kreuzberg 4.0.0.rc2 → 4.0.1

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 (446) 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 +543 -538
  6. data/Gemfile +8 -8
  7. data/Gemfile.lock +194 -6
  8. data/README.md +391 -426
  9. data/Rakefile +34 -25
  10. data/Steepfile +51 -47
  11. data/examples/async_patterns.rb +283 -341
  12. data/ext/kreuzberg_rb/extconf.rb +65 -45
  13. data/ext/kreuzberg_rb/native/.cargo/config.toml +23 -0
  14. data/ext/kreuzberg_rb/native/Cargo.lock +7619 -6535
  15. data/ext/kreuzberg_rb/native/Cargo.toml +75 -44
  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 +3802 -2998
  23. data/extconf.rb +60 -28
  24. data/kreuzberg.gemspec +199 -148
  25. data/lib/kreuzberg/api_proxy.rb +126 -142
  26. data/lib/kreuzberg/cache_api.rb +67 -46
  27. data/lib/kreuzberg/cli.rb +47 -55
  28. data/lib/kreuzberg/cli_proxy.rb +117 -127
  29. data/lib/kreuzberg/config.rb +936 -691
  30. data/lib/kreuzberg/error_context.rb +136 -32
  31. data/lib/kreuzberg/errors.rb +116 -118
  32. data/lib/kreuzberg/extraction_api.rb +313 -85
  33. data/lib/kreuzberg/mcp_proxy.rb +177 -186
  34. data/lib/kreuzberg/ocr_backend_protocol.rb +40 -113
  35. data/lib/kreuzberg/post_processor_protocol.rb +15 -86
  36. data/lib/kreuzberg/result.rb +334 -216
  37. data/lib/kreuzberg/setup_lib_path.rb +99 -80
  38. data/lib/kreuzberg/types.rb +170 -0
  39. data/lib/kreuzberg/validator_protocol.rb +16 -89
  40. data/lib/kreuzberg/version.rb +5 -5
  41. data/lib/kreuzberg.rb +96 -103
  42. data/lib/libpdfium.so +0 -0
  43. data/sig/kreuzberg/internal.rbs +184 -184
  44. data/sig/kreuzberg.rbs +561 -520
  45. data/spec/binding/async_operations_spec.rb +473 -0
  46. data/spec/binding/batch_operations_spec.rb +595 -0
  47. data/spec/binding/batch_spec.rb +359 -0
  48. data/spec/binding/cache_spec.rb +227 -227
  49. data/spec/binding/cli_proxy_spec.rb +85 -85
  50. data/spec/binding/cli_spec.rb +55 -55
  51. data/spec/binding/config_result_spec.rb +377 -0
  52. data/spec/binding/config_spec.rb +419 -345
  53. data/spec/binding/config_validation_spec.rb +377 -283
  54. data/spec/binding/embeddings_spec.rb +816 -0
  55. data/spec/binding/error_handling_spec.rb +399 -213
  56. data/spec/binding/error_recovery_spec.rb +488 -0
  57. data/spec/binding/errors_spec.rb +66 -66
  58. data/spec/binding/font_config_spec.rb +220 -0
  59. data/spec/binding/images_spec.rb +738 -0
  60. data/spec/binding/keywords_extraction_spec.rb +600 -0
  61. data/spec/binding/metadata_types_spec.rb +1228 -0
  62. data/spec/binding/pages_extraction_spec.rb +471 -0
  63. data/spec/binding/plugins/ocr_backend_spec.rb +307 -307
  64. data/spec/binding/plugins/postprocessor_spec.rb +269 -269
  65. data/spec/binding/plugins/validator_spec.rb +273 -274
  66. data/spec/binding/tables_spec.rb +641 -0
  67. data/spec/fixtures/config.toml +38 -39
  68. data/spec/fixtures/config.yaml +41 -41
  69. data/spec/fixtures/invalid_config.toml +3 -4
  70. data/spec/smoke/package_spec.rb +177 -178
  71. data/spec/spec_helper.rb +40 -42
  72. data/spec/unit/config/chunking_config_spec.rb +213 -0
  73. data/spec/unit/config/embedding_config_spec.rb +343 -0
  74. data/spec/unit/config/extraction_config_spec.rb +438 -0
  75. data/spec/unit/config/font_config_spec.rb +285 -0
