57
  }, async () => textResult(await audit(cwd)));
57
58
  server.registerTool("mimir_security_audit", {
58
59
  title: "Mimir Security Audit",
59
- description: "Show local privacy, network, redaction, MCP, and gitignore posture.",
60
+ description: "Show local privacy, provider, redaction, MCP, and gitignore posture.",
60
61
  inputSchema: z.object({}),
61
62
  }, async () => textResult(await securityAudit(cwd)));
62
63
  await server.connect(new StdioServerTransport());
package/dist/mcp.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"mcp.js","sourceRoot":"","sources":["../src/mcp.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAA;AACnE,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAA;AAChF,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAA;AACvB,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,KAAK,EAAE,MAAM,aAAa,CAAA;AACnC,OAAO,EAAE,GAAG,EAAE,MAAM,EAAE,MAAM,YAAY,CAAA;AACxC,OAAO,EAAE,aAAa,EAAE,MAAM,eAAe,CAAA;AAC7C,OAAO,EAAE,SAAS,EAAE,MAAM,YAAY,CAAA;AACtC,OAAO,EAAE,OAAO,EAAE,MAAM,cAAc,CAAA;AAEtC,MAAM,CAAC,KAAK,UAAU,QAAQ,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAChD,MAAM,MAAM,GAAG,IAAI,SAAS,CAAC;QAC3B,IAAI,EAAE,OAAO;QACb,OAAO,EAAE,OAAO;KACjB,CAAC,CAAA;IAEF,MAAM,CAAC,YAAY,CACjB,cAAc,EACd;QACE,KAAK,EAAE,cAAc;QACrB,WAAW,EAAE,0DAA0D;QACvE,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE;QACT,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;QACpC,MAAM,aAAa,GAAG,MAAM,SAAS,CAAC,MAAM,CAAC,CAAA;QAC7C,MAAM,MAAM,GAAG;YACb,WAAW,EAAE,MAAM,CAAC,WAAW;YAC/B,MAAM,EAAE,MAAM,CAAC,MAAM;YACrB,UAAU,EAAE,MAAM,CAAC,UAAU;YAC7B,WAAW,EAAE,MAAM,CAAC,WAAW;YAC/B,UAAU,EAAE,MAAM,CAAC,UAAU;YAC7B,QAAQ,EAAE,MAAM,CAAC,QAAQ;YACzB,aAAa,EAAE,MAAM,CAAC,aAAa;YACnC,gBAAgB,EAAE,MAAM,CAAC,SAAS,CAAC,OAAO;YAC1C,UAAU,EAAE,MAAM,CAAC,UAAU;YAC7B,aAAa;SACd,CAAA;QAED,OAAO,UAAU,CAAC,MAAM,CAAC,CAAA;IAC3B,CAAC,CACF,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,cAAc,EACd;QACE,KAAK,EAAE,cAAc;QACrB,WAAW,EAAE,iEAAiE;QAC9E,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;YACxB,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE;SAC7C,CAAC;KACH,EACD,KAAK,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE,EAAE,CAAC,UAAU,CAAC,MAAM,MAAM,CAAC,KAAK,EAAE,MAAM,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAC,CAAC,CAC3F,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,WAAW,EACX;QACE,KAAK,EAAE,WAAW;QAClB,WAAW,EACT,mFAAmF;QACrF,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;YACxB,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE;SAC7C,CAAC;KACH,EACD,KAAK,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE,EAAE,CAAC,UAAU,CAAC,MAAM,GAAG,CAAC,KAAK,EAAE,MAAM,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAC,CAAC,CACxF,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,aAAa,EACb;QACE,KAAK,EAAE,aAAa;QACpB,WAAW,EAAE,uEAAuE;QACpF,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE,CAAC,UAAU,CAAC,MAAM,KAAK,CAAC,GAAG,CAAC,CAAC,CACzC,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,sBAAsB,EACtB;QACE,KAAK,EAAE,sBAAsB;QAC7B,WAAW,EAAE,qEAAqE;QAClF,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE,CAAC,UAAU,CAAC,MAAM,aAAa,CAAC,GAAG,CAAC,CAAC,CACjD,CAAA;IAED,MAAM,MAAM,CAAC,OAAO,CAAC,IAAI,oBAAoB,EAAE,CAAC,CAAA;AAClD,CAAC;AAED,SAAS,UAAU,CAAC,KAAc;IAChC,OAAO;QACL,OAAO,EAAE;YACP;gBACE,IAAI,EAAE,MAAM;gBACZ,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;aACrC;SACF;KACF,CAAA;AACH,CAAC;AAED,KAAK,UAAU,aAAa,CAC1B,GAAW,EACX,IAAwB;IAExB,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;IACpC,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI,EAAE,MAAM,CAAC,UAAU,CAAC,CAAA;IACpE,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,WAAW,EAAE,CAAA;AACnC,CAAC"}
1
