kreuzberg 4.0.0.pre.rc.11 → 4.0.0.pre.rc.13

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 (369) hide show
  1. checksums.yaml +4 -4
  2. data/.gitignore +14 -14
  3. data/.rspec +3 -3
  4. data/.rubocop.yaml +1 -1
  5. data/.rubocop.yml +538 -538
  6. data/Gemfile +8 -8
  7. data/Gemfile.lock +2 -105
  8. data/README.md +454 -454
  9. data/Rakefile +25 -25
  10. data/Steepfile +47 -47
  11. data/examples/async_patterns.rb +341 -341
  12. data/ext/kreuzberg_rb/extconf.rb +45 -45
  13. data/ext/kreuzberg_rb/native/.cargo/config.toml +2 -2
  14. data/ext/kreuzberg_rb/native/Cargo.lock +6941 -6941
  15. data/ext/kreuzberg_rb/native/Cargo.toml +54 -54
  16. data/ext/kreuzberg_rb/native/README.md +425 -425
  17. data/ext/kreuzberg_rb/native/build.rs +15 -15
  18. data/ext/kreuzberg_rb/native/include/ieeefp.h +11 -11
  19. data/ext/kreuzberg_rb/native/include/msvc_compat/strings.h +14 -14
  20. data/ext/kreuzberg_rb/native/include/strings.h +20 -20
  21. data/ext/kreuzberg_rb/native/include/unistd.h +47 -47
  22. data/ext/kreuzberg_rb/native/src/lib.rs +3158 -3158
  23. data/extconf.rb +28 -28
  24. data/kreuzberg.gemspec +214 -214
  25. data/lib/kreuzberg/api_proxy.rb +142 -142
  26. data/lib/kreuzberg/cache_api.rb +81 -81
  27. data/lib/kreuzberg/cli.rb +55 -55
  28. data/lib/kreuzberg/cli_proxy.rb +127 -127
  29. data/lib/kreuzberg/config.rb +724 -724
  30. data/lib/kreuzberg/error_context.rb +80 -80
  31. data/lib/kreuzberg/errors.rb +118 -118
  32. data/lib/kreuzberg/extraction_api.rb +340 -340
  33. data/lib/kreuzberg/mcp_proxy.rb +186 -186
  34. data/lib/kreuzberg/ocr_backend_protocol.rb +113 -113
  35. data/lib/kreuzberg/post_processor_protocol.rb +86 -86
  36. data/lib/kreuzberg/result.rb +279 -279
  37. data/lib/kreuzberg/setup_lib_path.rb +80 -80
  38. data/lib/kreuzberg/validator_protocol.rb +89 -89
  39. data/lib/kreuzberg/version.rb +5 -5
  40. data/lib/kreuzberg.rb +109 -109
  41. data/lib/{libpdfium.dylib → pdfium.dll} +0 -0
  42. data/sig/kreuzberg/internal.rbs +184 -184
  43. data/sig/kreuzberg.rbs +546 -546
  44. data/spec/binding/cache_spec.rb +227 -227
  45. data/spec/binding/cli_proxy_spec.rb +85 -85
  46. data/spec/binding/cli_spec.rb +55 -55
  47. data/spec/binding/config_spec.rb +345 -345
  48. data/spec/binding/config_validation_spec.rb +283 -283
  49. data/spec/binding/error_handling_spec.rb +213 -213
  50. data/spec/binding/errors_spec.rb +66 -66
  51. data/spec/binding/plugins/ocr_backend_spec.rb +307 -307
  52. data/spec/binding/plugins/postprocessor_spec.rb +269 -269
  53. data/spec/binding/plugins/validator_spec.rb +274 -274
  54. data/spec/fixtures/config.toml +39 -39
  55. data/spec/fixtures/config.yaml +41 -41
  56. data/spec/fixtures/invalid_config.toml +4 -4
  57. data/spec/smoke/package_spec.rb +178 -178
  58. data/spec/spec_helper.rb +42 -42
  59. data/vendor/Cargo.toml +2 -1
  60. data/vendor/kreuzberg/Cargo.toml +2 -2
  61. data/vendor/kreuzberg/README.md +230 -230
  62. data/vendor/kreuzberg/benches/otel_overhead.rs +48 -48
  63. data/vendor/kreuzberg/build.rs +843 -843
  64. data/vendor/kreuzberg/src/api/error.rs +81 -81
  65. data/vendor/kreuzberg/src/api/handlers.rs +199 -199
  66. data/vendor/kreuzberg/src/api/mod.rs +79 -79
  67. data/vendor/kreuzberg/src/api/server.rs +353 -353
  68. data/vendor/kreuzberg/src/api/types.rs +170 -170
  69. data/vendor/kreuzberg/src/cache/mod.rs +1167 -1167
  70. data/vendor/kreuzberg/src/chunking/mod.rs +1877 -1877
  71. data/vendor/kreuzberg/src/chunking/processor.rs +220 -220
  72. data/vendor/kreuzberg/src/core/batch_mode.rs +95 -95
  73. data/vendor/kreuzberg/src/core/config.rs +1080 -1080
  74. data/vendor/kreuzberg/src/core/extractor.rs +1156 -1156
  75. data/vendor/kreuzberg/src/core/io.rs +329 -329
  76. data/vendor/kreuzberg/src/core/mime.rs +605 -605
  77. data/vendor/kreuzberg/src/core/mod.rs +47 -47
  78. data/vendor/kreuzberg/src/core/pipeline.rs +1184 -1184
  79. data/vendor/kreuzberg/src/embeddings.rs +500 -500
  80. data/vendor/kreuzberg/src/error.rs +431 -431
  81. data/vendor/kreuzberg/src/extraction/archive.rs +954 -954
  82. data/vendor/kreuzberg/src/extraction/docx.rs +398 -398
  83. data/vendor/kreuzberg/src/extraction/email.rs +854 -854
  84. data/vendor/kreuzberg/src/extraction/excel.rs +688 -688
  85. data/vendor/kreuzberg/src/extraction/html.rs +601 -601
  86. data/vendor/kreuzberg/src/extraction/image.rs +491 -491
  87. data/vendor/kreuzberg/src/extraction/libreoffice.rs +574 -562
  88. data/vendor/kreuzberg/src/extraction/markdown.rs +213 -213
  89. data/vendor/kreuzberg/src/extraction/mod.rs +81 -81
  90. data/vendor/kreuzberg/src/extraction/office_metadata/app_properties.rs +398 -398
  91. data/vendor/kreuzberg/src/extraction/office_metadata/core_properties.rs +247 -247
  92. data/vendor/kreuzberg/src/extraction/office_metadata/custom_properties.rs +240 -240
  93. data/vendor/kreuzberg/src/extraction/office_metadata/mod.rs +130 -130
  94. data/vendor/kreuzberg/src/extraction/office_metadata/odt_properties.rs +284 -284
  95. data/vendor/kreuzberg/src/extraction/pptx.rs +3100 -3100
  96. data/vendor/kreuzberg/src/extraction/structured.rs +490 -490
  97. data/vendor/kreuzberg/src/extraction/table.rs +328 -328
  98. data/vendor/kreuzberg/src/extraction/text.rs +269 -269
  99. data/vendor/kreuzberg/src/extraction/xml.rs +333 -333
  100. data/vendor/kreuzberg/src/extractors/archive.rs +447 -447
  101. data/vendor/kreuzberg/src/extractors/bibtex.rs +470 -470
  102. data/vendor/kreuzberg/src/extractors/docbook.rs +504 -504
  103. data/vendor/kreuzberg/src/extractors/docx.rs +400 -400
  104. data/vendor/kreuzberg/src/extractors/email.rs +157 -157
  105. data/vendor/kreuzberg/src/extractors/epub.rs +708 -708
  106. data/vendor/kreuzberg/src/extractors/excel.rs +345 -345
  107. data/vendor/kreuzberg/src/extractors/fictionbook.rs +492 -492
  108. data/vendor/kreuzberg/src/extractors/html.rs +407 -407
  109. data/vendor/kreuzberg/src/extractors/image.rs +219 -219
  110. data/vendor/kreuzberg/src/extractors/jats.rs +1054 -1054
  111. data/vendor/kreuzberg/src/extractors/jupyter.rs +368 -368
  112. data/vendor/kreuzberg/src/extractors/latex.rs +653 -653
  113. data/vendor/kreuzberg/src/extractors/markdown.rs +701 -701
  114. data/vendor/kreuzberg/src/extractors/mod.rs +429 -429
  115. data/vendor/kreuzberg/src/extractors/odt.rs +628 -628
  116. data/vendor/kreuzberg/src/extractors/opml.rs +635 -635
  117. data/vendor/kreuzberg/src/extractors/orgmode.rs +529 -529
  118. data/vendor/kreuzberg/src/extractors/pdf.rs +749 -722
  119. data/vendor/kreuzberg/src/extractors/pptx.rs +267 -267
  120. data/vendor/kreuzberg/src/extractors/rst.rs +577 -577
  121. data/vendor/kreuzberg/src/extractors/rtf.rs +809 -809
  122. data/vendor/kreuzberg/src/extractors/security.rs +484 -484
  123. data/vendor/kreuzberg/src/extractors/security_tests.rs +367 -367
  124. data/vendor/kreuzberg/src/extractors/structured.rs +142 -142
  125. data/vendor/kreuzberg/src/extractors/text.rs +265 -265
  126. data/vendor/kreuzberg/src/extractors/typst.rs +651 -651
  127. data/vendor/kreuzberg/src/extractors/xml.rs +147 -147
  128. data/vendor/kreuzberg/src/image/dpi.rs +164 -164
  129. data/vendor/kreuzberg/src/image/mod.rs +6 -6
  130. data/vendor/kreuzberg/src/image/preprocessing.rs +417 -417
  131. data/vendor/kreuzberg/src/image/resize.rs +89 -89
  132. data/vendor/kreuzberg/src/keywords/config.rs +154 -154
  133. data/vendor/kreuzberg/src/keywords/mod.rs +237 -237
  134. data/vendor/kreuzberg/src/keywords/processor.rs +275 -275
  135. data/vendor/kreuzberg/src/keywords/rake.rs +293 -293
  136. data/vendor/kreuzberg/src/keywords/types.rs +68 -68
  137. data/vendor/kreuzberg/src/keywords/yake.rs +163 -163
  138. data/vendor/kreuzberg/src/language_detection/mod.rs +985 -985
  139. data/vendor/kreuzberg/src/language_detection/processor.rs +219 -219
  140. data/vendor/kreuzberg/src/lib.rs +113 -113
  141. data/vendor/kreuzberg/src/mcp/mod.rs +35 -35
  142. data/vendor/kreuzberg/src/mcp/server.rs +2076 -2076
  143. data/vendor/kreuzberg/src/ocr/cache.rs +469 -469
  144. data/vendor/kreuzberg/src/ocr/error.rs +37 -37
  145. data/vendor/kreuzberg/src/ocr/hocr.rs +216 -216
  146. data/vendor/kreuzberg/src/ocr/mod.rs +58 -58
  147. data/vendor/kreuzberg/src/ocr/processor.rs +863 -863
  148. data/vendor/kreuzberg/src/ocr/table/mod.rs +4 -4
  149. data/vendor/kreuzberg/src/ocr/table/tsv_parser.rs +144 -144
  150. data/vendor/kreuzberg/src/ocr/tesseract_backend.rs +452 -452
  151. data/vendor/kreuzberg/src/ocr/types.rs +393 -393
  152. data/vendor/kreuzberg/src/ocr/utils.rs +47 -47
  153. data/vendor/kreuzberg/src/ocr/validation.rs +206 -206
  154. data/vendor/kreuzberg/src/panic_context.rs +154 -154
  155. data/vendor/kreuzberg/src/pdf/bindings.rs +44 -44
  156. data/vendor/kreuzberg/src/pdf/bundled.rs +346 -346
  157. data/vendor/kreuzberg/src/pdf/error.rs +130 -130
  158. data/vendor/kreuzberg/src/pdf/images.rs +139 -139
  159. data/vendor/kreuzberg/src/pdf/metadata.rs +489 -489
  160. data/vendor/kreuzberg/src/pdf/mod.rs +68 -68
  161. data/vendor/kreuzberg/src/pdf/rendering.rs +368 -368
  162. data/vendor/kreuzberg/src/pdf/table.rs +420 -420
  163. data/vendor/kreuzberg/src/pdf/text.rs +240 -240
  164. data/vendor/kreuzberg/src/plugins/extractor.rs +1044 -1044
  165. data/vendor/kreuzberg/src/plugins/mod.rs +212 -212
  166. data/vendor/kreuzberg/src/plugins/ocr.rs +639 -639
  167. data/vendor/kreuzberg/src/plugins/processor.rs +650 -650
  168. data/vendor/kreuzberg/src/plugins/registry.rs +1339 -1339
  169. data/vendor/kreuzberg/src/plugins/traits.rs +258 -258
  170. data/vendor/kreuzberg/src/plugins/validator.rs +967 -967
  171. data/vendor/kreuzberg/src/stopwords/mod.rs +1470 -1470
  172. data/vendor/kreuzberg/src/text/mod.rs +25 -25
  173. data/vendor/kreuzberg/src/text/quality.rs +697 -697
  174. data/vendor/kreuzberg/src/text/quality_processor.rs +219 -219
