kreuzberg 4.0.0.pre.rc.29 → 4.0.0.rc1

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 (321) hide show
  1. checksums.yaml +4 -4
  2. data/.gitignore +0 -6
  3. data/.rubocop.yaml +534 -1
  4. data/Gemfile +2 -1
  5. data/Gemfile.lock +28 -116
  6. data/README.md +269 -629
  7. data/Rakefile +0 -9
  8. data/Steepfile +4 -8
  9. data/examples/async_patterns.rb +58 -1
  10. data/ext/kreuzberg_rb/extconf.rb +5 -35
  11. data/ext/kreuzberg_rb/native/Cargo.toml +16 -55
  12. data/ext/kreuzberg_rb/native/build.rs +14 -12
  13. data/ext/kreuzberg_rb/native/include/ieeefp.h +1 -1
  14. data/ext/kreuzberg_rb/native/include/msvc_compat/strings.h +1 -1
  15. data/ext/kreuzberg_rb/native/include/strings.h +2 -2
  16. data/ext/kreuzberg_rb/native/include/unistd.h +1 -1
  17. data/ext/kreuzberg_rb/native/src/lib.rs +34 -897
  18. data/extconf.rb +6 -38
  19. data/kreuzberg.gemspec +20 -114
  20. data/lib/kreuzberg/api_proxy.rb +18 -2
  21. data/lib/kreuzberg/cache_api.rb +0 -22
  22. data/lib/kreuzberg/cli.rb +10 -2
  23. data/lib/kreuzberg/cli_proxy.rb +10 -0
  24. data/lib/kreuzberg/config.rb +22 -274
  25. data/lib/kreuzberg/errors.rb +7 -73
  26. data/lib/kreuzberg/extraction_api.rb +8 -237
  27. data/lib/kreuzberg/mcp_proxy.rb +11 -2
  28. data/lib/kreuzberg/ocr_backend_protocol.rb +73 -0
  29. data/lib/kreuzberg/post_processor_protocol.rb +71 -0
  30. data/lib/kreuzberg/result.rb +33 -151
  31. data/lib/kreuzberg/setup_lib_path.rb +2 -22
  32. data/lib/kreuzberg/validator_protocol.rb +73 -0
  33. data/lib/kreuzberg/version.rb +1 -1
  34. data/lib/kreuzberg.rb +13 -27
  35. data/pkg/kreuzberg-4.0.0.rc1.gem +0 -0
  36. data/sig/kreuzberg.rbs +12 -105
  37. data/spec/binding/cache_spec.rb +22 -22
  38. data/spec/binding/cli_proxy_spec.rb +4 -2
  39. data/spec/binding/cli_spec.rb +11 -12
  40. data/spec/binding/config_spec.rb +0 -74
  41. data/spec/binding/config_validation_spec.rb +6 -100
  42. data/spec/binding/error_handling_spec.rb +97 -283
  43. data/spec/binding/plugins/ocr_backend_spec.rb +8 -8
  44. data/spec/binding/plugins/postprocessor_spec.rb +11 -11
  45. data/spec/binding/plugins/validator_spec.rb +13 -12
  46. data/spec/examples.txt +104 -0
  47. data/spec/fixtures/config.toml +1 -0
  48. data/spec/fixtures/config.yaml +1 -0
  49. data/spec/fixtures/invalid_config.toml +1 -0
  50. data/spec/smoke/package_spec.rb +3 -2
  51. data/spec/spec_helper.rb +3 -1
  52. data/vendor/kreuzberg/Cargo.toml +67 -192
  53. data/vendor/kreuzberg/README.md +9 -97
  54. data/vendor/kreuzberg/build.rs +194 -516
  55. data/vendor/kreuzberg/src/api/handlers.rs +9 -130
  56. data/vendor/kreuzberg/src/api/mod.rs +3 -18
  57. data/vendor/kreuzberg/src/api/server.rs +71 -236
  58. data/vendor/kreuzberg/src/api/types.rs +7 -43
  59. data/vendor/kreuzberg/src/bin/profile_extract.rs +455 -0
  60. data/vendor/kreuzberg/src/cache/mod.rs +3 -27
  61. data/vendor/kreuzberg/src/chunking/mod.rs +79 -1705
  62. data/vendor/kreuzberg/src/core/batch_mode.rs +0 -60
  63. data/vendor/kreuzberg/src/core/config.rs +23 -905
  64. data/vendor/kreuzberg/src/core/extractor.rs +106 -403
  65. data/vendor/kreuzberg/src/core/io.rs +2 -4
  66. data/vendor/kreuzberg/src/core/mime.rs +12 -2
  67. data/vendor/kreuzberg/src/core/mod.rs +3 -22
  68. data/vendor/kreuzberg/src/core/pipeline.rs +78 -395
  69. data/vendor/kreuzberg/src/embeddings.rs +21 -169
  70. data/vendor/kreuzberg/src/error.rs +2 -2
  71. data/vendor/kreuzberg/src/extraction/archive.rs +31 -36
  72. data/vendor/kreuzberg/src/extraction/docx.rs +1 -365
  73. data/vendor/kreuzberg/src/extraction/email.rs +11 -12
  74. data/vendor/kreuzberg/src/extraction/excel.rs +129 -138
  75. data/vendor/kreuzberg/src/extraction/html.rs +170 -1447
  76. data/vendor/kreuzberg/src/extraction/image.rs +14 -138
  77. data/vendor/kreuzberg/src/extraction/libreoffice.rs +3 -13
  78. data/vendor/kreuzberg/src/extraction/mod.rs +5 -21