+ {"version":3,"file":"mcp.js","sourceRoot":"","sources":["../src/mcp.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAA;AACnE,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAA;AAChF,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAA;AACvB,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,KAAK,EAAE,MAAM,aAAa,CAAA;AACnC,OAAO,EAAE,GAAG,EAAE,MAAM,EAAE,MAAM,YAAY,CAAA;AACxC,OAAO,EAAE,aAAa,EAAE,MAAM,eAAe,CAAA;AAC7C,OAAO,EAAE,SAAS,EAAE,MAAM,YAAY,CAAA;AACtC,OAAO,EAAE,OAAO,EAAE,MAAM,cAAc,CAAA;AAEtC,MAAM,oBAAoB,GAAG,CAAC,CAAC,MAAM,CAAC;IACpC,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;IACxB,IAAI,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE;CAC7C,CAAC,CAAA;AAEF,MAAM,CAAC,KAAK,UAAU,QAAQ,CAAC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE;IAChD,MAAM,MAAM,GAAG,IAAI,SAAS,CAAC;QAC3B,IAAI,EAAE,OAAO;QACb,OAAO,EAAE,OAAO;KACjB,CAAC,CAAA;IAEF,MAAM,CAAC,YAAY,CACjB,cAAc,EACd;QACE,KAAK,EAAE,cAAc;QACrB,WAAW,EAAE,0DAA0D;QACvE,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE;QACT,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;QACpC,MAAM,aAAa,GAAG,MAAM,SAAS,CAAC,MAAM,CAAC,CAAA;QAC7C,MAAM,MAAM,GAAG;YACb,WAAW,EAAE,MAAM,CAAC,WAAW;YAC/B,MAAM,EAAE,MAAM,CAAC,MAAM;YACrB,UAAU,EAAE,MAAM,CAAC,UAAU;YAC7B,WAAW,EAAE,MAAM,CAAC,WAAW;YAC/B,iBAAiB,EAAE,MAAM,CAAC,iBAAiB;YAC3C,cAAc,EAAE,MAAM,CAAC,cAAc;YACrC,kBAAkB,EAAE,MAAM,CAAC,kBAAkB;YAC7C,6BAA6B,EAAE,MAAM,CAAC,6BAA6B;YACnE,aAAa,EAAE,KAAK;YACpB,gBAAgB,EAAE,MAAM,CAAC,SAAS,CAAC,OAAO;YAC1C,UAAU,EAAE,MAAM,CAAC,UAAU;YAC7B,iBAAiB,EAAE,MAAM,CAAC,iBAAiB;YAC3C,aAAa;SACd,CAAA;QAED,OAAO,UAAU,CAAC,MAAM,CAAC,CAAA;IAC3B,CAAC,CACF,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,cAAc,EACd;QACE,KAAK,EAAE,cAAc;QACrB,WAAW,EAAE,iEAAiE;QAC9E,WAAW,EAAE,oBAAoB;KAClC,EACD,KAAK,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE,EAAE,CAAC,UAAU,CAAC,MAAM,MAAM,CAAC,KAAK,EAAE,MAAM,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAC,CAAC,CAC3F,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,WAAW,EACX;QACE,KAAK,EAAE,WAAW;QAClB,WAAW,EAAE,uEAAuE;QACpF,WAAW,EAAE,oBAAoB;KAClC,EACD,KAAK,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE,EAAE,CAAC,UAAU,CAAC,MAAM,GAAG,CAAC,KAAK,EAAE,MAAM,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAC,CAAC,CACxF,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,aAAa,EACb;QACE,KAAK,EAAE,aAAa;QACpB,WAAW,EAAE,uEAAuE;QACpF,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE,CAAC,UAAU,CAAC,MAAM,KAAK,CAAC,GAAG,CAAC,CAAC,CACzC,CAAA;IAED,MAAM,CAAC,YAAY,CACjB,sBAAsB,EACtB;QACE,KAAK,EAAE,sBAAsB;QAC7B,WAAW,EAAE,sEAAsE;QACnF,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC;KAC1B,EACD,KAAK,IAAI,EAAE,CAAC,UAAU,CAAC,MAAM,aAAa,CAAC,GAAG,CAAC,CAAC,CACjD,CAAA;IAED,MAAM,MAAM,CAAC,OAAO,CAAC,IAAI,oBAAoB,EAAE,CAAC,CAAA;AAClD,CAAC;AAED,SAAS,UAAU,CAAC,KAAc;IAChC,OAAO;QACL,OAAO,EAAE;YACP;gBACE,IAAI,EAAE,MAAM;gBACZ,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;aACrC;SACF;KACF,CAAA;AACH,CAAC;AAED,KAAK,UAAU,aAAa,CAC1B,GAAW,EACX,IAAwB;IAExB,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,GAAG,CAAC,CAAA;IACpC,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI,EAAE,MAAM,CAAC,UAAU,CAAC,CAAA;IACpE,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,WAAW,EAAE,CAAA;AACnC,CAAC"}
@@ -1 +1 @@
1
- {"version":3,"file":"parsing.d.ts","sourceRoot":"","sources":["../src/parsing.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,YAAY,CAAA;AAE5D,wBAAsB,SAAS,CAAC,IAAI,EAAE,UAAU,GAAG,OAAO,CAAC,cAAc,CAAC,CA6BzE"}
1
+ {"version":3,"file":"parsing.d.ts","sourceRoot":"","sources":["../src/parsing.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,YAAY,CAAA;AAI5D,wBAAsB,SAAS,CAAC,IAAI,EAAE,UAAU,GAAG,OAAO,CAAC,cAAc,CAAC,CA8CzE"}
package/dist/parsing.js CHANGED
@@ -1,13 +1,29 @@
1
1
  import { readFile } from "node:fs/promises";
2