  76. data/spec/unit/config/hierarchy_config_spec.rb +314 -0
  77. data/spec/unit/config/image_extraction_config_spec.rb +209 -0
  78. data/spec/unit/config/image_preprocessing_config_spec.rb +249 -0
  79. data/spec/unit/config/keyword_config_spec.rb +229 -0
  80. data/spec/unit/config/language_detection_config_spec.rb +258 -0
  81. data/spec/unit/config/ocr_config_spec.rb +171 -0
  82. data/spec/unit/config/page_config_spec.rb +221 -0
  83. data/spec/unit/config/pdf_config_spec.rb +267 -0
  84. data/spec/unit/config/postprocessor_config_spec.rb +290 -0
  85. data/spec/unit/config/tesseract_config_spec.rb +181 -0
  86. data/spec/unit/config/token_reduction_config_spec.rb +251 -0
  87. data/test/metadata_types_test.rb +959 -0
  88. data/vendor/Cargo.toml +61 -0
  89. data/vendor/kreuzberg/Cargo.toml +259 -204
  90. data/vendor/kreuzberg/README.md +263 -175
  91. data/vendor/kreuzberg/build.rs +782 -474
  92. data/vendor/kreuzberg/examples/bench_fixes.rs +71 -0
  93. data/vendor/kreuzberg/examples/test_pdfium_fork.rs +62 -0
  94. data/vendor/kreuzberg/src/api/error.rs +81 -81
  95. data/vendor/kreuzberg/src/api/handlers.rs +320 -199
  96. data/vendor/kreuzberg/src/api/mod.rs +94 -79
  97. data/vendor/kreuzberg/src/api/server.rs +518 -353
  98. data/vendor/kreuzberg/src/api/types.rs +206 -170
  99. data/vendor/kreuzberg/src/cache/mod.rs +1167 -1167
  100. data/vendor/kreuzberg/src/chunking/mod.rs +2303 -677
  101. data/vendor/kreuzberg/src/chunking/processor.rs +219 -0
  102. data/vendor/kreuzberg/src/core/batch_mode.rs +95 -95
  103. data/vendor/kreuzberg/src/core/batch_optimizations.rs +385 -0
  104. data/vendor/kreuzberg/src/core/config.rs +1914 -1032
  105. data/vendor/kreuzberg/src/core/config_validation.rs +949 -0
  106. data/vendor/kreuzberg/src/core/extractor.rs +1200 -1024
  107. data/vendor/kreuzberg/src/core/formats.rs +235 -0
  108. data/vendor/kreuzberg/src/core/io.rs +329 -329
  109. data/vendor/kreuzberg/src/core/mime.rs +605 -605
  110. data/vendor/kreuzberg/src/core/mod.rs +61 -45
  111. data/vendor/kreuzberg/src/core/pipeline.rs +1223 -984
  112. data/vendor/kreuzberg/src/core/server_config.rs +1220 -0
  113. data/vendor/kreuzberg/src/embeddings.rs +471 -432
  114. data/vendor/kreuzberg/src/error.rs +431 -431
  115. data/vendor/kreuzberg/src/extraction/archive.rs +959 -954
  116. data/vendor/kreuzberg/src/extraction/capacity.rs +263 -0
  117. data/vendor/kreuzberg/src/extraction/docx.rs +404 -40
  118. data/vendor/kreuzberg/src/extraction/email.rs +855 -854
  119. data/vendor/kreuzberg/src/extraction/excel.rs +697 -688
  120. data/vendor/kreuzberg/src/extraction/html.rs +1830 -553
  121. data/vendor/kreuzberg/src/extraction/image.rs +492 -368
  122. data/vendor/kreuzberg/src/extraction/libreoffice.rs +574 -563
  123. data/vendor/kreuzberg/src/extraction/markdown.rs +216 -213
  124. data/vendor/kreuzberg/src/extraction/mod.rs +93 -81
  125. data/vendor/kreuzberg/src/extraction/office_metadata/app_properties.rs +398 -398
  126. data/vendor/kreuzberg/src/extraction/office_metadata/core_properties.rs +247 -247
  127. data/vendor/kreuzberg/src/extraction/office_metadata/custom_properties.rs +240 -240
  128. data/vendor/kreuzberg/src/extraction/office_metadata/mod.rs +130 -130