  79. data/vendor/kreuzberg/src/extraction/office_metadata/mod.rs +0 -2
  80. data/vendor/kreuzberg/src/extraction/pandoc/batch.rs +275 -0
  81. data/vendor/kreuzberg/src/extraction/pandoc/mime_types.rs +178 -0
  82. data/vendor/kreuzberg/src/extraction/pandoc/mod.rs +491 -0
  83. data/vendor/kreuzberg/src/extraction/pandoc/server.rs +496 -0
  84. data/vendor/kreuzberg/src/extraction/pandoc/subprocess.rs +1188 -0
  85. data/vendor/kreuzberg/src/extraction/pandoc/version.rs +162 -0
  86. data/vendor/kreuzberg/src/extraction/pptx.rs +94 -196
  87. data/vendor/kreuzberg/src/extraction/structured.rs +4 -5
  88. data/vendor/kreuzberg/src/extraction/table.rs +1 -2
  89. data/vendor/kreuzberg/src/extraction/text.rs +10 -18
  90. data/vendor/kreuzberg/src/extractors/archive.rs +0 -22
  91. data/vendor/kreuzberg/src/extractors/docx.rs +148 -69
  92. data/vendor/kreuzberg/src/extractors/email.rs +9 -37
  93. data/vendor/kreuzberg/src/extractors/excel.rs +40 -81
  94. data/vendor/kreuzberg/src/extractors/html.rs +173 -182
  95. data/vendor/kreuzberg/src/extractors/image.rs +8 -32
  96. data/vendor/kreuzberg/src/extractors/mod.rs +10 -171
  97. data/vendor/kreuzberg/src/extractors/pandoc.rs +201 -0
  98. data/vendor/kreuzberg/src/extractors/pdf.rs +64 -329
  99. data/vendor/kreuzberg/src/extractors/pptx.rs +34 -79
  100. data/vendor/kreuzberg/src/extractors/structured.rs +0 -16
  101. data/vendor/kreuzberg/src/extractors/text.rs +7 -30
  102. data/vendor/kreuzberg/src/extractors/xml.rs +8 -27
  103. data/vendor/kreuzberg/src/keywords/processor.rs +1 -9
  104. data/vendor/kreuzberg/src/keywords/rake.rs +1 -0
  105. data/vendor/kreuzberg/src/language_detection/mod.rs +51 -94
  106. data/vendor/kreuzberg/src/lib.rs +5 -17
  107. data/vendor/kreuzberg/src/mcp/mod.rs +1 -4
  108. data/vendor/kreuzberg/src/mcp/server.rs +21 -145
  109. data/vendor/kreuzberg/src/ocr/mod.rs +0 -2
  110. data/vendor/kreuzberg/src/ocr/processor.rs +8 -19
  111. data/vendor/kreuzberg/src/ocr/tesseract_backend.rs +0 -2
  112. data/vendor/kreuzberg/src/pdf/error.rs +1 -93
  113. data/vendor/kreuzberg/src/pdf/metadata.rs +100 -263
  114. data/vendor/kreuzberg/src/pdf/mod.rs +2 -33
  115. data/vendor/kreuzberg/src/pdf/rendering.rs +12 -12
  116. data/vendor/kreuzberg/src/pdf/table.rs +64 -61
  117. data/vendor/kreuzberg/src/pdf/text.rs +24 -416
  118. data/vendor/kreuzberg/src/plugins/extractor.rs +8 -40
  119. data/vendor/kreuzberg/src/plugins/mod.rs +0 -3
  120. data/vendor/kreuzberg/src/plugins/ocr.rs +14 -22
  121. data/vendor/kreuzberg/src/plugins/processor.rs +1 -10
  122. data/vendor/kreuzberg/src/plugins/registry.rs +0 -15
  123. data/vendor/kreuzberg/src/plugins/validator.rs +8 -20
  124. data/vendor/kreuzberg/src/stopwords/mod.rs +2 -2
  125. data/vendor/kreuzberg/src/text/mod.rs +0 -8
  126. data/vendor/kreuzberg/src/text/quality.rs +15 -28
  127. data/vendor/kreuzberg/src/text/string_utils.rs +10 -22
  128. data/vendor/kreuzberg/src/text/token_reduction/core.rs +50 -86
  129. data/vendor/kreuzberg/src/text/token_reduction/filters.rs +16 -37
  130. data/vendor/kreuzberg/src/text/token_reduction/simd_text.rs +1 -2
  131. data/vendor/kreuzberg/src/types.rs +67 -907
  132. data/vendor/kreuzberg/src/utils/mod.rs +0 -14
  133. data/vendor/kreuzberg/src/utils/quality.rs +3 -12
  134. data/vendor/kreuzberg/tests/api_tests.rs +0 -506
  135. data/vendor/kreuzberg/tests/archive_integration.rs +0 -2
  136. data/vendor/kreuzberg/tests/batch_orchestration.rs +12 -57
  137. data/vendor/kreuzberg/tests/batch_processing.rs +8 -32
  138. data/vendor/kreuzberg/tests/chunking_offset_demo.rs +92 -0
  139. data/vendor/kreuzberg/tests/concurrency_stress.rs +8 -40
  140. data/vendor/kreuzberg/tests/config_features.rs +1 -33
  141. data/vendor/kreuzberg/tests/config_loading_tests.rs +39 -16
  142. data/vendor/kreuzberg/tests/core_integration.rs +9 -35
  143. data/vendor/kreuzberg/tests/csv_integration.rs +81 -71