+ import { strFromU8, unzipSync } from "fflate";
2
3
  import { htmlToText } from "html-to-text";
3
4
  import { extractText, getDocumentProxy } from "unpdf";
4
5
  import YAML from "yaml";
6
+ const MAX_OFFICE_XML_ENTRY_BYTES = 25_000_000;
5
7
  export async function parseFile(file) {
6
8
  let text;
7
9
  switch (file.extension) {
8
10
  case ".pdf":
9
11
  text = await parsePdf(file.absolutePath);
10
12
  break;
13
+ case ".docx":
14
+ text = await parseDocx(file.absolutePath);
15
+ break;
16
+ case ".pptx":
17
+ text = await parsePptx(file.absolutePath);
18
+ break;
19
+ case ".xlsx":
20
+ text = await parseXlsx(file.absolutePath);
21
+ break;
22
+ case ".odt":
23
+ case ".ods":
24
+ case ".odp":
25
+ text = await parseOpenDocument(file.absolutePath);
26
+ break;
11
27
  case ".html":
12
28
  case ".htm":
13
29
  text = htmlToText(await readFile(file.absolutePath, "utf8"), {
@@ -25,11 +41,133 @@ export async function parseFile(file) {
25
41
  case ".yml":
26
42
  text = YAML.stringify(YAML.parse(await readFile(file.absolutePath, "utf8")));
27
43
  break;
44
+ case ".rtf":
45
+ text = stripRtf(await readFile(file.absolutePath, "utf8"));
46
+ break;
28
47
  default:
29
48
  text = await readFile(file.absolutePath, "utf8");
30
49
  }
31
50
  return { file, text: normalizeText(text) };
32
51
  }
52
+ async function parseDocx(filePath) {
53
+ const entries = unzipOfficeFile(await readFile(filePath));
54
+ return xmlEntriesToText(entries, [
55
+ /^word\/document\.xml$/u,
56
+ /^word\/header\d*\.xml$/u,
57
+ /^word\/footer\d*\.xml$/u,
58
+ /^word\/footnotes\.xml$/u,
59
+ /^word\/endnotes\.xml$/u,
60
+ /^word\/comments\.xml$/u,
61
+ ]);
62
+ }
63
+ async function parsePptx(filePath) {
64
+ const entries = unzipOfficeFile(await readFile(filePath));
65
+ return xmlEntriesToText(entries, [
66
+ /^ppt\/slides\/slide\d+\.xml$/u,
67
+ /^ppt\/notesSlides\/notesSlide\d+\.xml$/u,
68
+ ]);
69
+ }
70
+ async function parseXlsx(filePath) {
71
+ const entries = unzipOfficeFile(await readFile(filePath));
72
+ const sharedStrings = parseSharedStrings(entries.get("xl/sharedStrings.xml") ?? "");
73
+ const sheets = [...entries.entries()]
74
+ .filter(([name]) => /^xl\/worksheets\/sheet\d+\.xml$/u.test(name))
75
+ .sort(([a], [b]) => a.localeCompare(b));
76
+ const rows = [];
77
+ for (const [name, xml] of sheets) {
78
+ const values = parseSheetValues(xml, sharedStrings);
79
+ if (values.length > 0) {
80
+ rows.push(`# ${name}`, values.join("\n"));
81
+ }
82
+ }
83
+ return rows.join("\n\n");
84
+ }
85
+ async function parseOpenDocument(filePath) {
86
+ const entries = unzipOfficeFile(await readFile(filePath));
87
+ return xmlEntriesToText(entries, [/^content\.xml$/u, /^meta\.xml$/u]);
88
+ }
89
+ function unzipOfficeFile(buffer) {
90
+ const unzipped = unzipSync(new Uint8Array(buffer), {
91
+ filter: (file) => file.originalSize <= MAX_OFFICE_XML_ENTRY_BYTES,
92
+ });
93
+ const entries = new Map();
94
+ for (const [name, content] of Object.entries(unzipped)) {
95
+ if (name.endsWith(".xml")) {
96
+ entries.set(name, strFromU8(content));
97
+ }
98
+ }
99
+ return entries;
100
+ }
101
+ function xmlEntriesToText(entries, patterns) {
102
+ const parts = [];
103
+ for (const [name, xml] of [...entries.entries()].sort(([a], [b]) => a.localeCompare(b))) {
104
+ if (patterns.some((pattern) => pattern.test(name))) {
105
+ const text = xmlToText(xml);
106
+ if (text) {
107
+ parts.push(text);
108
+ }
109
+ }
110
+ }
111
+ return parts.join("\n\n");
112
+ }
113
+ function parseSharedStrings(xml) {
114
+ return [...xml.matchAll(/<si\b[\s\S]*?<\/si>/gu)].map(([item]) => xmlToText(item));
115
+ }
116
+ function parseSheetValues(xml, sharedStrings) {
117
+ const rows = [];
118