  175. data/vendor/kreuzberg/src/text/string_utils.rs +217 -217
  176. data/vendor/kreuzberg/src/text/token_reduction/cjk_utils.rs +164 -164
  177. data/vendor/kreuzberg/src/text/token_reduction/config.rs +100 -100
  178. data/vendor/kreuzberg/src/text/token_reduction/core.rs +796 -796
  179. data/vendor/kreuzberg/src/text/token_reduction/filters.rs +902 -902
  180. data/vendor/kreuzberg/src/text/token_reduction/mod.rs +160 -160
  181. data/vendor/kreuzberg/src/text/token_reduction/semantic.rs +619 -619
  182. data/vendor/kreuzberg/src/text/token_reduction/simd_text.rs +147 -147
  183. data/vendor/kreuzberg/src/types.rs +1055 -1055
  184. data/vendor/kreuzberg/src/utils/mod.rs +17 -17
  185. data/vendor/kreuzberg/src/utils/quality.rs +959 -959
  186. data/vendor/kreuzberg/src/utils/string_utils.rs +381 -381
  187. data/vendor/kreuzberg/stopwords/af_stopwords.json +53 -53
  188. data/vendor/kreuzberg/stopwords/ar_stopwords.json +482 -482
  189. data/vendor/kreuzberg/stopwords/bg_stopwords.json +261 -261
  190. data/vendor/kreuzberg/stopwords/bn_stopwords.json +400 -400
  191. data/vendor/kreuzberg/stopwords/br_stopwords.json +1205 -1205
  192. data/vendor/kreuzberg/stopwords/ca_stopwords.json +280 -280
  193. data/vendor/kreuzberg/stopwords/cs_stopwords.json +425 -425
  194. data/vendor/kreuzberg/stopwords/da_stopwords.json +172 -172
  195. data/vendor/kreuzberg/stopwords/de_stopwords.json +622 -622
  196. data/vendor/kreuzberg/stopwords/el_stopwords.json +849 -849
  197. data/vendor/kreuzberg/stopwords/en_stopwords.json +1300 -1300
  198. data/vendor/kreuzberg/stopwords/eo_stopwords.json +175 -175
  199. data/vendor/kreuzberg/stopwords/es_stopwords.json +734 -734
  200. data/vendor/kreuzberg/stopwords/et_stopwords.json +37 -37
  201. data/vendor/kreuzberg/stopwords/eu_stopwords.json +100 -100
  202. data/vendor/kreuzberg/stopwords/fa_stopwords.json +801 -801
  203. data/vendor/kreuzberg/stopwords/fi_stopwords.json +849 -849
  204. data/vendor/kreuzberg/stopwords/fr_stopwords.json +693 -693
  205. data/vendor/kreuzberg/stopwords/ga_stopwords.json +111 -111
  206. data/vendor/kreuzberg/stopwords/gl_stopwords.json +162 -162
  207. data/vendor/kreuzberg/stopwords/gu_stopwords.json +226 -226
  208. data/vendor/kreuzberg/stopwords/ha_stopwords.json +41 -41
  209. data/vendor/kreuzberg/stopwords/he_stopwords.json +196 -196
  210. data/vendor/kreuzberg/stopwords/hi_stopwords.json +227 -227
  211. data/vendor/kreuzberg/stopwords/hr_stopwords.json +181 -181
  212. data/vendor/kreuzberg/stopwords/hu_stopwords.json +791 -791
  213. data/vendor/kreuzberg/stopwords/hy_stopwords.json +47 -47
  214. data/vendor/kreuzberg/stopwords/id_stopwords.json +760 -760
  215. data/vendor/kreuzberg/stopwords/it_stopwords.json +634 -634
  216. data/vendor/kreuzberg/stopwords/ja_stopwords.json +136 -136
  217. data/vendor/kreuzberg/stopwords/kn_stopwords.json +84 -84
  218. data/vendor/kreuzberg/stopwords/ko_stopwords.json +681 -681
  219. data/vendor/kreuzberg/stopwords/ku_stopwords.json +64 -64
  220. data/vendor/kreuzberg/stopwords/la_stopwords.json +51 -51
  221. data/vendor/kreuzberg/stopwords/lt_stopwords.json +476 -476
  222. data/vendor/kreuzberg/stopwords/lv_stopwords.json +163 -163
  223. data/vendor/kreuzberg/stopwords/ml_stopwords.json +1 -1
  224. data/vendor/kreuzberg/stopwords/mr_stopwords.json +101 -101
  225. data/vendor/kreuzberg/stopwords/ms_stopwords.json +477 -477
  226. data/vendor/kreuzberg/stopwords/ne_stopwords.json +490 -490
  227. data/vendor/kreuzberg/stopwords/nl_stopwords.json +415 -415
  228. data/vendor/kreuzberg/stopwords/no_stopwords.json +223 -223
  229. data/vendor/kreuzberg/stopwords/pl_stopwords.json +331 -331
  230. data/vendor/kreuzberg/stopwords/pt_stopwords.json +562 -562
  231. data/vendor/kreuzberg/stopwords/ro_stopwords.json +436 -436
  232. data/vendor/kreuzberg/stopwords/ru_stopwords.json +561 -561
  233. data/vendor/kreuzberg/stopwords/si_stopwords.json +193 -193
  234. data/vendor/kreuzberg/stopwords/sk_stopwords.json +420 -420
  235. data/vendor/kreuzberg/stopwords/sl_stopwords.json +448 -448
  236. data/vendor/kreuzberg/stopwords/so_stopwords.json +32 -32
  237. data/vendor/kreuzberg/stopwords/st_stopwords.json +33 -33
  238. data/vendor/kreuzberg/stopwords/sv_stopwords.json +420 -420
  239. data/vendor/kreuzberg/stopwords/sw_stopwords.json +76 -76
  240. data/vendor/kreuzberg/stopwords/ta_stopwords.json +129 -129
  241. data/vendor/kreuzberg/stopwords/te_stopwords.json +54 -54
  242. data/vendor/kreuzberg/stopwords/th_stopwords.json +118 -118
  243. data/vendor/kreuzberg/stopwords/tl_stopwords.json +149 -149
  244. data/vendor/kreuzberg/stopwords/tr_stopwords.json +506 -506
  245. data/vendor/kreuzberg/stopwords/uk_stopwords.json +75 -75
  246. data/vendor/kreuzberg/stopwords/ur_stopwords.json +519 -519
  247. data/vendor/kreuzberg/stopwords/vi_stopwords.json +647 -647
  248. data/vendor/kreuzberg/stopwords/yo_stopwords.json +62 -62
  249. data/vendor/kreuzberg/stopwords/zh_stopwords.json +796 -796
  250. data/vendor/kreuzberg/stopwords/zu_stopwords.json +31 -31
  251. data/vendor/kreuzberg/tests/api_extract_multipart.rs +52 -52
  252. data/vendor/kreuzberg/tests/api_tests.rs +966 -966
  253. data/vendor/kreuzberg/tests/archive_integration.rs +545 -545
  254. data/vendor/kreuzberg/tests/batch_orchestration.rs +556 -556
  255. data/vendor/kreuzberg/tests/batch_processing.rs +318 -318
  256. data/vendor/kreuzberg/tests/bibtex_parity_test.rs +421 -421
  257. data/vendor/kreuzberg/tests/concurrency_stress.rs +533 -533
  258. data/vendor/kreuzberg/tests/config_features.rs +612 -612
  259. data/vendor/kreuzberg/tests/config_loading_tests.rs +416 -416
  260. data/vendor/kreuzberg/tests/core_integration.rs +510 -510
  261. data/vendor/kreuzberg/tests/csv_integration.rs +414 -414
  262. data/vendor/kreuzberg/tests/docbook_extractor_tests.rs +500 -500
  263. data/vendor/kreuzberg/tests/docx_metadata_extraction_test.rs +122 -122
  264. data/vendor/kreuzberg/tests/docx_vs_pandoc_comparison.rs +370 -370
  265. data/vendor/kreuzberg/tests/email_integration.rs +327 -327
  266. data/vendor/kreuzberg/tests/epub_native_extractor_tests.rs +275 -275
  267. data/vendor/kreuzberg/tests/error_handling.rs +402 -402
  268. data/vendor/kreuzberg/tests/fictionbook_extractor_tests.rs +228 -228
  269. data/vendor/kreuzberg/tests/format_integration.rs +164 -164
  270. data/vendor/kreuzberg/tests/helpers/mod.rs +142 -142
  271. data/vendor/kreuzberg/tests/html_table_test.rs +551 -551
  272. data/vendor/kreuzberg/tests/image_integration.rs +255 -255
  273. data/vendor/kreuzberg/tests/instrumentation_test.rs +139 -139
  274. data/vendor/kreuzberg/tests/jats_extractor_tests.rs +639 -639
  275. data/vendor/kreuzberg/tests/jupyter_extractor_tests.rs +704 -704
  276. data/vendor/kreuzberg/tests/keywords_integration.rs +479 -479
  277. data/vendor/kreuzberg/tests/keywords_quality.rs +509 -509
  278. data/vendor/kreuzberg/tests/latex_extractor_tests.rs +496 -496
  279. data/vendor/kreuzberg/tests/markdown_extractor_tests.rs +490 -490
  280. data/vendor/kreuzberg/tests/mime_detection.rs +429 -429
  281. data/vendor/kreuzberg/tests/ocr_configuration.rs +514 -514
  282. data/vendor/kreuzberg/tests/ocr_errors.rs +698 -698
  283. data/vendor/kreuzberg/tests/ocr_quality.rs +629 -629
  284. data/vendor/kreuzberg/tests/ocr_stress.rs +469 -469
  285. data/vendor/kreuzberg/tests/odt_extractor_tests.rs +674 -674
  286. data/vendor/kreuzberg/tests/opml_extractor_tests.rs +616 -616
  287. data/vendor/kreuzberg/tests/orgmode_extractor_tests.rs +822 -822
  288. data/vendor/kreuzberg/tests/pdf_integration.rs +45 -45
  289. data/vendor/kreuzberg/tests/pdfium_linking.rs +374 -374
  290. data/vendor/kreuzberg/tests/pipeline_integration.rs +1436 -1436
  291. data/vendor/kreuzberg/tests/plugin_ocr_backend_test.rs +776 -776
  292. data/vendor/kreuzberg/tests/plugin_postprocessor_test.rs +560 -560
  293. data/vendor/kreuzberg/tests/plugin_system.rs +927 -927
  294. data/vendor/kreuzberg/tests/plugin_validator_test.rs +783 -783
  295. data/vendor/kreuzberg/tests/registry_integration_tests.rs +587 -587
  296. data/vendor/kreuzberg/tests/rst_extractor_tests.rs +694 -694
  297. data/vendor/kreuzberg/tests/rtf_extractor_tests.rs +775 -775
  298. data/vendor/kreuzberg/tests/security_validation.rs +416 -416
  299. data/vendor/kreuzberg/tests/stopwords_integration_test.rs +888 -888
  300. data/vendor/kreuzberg/tests/test_fastembed.rs +631 -631
  301. data/vendor/kreuzberg/tests/typst_behavioral_tests.rs +1260 -1260
  302. data/vendor/kreuzberg/tests/typst_extractor_tests.rs +648 -648
  303. data/vendor/kreuzberg/tests/xlsx_metadata_extraction_test.rs +87 -87
  304. data/vendor/kreuzberg-ffi/Cargo.toml +3 -3
  305. data/vendor/kreuzberg-ffi/README.md +851 -851
  306. data/vendor/kreuzberg-ffi/build.rs +176 -176
  307. data/vendor/kreuzberg-ffi/cbindgen.toml +27 -27
  308. data/vendor/kreuzberg-ffi/kreuzberg-ffi-install.pc +12 -12
  309. data/vendor/kreuzberg-ffi/kreuzberg-ffi.pc.in +12 -12
  310. data/vendor/kreuzberg-ffi/kreuzberg.h +1087 -1087
  311. data/vendor/kreuzberg-ffi/src/lib.rs +3616 -3616
  312. data/vendor/kreuzberg-ffi/src/panic_shield.rs +247 -247
  313. data/vendor/kreuzberg-ffi/tests.disabled/README.md +48 -48
  314. data/vendor/kreuzberg-ffi/tests.disabled/config_loading_tests.rs +299 -299
  315. data/vendor/kreuzberg-ffi/tests.disabled/config_tests.rs +346 -346
  316. data/vendor/kreuzberg-ffi/tests.disabled/extractor_tests.rs +232 -232
  317. data/vendor/kreuzberg-ffi/tests.disabled/plugin_registration_tests.rs +470 -470
  318. data/vendor/kreuzberg-tesseract/.commitlintrc.json +13 -13
  319. data/vendor/kreuzberg-tesseract/.crate-ignore +2 -2
  320. data/vendor/kreuzberg-tesseract/Cargo.lock +2933 -2933
  321. data/vendor/kreuzberg-tesseract/Cargo.toml +2 -2