  144. data/vendor/kreuzberg/tests/docx_metadata_extraction_test.rs +25 -23
  145. data/vendor/kreuzberg/tests/email_integration.rs +1 -3
  146. data/vendor/kreuzberg/tests/error_handling.rs +34 -43
  147. data/vendor/kreuzberg/tests/format_integration.rs +1 -7
  148. data/vendor/kreuzberg/tests/helpers/mod.rs +0 -60
  149. data/vendor/kreuzberg/tests/image_integration.rs +0 -2
  150. data/vendor/kreuzberg/tests/mime_detection.rs +16 -17
  151. data/vendor/kreuzberg/tests/ocr_configuration.rs +0 -4
  152. data/vendor/kreuzberg/tests/ocr_errors.rs +0 -22
  153. data/vendor/kreuzberg/tests/ocr_quality.rs +0 -2
  154. data/vendor/kreuzberg/tests/pandoc_integration.rs +503 -0
  155. data/vendor/kreuzberg/tests/pdf_integration.rs +0 -2
  156. data/vendor/kreuzberg/tests/pipeline_integration.rs +2 -36
  157. data/vendor/kreuzberg/tests/plugin_ocr_backend_test.rs +0 -5
  158. data/vendor/kreuzberg/tests/plugin_postprocessor_test.rs +1 -17
  159. data/vendor/kreuzberg/tests/plugin_system.rs +0 -6
  160. data/vendor/kreuzberg/tests/registry_integration_tests.rs +22 -2
  161. data/vendor/kreuzberg/tests/security_validation.rs +1 -13
  162. data/vendor/kreuzberg/tests/test_fastembed.rs +23 -45
  163. metadata +25 -171
  164. data/.rubocop.yml +0 -543
  165. data/ext/kreuzberg_rb/native/.cargo/config.toml +0 -23
  166. data/ext/kreuzberg_rb/native/Cargo.lock +0 -7619
  167. data/lib/kreuzberg/error_context.rb +0 -136
  168. data/lib/kreuzberg/types.rb +0 -170
  169. data/lib/libpdfium.so +0 -0
  170. data/spec/binding/async_operations_spec.rb +0 -473
  171. data/spec/binding/batch_operations_spec.rb +0 -595
  172. data/spec/binding/batch_spec.rb +0 -359
  173. data/spec/binding/config_result_spec.rb +0 -377
  174. data/spec/binding/embeddings_spec.rb +0 -816
  175. data/spec/binding/error_recovery_spec.rb +0 -488
  176. data/spec/binding/font_config_spec.rb +0 -220
  177. data/spec/binding/images_spec.rb +0 -738
  178. data/spec/binding/keywords_extraction_spec.rb +0 -600
  179. data/spec/binding/metadata_types_spec.rb +0 -1228
  180. data/spec/binding/pages_extraction_spec.rb +0 -471
  181. data/spec/binding/tables_spec.rb +0 -641
  182. data/spec/unit/config/chunking_config_spec.rb +0 -213
  183. data/spec/unit/config/embedding_config_spec.rb +0 -343
  184. data/spec/unit/config/extraction_config_spec.rb +0 -438
  185. data/spec/unit/config/font_config_spec.rb +0 -285
  186. data/spec/unit/config/hierarchy_config_spec.rb +0 -314
  187. data/spec/unit/config/image_extraction_config_spec.rb +0 -209
  188. data/spec/unit/config/image_preprocessing_config_spec.rb +0 -249
  189. data/spec/unit/config/keyword_config_spec.rb +0 -229
  190. data/spec/unit/config/language_detection_config_spec.rb +0 -258
  191. data/spec/unit/config/ocr_config_spec.rb +0 -171
  192. data/spec/unit/config/page_config_spec.rb +0 -221
  193. data/spec/unit/config/pdf_config_spec.rb +0 -267
  194. data/spec/unit/config/postprocessor_config_spec.rb +0 -290
  195. data/spec/unit/config/tesseract_config_spec.rb +0 -181
  196. data/spec/unit/config/token_reduction_config_spec.rb +0 -251
  197. data/test/metadata_types_test.rb +0 -959
  198. data/vendor/Cargo.toml +0 -61
  199. data/vendor/kreuzberg/examples/bench_fixes.rs +0 -71
  200. data/vendor/kreuzberg/examples/test_pdfium_fork.rs +0 -62
  201. data/vendor/kreuzberg/src/chunking/processor.rs +0 -219
  202. data/vendor/kreuzberg/src/core/batch_optimizations.rs +0 -385
  203. data/vendor/kreuzberg/src/core/config_validation.rs +0 -949
  204. data/vendor/kreuzberg/src/core/formats.rs +0 -235
  205. data/vendor/kreuzberg/src/core/server_config.rs +0 -1220
  206. data/vendor/kreuzberg/src/extraction/capacity.rs +0 -263
  207. data/vendor/kreuzberg/src/extraction/markdown.rs +0 -216
  208. data/vendor/kreuzberg/src/extraction/office_metadata/odt_properties.rs +0 -284
  209. data/vendor/kreuzberg/src/extractors/bibtex.rs +0 -470
  210. data/vendor/kreuzberg/src/extractors/docbook.rs +0 -504