  129. data/vendor/kreuzberg/src/extraction/office_metadata/odt_properties.rs +284 -287
  130. data/vendor/kreuzberg/src/extraction/pptx.rs +3102 -3000
  131. data/vendor/kreuzberg/src/extraction/structured.rs +491 -490
  132. data/vendor/kreuzberg/src/extraction/table.rs +329 -328
  133. data/vendor/kreuzberg/src/extraction/text.rs +277 -269
  134. data/vendor/kreuzberg/src/extraction/xml.rs +333 -333
  135. data/vendor/kreuzberg/src/extractors/archive.rs +447 -446
  136. data/vendor/kreuzberg/src/extractors/bibtex.rs +470 -469
  137. data/vendor/kreuzberg/src/extractors/docbook.rs +504 -502
  138. data/vendor/kreuzberg/src/extractors/docx.rs +400 -367
  139. data/vendor/kreuzberg/src/extractors/email.rs +157 -143
  140. data/vendor/kreuzberg/src/extractors/epub.rs +696 -707
  141. data/vendor/kreuzberg/src/extractors/excel.rs +385 -343
  142. data/vendor/kreuzberg/src/extractors/fictionbook.rs +492 -491
  143. data/vendor/kreuzberg/src/extractors/html.rs +419 -393
  144. data/vendor/kreuzberg/src/extractors/image.rs +219 -198
  145. data/vendor/kreuzberg/src/extractors/jats.rs +1054 -1051
  146. data/vendor/kreuzberg/src/extractors/jupyter.rs +368 -367
  147. data/vendor/kreuzberg/src/extractors/latex.rs +653 -652
  148. data/vendor/kreuzberg/src/extractors/markdown.rs +701 -700
  149. data/vendor/kreuzberg/src/extractors/mod.rs +429 -365
  150. data/vendor/kreuzberg/src/extractors/odt.rs +628 -628
  151. data/vendor/kreuzberg/src/extractors/opml.rs +635 -634
  152. data/vendor/kreuzberg/src/extractors/orgmode.rs +529 -528
  153. data/vendor/kreuzberg/src/extractors/pdf.rs +761 -493
  154. data/vendor/kreuzberg/src/extractors/pptx.rs +279 -248
  155. data/vendor/kreuzberg/src/extractors/rst.rs +577 -576
  156. data/vendor/kreuzberg/src/extractors/rtf.rs +809 -810
  157. data/vendor/kreuzberg/src/extractors/security.rs +484 -484
  158. data/vendor/kreuzberg/src/extractors/security_tests.rs +367 -367
  159. data/vendor/kreuzberg/src/extractors/structured.rs +142 -140
  160. data/vendor/kreuzberg/src/extractors/text.rs +265 -260
  161. data/vendor/kreuzberg/src/extractors/typst.rs +651 -650
  162. data/vendor/kreuzberg/src/extractors/xml.rs +147 -135
  163. data/vendor/kreuzberg/src/image/dpi.rs +164 -164
  164. data/vendor/kreuzberg/src/image/mod.rs +6 -6
  165. data/vendor/kreuzberg/src/image/preprocessing.rs +417 -417
  166. data/vendor/kreuzberg/src/image/resize.rs +89 -89
  167. data/vendor/kreuzberg/src/keywords/config.rs +154 -154
  168. data/vendor/kreuzberg/src/keywords/mod.rs +237 -237
  169. data/vendor/kreuzberg/src/keywords/processor.rs +275 -267
  170. data/vendor/kreuzberg/src/keywords/rake.rs +293 -293
  171. data/vendor/kreuzberg/src/keywords/types.rs +68 -68
  172. data/vendor/kreuzberg/src/keywords/yake.rs +163 -163
  173. data/vendor/kreuzberg/src/language_detection/mod.rs +985 -942
  174. data/vendor/kreuzberg/src/language_detection/processor.rs +218 -0
  175. data/vendor/kreuzberg/src/lib.rs +114 -105
  176. data/vendor/kreuzberg/src/mcp/mod.rs +35 -32
  177. data/vendor/kreuzberg/src/mcp/server.rs +2090 -1968
  178. data/vendor/kreuzberg/src/ocr/cache.rs +469 -469
  179. data/vendor/kreuzberg/src/ocr/error.rs +37 -37
  180. data/vendor/kreuzberg/src/ocr/hocr.rs +216 -216
  181. data/vendor/kreuzberg/src/ocr/language_registry.rs +520 -0