  322. data/vendor/kreuzberg-tesseract/LICENSE +22 -22
  323. data/vendor/kreuzberg-tesseract/README.md +399 -399
  324. data/vendor/kreuzberg-tesseract/build.rs +1354 -1354
  325. data/vendor/kreuzberg-tesseract/patches/README.md +71 -71
  326. data/vendor/kreuzberg-tesseract/patches/tesseract.diff +199 -199
  327. data/vendor/kreuzberg-tesseract/src/api.rs +1371 -1371
  328. data/vendor/kreuzberg-tesseract/src/choice_iterator.rs +77 -77
  329. data/vendor/kreuzberg-tesseract/src/enums.rs +297 -297
  330. data/vendor/kreuzberg-tesseract/src/error.rs +81 -81
  331. data/vendor/kreuzberg-tesseract/src/lib.rs +145 -145
  332. data/vendor/kreuzberg-tesseract/src/monitor.rs +57 -57
  333. data/vendor/kreuzberg-tesseract/src/mutable_iterator.rs +197 -197
  334. data/vendor/kreuzberg-tesseract/src/page_iterator.rs +253 -253
  335. data/vendor/kreuzberg-tesseract/src/result_iterator.rs +286 -286
  336. data/vendor/kreuzberg-tesseract/src/result_renderer.rs +183 -183
  337. data/vendor/kreuzberg-tesseract/tests/integration_test.rs +211 -211
  338. data/vendor/rb-sys/.cargo_vcs_info.json +5 -5
  339. data/vendor/rb-sys/Cargo.lock +393 -393
  340. data/vendor/rb-sys/Cargo.toml +70 -70
  341. data/vendor/rb-sys/Cargo.toml.orig +57 -57
  342. data/vendor/rb-sys/LICENSE-APACHE +190 -190
  343. data/vendor/rb-sys/LICENSE-MIT +21 -21
  344. data/vendor/rb-sys/build/features.rs +111 -111
  345. data/vendor/rb-sys/build/main.rs +286 -286
  346. data/vendor/rb-sys/build/stable_api_config.rs +155 -155
  347. data/vendor/rb-sys/build/version.rs +50 -50
  348. data/vendor/rb-sys/readme.md +36 -36
  349. data/vendor/rb-sys/src/bindings.rs +21 -21
  350. data/vendor/rb-sys/src/hidden.rs +11 -11
  351. data/vendor/rb-sys/src/lib.rs +35 -35
  352. data/vendor/rb-sys/src/macros.rs +371 -371
  353. data/vendor/rb-sys/src/memory.rs +53 -53
  354. data/vendor/rb-sys/src/ruby_abi_version.rs +38 -38
  355. data/vendor/rb-sys/src/special_consts.rs +31 -31
  356. data/vendor/rb-sys/src/stable_api/compiled.c +179 -179
  357. data/vendor/rb-sys/src/stable_api/compiled.rs +257 -257
  358. data/vendor/rb-sys/src/stable_api/ruby_2_7.rs +324 -324
  359. data/vendor/rb-sys/src/stable_api/ruby_3_0.rs +332 -332
  360. data/vendor/rb-sys/src/stable_api/ruby_3_1.rs +325 -325
  361. data/vendor/rb-sys/src/stable_api/ruby_3_2.rs +323 -323
  362. data/vendor/rb-sys/src/stable_api/ruby_3_3.rs +339 -339
  363. data/vendor/rb-sys/src/stable_api/ruby_3_4.rs +339 -339
  364. data/vendor/rb-sys/src/stable_api.rs +260 -260
  365. data/vendor/rb-sys/src/symbol.rs +31 -31
  366. data/vendor/rb-sys/src/tracking_allocator.rs +330 -330
  367. data/vendor/rb-sys/src/utils.rs +89 -89
  368. data/vendor/rb-sys/src/value_type.rs +7 -7
  369. metadata +7 -80
@@ -1,967 +1,967 @@
1
- //! Validator plugin trait.
2
- //!
3
- //! This module defines the trait for implementing custom validation logic.
4
-
5
- use crate::Result;
6
- use crate::core::config::ExtractionConfig;
7
- use crate::plugins::Plugin;
8
- use crate::types::ExtractionResult;
9
- use async_trait::async_trait;
10
- use std::sync::Arc;
11
-
12
- /// Trait for validator plugins.
13
- ///
14
- /// Validators check extraction results for quality, completeness, or correctness.
15
- /// Unlike post-processors, validator errors **fail fast** - if a validator returns
16
- /// an error, the extraction fails immediately.
17
- ///
18
- /// # Use Cases
19
- ///
20
- /// - **Quality Gates**: Ensure extracted content meets minimum quality standards
21
- /// - **Compliance**: Verify content meets regulatory requirements
22
- /// - **Content Filtering**: Reject documents containing unwanted content
23
- /// - **Format Validation**: Verify extracted content structure
24
- /// - **Security Checks**: Scan for malicious content
25
- ///
26
- /// # Error Handling
27
- ///
28
- /// Validator errors are **fatal** - they cause the extraction to fail and bubble up
29
- /// to the caller. Use validators for hard requirements that must be met.
30
- ///
31
- /// For non-fatal checks, use post-processors instead.
32
- ///
33
- /// # Thread Safety
34
- ///
35
- /// Validators must be thread-safe (`Send + Sync`).
36
- ///
37
- /// # Example
38
- ///
39
- /// ```rust
40
- /// use kreuzberg::plugins::{Plugin, Validator};
41
- /// use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
42
- /// use async_trait::async_trait;
43
- ///
44
- /// /// Validate that extracted content has minimum length
45
- /// struct MinimumLengthValidator {
46
- /// min_length: usize,
47
- /// }
48
- ///
49
- /// impl Plugin for MinimumLengthValidator {
50
- /// fn name(&self) -> &str { "min-length-validator" }
51
- /// fn version(&self) -> String { "1.0.0".to_string() }
52
- /// fn initialize(&self) -> Result<()> { Ok(()) }
53
- /// fn shutdown(&self) -> Result<()> { Ok(()) }
54
- /// }
55
- ///
56
- /// #[async_trait]
57
- /// impl Validator for MinimumLengthValidator {
58
- /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
59
- /// -> Result<()> {
60
- /// if result.content.len() < self.min_length {
61
- /// return Err(KreuzbergError::validation(format!(
62
- /// "Content too short: {} < {} characters",
63
- /// result.content.len(),
64
- /// self.min_length
65
- /// )));
66
- /// }
67
- /// Ok(())
68
- /// }
69
- /// }
70
- /// ```
71
- #[cfg_attr(not(target_arch = "wasm32"), async_trait)]
72
- #[cfg_attr(target_arch = "wasm32", async_trait(?Send))]
73
- pub trait Validator: Plugin {
74
- /// Validate an extraction result.
75
- ///
76
- /// Check the extraction result and return `Ok(())` if valid, or an error
77
- /// if validation fails.
78
- ///
79
- /// # Arguments
80
- ///
81
- /// * `result` - The extraction result to validate
82
- /// * `config` - Extraction configuration
83
- ///
84
- /// # Returns
85
- ///
86
- /// - `Ok(())` if validation passes
87
- /// - `Err(...)` if validation fails (extraction will fail)
88
- ///
89
- /// # Errors
90
- ///
91
- /// - `KreuzbergError::Validation` - Validation failed
92
- /// - Any other error type appropriate for the failure
93
- ///
94
- /// # Example - Content Length Validation
95
- ///
96
- /// ```rust
97
- /// # use kreuzberg::plugins::{Plugin, Validator};
98
- /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
99
- /// # use async_trait::async_trait;
100
- /// # struct ContentLengthValidator { min: usize, max: usize }
101
- /// # impl Plugin for ContentLengthValidator {
102
- /// # fn name(&self) -> &str { "length-validator" }
103
- /// # fn version(&self) -> String { "1.0.0".to_string() }
104
- /// # fn initialize(&self) -> Result<()> { Ok(()) }
105
- /// # fn shutdown(&self) -> Result<()> { Ok(()) }
106
- /// # }
107
- /// # #[async_trait]
108
- /// # impl Validator for ContentLengthValidator {
109
- /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
110
- /// -> Result<()> {
111
- /// let length = result.content.len();
112
- ///
113
- /// if length < self.min {
114
- /// return Err(KreuzbergError::validation(format!(
115
- /// "Content too short: {} < {} characters",
116
- /// length, self.min
117
- /// )));
118
- /// }
119
- ///
120
- /// if length > self.max {
121
- /// return Err(KreuzbergError::validation(format!(
122
- /// "Content too long: {} > {} characters",
123
- /// length, self.max
124
- /// )));
125
- /// }
126
- ///
127
- /// Ok(())
128
- /// }
129
- /// # }
130
- /// ```
131
- ///
132
- /// # Example - Quality Score Validation
133
- ///
134
- /// ```rust
135
- /// # use kreuzberg::plugins::{Plugin, Validator};
136
- /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
137
- /// # use async_trait::async_trait;
138
- /// # struct QualityValidator { min_score: f64 }
139
- /// # impl Plugin for QualityValidator {
140
- /// # fn name(&self) -> &str { "quality-validator" }
141
- /// # fn version(&self) -> String { "1.0.0".to_string() }
142
- /// # fn initialize(&self) -> Result<()> { Ok(()) }
143
- /// # fn shutdown(&self) -> Result<()> { Ok(()) }
144
- /// # }
145
- /// # #[async_trait]
146
- /// # impl Validator for QualityValidator {
147
- /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
148
- /// -> Result<()> {
149
- /// // Check if quality_score exists in metadata
150
- /// let score = result.metadata
151
- /// .additional
152
- /// .get("quality_score")
153
- /// .and_then(|v| v.as_f64())
154
- /// .unwrap_or(0.0);
155
- ///
156
- /// if score < self.min_score {
157
- /// return Err(KreuzbergError::validation(format!(
158
- /// "Quality score too low: {} < {}",
159
- /// score, self.min_score
160
- /// )));
161
- /// }
162
- ///
163
- /// Ok(())
164
- /// }
165
- /// # }
166
- /// ```
167
- ///
168
- /// # Example - Security Validation
169
- ///
170
- /// ```rust
171
- /// # use kreuzberg::plugins::{Plugin, Validator};
172
- /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
173
- /// # use async_trait::async_trait;
174
- /// # struct SecurityValidator { blocked_patterns: Vec<String> }
175
- /// # impl Plugin for SecurityValidator {
176
- /// # fn name(&self) -> &str { "security-validator" }
177
- /// # fn version(&self) -> String { "1.0.0".to_string() }
178
- /// # fn initialize(&self) -> Result<()> { Ok(()) }
179
- /// # fn shutdown(&self) -> Result<()> { Ok(()) }
180
- /// # }
181
- /// # #[async_trait]
182
- /// # impl Validator for SecurityValidator {
183
- /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
184
- /// -> Result<()> {
185
- /// // Check for blocked patterns
186
- /// for pattern in &self.blocked_patterns {
187
- /// if result.content.contains(pattern) {
188
- /// return Err(KreuzbergError::validation(format!(
189
- /// "Content contains blocked pattern: {}",
190
- /// pattern
191
- /// )));
192
- /// }
193
- /// }
194
- ///
195
- /// Ok(())
196
- /// }
197
- /// # }
198
- /// ```
199
- async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig) -> Result<()>;
200
-
201
- /// Optional: Check if this validator should run for a given result.