  211. data/vendor/kreuzberg/src/extractors/epub.rs +0 -696
  212. data/vendor/kreuzberg/src/extractors/fictionbook.rs +0 -492
  213. data/vendor/kreuzberg/src/extractors/jats.rs +0 -1054
  214. data/vendor/kreuzberg/src/extractors/jupyter.rs +0 -368
  215. data/vendor/kreuzberg/src/extractors/latex.rs +0 -653
  216. data/vendor/kreuzberg/src/extractors/markdown.rs +0 -701
  217. data/vendor/kreuzberg/src/extractors/odt.rs +0 -628
  218. data/vendor/kreuzberg/src/extractors/opml.rs +0 -635
  219. data/vendor/kreuzberg/src/extractors/orgmode.rs +0 -529
  220. data/vendor/kreuzberg/src/extractors/rst.rs +0 -577
  221. data/vendor/kreuzberg/src/extractors/rtf.rs +0 -809
  222. data/vendor/kreuzberg/src/extractors/security.rs +0 -484
  223. data/vendor/kreuzberg/src/extractors/security_tests.rs +0 -367
  224. data/vendor/kreuzberg/src/extractors/typst.rs +0 -651
  225. data/vendor/kreuzberg/src/language_detection/processor.rs +0 -218
  226. data/vendor/kreuzberg/src/ocr/language_registry.rs +0 -520
  227. data/vendor/kreuzberg/src/panic_context.rs +0 -154
  228. data/vendor/kreuzberg/src/pdf/bindings.rs +0 -306
  229. data/vendor/kreuzberg/src/pdf/bundled.rs +0 -408
  230. data/vendor/kreuzberg/src/pdf/fonts.rs +0 -358
  231. data/vendor/kreuzberg/src/pdf/hierarchy.rs +0 -903
  232. data/vendor/kreuzberg/src/text/quality_processor.rs +0 -231
  233. data/vendor/kreuzberg/src/text/utf8_validation.rs +0 -193
  234. data/vendor/kreuzberg/src/utils/pool.rs +0 -503
  235. data/vendor/kreuzberg/src/utils/pool_sizing.rs +0 -364
  236. data/vendor/kreuzberg/src/utils/string_pool.rs +0 -761
  237. data/vendor/kreuzberg/tests/api_embed.rs +0 -360
  238. data/vendor/kreuzberg/tests/api_extract_multipart.rs +0 -52
  239. data/vendor/kreuzberg/tests/api_large_pdf_extraction.rs +0 -471
  240. data/vendor/kreuzberg/tests/api_large_pdf_extraction_diagnostics.rs +0 -289
  241. data/vendor/kreuzberg/tests/batch_pooling_benchmark.rs +0 -154
  242. data/vendor/kreuzberg/tests/bibtex_parity_test.rs +0 -421
  243. data/vendor/kreuzberg/tests/config_integration_test.rs +0 -753
  244. data/vendor/kreuzberg/tests/data/hierarchy_ground_truth.json +0 -294
  245. data/vendor/kreuzberg/tests/docbook_extractor_tests.rs +0 -500
  246. data/vendor/kreuzberg/tests/docx_vs_pandoc_comparison.rs +0 -370
  247. data/vendor/kreuzberg/tests/epub_native_extractor_tests.rs +0 -275
  248. data/vendor/kreuzberg/tests/fictionbook_extractor_tests.rs +0 -228
  249. data/vendor/kreuzberg/tests/html_table_test.rs +0 -551
  250. data/vendor/kreuzberg/tests/instrumentation_test.rs +0 -139
  251. data/vendor/kreuzberg/tests/jats_extractor_tests.rs +0 -639
  252. data/vendor/kreuzberg/tests/jupyter_extractor_tests.rs +0 -704
  253. data/vendor/kreuzberg/tests/latex_extractor_tests.rs +0 -496
  254. data/vendor/kreuzberg/tests/markdown_extractor_tests.rs +0 -490
  255. data/vendor/kreuzberg/tests/ocr_language_registry.rs +0 -191
  256. data/vendor/kreuzberg/tests/odt_extractor_tests.rs +0 -674
  257. data/vendor/kreuzberg/tests/opml_extractor_tests.rs +0 -616
  258. data/vendor/kreuzberg/tests/orgmode_extractor_tests.rs +0 -822
  259. data/vendor/kreuzberg/tests/page_markers.rs +0 -297
  260. data/vendor/kreuzberg/tests/pdf_hierarchy_detection.rs +0 -301
  261. data/vendor/kreuzberg/tests/pdf_hierarchy_quality.rs +0 -589
  262. data/vendor/kreuzberg/tests/pdf_ocr_triggering.rs +0 -301
  263. data/vendor/kreuzberg/tests/pdf_text_merging.rs +0 -475
  264. data/vendor/kreuzberg/tests/pdfium_linking.rs +0 -340
  265. data/vendor/kreuzberg/tests/rst_extractor_tests.rs +0 -694
  266. data/vendor/kreuzberg/tests/rtf_extractor_tests.rs +0 -775
  267. data/vendor/kreuzberg/tests/typst_behavioral_tests.rs +0 -1260
  268. data/vendor/kreuzberg/tests/typst_extractor_tests.rs +0 -648
  269. data/vendor/kreuzberg-ffi/Cargo.toml +0 -67
  270. data/vendor/kreuzberg-ffi/README.md +0 -851
  271. data/vendor/kreuzberg-ffi/benches/result_view_benchmark.rs +0 -227