+ for (const rowMatch of xml.matchAll(/<row\b[\s\S]*?<\/row>/gu)) {
119
+ const rowXml = rowMatch[0];
120
+ const values = [...rowXml.matchAll(/<c\b([^>]*)>([\s\S]*?)<\/c>/gu)]
121
+ .map((cellMatch) => {
122
+ const attrs = cellMatch[1] ?? "";
123
+ const cellXml = cellMatch[2] ?? "";
124
+ const inline = firstMatch(cellXml, /<is\b[\s\S]*?<\/is>/u);
125
+ if (inline) {
126
+ return xmlToText(inline);
127
+ }
128
+ const rawValue = firstMatch(cellXml, /<v>([\s\S]*?)<\/v>/u);
129
+ if (!rawValue) {
130
+ return "";
131
+ }
132
+ if (/\bt="s"/u.test(attrs)) {
133
+ return sharedStrings[Number.parseInt(rawValue, 10)] ?? "";
134
+ }
135
+ return decodeXmlEntities(rawValue);
136
+ })
137
+ .filter(Boolean);
138
+ if (values.length > 0) {
139
+ rows.push(values.join("\t"));
140
+ }
141
+ }
142
+ return rows;
143
+ }
144
+ function firstMatch(input, pattern) {
145
+ const match = input.match(pattern);
146
+ return match?.[1] ?? match?.[0] ?? "";
147
+ }
148
+ function xmlToText(xml) {
149
+ return normalizeText(decodeXmlEntities(xml
150
+ .replace(/<w:tab\/>/gu, " ")
151
+ .replace(/<w:br\/>/gu, "\n")
152
+ .replace(/<\/(?:w:p|a:p|text:p|text:h|table:table-row)>/gu, "\n")
153
+ .replace(/<[^>]+>/gu, " ")
154
+ .replace(/[ \t]{2,}/gu, " ")));
155
+ }
156
+ function stripRtf(input) {
157
+ return input
158
+ .replace(/\\par[d]?/gu, "\n")
159
+ .replace(/\\'[0-9a-fA-F]{2}/gu, " ")
160
+ .replace(/\\[a-zA-Z]+-?\d* ?/gu, " ")
161
+ .replace(/[{}]/gu, " ");
162
+ }
163
+ function decodeXmlEntities(input) {
164
+ return input
165
+ .replace(/&lt;/gu, "<")
166
+ .replace(/&gt;/gu, ">")
167
+ .replace(/&quot;/gu, '"')
168
+ .replace(/&apos;/gu, "'")
169
+ .replace(/&amp;/gu, "&");
170
+ }
33
171
  async function parsePdf(filePath) {
34
172
  const buffer = await readFile(filePath);
35
173
  const pdf = await getDocumentProxy(new Uint8Array(buffer));
@@ -1 +1 @@
1
- {"version":3,"file":"parsing.js","sourceRoot":"","sources":["../src/parsing.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAA;AAC3C,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAA;AACzC,OAAO,EAAE,WAAW,EAAE,gBAAgB,EAAE,MAAM,OAAO,CAAA;AACrD,OAAO,IAAI,MAAM,MAAM,CAAA;AAGvB,MAAM,CAAC,KAAK,UAAU,SAAS,CAAC,IAAgB;IAC9C,IAAI,IAAY,CAAA;IAEhB,QAAQ,IAAI,CAAC,SAAS,EAAE,CAAC;QACvB,KAAK,MAAM;YACT,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACxC,MAAK;QACP,KAAK,OAAO,CAAC;QACb,KAAK,MAAM;YACT,IAAI,GAAG,UAAU,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,EAAE;gBAC3D,QAAQ,EAAE,KAAK;gBACf,SAAS,EAAE;oBACT,EAAE,QAAQ,EAAE,GAAG,EAAE,OAAO,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,EAAE;oBAChD,EAAE,QAAQ,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE;iBACpC;aACF,CAAC,CAAA;YACF,MAAK;QACP,KAAK,OAAO;YACV,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,CAAA;YACrF,MAAK;QACP,KAAK,OAAO,CAAC;QACb,KAAK,MAAM;YACT,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,CAAC,CAAA;YAC5E,MAAK;QACP;YACE,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAA;IACpD,CAAC;IAED,OAAO,EAAE,IAAI,EAAE,IAAI,EAAE,aAAa,CAAC,IAAI,CAAC,EAAE,CAAA;AAC5C,CAAC;AAED,KAAK,UAAU,QAAQ,CAAC,QAAgB;IACtC,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAA;IACvC,MAAM,GAAG,GAAG,MAAM,gBAAgB,CAAC,IAAI,UAAU,CAAC,MAAM,CAAC,CAAC,CAAA;IAC1D,MAAM,MAAM,GAAG,MAAM,WAAW,CAAC,GAAG,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;IAC3D,OAAO,MAAM,CAAC,IAAI,CAAA;AACpB,CAAC;AAED,SAAS,aAAa,CAAC,KAAa;IAClC,OAAO,KAAK;SACT,OAAO,CAAC,OAAO,EAAE,IAAI,CAAC;SACtB,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC;SAC1B,OAAO,CAAC,SAAS,EAAE,QAAQ,CAAC;SAC5B,IAAI,EAAE,CAAA;AACX,CAAC"}
1