  182. data/vendor/kreuzberg/src/ocr/mod.rs +60 -58
  183. data/vendor/kreuzberg/src/ocr/processor.rs +858 -863
  184. data/vendor/kreuzberg/src/ocr/table/mod.rs +4 -4
  185. data/vendor/kreuzberg/src/ocr/table/tsv_parser.rs +144 -144
  186. data/vendor/kreuzberg/src/ocr/tesseract_backend.rs +456 -450
  187. data/vendor/kreuzberg/src/ocr/types.rs +393 -393
  188. data/vendor/kreuzberg/src/ocr/utils.rs +47 -47
  189. data/vendor/kreuzberg/src/ocr/validation.rs +206 -206
  190. data/vendor/kreuzberg/src/panic_context.rs +154 -154
  191. data/vendor/kreuzberg/src/pdf/bindings.rs +306 -0
  192. data/vendor/kreuzberg/src/pdf/bundled.rs +408 -0
  193. data/vendor/kreuzberg/src/pdf/error.rs +214 -122
  194. data/vendor/kreuzberg/src/pdf/fonts.rs +358 -0
  195. data/vendor/kreuzberg/src/pdf/hierarchy.rs +903 -0
  196. data/vendor/kreuzberg/src/pdf/images.rs +139 -139
  197. data/vendor/kreuzberg/src/pdf/metadata.rs +509 -346
  198. data/vendor/kreuzberg/src/pdf/mod.rs +81 -50
  199. data/vendor/kreuzberg/src/pdf/rendering.rs +369 -369
  200. data/vendor/kreuzberg/src/pdf/table.rs +417 -393
  201. data/vendor/kreuzberg/src/pdf/text.rs +553 -158
  202. data/vendor/kreuzberg/src/plugins/extractor.rs +1042 -1013
  203. data/vendor/kreuzberg/src/plugins/mod.rs +212 -209
  204. data/vendor/kreuzberg/src/plugins/ocr.rs +637 -620
  205. data/vendor/kreuzberg/src/plugins/processor.rs +650 -642
  206. data/vendor/kreuzberg/src/plugins/registry.rs +1339 -1337
  207. data/vendor/kreuzberg/src/plugins/traits.rs +258 -258
  208. data/vendor/kreuzberg/src/plugins/validator.rs +967 -956
  209. data/vendor/kreuzberg/src/stopwords/mod.rs +1470 -1470
  210. data/vendor/kreuzberg/src/text/mod.rs +27 -19
  211. data/vendor/kreuzberg/src/text/quality.rs +710 -697
  212. data/vendor/kreuzberg/src/text/quality_processor.rs +231 -0
  213. data/vendor/kreuzberg/src/text/string_utils.rs +229 -217
  214. data/vendor/kreuzberg/src/text/token_reduction/cjk_utils.rs +164 -164
  215. data/vendor/kreuzberg/src/text/token_reduction/config.rs +100 -100
  216. data/vendor/kreuzberg/src/text/token_reduction/core.rs +832 -796
  217. data/vendor/kreuzberg/src/text/token_reduction/filters.rs +923 -902
  218. data/vendor/kreuzberg/src/text/token_reduction/mod.rs +160 -160
  219. data/vendor/kreuzberg/src/text/token_reduction/semantic.rs +619 -619
  220. data/vendor/kreuzberg/src/text/token_reduction/simd_text.rs +148 -147
  221. data/vendor/kreuzberg/src/text/utf8_validation.rs +193 -0
  222. data/vendor/kreuzberg/src/types.rs +1713 -903
  223. data/vendor/kreuzberg/src/utils/mod.rs +31 -17
  224. data/vendor/kreuzberg/src/utils/pool.rs +503 -0
  225. data/vendor/kreuzberg/src/utils/pool_sizing.rs +364 -0
  226. data/vendor/kreuzberg/src/utils/quality.rs +968 -959
  227. data/vendor/kreuzberg/src/utils/string_pool.rs +761 -0
  228. data/vendor/kreuzberg/src/utils/string_utils.rs +381 -381
  229. data/vendor/kreuzberg/stopwords/af_stopwords.json +53 -53
  230. data/vendor/kreuzberg/stopwords/ar_stopwords.json +482 -482
  231. data/vendor/kreuzberg/stopwords/bg_stopwords.json +261 -261
  232. data/vendor/kreuzberg/stopwords/bn_stopwords.json +400 -400
  233. data/vendor/kreuzberg/stopwords/br_stopwords.json +1205 -1205
  234. data/vendor/kreuzberg/stopwords/ca_stopwords.json +280 -280