  272. data/vendor/kreuzberg-ffi/build.rs +0 -168
  273. data/vendor/kreuzberg-ffi/cbindgen.toml +0 -37
  274. data/vendor/kreuzberg-ffi/kreuzberg-ffi.pc.in +0 -12
  275. data/vendor/kreuzberg-ffi/kreuzberg.h +0 -3012
  276. data/vendor/kreuzberg-ffi/src/batch_streaming.rs +0 -588
  277. data/vendor/kreuzberg-ffi/src/config.rs +0 -1341
  278. data/vendor/kreuzberg-ffi/src/error.rs +0 -901
  279. data/vendor/kreuzberg-ffi/src/extraction.rs +0 -555
  280. data/vendor/kreuzberg-ffi/src/helpers.rs +0 -879
  281. data/vendor/kreuzberg-ffi/src/lib.rs +0 -977
  282. data/vendor/kreuzberg-ffi/src/memory.rs +0 -493
  283. data/vendor/kreuzberg-ffi/src/mime.rs +0 -329
  284. data/vendor/kreuzberg-ffi/src/panic_shield.rs +0 -265
  285. data/vendor/kreuzberg-ffi/src/plugins/document_extractor.rs +0 -442
  286. data/vendor/kreuzberg-ffi/src/plugins/mod.rs +0 -14
  287. data/vendor/kreuzberg-ffi/src/plugins/ocr_backend.rs +0 -628
  288. data/vendor/kreuzberg-ffi/src/plugins/post_processor.rs +0 -438
  289. data/vendor/kreuzberg-ffi/src/plugins/validator.rs +0 -329
  290. data/vendor/kreuzberg-ffi/src/result.rs +0 -510
  291. data/vendor/kreuzberg-ffi/src/result_pool.rs +0 -639
  292. data/vendor/kreuzberg-ffi/src/result_view.rs +0 -773
  293. data/vendor/kreuzberg-ffi/src/string_intern.rs +0 -568
  294. data/vendor/kreuzberg-ffi/src/types.rs +0 -363
  295. data/vendor/kreuzberg-ffi/src/util.rs +0 -210
  296. data/vendor/kreuzberg-ffi/src/validation.rs +0 -848
  297. data/vendor/kreuzberg-ffi/tests.disabled/README.md +0 -48
  298. data/vendor/kreuzberg-ffi/tests.disabled/config_loading_tests.rs +0 -299
  299. data/vendor/kreuzberg-ffi/tests.disabled/config_tests.rs +0 -346
  300. data/vendor/kreuzberg-ffi/tests.disabled/extractor_tests.rs +0 -232
  301. data/vendor/kreuzberg-ffi/tests.disabled/plugin_registration_tests.rs +0 -470
  302. data/vendor/kreuzberg-tesseract/.commitlintrc.json +0 -13
  303. data/vendor/kreuzberg-tesseract/.crate-ignore +0 -2
  304. data/vendor/kreuzberg-tesseract/Cargo.lock +0 -2933
  305. data/vendor/kreuzberg-tesseract/Cargo.toml +0 -57
  306. data/vendor/kreuzberg-tesseract/LICENSE +0 -22
  307. data/vendor/kreuzberg-tesseract/README.md +0 -399
  308. data/vendor/kreuzberg-tesseract/build.rs +0 -1127
  309. data/vendor/kreuzberg-tesseract/patches/README.md +0 -71
  310. data/vendor/kreuzberg-tesseract/patches/tesseract.diff +0 -199
  311. data/vendor/kreuzberg-tesseract/src/api.rs +0 -1371
  312. data/vendor/kreuzberg-tesseract/src/choice_iterator.rs +0 -77
  313. data/vendor/kreuzberg-tesseract/src/enums.rs +0 -297
  314. data/vendor/kreuzberg-tesseract/src/error.rs +0 -81
  315. data/vendor/kreuzberg-tesseract/src/lib.rs +0 -145
  316. data/vendor/kreuzberg-tesseract/src/monitor.rs +0 -57
  317. data/vendor/kreuzberg-tesseract/src/mutable_iterator.rs +0 -197
  318. data/vendor/kreuzberg-tesseract/src/page_iterator.rs +0 -253
  319. data/vendor/kreuzberg-tesseract/src/result_iterator.rs +0 -286
  320. data/vendor/kreuzberg-tesseract/src/result_renderer.rs +0 -183
  321. data/vendor/kreuzberg-tesseract/tests/integration_test.rs +0 -211
@@ -1,438 +0,0 @@
1
- //! PostProcessor plugin system FFI bindings
2
- //!
3
- //! Provides FFI functions for registering, managing, and executing custom post-processors
4
- //! from C/Java/other FFI languages.
5
-
6
- use std::ffi::{CStr, CString};
7
- use std::os::raw::c_char;
8
- use std::ptr;
9
- use std::sync::Arc;
10
-
11
- use async_trait::async_trait;
12
- use kreuzberg::core::config::ExtractionConfig;
13
- use kreuzberg::plugins::{Plugin, ProcessingStage};
14
- use kreuzberg::types::ExtractionResult;
15
- use kreuzberg::{KreuzbergError, Result};
16
-
17
- use crate::helpers::{clear_last_error, set_last_error};
18
- use crate::memory::kreuzberg_free_string;
19
- use crate::{ffi_panic_guard, ffi_panic_guard_bool};
20
-
21
- /// Type alias for the PostProcessor callback function.