202
- ///
203
- /// Allows conditional validation based on MIME type, metadata, or content.
204
- /// Defaults to `true` (always run).
205
- ///
206
- /// # Arguments
207
- ///
208
- /// * `result` - The extraction result to check
209
- /// * `config` - Extraction configuration
210
- ///
211
- /// # Returns
212
- ///
213
- /// `true` if the validator should run, `false` to skip.
214
- ///
215
- /// # Example
216
- ///
217
- /// ```rust
218
- /// # use kreuzberg::plugins::{Plugin, Validator};
219
- /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig};
220
- /// # use async_trait::async_trait;
221
- /// # struct PdfValidator;
222
- /// # impl Plugin for PdfValidator {
223
- /// # fn name(&self) -> &str { "pdf-validator" }
224
- /// # fn version(&self) -> String { "1.0.0".to_string() }
225
- /// # fn initialize(&self) -> Result<()> { Ok(()) }
226
- /// # fn shutdown(&self) -> Result<()> { Ok(()) }
227
- /// # }
228
- /// # #[async_trait]
229
- /// # impl Validator for PdfValidator {
230
- /// # async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> { Ok(()) }
231
- /// /// Only validate PDF documents
232
- /// fn should_validate(&self, result: &ExtractionResult, config: &ExtractionConfig) -> bool {
233
- /// result.mime_type == "application/pdf"
234
- /// }
235
- /// # }
236
- /// ```
237
- fn should_validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> bool {
238
- true
239
- }
240
-
241
- /// Optional: Get the validation priority.
242
- ///
243
- /// Higher priority validators run first. Useful for ordering validation checks
244
- /// (e.g., run cheap validations before expensive ones).
245
- ///
246
- /// Default priority is 50.
247
- ///
248
- /// # Returns
249
- ///
250
- /// Priority value (higher = runs earlier).
251
- ///
252
- /// # Example
253
- ///
254
- /// ```rust
255
- /// # use kreuzberg::plugins::{Plugin, Validator};
256
- /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig};
257
- /// # use async_trait::async_trait;
258
- /// # struct FastValidator;
259
- /// # impl Plugin for FastValidator {
260
- /// # fn name(&self) -> &str { "fast-validator" }
261
- /// # fn version(&self) -> String { "1.0.0".to_string() }
262
- /// # fn initialize(&self) -> Result<()> { Ok(()) }
263
- /// # fn shutdown(&self) -> Result<()> { Ok(()) }
264
- /// # }
265
- /// # #[async_trait]
266
- /// # impl Validator for FastValidator {
267
- /// # async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> { Ok(()) }
268
- /// /// Run this validator first (it's fast)
269
- /// fn priority(&self) -> i32 {
270
- /// 100
271
- /// }
272
- /// # }
273
- /// ```
274
- fn priority(&self) -> i32 {
275
- 50
276
- }
277
- }
278
-
279
- /// Register a validator with the global registry.
280
- ///
281
- /// The validator will be registered with its default priority and will be called
282
- /// during extraction validation. The validator's `name()` method is used as the
283
- /// registration name.
284
- ///
285
- /// # Arguments
286
- ///
287
- /// * `validator` - The validator implementation wrapped in Arc
288
- ///
289
- /// # Returns
290
- ///
291
- /// - `Ok(())` if registration succeeded
292
- /// - `Err(...)` if validation failed or initialization failed
293
- ///
294
- /// # Errors
295
- ///
296
- /// - `KreuzbergError::Validation` - Invalid validator name (empty or contains whitespace)
297
- /// - Any error from the validator's `initialize()` method
298
- ///
299
- /// # Example
300
- ///
301
- /// ```rust
302
- /// use kreuzberg::plugins::{Plugin, Validator, register_validator};
303
- /// use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
304
- /// use async_trait::async_trait;
305
- /// use std::sync::Arc;
306
- ///
307
- /// struct MinLengthValidator { min_length: usize }
308
- ///
309
- /// impl Plugin for MinLengthValidator {
310
- /// fn name(&self) -> &str { "min-length" }
311
- /// fn version(&self) -> String { "1.0.0".to_string() }
312
- /// fn initialize(&self) -> Result<()> { Ok(()) }
313
- /// fn shutdown(&self) -> Result<()> { Ok(()) }
314
- /// }
315
- ///
316
- /// #[async_trait]
317
- /// impl Validator for MinLengthValidator {
318
- /// async fn validate(&self, result: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
319
- /// if result.content.len() < self.min_length {
320
- /// return Err(KreuzbergError::validation(
321
- /// format!("Content too short: {} < {}", result.content.len(), self.min_length)
322
- /// ));
323
- /// }
324
- /// Ok(())
325
- /// }
326
- /// }
327
- ///
328
- /// # tokio_test::block_on(async {
329
- /// let validator = Arc::new(MinLengthValidator { min_length: 10 });
330
- /// register_validator(validator)?;
331
- /// # Ok::<(), KreuzbergError>(())
332
- /// # });
333
- /// ```
334
- pub fn register_validator(validator: Arc<dyn Validator>) -> crate::Result<()> {
335
- use crate::plugins::registry::get_validator_registry;
336
-
337
- let registry = get_validator_registry();
338
- let mut registry = registry
339
- .write()
340
- .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
341
-
342
- registry.register(validator)
343
- }
344
-
345
- /// Unregister a validator by name.
346
- ///
347
- /// Removes the validator from the global registry and calls its `shutdown()` method.
348
- ///
349
- /// # Arguments
350
- ///
351
- /// * `name` - Name of the validator to unregister
352
- ///
353
- /// # Returns
354
- ///
355
- /// - `Ok(())` if the validator was unregistered or didn't exist
356
- /// - `Err(...)` if the shutdown method failed
357
- ///
358
- /// # Example
359
- ///
360
- /// ```rust
361
- /// use kreuzberg::plugins::unregister_validator;
362
- ///
363
- /// # tokio_test::block_on(async {
364
- /// unregister_validator("min-length")?;
365
- /// # Ok::<(), kreuzberg::KreuzbergError>(())
366
- /// # });
367
- /// ```
368
- pub fn unregister_validator(name: &str) -> crate::Result<()> {
369
- use crate::plugins::registry::get_validator_registry;
370
-
371
- let registry = get_validator_registry();
372
- let mut registry = registry
373
- .write()
374
- .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
375
-
376
- registry.remove(name)
377
- }
378
-
379
- /// List all registered validators.
380
- ///
381
- /// Returns the names of all validators currently registered in the global registry.
382
- ///
383
- /// # Returns
384
- ///
385
- /// A vector of validator names.
386
- ///
387
- /// # Example
388
- ///
389
- /// ```rust
390
- /// use kreuzberg::plugins::list_validators;
391
- ///
392
- /// # tokio_test::block_on(async {
393
- /// let validators = list_validators()?;
394
- /// for name in validators {
395
- /// println!("Registered validator: {}", name);
396
- /// }
397
- /// # Ok::<(), kreuzberg::KreuzbergError>(())
398
- /// # });
399
- /// ```
400
- pub fn list_validators() -> crate::Result<Vec<String>> {
401
- use crate::plugins::registry::get_validator_registry;
402
-
403
- let registry = get_validator_registry();
404
- let registry = registry
405
- .read()
406
- .expect("~keep Failed to acquire read lock on validator registry"); // ~keep
407
-
408
- Ok(registry.list())
409
- }
410
-
411
- /// Clear all validators from the global registry.
412
- ///
413
- /// Removes all validators and calls their `shutdown()` methods.