+ {"version":3,"file":"parsing.js","sourceRoot":"","sources":["../src/parsing.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAA;AAC3C,OAAO,EAAE,SAAS,EAAE,SAAS,EAAE,MAAM,QAAQ,CAAA;AAC7C,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAA;AACzC,OAAO,EAAE,WAAW,EAAE,gBAAgB,EAAE,MAAM,OAAO,CAAA;AACrD,OAAO,IAAI,MAAM,MAAM,CAAA;AAGvB,MAAM,0BAA0B,GAAG,UAAU,CAAA;AAE7C,MAAM,CAAC,KAAK,UAAU,SAAS,CAAC,IAAgB;IAC9C,IAAI,IAAY,CAAA;IAEhB,QAAQ,IAAI,CAAC,SAAS,EAAE,CAAC;QACvB,KAAK,MAAM;YACT,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACxC,MAAK;QACP,KAAK,OAAO;YACV,IAAI,GAAG,MAAM,SAAS,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACzC,MAAK;QACP,KAAK,OAAO;YACV,IAAI,GAAG,MAAM,SAAS,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACzC,MAAK;QACP,KAAK,OAAO;YACV,IAAI,GAAG,MAAM,SAAS,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACzC,MAAK;QACP,KAAK,MAAM,CAAC;QACZ,KAAK,MAAM,CAAC;QACZ,KAAK,MAAM;YACT,IAAI,GAAG,MAAM,iBAAiB,CAAC,IAAI,CAAC,YAAY,CAAC,CAAA;YACjD,MAAK;QACP,KAAK,OAAO,CAAC;QACb,KAAK,MAAM;YACT,IAAI,GAAG,UAAU,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,EAAE;gBAC3D,QAAQ,EAAE,KAAK;gBACf,SAAS,EAAE;oBACT,EAAE,QAAQ,EAAE,GAAG,EAAE,OAAO,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,EAAE;oBAChD,EAAE,QAAQ,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE;iBACpC;aACF,CAAC,CAAA;YACF,MAAK;QACP,KAAK,OAAO;YACV,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,CAAA;YACrF,MAAK;QACP,KAAK,OAAO,CAAC;QACb,KAAK,MAAM;YACT,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,CAAC,CAAA;YAC5E,MAAK;QACP,KAAK,MAAM;YACT,IAAI,GAAG,QAAQ,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,CAAA;YAC1D,MAAK;QACP;YACE,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAA;IACpD,CAAC;IAED,OAAO,EAAE,IAAI,EAAE,IAAI,EAAE,aAAa,CAAC,IAAI,CAAC,EAAE,CAAA;AAC5C,CAAC;AAED,KAAK,UAAU,SAAS,CAAC,QAAgB;IACvC,MAAM,OAAO,GAAG,eAAe,CAAC,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC,CAAA;IACzD,OAAO,gBAAgB,CAAC,OAAO,EAAE;QAC/B,wBAAwB;QACxB,yBAAyB;QACzB,yBAAyB;QACzB,yBAAyB;QACzB,wBAAwB;QACxB,wBAAwB;KACzB,CAAC,CAAA;AACJ,CAAC;AAED,KAAK,UAAU,SAAS,CAAC,QAAgB;IACvC,MAAM,OAAO,GAAG,eAAe,CAAC,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC,CAAA;IACzD,OAAO,gBAAgB,CAAC,OAAO,EAAE;QAC/B,+BAA+B;QAC/B,yCAAyC;KAC1C,CAAC,CAAA;AACJ,CAAC;AAED,KAAK,UAAU,SAAS,CAAC,QAAgB;IACvC,MAAM,OAAO,GAAG,eAAe,CAAC,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC,CAAA;IACzD,MAAM,aAAa,GAAG,kBAAkB,CAAC,OAAO,CAAC,GAAG,CAAC,sBAAsB,CAAC,IAAI,EAAE,CAAC,CAAA;IACnF,MAAM,MAAM,GAAG,CAAC,GAAG,OAAO,CAAC,OAAO,EAAE,CAAC;SAClC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,EAAE,CAAC,kCAAkC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;SACjE,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,CAAA;IAEzC,MAAM,IAAI,GAAa,EAAE,CAAA;IACzB,KAAK,MAAM,CAAC,IAAI,EAAE,GAAG,CAAC,IAAI,MAAM,EAAE,CAAC;QACjC,MAAM,MAAM,GAAG,gBAAgB,CAAC,GAAG,EAAE,aAAa,CAAC,CAAA;QACnD,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,IAAI,CAAC,IAAI,CAAC,KAAK,IAAI,EAAE,EAAE,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,CAAA;QAC3C,CAAC;IACH,CAAC;IACD,OAAO,IAAI,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;AAC1B,CAAC;AAED,KAAK,UAAU,iBAAiB,CAAC,QAAgB;IAC/C,MAAM,OAAO,GAAG,eAAe,CAAC,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC,CAAA;IACzD,OAAO,gBAAgB,CAAC,OAAO,EAAE,CAAC,iBAAiB,EAAE,cAAc,CAAC,CAAC,CAAA;AACvE,CAAC;AAED,SAAS,eAAe,CAAC,MAAc;IACrC,MAAM,QAAQ,GAAG,SAAS,CAAC,IAAI,UAAU,CAAC,MAAM,CAAC,EAAE;QACjD,MAAM,EAAE,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,YAAY,IAAI,0BAA0B;KAClE,CAAC,CAAA;IACF,MAAM,OAAO,GAAG,IAAI,GAAG,EAAkB,CAAA;IACzC,KAAK,MAAM,CAAC,IAAI,EAAE,OAAO,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,QAAQ,CAAC,EAAE,CAAC;QACvD,IAAI,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,CAAC;YAC1B,OAAO,CAAC,GAAG,CAAC,IAAI,EAAE,SAAS,CAAC,OAAO,CAAC,CAAC,CAAA;QACvC,CAAC;IACH,CAAC;IACD,OAAO,OAAO,CAAA;AAChB,CAAC;AAED,SAAS,gBAAgB,CAAC,OAA4B,EAAE,QAAkB;IACxE,MAAM,KAAK,GAAa,EAAE,CAAA;IAC1B,KAAK,MAAM,CAAC,IAAI,EAAE,GAAG,CAAC,IAAI,CAAC,GAAG,OAAO,CAAC,OAAO,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC;QACxF,IAAI,QAAQ,CAAC,IAAI,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC;YACnD,MAAM,IAAI,GAAG,SAAS,CAAC,GAAG,CAAC,CAAA;YAC3B,IAAI,IAAI,EAAE,CAAC;gBACT,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;YAClB,CAAC;QACH,CAAC;IACH,CAAC;IACD,OAAO,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;AAC3B,CAAC;AAED,SAAS,kBAAkB,CAAC,GAAW;IACrC,OAAO,CAAC,GAAG,GAAG,CAAC,QAAQ,CAAC,uBAAuB,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,EAAE,CAAC,SAAS,CAAC,IAAI,CAAC,CAAC,CAAA;AACpF,CAAC;AAED,SAAS,gBAAgB,CAAC,GAAW,EAAE,aAAuB;IAC5D,MAAM,IAAI,GAAa,EAAE,CAAA;IACzB,KAAK,MAAM,QAAQ,IAAI,GAAG,CAAC,QAAQ,CAAC,yBAAyB,CAAC,EAAE,CAAC;QAC/D,MAAM,MAAM,GAAG,QAAQ,CAAC,CAAC,CAAC,CAAA;QAC1B,MAAM,MAAM,GAAG,CAAC,GAAG,MAAM,CAAC,QAAQ,CAAC,+BAA+B,CAAC,CAAC;aACjE,GAAG,CAAC,CAAC,SAAS,EAAE,EAAE;YACjB,MAAM,KAAK,GAAG,SAAS,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;YAChC,MAAM,OAAO,GAAG,SAAS,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;YAClC,MAAM,MAAM,GAAG,UAAU,CAAC,OAAO,EAAE,sBAAsB,CAAC,CAAA;YAC1D,IAAI,MAAM,EAAE,CAAC;gBACX,OAAO,SAAS,CAAC,MAAM,CAAC,CAAA;YAC1B,CAAC;YAED,MAAM,QAAQ,GAAG,UAAU,CAAC,OAAO,EAAE,qBAAqB,CAAC,CAAA;YAC3D,IAAI,CAAC,QAAQ,EAAE,CAAC;gBACd,OAAO,EAAE,CAAA;YACX,CAAC;YAED,IAAI,UAAU,CAAC,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC3B,OAAO,aAAa,CAAC,MAAM,CAAC,QAAQ,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAC3D,CAAC;YACD,OAAO,iBAAiB,CAAC,QAAQ,CAAC,CAAA;QACpC,CAAC,CAAC;aACD,MAAM,CAAC,OAAO,CAAC,CAAA;QAElB,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,IAAI,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,CAAA;QAC9B,CAAC;IACH,CAAC;IACD,OAAO,IAAI,CAAA;AACb,CAAC;AAED,SAAS,UAAU,CAAC,KAAa,EAAE,OAAe;IAChD,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,OAAO,CAAC,CAAA;IAClC,OAAO,KAAK,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;AACvC,CAAC;AAED,SAAS,SAAS,CAAC,GAAW;IAC5B,OAAO,aAAa,CAClB,iBAAiB,CACf,GAAG;SACA,OAAO,CAAC,aAAa,EAAE,GAAG,CAAC;SAC3B,OAAO,CAAC,YAAY,EAAE,IAAI,CAAC;SAC3B,OAAO,CAAC,iDAAiD,EAAE,IAAI,CAAC;SAChE,OAAO,CAAC,WAAW,EAAE,GAAG,CAAC;SACzB,OAAO,CAAC,aAAa,EAAE,GAAG,CAAC,CAC/B,CACF,CAAA;AACH,CAAC;AAED,SAAS,QAAQ,CAAC,KAAa;IAC7B,OAAO,KAAK;SACT,OAAO,CAAC,aAAa,EAAE,IAAI,CAAC;SAC5B,OAAO,CAAC,qBAAqB,EAAE,GAAG,CAAC;SACnC,OAAO,CAAC,sBAAsB,EAAE,GAAG,CAAC;SACpC,OAAO,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAA;AAC3B,CAAC;AAED,SAAS,iBAAiB,CAAC,KAAa;IACtC,OAAO,KAAK;SACT,OAAO,CAAC,QAAQ,EAAE,GAAG,CAAC;SACtB,OAAO,CAAC,QAAQ,EAAE,GAAG,CAAC;SACtB,OAAO,CAAC,UAAU,EAAE,GAAG,CAAC;SACxB,OAAO,CAAC,UAAU,EAAE,GAAG,CAAC;SACxB,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC,CAAA;AAC5B,CAAC;AAED,KAAK,UAAU,QAAQ,CAAC,QAAgB;IACtC,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAA;IACvC,MAAM,GAAG,GAAG,MAAM,gBAAgB,CAAC,IAAI,UAAU,CAAC,MAAM,CAAC,CAAC,CAAA;IAC1D,MAAM,MAAM,GAAG,MAAM,WAAW,CAAC,GAAG,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;IAC3D,OAAO,MAAM,CAAC,IAAI,CAAA;AACpB,CAAC;AAED,SAAS,aAAa,CAAC,KAAa;IAClC,OAAO,KAAK;SACT,OAAO,CAAC,OAAO,EAAE,IAAI,CAAC;SACtB,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC;SAC1B,OAAO,CAAC,SAAS,EAAE,QAAQ,CAAC;SAC5B,IAAI,EAAE,CAAA;AACX,CAAC"}
@@ -1 +1 @@
1
- {"version":3,"file":"query.d.ts","sourceRoot":"","sources":["../src/query.ts"],"names":[],"mappings":"AAMA,OAAO,KAAK,EAAE,SAAS,EAAE,aAAa,EAAE,YAAY,EAAE,MAAM,YAAY,CAAA;AAUxE,wBAAsB,MAAM,CAAC,KAAK,EAAE,MAAM,EAAE,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,YAAY,EAAE,CAAC,CA2BhG;AAED,wBAAsB,GAAG,CAAC,KAAK,EAAE,MAAM,EAAE,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,SAAS,CAAC,CA+CxF"}
1