22
- ///
23
- /// # Parameters
24
- ///
25
- /// - `result_json`: JSON-encoded ExtractionResult (null-terminated string)
26
- ///
27
- /// # Returns
28
- ///
29
- /// Null-terminated JSON string containing the processed ExtractionResult
30
- /// (must be freed by Rust via kreuzberg_free_string), or NULL on error.
31
- ///
32
- /// # Safety
33
- ///
34
- /// The callback must:
35
- /// - Not store the result_json pointer (it's only valid for the duration of the call)
36
- /// - Return a valid null-terminated UTF-8 JSON string allocated by the caller
37
- /// - Return NULL on error (error message should be retrievable separately)
38
- pub type PostProcessorCallback = unsafe extern "C" fn(result_json: *const c_char) -> *mut c_char;
39
-
40
- /// FFI wrapper for custom PostProcessors registered from Java/C.
41
- ///
42
- /// This struct wraps a C function pointer and implements the PostProcessor trait,
43
- /// allowing custom post-processing implementations from FFI languages to be registered
44
- /// and used within the Rust extraction pipeline.
45
- struct FfiPostProcessor {
46
- name: String,
47
- callback: PostProcessorCallback,
48
- stage: ProcessingStage,
49
- }
50
-
51
- impl FfiPostProcessor {
52
- fn new(name: String, callback: PostProcessorCallback, stage: ProcessingStage) -> Self {
53
- Self { name, callback, stage }
54
- }
55
- }
56
-
57
- impl Plugin for FfiPostProcessor {
58
- fn name(&self) -> &str {
59
- &self.name
60
- }
61
-
62
- fn version(&self) -> String {
63
- "ffi-1.0.0".to_string()
64
- }
65
-
66
- fn initialize(&self) -> Result<()> {
67
- Ok(())
68
- }
69
-
70
- fn shutdown(&self) -> Result<()> {
71
- Ok(())
72
- }
73
- }
74
-
75
- #[async_trait]
76
- impl kreuzberg::plugins::PostProcessor for FfiPostProcessor {
77
- async fn process(&self, result: &mut ExtractionResult, _config: &ExtractionConfig) -> Result<()> {
78
- let result_json = serde_json::to_string(&*result).map_err(|e| KreuzbergError::Validation {
79
- message: format!("Failed to serialize ExtractionResult: {}", e),
80
- source: Some(Box::new(e)),
81
- })?;
82
-
83
- let callback = self.callback;
84
- let processor_name = self.name.clone();
85
- let result_json_owned = result_json.clone();
86
-
87
- let processed_json = tokio::task::spawn_blocking(move || {
88
- let result_cstring = CString::new(result_json_owned).map_err(|e| KreuzbergError::Validation {
89
- message: format!("Failed to create C string from result JSON: {}", e),
90
- source: Some(Box::new(e)),
91
- })?;
92
-
93
- let processed_ptr = unsafe { callback(result_cstring.as_ptr()) };
94
-
95
- if processed_ptr.is_null() {
96
- return Err(KreuzbergError::Plugin {
97
- message: "PostProcessor returned NULL (operation failed)".to_string(),
98
- plugin_name: processor_name.clone(),
99
- });
100
- }
101
-
102
- let processed_cstr = unsafe { CStr::from_ptr(processed_ptr) };
103
- let json = processed_cstr
104
- .to_str()
105
- .map_err(|e| KreuzbergError::Plugin {
106
- message: format!("PostProcessor returned invalid UTF-8: {}", e),
107
- plugin_name: processor_name.clone(),
108
- })?
109
- .to_string();
110
-
111
- unsafe { kreuzberg_free_string(processed_ptr) };
112
-
113
- Ok(json)
114
- })
115
- .await
116
- .map_err(|e| KreuzbergError::Plugin {
117
- message: format!("PostProcessor task panicked: {}", e),
118
- plugin_name: self.name.clone(),
119
- })??;
120
-
121
- let processed_result: ExtractionResult =
122
- serde_json::from_str(&processed_json).map_err(|e| KreuzbergError::Plugin {
123
- message: format!("Failed to deserialize processed result: {}", e),
124
- plugin_name: self.name.clone(),
125
- })?;
126
-
127
- *result = processed_result;
128
-
129
- Ok(())
130
- }
131
-
132
- fn processing_stage(&self) -> kreuzberg::plugins::ProcessingStage {
133
- self.stage
134
- }
135
- }
136
-
137
- fn parse_processing_stage(stage: Option<&str>) -> std::result::Result<ProcessingStage, String> {
138
- match stage {
139
- Some(value) => match value.to_lowercase().as_str() {
140
- "early" => Ok(ProcessingStage::Early),
141
- "middle" => Ok(ProcessingStage::Middle),
142
- "late" => Ok(ProcessingStage::Late),
143
- other => Err(format!(
144
- "Invalid processing stage '{}'. Expected one of: early, middle, late",
145
- other
146
- )),
147
- },
148
- None => Ok(ProcessingStage::Middle),
149
- }
150
- }
151
-
152
- /// Register a custom PostProcessor via FFI callback.