414
- ///
415
- /// # Returns
416
- ///
417
- /// - `Ok(())` if all validators were cleared successfully
418
- /// - `Err(...)` if any shutdown method failed
419
- ///
420
- /// # Example
421
- ///
422
- /// ```rust
423
- /// use kreuzberg::plugins::clear_validators;
424
- ///
425
- /// # tokio_test::block_on(async {
426
- /// clear_validators()?;
427
- /// # Ok::<(), kreuzberg::KreuzbergError>(())
428
- /// # });
429
- /// ```
430
- pub fn clear_validators() -> crate::Result<()> {
431
- use crate::plugins::registry::get_validator_registry;
432
-
433
- let registry = get_validator_registry();
434
- let mut registry = registry
435
- .write()
436
- .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
437
-
438
- registry.shutdown_all()
439
- }
440
-
441
- #[cfg(test)]
442
- mod tests {
443
- use super::*;
444
- use crate::KreuzbergError;
445
- use std::collections::HashMap;
446
-
447
- struct MockValidator {
448
- should_fail: bool,
449
- }
450
-
451
- impl Plugin for MockValidator {
452
- fn name(&self) -> &str {
453
- "mock-validator"
454
- }
455
-
456
- fn version(&self) -> String {
457
- "1.0.0".to_string()
458
- }
459
-
460
- fn initialize(&self) -> Result<()> {
461
- Ok(())
462
- }
463
-
464
- fn shutdown(&self) -> Result<()> {
465
- Ok(())
466
- }
467
- }
468
-
469
- #[async_trait]
470
- impl Validator for MockValidator {
471
- async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
472
- if self.should_fail {
473
- Err(KreuzbergError::validation("Validation failed".to_string()))
474
- } else {
475
- Ok(())
476
- }
477
- }
478
- }
479
-
480
- #[tokio::test]
481
- async fn test_validator_success() {
482
- let validator = MockValidator { should_fail: false };
483
-
484
- let result = ExtractionResult {
485
- content: "test content".to_string(),
486
- mime_type: "text/plain".to_string(),
487
- metadata: crate::types::Metadata::default(),
488
- tables: vec![],
489
- detected_languages: None,
490
- chunks: None,
491
- images: None,
492
- pages: None,
493
- };
494
-
495
- let config = ExtractionConfig::default();
496
- assert!(validator.validate(&result, &config).await.is_ok());
497
- }
498
-
499
- #[tokio::test]
500
- async fn test_validator_failure() {
501
- let validator = MockValidator { should_fail: true };
502
-
503
- let result = ExtractionResult {
504
- content: "test content".to_string(),
505
- mime_type: "text/plain".to_string(),
506
- metadata: crate::types::Metadata::default(),
507
- tables: vec![],
508
- detected_languages: None,
509
- chunks: None,
510
- images: None,
511
- pages: None,
512
- };
513
-
514
- let config = ExtractionConfig::default();
515
- let validation_result = validator.validate(&result, &config).await;
516
-
517
- assert!(matches!(validation_result, Err(KreuzbergError::Validation { .. })));
518
- }
519
-
520
- #[test]
521
- fn test_validator_should_validate_default() {
522
- let validator = MockValidator { should_fail: false };
523
-
524
- let result = ExtractionResult {
525
- content: "test".to_string(),
526
- mime_type: "text/plain".to_string(),
527
- metadata: crate::types::Metadata::default(),
528
- tables: vec![],
529
- detected_languages: None,
530
- chunks: None,
531
- images: None,
532
- pages: None,
533
- };
534
-
535
- let config = ExtractionConfig::default();
536
-
537
- assert!(validator.should_validate(&result, &config));
538
- }
539
-
540
- #[test]
541
- fn test_validator_priority_default() {
542
- let validator = MockValidator { should_fail: false };
543
- assert_eq!(validator.priority(), 50);
544
- }
545
-
546
- #[tokio::test]
547
- async fn test_validator_plugin_interface() {
548
- let validator = MockValidator { should_fail: false };
549
-
550
- assert_eq!(validator.name(), "mock-validator");
551
- assert_eq!(validator.version(), "1.0.0");
552
- assert!(validator.initialize().is_ok());
553
- assert!(validator.shutdown().is_ok());
554
- }
555
-
556
- #[tokio::test]
557
- async fn test_validator_empty_content() {
558
- let validator = MockValidator { should_fail: false };
559
-
560
- let result = ExtractionResult {
561
- content: String::new(),
562
- mime_type: "text/plain".to_string(),
563
- metadata: crate::types::Metadata::default(),
564
- tables: vec![],
565
- detected_languages: None,
566
- chunks: None,
567
- images: None,
568
- pages: None,
569
- };
570
-
571
- let config = ExtractionConfig::default();
572
- assert!(validator.validate(&result, &config).await.is_ok());
573
- }
574
-
575
- #[test]
576
- fn test_validator_should_validate_conditional() {
577
- struct PdfOnlyValidator;
578
-
579
- impl Plugin for PdfOnlyValidator {
580
- fn name(&self) -> &str {
581
- "pdf-only"
582
- }
583
- fn version(&self) -> String {
584
- "1.0.0".to_string()
585
- }
586
- fn initialize(&self) -> Result<()> {
587
- Ok(())
588
- }
589
- fn shutdown(&self) -> Result<()> {
590
- Ok(())
591
- }
592
- }
593
-
594
- #[async_trait]
595
- impl Validator for PdfOnlyValidator {
596
- async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
597
- Ok(())
598
- }
599
-
600
- fn should_validate(&self, result: &ExtractionResult, _config: &ExtractionConfig) -> bool {
601
- result.mime_type == "application/pdf"
602
- }
603
- }
604
-
605
- let validator = PdfOnlyValidator;
606
- let config = ExtractionConfig::default();
607
-
608
- let pdf_result = ExtractionResult {
609
- content: "test".to_string(),
610
- mime_type: "application/pdf".to_string(),
611
- metadata: crate::types::Metadata::default(),
612
- tables: vec![],
613
- detected_languages: None,
614
- chunks: None,
615
- images: None,
616
- pages: None,
617
- };
618
-
619
- let txt_result = ExtractionResult {
620
- content: "test".to_string(),
621
- mime_type: "text/plain".to_string(),
622
- metadata: crate::types::Metadata::default(),
623
- tables: vec![],
624
- detected_languages: None,
625
- chunks: None,
626
- images: None,
627
- pages: None,
628
- };
629
-
630
- assert!(validator.should_validate(&pdf_result, &config));
631
- assert!(!validator.should_validate(&txt_result, &config));
632
- }
633
-
634
- #[test]
635
- fn test_validator_priority_ranges() {
636
- struct HighPriorityValidator;
637
- struct LowPriorityValidator;
638
-
639
- impl Plugin for HighPriorityValidator {
640
- fn name(&self) -> &str {
641
- "high-priority"
642
- }
643
- fn version(&self) -> String {
644
- "1.0.0".to_string()
645
- }
646
- fn initialize(&self) -> Result<()> {
647
- Ok(())
648
- }
649
- fn shutdown(&self) -> Result<()> {
650
- Ok(())
651
- }
652
- }
653
-
654
- impl Plugin for LowPriorityValidator {
655
- fn name(&self) -> &str {
656
- "low-priority"
657
- }
658
- fn version(&self) -> String {
659
- "1.0.0".to_string()
660
- }
661
- fn initialize(&self) -> Result<()> {
662
- Ok(())
663
- }
664
- fn shutdown(&self) -> Result<()> {
665
- Ok(())
666
- }
667
- }
668
-
669
- #[async_trait]
670
- impl Validator for HighPriorityValidator {
671
- async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
672
- Ok(())
673
- }
674
-
675
- fn priority(&self) -> i32 {
676
- 100
677
- }
678
- }
679
-
680
- #[async_trait]
681
- impl Validator for LowPriorityValidator {
682
- async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
683
- Ok(())
684
- }
685
-
686
- fn priority(&self) -> i32 {
687
- 10
688
- }
689
- }
690
-
691
- let high = HighPriorityValidator;
692
- let low = LowPriorityValidator;
693
-
694
- assert_eq!(high.priority(), 100);
695
- assert_eq!(low.priority(), 10);
696
- assert!(high.priority() > low.priority());
697
- }
698
-
699
- #[tokio::test]
700
- async fn test_validator_error_message() {
701
- let validator = MockValidator { should_fail: true };
702
-
703
- let result = ExtractionResult {
704
- content: "test".to_string(),
705
- mime_type: "text/plain".to_string(),
706
- metadata: crate::types::Metadata::default(),
707
- tables: vec![],
708
- detected_languages: None,
709
- chunks: None,
710
- images: None,
711
- pages: None,
712
- };
713
-
714
- let config = ExtractionConfig::default();
715
- let err = validator.validate(&result, &config).await.unwrap_err();
716
-
717
- match err {
718
- KreuzbergError::Validation { message: msg, .. } => {
719
- assert_eq!(msg, "Validation failed");
720
- }
721
- _ => panic!("Expected Validation error"),
722
- }
723
- }
724
-
725
- #[tokio::test]
726
- async fn test_validator_with_metadata() {
727
- let validator = MockValidator { should_fail: false };
728
-
729
- let mut additional = HashMap::new();
730
- additional.insert("quality_score".to_string(), serde_json::json!(0.95));
731
-
732
- let result = ExtractionResult {
733
- content: "test".to_string(),
734
- mime_type: "text/plain".to_string(),
735
- metadata: crate::types::Metadata {
736
- additional,
737
- ..Default::default()
738
- },
739
- pages: None,
740
- tables: vec![],
741
- detected_languages: None,
742
- chunks: None,
743
- images: None,
744
- };
745
-
746
- let config = ExtractionConfig::default();
747
- assert!(validator.validate(&result, &config).await.is_ok());
748
- }
749
-
750
- #[tokio::test]
751
- async fn test_validator_with_tables() {
752
- use crate::types::Table;
753
-
754
- let validator = MockValidator { should_fail: false };
755
-
756
- let table = Table {
757
- cells: vec![vec!["A".to_string(), "B".to_string()]],
758
- markdown: "| A | B |".to_string(),
759
- page_number: 0,
760
- };
761
-
762
- let result = ExtractionResult {
763
- content: "test".to_string(),
764
- mime_type: "text/plain".to_string(),
765
- metadata: crate::types::Metadata::default(),
766
- tables: vec![table],
767
- detected_languages: None,
768
- chunks: None,
769
- images: None,
770
- pages: None,
771
- };
772
-
773
- let config = ExtractionConfig::default();
774
- assert!(validator.validate(&result, &config).await.is_ok());
775
- }
776
-
777
- #[tokio::test]
778
- async fn test_validator_different_mime_types() {
779
- let validator = MockValidator { should_fail: false };
780
- let config = ExtractionConfig::default();
781
-
782
- let mime_types = vec![
783
- "text/plain",
784
- "application/pdf",
785
- "application/json",
786
- "text/html",
787
- "image/png",
788
- ];
789
-
790
- for mime_type in mime_types {
791
- let result = ExtractionResult {
792
- content: "test".to_string(),
793
- mime_type: mime_type.to_string(),
794
- metadata: crate::types::Metadata::default(),
795
- tables: vec![],
796
- detected_languages: None,
797
- chunks: None,
798
- images: None,
799
- pages: None,
800
- };
801
-
802
- assert!(validator.validate(&result, &config).await.is_ok());
803
- }
804
- }
805
-
806
- #[tokio::test]
807
- async fn test_validator_long_content() {
808
- let validator = MockValidator { should_fail: false };
809
-
810
- let result = ExtractionResult {
811
- content: "test content ".repeat(10000),
812
- mime_type: "text/plain".to_string(),
813
- metadata: crate::types::Metadata::default(),
814
- tables: vec![],
815
- detected_languages: None,
816
- chunks: None,
817
- images: None,
818
- pages: None,
819
- };
820
-
821
- let config = ExtractionConfig::default();
822
- assert!(validator.validate(&result, &config).await.is_ok());
823
- }
824
-
825
- #[test]
826
- #[serial_test::serial]
827
- fn test_register_validator() {
828
- use std::sync::Arc;
829
-
830
- let validator = Arc::new(MockValidator { should_fail: false });
831
- let result = super::register_validator(validator);
832
- assert!(result.is_ok());
833
-
834
- let _ = super::unregister_validator("mock-validator");
835
- }
836
-
837
- #[test]
838
- #[serial_test::serial]
839
- fn test_unregister_validator() {
840
- use std::sync::Arc;
841
-
842
- let validator = Arc::new(MockValidator { should_fail: false });
843
- super::register_validator(validator).unwrap();
844
-
845
- let result = super::unregister_validator("mock-validator");
846
- assert!(result.is_ok());
847
- }
848
-
849
- #[test]
850
- #[serial_test::serial]
851
- fn test_unregister_nonexistent_validator() {
852
- let result = super::unregister_validator("nonexistent-validator-xyz");
853
- assert!(result.is_ok());
854
- }
855
-
856
- #[test]
857
- #[serial_test::serial]
858
- fn test_list_validators() {
859
- use std::sync::Arc;
860
-
861
- super::clear_validators().unwrap();
862
-
863
- let validator1 = Arc::new(MockValidator { should_fail: false });
864
- let validator2 = Arc::new(MockValidator { should_fail: false });
865
-
866
- let list_before = super::list_validators().unwrap();
867
- assert_eq!(list_before.len(), 0);
868
-
869
- super::register_validator(validator1).unwrap();
870
- super::register_validator(validator2).unwrap();
871
-
872
- let list = super::list_validators().unwrap();
873
- assert_eq!(list.len(), 1);
874
- assert!(list.contains(&"mock-validator".to_string()));
875
-
876
- super::unregister_validator("mock-validator").unwrap();
877
- }
878
-
879
- #[test]
880
- #[serial_test::serial]
881
- fn test_clear_validators() {
882
- use std::sync::Arc;
883
-
884
- super::clear_validators().unwrap();
885
-
886
- let validator1 = Arc::new(MockValidator { should_fail: false });
887
- let validator2 = Arc::new(MockValidator { should_fail: false });
888
-
889
- super::register_validator(validator1).unwrap();
890
- super::register_validator(validator2).unwrap();
891
-
892
- let list_before = super::list_validators().unwrap();
893
- assert!(!list_before.is_empty());
894
-
895
- let result = super::clear_validators();
896
- assert!(result.is_ok());
897
-
898
- let list = super::list_validators().unwrap();
899
- assert_eq!(list.len(), 0);
900
- }
901
-
902
- #[test]
903
- #[serial_test::serial]
904
- fn test_register_validator_with_invalid_name() {
905
- use std::sync::Arc;
906
-
907
- struct InvalidNameValidator;
908
- impl Plugin for InvalidNameValidator {
909
- fn name(&self) -> &str {
910
- "invalid name with spaces"
911
- }
912
- fn version(&self) -> String {
913
- "1.0.0".to_string()
914
- }
915
- fn initialize(&self) -> Result<()> {
916
- Ok(())
917
- }
918
- fn shutdown(&self) -> Result<()> {
919
- Ok(())
920
- }
921
- }
922
-
923
- #[async_trait]
924
- impl Validator for InvalidNameValidator {
925
- async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
926
- Ok(())
927
- }
928
- }
929
-
930
- let validator = Arc::new(InvalidNameValidator);
931
- let result = super::register_validator(validator);
932
- assert!(matches!(result, Err(KreuzbergError::Validation { .. })));
933
- }
934
-
935
- #[test]
936
- #[serial_test::serial]
937
- fn test_register_validator_with_empty_name() {
938
- use std::sync::Arc;
939
-
940
- struct EmptyNameValidator;
941
- impl Plugin for EmptyNameValidator {
942
- fn name(&self) -> &str {
943
- ""
944
- }
945
- fn version(&self) -> String {
946
- "1.0.0".to_string()
947
- }
948
- fn initialize(&self) -> Result<()> {
949
- Ok(())
950
- }
951
- fn shutdown(&self) -> Result<()> {
952
- Ok(())
953
- }
954
- }
955
-
956
- #[async_trait]
957
- impl Validator for EmptyNameValidator {
958
- async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
959
- Ok(())
960
- }
961
- }
962
-
963
- let validator = Arc::new(EmptyNameValidator);
964
- let result = super::register_validator(validator);
965
- assert!(matches!(result, Err(KreuzbergError::Validation { .. })));
966
- }
967
- }
1
+ //! Validator plugin trait.