+ {"version":3,"file":"query.d.ts","sourceRoot":"","sources":["../src/query.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,SAAS,EAAE,aAAa,EAAE,YAAY,EAAE,MAAM,YAAY,CAAA;AAUxE,wBAAsB,MAAM,CAAC,KAAK,EAAE,MAAM,EAAE,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,YAAY,EAAE,CAAC,CA2BhG;AAED,wBAAsB,GAAG,CAAC,KAAK,EAAE,MAAM,EAAE,OAAO,GAAE,aAAkB,GAAG,OAAO,CAAC,SAAS,CAAC,CAsBxF"}
package/dist/query.js CHANGED
@@ -1,8 +1,6 @@
1
- import { Ollama } from "ollama";
2
1
  import { recordAccess } from "./access-log.js";
3
2
  import { loadConfig } from "./config.js";
4
3
  import { embedText } from "./embeddings.js";
5
- import { assertNetworkPolicy } from "./network.js";
6
4
  import { openRowsTable } from "./store.js";
7
5
  export async function search(query, options = {}) {
8
6
  const config = await loadConfig(String(options.cwd ?? process.cwd()));
@@ -39,25 +37,6 @@ export async function ask(query, options = {}) {
39
37
  sources,
40
38
  };
41
39
  }
42
- const context = sources
43
- .map((source, index) => `[${index + 1}] ${source.relativePath}#${source.chunkIndex}\n${source.text}`)
44
- .join("\n\n---\n\n");
45
- assertNetworkPolicy(config);
46
- const client = new Ollama({ host: config.ollamaHost });
47
- const response = await client.chat({
48
- model: config.llmModel,
49
- messages: [
50
- {
51
- role: "system",
52
- content: "Answer only from the provided context. If the context is insufficient, say what is missing. Cite sources with [1], [2], etc.",
53
- },
54
- {
55
- role: "user",
56
- content: `Question:\n${query}\n\nContext:\n${context}`,
57
- },
58
- ],
59
- stream: false,
60
- });
61
40
  await recordAccess(config, {
62
41
  action: "ask",
63
42
  query,
@@ -65,8 +44,21 @@ export async function ask(query, options = {}) {
65
44
  resultCount: sources.length,
66
45
  });
67
46
  return {
68
- answer: response.message.content,
47
+ answer: retrievalOnlyAnswer(sources),
69
48
  sources,
70
49
  };
71
50
  }
51
+ function retrievalOnlyAnswer(sources) {
52
+ const snippets = sources
53
+ .map((source, index) => {
54
+ const text = source.text.replace(/\s+/gu, " ").trim();
55
+ return `[${index + 1}] ${source.relativePath}#${source.chunkIndex}: ${text}`;
56
+ })
57
+ .join("\n\n");
58
+ return [
59
+ "Mimir returns retrieval context only. Use these passages as grounded context for your agent or LLM:",
60
+ "",
61
+ snippets,
62
+ ].join("\n");
63
+ }
72
64
  //# sourceMappingURL=query.js.map
package/dist/query.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"query.js","sourceRoot":"","sources":["../src/query.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,MAAM,EAAE,MAAM,QAAQ,CAAA;AAC/B,OAAO,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAA;AAC9C,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,SAAS,EAAE,MAAM,iBAAiB,CAAA;AAC3C,OAAO,EAAE,mBAAmB,EAAE,MAAM,cAAc,CAAA;AAClD,OAAO,EAAE,aAAa,EAAE,MAAM,YAAY,CAAA;AAW1C,MAAM,CAAC,KAAK,UAAU,MAAM,CAAC,KAAa,EAAE,UAAyB,EAAE;IACrE,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,KAAK,GAAG,MAAM,aAAa,CAAC,MAAM,CAAC,CAAA;IACzC,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,EAAE,CAAA;IACX,CAAC;IAED,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,KAAK,EAAE,MAAM,CAAC,CAAA;IAC7C,MAAM,IAAI,GAAG,CAAC,MAAM,KAAK;SACtB,YAAY,CAAC,MAAM,CAAC;SACpB,KAAK,CAAC,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI,CAAC;SAClC,OAAO,EAAE,CAAgB,CAAA;IAE5B,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC;QACjC,MAAM,EAAE,GAAG,CAAC,MAAM;QAClB,YAAY,EAAE,GAAG,CAAC,YAAY;QAC9B,UAAU,EAAE,GAAG,CAAC,UAAU;QAC1B,IAAI,EAAE,GAAG,CAAC,IAAI;QACd,QAAQ,EAAE,OAAO,GAAG,CAAC,SAAS,KAAK,QAAQ,CAAC,CAAC,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI;KACnE,CAAC,CAAC,CAAA;IACH,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,QAAQ;QAChB,KAAK;QACL,IAAI,EAAE,