153
- ///
154
- /// # Safety
155
- ///
156
- /// - `name` must be a valid null-terminated C string
157
- /// - `callback` must be a valid function pointer that:
158
- /// - Does not store the result_json pointer
159
- /// - Returns a null-terminated UTF-8 JSON string or NULL on error
160
- /// - The returned string must be freeable by kreuzberg_free_string
161
- /// - `priority` determines the order of execution (higher priority runs first)
162
- /// - Returns true on success, false on error (check kreuzberg_last_error)
163
- ///
164
- /// # Example (C)
165
- ///
166
- /// ```c
167
- /// char* my_post_processor(const char* result_json) {
168
- /// // Parse result_json, modify it, return JSON string
169
- /// return strdup("{\"content\":\"PROCESSED\"}");
170
- /// }
171
- ///
172
- /// bool success = kreuzberg_register_post_processor("my-processor", my_post_processor, 100);
173
- /// if (!success) {
174
- /// const char* error = kreuzberg_last_error();
175
- /// printf("Failed to register: %s\n", error);
176
- /// }
177
- /// ```
178
- #[unsafe(no_mangle)]
179
- pub unsafe extern "C" fn kreuzberg_register_post_processor(
180
- name: *const c_char,
181
- callback: PostProcessorCallback,
182
- priority: i32,
183
- ) -> bool {
184
- ffi_panic_guard_bool!("kreuzberg_register_post_processor", {
185
- clear_last_error();
186
-
187
- if name.is_null() {
188
- set_last_error("PostProcessor name cannot be NULL".to_string());
189
- return false;
190
- }
191
-
192
- let name_str = match unsafe { CStr::from_ptr(name) }.to_str() {
193
- Ok(s) => s,
194
- Err(e) => {
195
- set_last_error(format!("Invalid UTF-8 in PostProcessor name: {}", e));
196
- return false;
197
- }
198
- };
199
-
200
- if name_str.is_empty() {
201
- set_last_error("Plugin name cannot be empty".to_string());
202
- return false;
203
- }
204
-
205
- if name_str.chars().any(|c| c.is_whitespace()) {
206
- set_last_error("Plugin name cannot contain whitespace".to_string());
207
- return false;
208
- }
209
-
210
- let processor = Arc::new(FfiPostProcessor::new(
211
- name_str.to_string(),
212
- callback,
213
- ProcessingStage::Middle,
214
- ));
215
-
216
- let registry = kreuzberg::plugins::registry::get_post_processor_registry();
217
- let mut registry_guard = match registry.write() {
218
- Ok(guard) => guard,
219
- Err(e) => {
220
- // ~keep: Lock poisoning indicates a panic in another thread holding the lock.
221
- set_last_error(format!("Failed to acquire registry write lock: {}", e));
222
- return false;
223
- }
224
- };
225
-
226
- match registry_guard.register(processor, priority) {
227
- Ok(()) => true,
228
- Err(e) => {
229
- set_last_error(format!("Failed to register PostProcessor: {}", e));
230
- false
231
- }
232
- }
233
- })
234
- }
235
-
236
- /// Register a custom PostProcessor with an explicit processing stage.
237
- ///
238
- /// # Safety
239
- ///
240
- /// - `name` must be a valid null-terminated C string
241
- /// - `stage` must be a valid null-terminated C string containing "early", "middle", or "late"
242
- /// - `callback` must be a valid function pointer that:
243
- /// - Does not store the result_json pointer
244
- /// - Returns a null-terminated UTF-8 JSON string or NULL on error
245
- /// - The returned string must be freeable by kreuzberg_free_string
246
- /// - `priority` determines the order of execution within the stage (higher priority runs first)
247
- /// - Returns true on success, false on error (check kreuzberg_last_error)
248
- #[unsafe(no_mangle)]
249
- pub unsafe extern "C" fn kreuzberg_register_post_processor_with_stage(
250
- name: *const c_char,
251
- callback: PostProcessorCallback,
252
- priority: i32,
253
- stage: *const c_char,
254
- ) -> bool {
255
- ffi_panic_guard_bool!("kreuzberg_register_post_processor_with_stage", {
256
- clear_last_error();
257
-
258
- if name.is_null() {
259
- set_last_error("PostProcessor name cannot be NULL".to_string());
260
- return false;
261
- }
262
-
263
- let name_str = match unsafe { CStr::from_ptr(name) }.to_str() {
264
- Ok(s) => s,
265
- Err(e) => {
266
- set_last_error(format!("Invalid UTF-8 in PostProcessor name: {}", e));
267
- return false;
268
- }
269
- };
270
-
271
- if name_str.is_empty() {
272
- set_last_error("Plugin name cannot be empty".to_string());
273
- return false;
274
- }
275
-
276
- if name_str.chars().any(|c| c.is_whitespace()) {
277
- set_last_error("Plugin name cannot contain whitespace".to_string());
278
- return false;
279
- }
280
-
281
- let stage_str = if stage.is_null() {
282
- None
283
- } else {
284
- match unsafe { CStr::from_ptr(stage) }.to_str() {
285
- Ok(s) => Some(s),
286
- Err(e) => {
287
- set_last_error(format!("Invalid UTF-8 in processing stage: {}", e));
288
- return false;
289
- }
290
- }
291
- };
292
-
293
- let stage = match parse_processing_stage(stage_str) {
294
- Ok(stage) => stage,
295
- Err(e) => {
296
- set_last_error(e);
297
- return false;
298
- }
299
- };
300
-
301
- let processor = Arc::new(FfiPostProcessor::new(name_str.to_string(), callback, stage));
302
-
303
- let registry = kreuzberg::plugins::registry::get_post_processor_registry();
304
- let mut registry_guard = match registry.write() {
305
- Ok(guard) => guard,
306
- Err(e) => {
307
- // ~keep: Lock poisoning indicates a panic in another thread holding the lock.