2
+ //!
3
+ //! This module defines the trait for implementing custom validation logic.
4
+
5
+ use crate::Result;
6
+ use crate::core::config::ExtractionConfig;
7
+ use crate::plugins::Plugin;
8
+ use crate::types::ExtractionResult;
9
+ use async_trait::async_trait;
10
+ use std::sync::Arc;
11
+
12
+ /// Trait for validator plugins.
13
+ ///
14
+ /// Validators check extraction results for quality, completeness, or correctness.
15
+ /// Unlike post-processors, validator errors **fail fast** - if a validator returns
16
+ /// an error, the extraction fails immediately.
17
+ ///
18
+ /// # Use Cases
19
+ ///
20
+ /// - **Quality Gates**: Ensure extracted content meets minimum quality standards
21
+ /// - **Compliance**: Verify content meets regulatory requirements
22
+ /// - **Content Filtering**: Reject documents containing unwanted content
23
+ /// - **Format Validation**: Verify extracted content structure
24
+ /// - **Security Checks**: Scan for malicious content
25
+ ///
26
+ /// # Error Handling
27
+ ///
28
+ /// Validator errors are **fatal** - they cause the extraction to fail and bubble up
29
+ /// to the caller. Use validators for hard requirements that must be met.
30
+ ///
31
+ /// For non-fatal checks, use post-processors instead.
32
+ ///
33
+ /// # Thread Safety
34
+ ///
35
+ /// Validators must be thread-safe (`Send + Sync`).
36
+ ///
37
+ /// # Example
38
+ ///
39
+ /// ```rust
40
+ /// use kreuzberg::plugins::{Plugin, Validator};
41
+ /// use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
42
+ /// use async_trait::async_trait;
43
+ ///
44
+ /// /// Validate that extracted content has minimum length
45
+ /// struct MinimumLengthValidator {
46
+ /// min_length: usize,
47
+ /// }
48
+ ///
49
+ /// impl Plugin for MinimumLengthValidator {
50
+ /// fn name(&self) -> &str { "min-length-validator" }
51
+ /// fn version(&self) -> String { "1.0.0".to_string() }
52
+ /// fn initialize(&self) -> Result<()> { Ok(()) }
53
+ /// fn shutdown(&self) -> Result<()> { Ok(()) }
54
+ /// }
55
+ ///
56
+ /// #[async_trait]
57
+ /// impl Validator for MinimumLengthValidator {
58
+ /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
59
+ /// -> Result<()> {
60
+ /// if result.content.len() < self.min_length {
61
+ /// return Err(KreuzbergError::validation(format!(
62
+ /// "Content too short: {} < {} characters",
63
+ /// result.content.len(),
64
+ /// self.min_length
65
+ /// )));
66
+ /// }
67
+ /// Ok(())
68
+ /// }
69
+ /// }
70
+ /// ```
71
+ #[cfg_attr(not(target_arch = "wasm32"), async_trait)]
72
+ #[cfg_attr(target_arch = "wasm32", async_trait(?Send))]
73
+ pub trait Validator: Plugin {
74
+ /// Validate an extraction result.
75
+ ///
76
+ /// Check the extraction result and return `Ok(())` if valid, or an error
77
+ /// if validation fails.
78
+ ///
79
+ /// # Arguments
80
+ ///
81
+ /// * `result` - The extraction result to validate
82
+ /// * `config` - Extraction configuration
83
+ ///
84
+ /// # Returns
85
+ ///
86
+ /// - `Ok(())` if validation passes
87
+ /// - `Err(...)` if validation fails (extraction will fail)
88
+ ///
89
+ /// # Errors
90
+ ///
91
+ /// - `KreuzbergError::Validation` - Validation failed
92
+ /// - Any other error type appropriate for the failure
93
+ ///
94
+ /// # Example - Content Length Validation
95
+ ///
96
+ /// ```rust
97
+ /// # use kreuzberg::plugins::{Plugin, Validator};
98
+ /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
99
+ /// # use async_trait::async_trait;
100
+ /// # struct ContentLengthValidator { min: usize, max: usize }
101
+ /// # impl Plugin for ContentLengthValidator {
102
+ /// # fn name(&self) -> &str { "length-validator" }
103
+ /// # fn version(&self) -> String { "1.0.0".to_string() }
104
+ /// # fn initialize(&self) -> Result<()> { Ok(()) }
105
+ /// # fn shutdown(&self) -> Result<()> { Ok(()) }
106
+ /// # }
107
+ /// # #[async_trait]
108
+ /// # impl Validator for ContentLengthValidator {
109
+ /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
110
+ /// -> Result<()> {
111
+ /// let length = result.content.len();
112
+ ///
113
+ /// if length < self.min {
114
+ /// return Err(KreuzbergError::validation(format!(
115
+ /// "Content too short: {} < {} characters",
116
+ /// length, self.min
117
+ /// )));
118
+ /// }
119
+ ///
120
+ /// if length > self.max {
121
+ /// return Err(KreuzbergError::validation(format!(
122
+ /// "Content too long: {} > {} characters",
123
+ /// length, self.max
124
+ /// )));
125
+ /// }
126
+ ///
127
+ /// Ok(())
128
+ /// }
129
+ /// # }
130
+ /// ```
131
+ ///
132
+ /// # Example - Quality Score Validation
133
+ ///
134
+ /// ```rust
135
+ /// # use kreuzberg::plugins::{Plugin, Validator};
136
+ /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
137
+ /// # use async_trait::async_trait;
138
+ /// # struct QualityValidator { min_score: f64 }
139
+ /// # impl Plugin for QualityValidator {
140
+ /// # fn name(&self) -> &str { "quality-validator" }
141
+ /// # fn version(&self) -> String { "1.0.0".to_string() }
142
+ /// # fn initialize(&self) -> Result<()> { Ok(()) }
143
+ /// # fn shutdown(&self) -> Result<()> { Ok(()) }
144
+ /// # }
145
+ /// # #[async_trait]
146
+ /// # impl Validator for QualityValidator {
147
+ /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
148
+ /// -> Result<()> {
149
+ /// // Check if quality_score exists in metadata
150
+ /// let score = result.metadata
151
+ /// .additional
152
+ /// .get("quality_score")
153
+ /// .and_then(|v| v.as_f64())
154
+ /// .unwrap_or(0.0);
155
+ ///
156
+ /// if score < self.min_score {
157
+ /// return Err(KreuzbergError::validation(format!(
158
+ /// "Quality score too low: {} < {}",
159
+ /// score, self.min_score
160
+ /// )));
161
+ /// }
162
+ ///
163
+ /// Ok(())
164
+ /// }
165
+ /// # }
166
+ /// ```
167
+ ///
168
+ /// # Example - Security Validation
169
+ ///
170
+ /// ```rust
171
+ /// # use kreuzberg::plugins::{Plugin, Validator};
172
+ /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
173
+ /// # use async_trait::async_trait;
174
+ /// # struct SecurityValidator { blocked_patterns: Vec<String> }
175
+ /// # impl Plugin for SecurityValidator {
176
+ /// # fn name(&self) -> &str { "security-validator" }
177
+ /// # fn version(&self) -> String { "1.0.0".to_string() }
178
+ /// # fn initialize(&self) -> Result<()> { Ok(()) }
179
+ /// # fn shutdown(&self) -> Result<()> { Ok(()) }
180
+ /// # }
181
+ /// # #[async_trait]
182
+ /// # impl Validator for SecurityValidator {
183
+ /// async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig)
184
+ /// -> Result<()> {
185
+ /// // Check for blocked patterns
186
+ /// for pattern in &self.blocked_patterns {
187
+ /// if result.content.contains(pattern) {
188
+ /// return Err(KreuzbergError::validation(format!(
189
+ /// "Content contains blocked pattern: {}",
190
+ /// pattern
191
+ /// )));
192
+ /// }
193
+ /// }
194
+ ///
195
+ /// Ok(())
196
+ /// }
197
+ /// # }
198
+ /// ```
199
+ async fn validate(&self, result: &ExtractionResult, config: &ExtractionConfig) -> Result<()>;
200
+
201
+ /// Optional: Check if this validator should run for a given result.
202
+ ///
203
+ /// Allows conditional validation based on MIME type, metadata, or content.
204
+ /// Defaults to `true` (always run).
205
+ ///
206
+ /// # Arguments
207
+ ///
208
+ /// * `result` - The extraction result to check
209
+ /// * `config` - Extraction configuration
210
+ ///
211
+ /// # Returns
212
+ ///
213
+ /// `true` if the validator should run, `false` to skip.
214
+ ///
215
+ /// # Example
216
+ ///
217
+ /// ```rust
218
+ /// # use kreuzberg::plugins::{Plugin, Validator};
219
+ /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig};
220
+ /// # use async_trait::async_trait;
221
+ /// # struct PdfValidator;
222
+ /// # impl Plugin for PdfValidator {
223
+ /// # fn name(&self) -> &str { "pdf-validator" }
224
+ /// # fn version(&self) -> String { "1.0.0".to_string() }
225
+ /// # fn initialize(&self) -> Result<()> { Ok(()) }
226
+ /// # fn shutdown(&self) -> Result<()> { Ok(()) }
227
+ /// # }
228
+ /// # #[async_trait]
229
+ /// # impl Validator for PdfValidator {
230
+ /// # async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> { Ok(()) }
231
+ /// /// Only validate PDF documents
232
+ /// fn should_validate(&self, result: &ExtractionResult, config: &ExtractionConfig) -> bool {
233
+ /// result.mime_type == "application/pdf"
234
+ /// }
235
+ /// # }
236
+ /// ```
237
+ fn should_validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> bool {
238
+ true
239
+ }
240
+
241
+ /// Optional: Get the validation priority.
242
+ ///
243
+ /// Higher priority validators run first. Useful for ordering validation checks
244
+ /// (e.g., run cheap validations before expensive ones).
245
+ ///
246
+ /// Default priority is 50.
247
+ ///
248
+ /// # Returns
249
+ ///
250
+ /// Priority value (higher = runs earlier).