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI;QACjC,WAAW,EAAE,OAAO,CAAC,MAAM;KAC5B,CAAC,CAAA;IACF,OAAO,OAAO,CAAA;AAChB,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,GAAG,CAAC,KAAa,EAAE,UAAyB,EAAE;IAClE,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,OAAO,GAAG,MAAM,MAAM,CAAC,KAAK,EAAE,OAAO,CAAC,CAAA;IAE5C,IAAI,OAAO,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACzB,OAAO;YACL,MAAM,EAAE,2EAA2E;YACnF,OAAO;SACR,CAAA;IACH,CAAC;IAED,MAAM,OAAO,GAAG,OAAO;SACpB,GAAG,CACF,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAChB,IAAI,KAAK,GAAG,CAAC,KAAK,MAAM,CAAC,YAAY,IAAI,MAAM,CAAC,UAAU,KAAK,MAAM,CAAC,IAAI,EAAE,CAC/E;SACA,IAAI,CAAC,aAAa,CAAC,CAAA;IAEtB,mBAAmB,CAAC,MAAM,CAAC,CAAA;IAC3B,MAAM,MAAM,GAAG,IAAI,MAAM,CAAC,EAAE,IAAI,EAAE,MAAM,CAAC,UAAU,EAAE,CAAC,CAAA;IACtD,MAAM,QAAQ,GAAG,MAAM,MAAM,CAAC,IAAI,CAAC;QACjC,KAAK,EAAE,MAAM,CAAC,QAAQ;QACtB,QAAQ,EAAE;YACR;gBACE,IAAI,EAAE,QAAQ;gBACd,OAAO,EACL,8HAA8H;aACjI;YACD;gBACE,IAAI,EAAE,MAAM;gBACZ,OAAO,EAAE,cAAc,KAAK,iBAAiB,OAAO,EAAE;aACvD;SACF;QACD,MAAM,EAAE,KAAK;KACd,CAAC,CAAA;IAEF,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,KAAK;QACb,KAAK;QACL,IAAI,EAAE,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI;QACjC,WAAW,EAAE,OAAO,CAAC,MAAM;KAC5B,CAAC,CAAA;IAEF,OAAO;QACL,MAAM,EAAE,QAAQ,CAAC,OAAO,CAAC,OAAO;QAChC,OAAO;KACR,CAAA;AACH,CAAC"}
1
+ {"version":3,"file":"query.js","sourceRoot":"","sources":["../src/query.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAA;AAC9C,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AACxC,OAAO,EAAE,SAAS,EAAE,MAAM,iBAAiB,CAAA;AAC3C,OAAO,EAAE,aAAa,EAAE,MAAM,YAAY,CAAA;AAW1C,MAAM,CAAC,KAAK,UAAU,MAAM,CAAC,KAAa,EAAE,UAAyB,EAAE;IACrE,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,KAAK,GAAG,MAAM,aAAa,CAAC,MAAM,CAAC,CAAA;IACzC,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,EAAE,CAAA;IACX,CAAC;IAED,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,KAAK,EAAE,MAAM,CAAC,CAAA;IAC7C,MAAM,IAAI,GAAG,CAAC,MAAM,KAAK;SACtB,YAAY,CAAC,MAAM,CAAC;SACpB,KAAK,CAAC,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI,CAAC;SAClC,OAAO,EAAE,CAAgB,CAAA;IAE5B,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC;QACjC,MAAM,EAAE,GAAG,CAAC,MAAM;QAClB,YAAY,EAAE,GAAG,CAAC,YAAY;QAC9B,UAAU,EAAE,GAAG,CAAC,UAAU;QAC1B,IAAI,EAAE,GAAG,CAAC,IAAI;QACd,QAAQ,EAAE,OAAO,GAAG,CAAC,SAAS,KAAK,QAAQ,CAAC,CAAC,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI;KACnE,CAAC,CAAC,CAAA;IACH,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,QAAQ;QAChB,KAAK;QACL,IAAI,EAAE,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI;QACjC,WAAW,EAAE,OAAO,CAAC,MAAM;KAC5B,CAAC,CAAA;IACF,OAAO,OAAO,CAAA;AAChB,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,GAAG,CAAC,KAAa,EAAE,UAAyB,EAAE;IAClE,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,MAAM,CAAC,OAAO,CAAC,GAAG,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;IACrE,MAAM,OAAO,GAAG,MAAM,MAAM,CAAC,KAAK,EAAE,OAAO,CAAC,CAAA;IAE5C,IAAI,OAAO,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACzB,OAAO;YACL,MAAM,EAAE,2EAA2E;YACnF,OAAO;SACR,CAAA;IACH,CAAC;IAED,MAAM,YAAY,CAAC,MAAM,EAAE;QACzB,MAAM,EAAE,KAAK;QACb,KAAK;QACL,IAAI,EAAE,OAAO,CAAC,IAAI,IAAI,MAAM,CAAC,IAAI;QACjC,WAAW,EAAE,OAAO,CAAC,MAAM;KAC5B,CAAC,CAAA;IAEF,OAAO;QACL,MAAM,EAAE,mBAAmB,CAAC,OAAO,CAAC;QACpC,OAAO;KACR,CAAA;AACH,CAAC;AAED,SAAS,mBAAmB,CAAC,OAAuB;IAClD,MAAM,QAAQ,GAAG,OAAO;SACrB,GAAG,CAAC,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE;QACrB,MAAM,IAAI,GAAG,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;QACrD,OAAO,IAAI,KAAK,GAAG,CAAC,KAAK,MAAM,CAAC,YAAY,IAAI,MAAM,CAAC,UAAU,KAAK,IAAI,EAAE,CAAA;IAC9E,CAAC,CAAC;SACD,IAAI,CAAC,MAAM,CAAC,CAAA;IAEf,OAAO;QACL,qGAAqG;QACrG,EAAE;QACF,QAAQ;KACT,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;AACd,CAAC"}