308
- set_last_error(format!("Failed to acquire registry write lock: {}", e));
309
- return false;
310
- }
311
- };
312
-
313
- match registry_guard.register(processor, priority) {
314
- Ok(()) => true,
315
- Err(e) => {
316
- set_last_error(format!("Failed to register PostProcessor: {}", e));
317
- false
318
- }
319
- }
320
- })
321
- }
322
-
323
- /// Unregister a PostProcessor by name.
324
- ///
325
- /// # Safety
326
- ///
327
- /// - `name` must be a valid null-terminated C string
328
- /// - Returns true on success, false on error (check kreuzberg_last_error)
329
- ///
330
- /// # Example (C)
331
- ///
332
- /// ```c
333
- /// bool success = kreuzberg_unregister_post_processor("my-processor");
334
- /// if (!success) {
335
- /// const char* error = kreuzberg_last_error();
336
- /// printf("Failed to unregister: %s\n", error);
337
- /// }
338
- /// ```
339
- #[unsafe(no_mangle)]
340
- pub unsafe extern "C" fn kreuzberg_unregister_post_processor(name: *const c_char) -> bool {
341
- ffi_panic_guard_bool!("kreuzberg_unregister_post_processor", {
342
- clear_last_error();
343
-
344
- if name.is_null() {
345
- set_last_error("PostProcessor name cannot be NULL".to_string());
346
- return false;
347
- }
348
-
349
- let name_str = match unsafe { CStr::from_ptr(name) }.to_str() {
350
- Ok(s) => s,
351
- Err(e) => {
352
- set_last_error(format!("Invalid UTF-8 in PostProcessor name: {}", e));
353
- return false;
354
- }
355
- };
356
-
357
- let registry = kreuzberg::plugins::registry::get_post_processor_registry();
358
- let mut registry_guard = match registry.write() {
359
- Ok(guard) => guard,
360
- Err(e) => {
361
- // ~keep: Lock poisoning indicates a panic in another thread holding the lock.
362
- set_last_error(format!("Failed to acquire registry write lock: {}", e));
363
- return false;
364
- }
365
- };
366
-
367
- match registry_guard.remove(name_str) {
368
- Ok(()) => true,
369
- Err(e) => {
370
- set_last_error(format!("Failed to remove PostProcessor: {}", e));
371
- false
372
- }
373
- }
374
- })
375
- }
376
-
377
- /// Clear all registered PostProcessors.
378
- ///
379
- /// # Safety
380
- ///
381
- /// - Removes all registered processors. Subsequent extractions will run without them.
382
- /// - Returns true on success, false on error.
383
- #[unsafe(no_mangle)]
384
- pub unsafe extern "C" fn kreuzberg_clear_post_processors() -> bool {
385
- ffi_panic_guard_bool!("kreuzberg_clear_post_processors", {
386
- clear_last_error();
387
-
388
- let registry = kreuzberg::plugins::registry::get_post_processor_registry();
389
- let mut registry_guard = match registry.write() {
390
- Ok(guard) => guard,
391
- Err(e) => {
392
- // ~keep: Lock poisoning indicates a panic in another thread holding the lock.
393
- set_last_error(format!("Failed to acquire registry write lock: {}", e));
394
- return false;
395
- }
396
- };
397
-
398
- *registry_guard = Default::default();
399
- true
400
- })
401
- }
402
-
403
- /// List all registered PostProcessors as a JSON array of names.
404
- ///
405
- /// # Safety
406
- ///
407
- /// - Returned string must be freed with `kreuzberg_free_string`.
408
- /// - Returns NULL on error (check `kreuzberg_last_error`).
409
- #[unsafe(no_mangle)]
410
- pub unsafe extern "C" fn kreuzberg_list_post_processors() -> *mut c_char {
411
- ffi_panic_guard!("kreuzberg_list_post_processors", {
412
- clear_last_error();
413
-
414
- let registry = kreuzberg::plugins::registry::get_post_processor_registry();
415
- let registry_guard = match registry.read() {
416
- Ok(guard) => guard,
417
- Err(e) => {
418
- // ~keep: Lock poisoning indicates a panic in another thread holding the lock.
419
- set_last_error(format!("Failed to acquire registry read lock: {}", e));
420
- return ptr::null_mut();
421
- }
422
- };
423
-
424
- match serde_json::to_string(&registry_guard.list()) {
425
- Ok(json) => match CString::new(json) {
426
- Ok(cstr) => cstr.into_raw(),
427
- Err(e) => {
428
- set_last_error(format!("Failed to create C string: {}", e));
429
- ptr::null_mut()
430
- }
431
- },
432
- Err(e) => {
433
- set_last_error(format!("Failed to serialize PostProcessor list: {}", e));
434
- ptr::null_mut()
435
- }
436
- }
437
- })
438
- }