251
+ ///
252
+ /// # Example
253
+ ///
254
+ /// ```rust
255
+ /// # use kreuzberg::plugins::{Plugin, Validator};
256
+ /// # use kreuzberg::{Result, ExtractionResult, ExtractionConfig};
257
+ /// # use async_trait::async_trait;
258
+ /// # struct FastValidator;
259
+ /// # impl Plugin for FastValidator {
260
+ /// # fn name(&self) -> &str { "fast-validator" }
261
+ /// # fn version(&self) -> String { "1.0.0".to_string() }
262
+ /// # fn initialize(&self) -> Result<()> { Ok(()) }
263
+ /// # fn shutdown(&self) -> Result<()> { Ok(()) }
264
+ /// # }
265
+ /// # #[async_trait]
266
+ /// # impl Validator for FastValidator {
267
+ /// # async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> { Ok(()) }
268
+ /// /// Run this validator first (it's fast)
269
+ /// fn priority(&self) -> i32 {
270
+ /// 100
271
+ /// }
272
+ /// # }
273
+ /// ```
274
+ fn priority(&self) -> i32 {
275
+ 50
276
+ }
277
+ }
278
+
279
+ /// Register a validator with the global registry.
280
+ ///
281
+ /// The validator will be registered with its default priority and will be called
282
+ /// during extraction validation. The validator's `name()` method is used as the
283
+ /// registration name.
284
+ ///
285
+ /// # Arguments
286
+ ///
287
+ /// * `validator` - The validator implementation wrapped in Arc
288
+ ///
289
+ /// # Returns
290
+ ///
291
+ /// - `Ok(())` if registration succeeded
292
+ /// - `Err(...)` if validation failed or initialization failed
293
+ ///
294
+ /// # Errors
295
+ ///
296
+ /// - `KreuzbergError::Validation` - Invalid validator name (empty or contains whitespace)
297
+ /// - Any error from the validator's `initialize()` method
298
+ ///
299
+ /// # Example
300
+ ///
301
+ /// ```rust
302
+ /// use kreuzberg::plugins::{Plugin, Validator, register_validator};
303
+ /// use kreuzberg::{Result, ExtractionResult, ExtractionConfig, KreuzbergError};
304
+ /// use async_trait::async_trait;
305
+ /// use std::sync::Arc;
306
+ ///
307
+ /// struct MinLengthValidator { min_length: usize }
308
+ ///
309
+ /// impl Plugin for MinLengthValidator {
310
+ /// fn name(&self) -> &str { "min-length" }
311
+ /// fn version(&self) -> String { "1.0.0".to_string() }
312
+ /// fn initialize(&self) -> Result<()> { Ok(()) }
313
+ /// fn shutdown(&self) -> Result<()> { Ok(()) }
314
+ /// }
315
+ ///
316
+ /// #[async_trait]
317
+ /// impl Validator for MinLengthValidator {
318
+ /// async fn validate(&self, result: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
319
+ /// if result.content.len() < self.min_length {
320
+ /// return Err(KreuzbergError::validation(
321
+ /// format!("Content too short: {} < {}", result.content.len(), self.min_length)
322
+ /// ));
323
+ /// }
324
+ /// Ok(())
325
+ /// }
326
+ /// }
327
+ ///
328
+ /// # tokio_test::block_on(async {
329
+ /// let validator = Arc::new(MinLengthValidator { min_length: 10 });
330
+ /// register_validator(validator)?;
331
+ /// # Ok::<(), KreuzbergError>(())
332
+ /// # });
333
+ /// ```
334
+ pub fn register_validator(validator: Arc<dyn Validator>) -> crate::Result<()> {
335
+ use crate::plugins::registry::get_validator_registry;
336
+
337
+ let registry = get_validator_registry();
338
+ let mut registry = registry
339
+ .write()
340
+ .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
341
+
342
+ registry.register(validator)
343
+ }
344
+
345
+ /// Unregister a validator by name.
346
+ ///
347
+ /// Removes the validator from the global registry and calls its `shutdown()` method.
348
+ ///
349
+ /// # Arguments
350
+ ///
351
+ /// * `name` - Name of the validator to unregister
352
+ ///
353
+ /// # Returns
354
+ ///
355
+ /// - `Ok(())` if the validator was unregistered or didn't exist
356
+ /// - `Err(...)` if the shutdown method failed
357
+ ///
358
+ /// # Example
359
+ ///
360
+ /// ```rust
361
+ /// use kreuzberg::plugins::unregister_validator;
362
+ ///
363
+ /// # tokio_test::block_on(async {
364
+ /// unregister_validator("min-length")?;
365
+ /// # Ok::<(), kreuzberg::KreuzbergError>(())
366
+ /// # });
367
+ /// ```
368
+ pub fn unregister_validator(name: &str) -> crate::Result<()> {
369
+ use crate::plugins::registry::get_validator_registry;
370
+
371
+ let registry = get_validator_registry();
372
+ let mut registry = registry
373
+ .write()
374
+ .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
375
+
376
+ registry.remove(name)
377
+ }
378
+
379
+ /// List all registered validators.
380
+ ///
381
+ /// Returns the names of all validators currently registered in the global registry.
382
+ ///
383
+ /// # Returns
384
+ ///
385
+ /// A vector of validator names.
386
+ ///
387
+ /// # Example
388
+ ///
389
+ /// ```rust
390
+ /// use kreuzberg::plugins::list_validators;
391
+ ///
392
+ /// # tokio_test::block_on(async {
393
+ /// let validators = list_validators()?;
394
+ /// for name in validators {
395
+ /// println!("Registered validator: {}", name);
396
+ /// }
397
+ /// # Ok::<(), kreuzberg::KreuzbergError>(())
398
+ /// # });
399
+ /// ```
400
+ pub fn list_validators() -> crate::Result<Vec<String>> {
401
+ use crate::plugins::registry::get_validator_registry;
402
+
403
+ let registry = get_validator_registry();
404
+ let registry = registry
405
+ .read()
406
+ .expect("~keep Failed to acquire read lock on validator registry"); // ~keep
407
+
408
+ Ok(registry.list())
409
+ }
410
+
411
+ /// Clear all validators from the global registry.
412
+ ///
413
+ /// Removes all validators and calls their `shutdown()` methods.
414
+ ///
415
+ /// # Returns
416
+ ///
417
+ /// - `Ok(())` if all validators were cleared successfully
418
+ /// - `Err(...)` if any shutdown method failed
419
+ ///
420
+ /// # Example
421
+ ///
422
+ /// ```rust
423
+ /// use kreuzberg::plugins::clear_validators;
424
+ ///
425
+ /// # tokio_test::block_on(async {
426
+ /// clear_validators()?;
427
+ /// # Ok::<(), kreuzberg::KreuzbergError>(())
428
+ /// # });
429
+ /// ```
430
+ pub fn clear_validators() -> crate::Result<()> {
431
+ use crate::plugins::registry::get_validator_registry;
432
+
433
+ let registry = get_validator_registry();
434
+ let mut registry = registry
435
+ .write()
436
+ .expect("~keep Failed to acquire write lock on validator registry"); // ~keep
437
+
438
+ registry.shutdown_all()
439
+ }
440
+
441
+ #[cfg(test)]
442
+ mod tests {
443
+ use super::*;
444
+ use crate::KreuzbergError;
445
+ use std::collections::HashMap;
446
+
447
+ struct MockValidator {
448
+ should_fail: bool,
449
+ }
450
+
451
+ impl Plugin for MockValidator {
452
+ fn name(&self) -> &str {
453
+ "mock-validator"
454
+ }
455
+
456
+ fn version(&self) -> String {
457
+ "1.0.0".to_string()
458
+ }
459
+
460
+ fn initialize(&self) -> Result<()> {
461
+ Ok(())
462
+ }
463
+
464
+ fn shutdown(&self) -> Result<()> {
465
+ Ok(())
466
+ }
467
+ }
468
+
469
+ #[async_trait]
470
+ impl Validator for MockValidator {
471
+ async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
472
+ if self.should_fail {
473
+ Err(KreuzbergError::validation("Validation failed".to_string()))
474
+ } else {
475
+ Ok(())
476
+ }
477
+ }
478
+ }
479
+
480
+ #[tokio::test]
481
+ async fn test_validator_success() {
482
+ let validator = MockValidator { should_fail: false };
483
+
484
+ let result = ExtractionResult {
485
+ content: "test content".to_string(),
486
+ mime_type: "text/plain".to_string(),
487
+ metadata: crate::types::Metadata::default(),
488
+ tables: vec![],
489
+ detected_languages: None,
490
+ chunks: None,
491
+ images: None,
492
+ pages: None,
493
+ };
494
+
495
+ let config = ExtractionConfig::default();
496
+ assert!(validator.validate(&result, &config).await.is_ok());
497
+ }
498
+
499
+ #[tokio::test]
500
+ async fn test_validator_failure() {
501
+ let validator = MockValidator { should_fail: true };
502
+
503
+ let result = ExtractionResult {
504
+ content: "test content".to_string(),
505
+ mime_type: "text/plain".to_string(),
506
+ metadata: crate::types::Metadata::default(),
507
+ tables: vec![],
508
+ detected_languages: None,
509
+ chunks: None,
510
+ images: None,
511
+ pages: None,
512
+ };
513
+
514
+ let config = ExtractionConfig::default();
515
+ let validation_result = validator.validate(&result, &config).await;
516
+
517
+ assert!(matches!(validation_result, Err(KreuzbergError::Validation { .. })));
518
+ }
519
+
520
+ #[test]
521
+ fn test_validator_should_validate_default() {
522
+ let validator = MockValidator { should_fail: false };
523
+
524
+ let result = ExtractionResult {
525
+ content: "test".to_string(),
526
+ mime_type: "text/plain".to_string(),
527
+ metadata: crate::types::Metadata::default(),
528
+ tables: vec![],
529
+ detected_languages: None,
530
+ chunks: None,
531
+ images: None,
532
+ pages: None,
533
+ };
534
+
535
+ let config = ExtractionConfig::default();
536
+
537
+ assert!(validator.should_validate(&result, &config));
538
+ }
539
+
540
+ #[test]
541
+ fn test_validator_priority_default() {
542
+ let validator = MockValidator { should_fail: false };
543
+ assert_eq!(validator.priority(), 50);
544
+ }
545
+
546
+ #[tokio::test]
547
+ async fn test_validator_plugin_interface() {
548
+ let validator = MockValidator { should_fail: false };
549
+
550
+ assert_eq!(validator.name(), "mock-validator");
551
+ assert_eq!(validator.version(), "1.0.0");
552
+ assert!(validator.initialize().is_ok());
553
+ assert!(validator.shutdown().is_ok());
554
+ }
555
+
556
+ #[tokio::test]
557
+ async fn test_validator_empty_content() {
558
+ let validator = MockValidator { should_fail: false };
559
+
560
+ let result = ExtractionResult {
561
+ content: String::new(),
562
+ mime_type: "text/plain".to_string(),
563
+ metadata: crate::types::Metadata::default(),
564
+ tables: vec![],
565
+ detected_languages: None,
566
+ chunks: None,
567
+ images: None,
568
+ pages: None,
569
+ };
570
+
571
+ let config = ExtractionConfig::default();
572
+ assert!(validator.validate(&result, &config).await.is_ok());
573
+ }
574
+
575
+ #[test]
576
+ fn test_validator_should_validate_conditional() {
577
+ struct PdfOnlyValidator;
578
+
579
+ impl Plugin for PdfOnlyValidator {
580
+ fn name(&self) -> &str {
581
+ "pdf-only"
582
+ }
583
+ fn version(&self) -> String {
584
+ "1.0.0".to_string()
585
+ }
586
+ fn initialize(&self) -> Result<()> {
587
+ Ok(())
588
+ }
589
+ fn shutdown(&self) -> Result<()> {
590
+ Ok(())
591
+ }
592
+ }
593
+
594
+ #[async_trait]
595
+ impl Validator for PdfOnlyValidator {
596
+ async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
597
+ Ok(())
598
+ }
599
+
600
+ fn should_validate(&self, result: &ExtractionResult, _config: &ExtractionConfig) -> bool {
601
+ result.mime_type == "application/pdf"
602
+ }
603
+ }
604
+
605
+ let validator = PdfOnlyValidator;
606
+ let config = ExtractionConfig::default();
607
+
608
+ let pdf_result = ExtractionResult {
609
+ content: "test".to_string(),
610
+ mime_type: "application/pdf".to_string(),
611
+ metadata: crate::types::Metadata::default(),
612
+ tables: vec![],
613
+ detected_languages: None,
614
+ chunks: None,
615
+ images: None,
616
+ pages: None,
617
+ };
618
+
619
+ let txt_result = ExtractionResult {
620
+ content: "test".to_string(),
621
+ mime_type: "text/plain".to_string(),
622
+ metadata: crate::types::Metadata::default(),
623
+ tables: vec![],
624
+ detected_languages: None,
625
+ chunks: None,
626
+ images: None,
627
+ pages: None,
628
+ };
629
+
630
+ assert!(validator.should_validate(&pdf_result, &config));
631
+ assert!(!validator.should_validate(&txt_result, &config));
632
+ }
633
+
634
+ #[test]
635
+ fn test_validator_priority_ranges() {
636
+ struct HighPriorityValidator;
637
+ struct LowPriorityValidator;
638
+
639
+ impl Plugin for HighPriorityValidator {
640
+ fn name(&self) -> &str {
641
+ "high-priority"
642
+ }
643
+ fn version(&self) -> String {
644
+ "1.0.0".to_string()
645
+ }
646
+ fn initialize(&self) -> Result<()> {
647
+ Ok(())
648
+ }
649
+ fn shutdown(&self) -> Result<()> {
650
+ Ok(())
651
+ }
652
+ }
653
+
654
+ impl Plugin for LowPriorityValidator {
655
+ fn name(&self) -> &str {
656
+ "low-priority"
657
+ }
658
+ fn version(&self) -> String {
659
+ "1.0.0".to_string()
660
+ }
661
+ fn initialize(&self) -> Result<()> {
662
+ Ok(())
663
+ }
664
+ fn shutdown(&self) -> Result<()> {
665
+ Ok(())
666
+ }
667
+ }
668
+
669
+ #[async_trait]
670
+ impl Validator for HighPriorityValidator {
671
+ async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
672
+ Ok(())
673
+ }
674
+
675
+ fn priority(&self) -> i32 {
676
+ 100
677
+ }
678
+ }
679
+
680
+ #[async_trait]
681
+ impl Validator for LowPriorityValidator {
682
+ async fn validate(&self, _result: &ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
683
+ Ok(())
684
+ }
685
+
686
+ fn priority(&self) -> i32 {
687
+ 10
688
+ }
689
+ }
690
+
691
+ let high = HighPriorityValidator;
692
+ let low = LowPriorityValidator;
693
+
694
+ assert_eq!(high.priority(), 100);
695
+ assert_eq!(low.priority(), 10);
696
+ assert!(high.priority() > low.priority());
697
+ }
698
+
699
+ #[tokio::test]
700
+ async fn test_validator_error_message() {
701
+ let validator = MockValidator { should_fail: true };
702
+
703
+ let result = ExtractionResult {
704
+ content: "test".to_string(),
705
+ mime_type: "text/plain".to_string(),
706
+ metadata: crate::types::Metadata::default(),
707
+ tables: vec![],
708
+ detected_languages: None,
709
+ chunks: None,
710
+ images: None,
711
+ pages: None,
712
+ };
713
+
714
+ let config = ExtractionConfig::default();
715
+ let err = validator.validate(&result, &config).await.unwrap_err();
716
+
717
+ match err {
718
+ KreuzbergError::Validation { message: msg, .. } => {
719
+ assert_eq!(msg, "Validation failed");
720
+ }
721
+ _ => panic!("Expected Validation error"),
722
+ }
723
+ }
724
+
725
+ #[tokio::test]
726
+ async fn test_validator_with_metadata() {
727
+ let validator = MockValidator { should_fail: false };
728
+
729
+ let mut additional = HashMap::new();
730
+ additional.insert("quality_score".to_string(), serde_json::json!(0.95));
731
+
732
+ let result = ExtractionResult {
733
+ content: "test".to_string(),
734
+ mime_type: "text/plain".to_string(),
735
+ metadata: crate::types::Metadata {
736
+ additional,
737
+ ..Default::default()
738
+ },
739
+ pages: None,
740
+ tables: vec![],
741
+ detected_languages: None,
742
+ chunks: None,
743
+ images: None,
744
+ };
745
+
746
+ let config = ExtractionConfig::default();
747
+ assert!(validator.validate(&result, &config).await.is_ok());
748
+ }
749
+
750
+ #[tokio::test]
751
+ async fn test_validator_with_tables() {
752
+ use crate::types::Table;
753
+
754
+ let validator = MockValidator { should_fail: false };
755
+
756
+ let table = Table {
757
+ cells: vec![vec!["A".to_string(), "B".to_string()]],
758
+ markdown: "| A | B |".to_string(),
759
+ page_number: 0,
760
+ };
761
+
762
+ let result = ExtractionResult {
763
+ content: "test".to_string(),
764
+ mime_type: "text/plain".to_string(),
765
+ metadata: crate::types::Metadata::default(),
766
+ tables: vec![table],
767
+ detected_languages: None,
768
+ chunks: None,
769
+ images: None,
770
+ pages: None,
771
+ };
772
+
773
+ let config = ExtractionConfig::default();
774
+ assert!(validator.validate(&result, &config).await.is_ok());
775
+ }
776
+
777
+ #[tokio::test]
778
+ async fn test_validator_different_mime_types() {
779
+ let validator = MockValidator { should_fail: false };
780
+ let config = ExtractionConfig::default();
781
+
782
+ let mime_types = vec![
783
+ "text/plain",
784
+ "application/pdf",
785
+ "application/json",
786
+ "text/html",
787
+ "image/png",
788
+ ];
789
+
790
+ for mime_type in mime_types {
791
+ let result = ExtractionResult {
792
+ content: "test".to_string(),
793
+ mime_type: mime_type.to_string(),
794
+ metadata: crate::types::Metadata::default(),
795
+ tables: vec![],
796
+ detected_languages: None,
797
+ chunks: None,
798
+ images: None,
799
+ pages: None,
800
+ };
801
+
802
+ assert!(validator.validate(&result, &config).await.is_ok());
803
+ }
804
+ }
805
+
806
+ #[tokio::test]
807
+ async fn test_validator_long_content() {
808
+ let validator = MockValidator { should_fail: false };
809
+
810
+ let result = ExtractionResult {
811
+ content: "test content ".repeat(10000),
812
+ mime_type: "text/plain".to_string(),
813
+ metadata: crate::types::Metadata::default(),
814
+ tables: vec![],
815
+ detected_languages: None,
816
+ chunks: None,
817
+ images: None,
818
+ pages: None,
819
+ };
820
+
821
+ let config = ExtractionConfig::default();
822
+ assert!(validator.validate(&result, &config).await.is_ok());
823
+ }
824
+
825
+ #[test]
826
+ #[serial_test::serial]
827
+ fn test_register_validator() {
828
+ use std::sync::Arc;
829
+
830
+ let validator = Arc::new(MockValidator { should_fail: false });
831
+ let result = super::register_validator(validator);
832
+ assert!(result.is_ok());
833
+
834
+ let _ = super::unregister_validator("mock-validator");
835
+ }
836
+
837
+ #[test]
838
+ #[serial_test::serial]
839
+ fn test_unregister_validator() {
840
+ use std::sync::Arc;
841
+
842
+ let validator = Arc::new(MockValidator { should_fail: false });
843
+ super::register_validator(validator).unwrap();
844
+
845
+ let result = super::unregister_validator("mock-validator");
846
+ assert!(result.is_ok());
847
+ }
848
+
849
+ #[test]
850
+ #[serial_test::serial]
851
+ fn test_unregister_nonexistent_validator() {
852
+ let result = super::unregister_validator("nonexistent-validator-xyz");
853
+ assert!(result.is_ok());
854
+ }
855
+
856
+ #[test]
857
+ #[serial_test::serial]
858
+ fn test_list_validators() {
859
+ use std::sync::Arc;
860
+
861
+ super::clear_validators().unwrap();
862
+
863
+ let validator1 = Arc::new(MockValidator { should_fail: false });
864
+ let validator2 = Arc::new(MockValidator { should_fail: false });
865
+
866
+ let list_before = super::list_validators().unwrap();
867
+ assert_eq!(list_before.len(), 0);
868
+
869
+ super::register_validator(validator1).unwrap();
870
+ super::register_validator(validator2).unwrap();
871
+
872
+ let list = super::list_validators().unwrap();
873
+ assert_eq!(list.len(), 1);
874
+ assert!(list.contains(&"mock-validator".to_string()));
875
+
876
+ super::unregister_validator("mock-validator").unwrap();
877
+ }
878
+
879
+ #[test]
880
+ #[serial_test::serial]
881
+ fn test_clear_validators() {
882
+ use std::sync::Arc;
883
+
884
+ super::clear_validators().unwrap();
885
+
886
+ let validator1 = Arc::new(MockValidator { should_fail: false });
887
+ let validator2 = Arc::new(MockValidator { should_fail: false });
888
+
889
+ super::register_validator(validator1).unwrap();
890
+ super::register_validator(validator2).unwrap();
891
+
892
+ let list_before = super::list_validators().unwrap();
893
+ assert!(!list_before.is_empty());
894
+
895
+ let result = super::clear_validators();
896
+ assert!(result.is_ok());
897
+
898
+ let list = super::list_validators().unwrap();
899
+ assert_eq!(list.len(), 0);
900
+ }
901
+
902
+ #[test]
903
+ #[serial_test::serial]
904
+ fn test_register_validator_with_invalid_name() {
905
+ use std::sync::Arc;
906
+
907
+ struct InvalidNameValidator;
908
+ impl Plugin for InvalidNameValidator {
909
+ fn name(&self) -> &str {
910
+ "invalid name with spaces"
911
+ }
912
+ fn version(&self) -> String {
913
+ "1.0.0".to_string()
914
+ }
915
+ fn initialize(&self) -> Result<()> {
916
+ Ok(())
917
+ }
918
+ fn shutdown(&self) -> Result<()> {
919
+ Ok(())
920
+ }
921
+ }
922
+
923
+ #[async_trait]
924
+ impl Validator for InvalidNameValidator {
925
+ async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
926
+ Ok(())
927
+ }
928
+ }
929
+
930
+ let validator = Arc::new(InvalidNameValidator);
931
+ let result = super::register_validator(validator);
932
+ assert!(matches!(result, Err(KreuzbergError::Validation { .. })));
933
+ }
934
+
935
+ #[test]
936
+ #[serial_test::serial]
937
+ fn test_register_validator_with_empty_name() {
938
+ use std::sync::Arc;
939
+
940
+ struct EmptyNameValidator;
941
+ impl Plugin for EmptyNameValidator {
942
+ fn name(&self) -> &str {
943
+ ""
944
+ }
945
+ fn version(&self) -> String {
946
+ "1.0.0".to_string()
947
+ }
948
+ fn initialize(&self) -> Result<()> {
949
+ Ok(())
950
+ }
951
+ fn shutdown(&self) -> Result<()> {
952
+ Ok(())
953
+ }
954
+ }
955
+
956
+ #[async_trait]
957
+ impl Validator for EmptyNameValidator {
958
+ async fn validate(&self, _: &ExtractionResult, _: &ExtractionConfig) -> Result<()> {
959
+ Ok(())
960
+ }
961
+ }
962
+
963
+ let validator = Arc::new(EmptyNameValidator);
964
+ let result = super::register_validator(validator);
965
+ assert!(matches!(result, Err(KreuzbergError::Validation { .. })));
966
+ }
967
+ }