  235. data/vendor/kreuzberg/stopwords/cs_stopwords.json +425 -425
  236. data/vendor/kreuzberg/stopwords/da_stopwords.json +172 -172
  237. data/vendor/kreuzberg/stopwords/de_stopwords.json +622 -622
  238. data/vendor/kreuzberg/stopwords/el_stopwords.json +849 -849
  239. data/vendor/kreuzberg/stopwords/en_stopwords.json +1300 -1300
  240. data/vendor/kreuzberg/stopwords/eo_stopwords.json +175 -175
  241. data/vendor/kreuzberg/stopwords/es_stopwords.json +734 -734
  242. data/vendor/kreuzberg/stopwords/et_stopwords.json +37 -37
  243. data/vendor/kreuzberg/stopwords/eu_stopwords.json +100 -100
  244. data/vendor/kreuzberg/stopwords/fa_stopwords.json +801 -801
  245. data/vendor/kreuzberg/stopwords/fi_stopwords.json +849 -849
  246. data/vendor/kreuzberg/stopwords/fr_stopwords.json +693 -693
  247. data/vendor/kreuzberg/stopwords/ga_stopwords.json +111 -111
  248. data/vendor/kreuzberg/stopwords/gl_stopwords.json +162 -162
  249. data/vendor/kreuzberg/stopwords/gu_stopwords.json +226 -226
  250. data/vendor/kreuzberg/stopwords/ha_stopwords.json +41 -41
  251. data/vendor/kreuzberg/stopwords/he_stopwords.json +196 -196
  252. data/vendor/kreuzberg/stopwords/hi_stopwords.json +227 -227
  253. data/vendor/kreuzberg/stopwords/hr_stopwords.json +181 -181
  254. data/vendor/kreuzberg/stopwords/hu_stopwords.json +791 -791
  255. data/vendor/kreuzberg/stopwords/hy_stopwords.json +47 -47
  256. data/vendor/kreuzberg/stopwords/id_stopwords.json +760 -760
  257. data/vendor/kreuzberg/stopwords/it_stopwords.json +634 -634
  258. data/vendor/kreuzberg/stopwords/ja_stopwords.json +136 -136
  259. data/vendor/kreuzberg/stopwords/kn_stopwords.json +84 -84
  260. data/vendor/kreuzberg/stopwords/ko_stopwords.json +681 -681
  261. data/vendor/kreuzberg/stopwords/ku_stopwords.json +64 -64
  262. data/vendor/kreuzberg/stopwords/la_stopwords.json +51 -51
  263. data/vendor/kreuzberg/stopwords/lt_stopwords.json +476 -476
  264. data/vendor/kreuzberg/stopwords/lv_stopwords.json +163 -163
  265. data/vendor/kreuzberg/stopwords/ml_stopwords.json +1 -1
  266. data/vendor/kreuzberg/stopwords/mr_stopwords.json +101 -101
  267. data/vendor/kreuzberg/stopwords/ms_stopwords.json +477 -477
  268. data/vendor/kreuzberg/stopwords/ne_stopwords.json +490 -490
  269. data/vendor/kreuzberg/stopwords/nl_stopwords.json +415 -415
  270. data/vendor/kreuzberg/stopwords/no_stopwords.json +223 -223
  271. data/vendor/kreuzberg/stopwords/pl_stopwords.json +331 -331
  272. data/vendor/kreuzberg/stopwords/pt_stopwords.json +562 -562
  273. data/vendor/kreuzberg/stopwords/ro_stopwords.json +436 -436
  274. data/vendor/kreuzberg/stopwords/ru_stopwords.json +561 -561
  275. data/vendor/kreuzberg/stopwords/si_stopwords.json +193 -193
  276. data/vendor/kreuzberg/stopwords/sk_stopwords.json +420 -420
  277. data/vendor/kreuzberg/stopwords/sl_stopwords.json +448 -448
  278. data/vendor/kreuzberg/stopwords/so_stopwords.json +32 -32
  279. data/vendor/kreuzberg/stopwords/st_stopwords.json +33 -33
  280. data/vendor/kreuzberg/stopwords/sv_stopwords.json +420 -420
  281. data/vendor/kreuzberg/stopwords/sw_stopwords.json +76 -76
  282. data/vendor/kreuzberg/stopwords/ta_stopwords.json +129 -129
  283. data/vendor/kreuzberg/stopwords/te_stopwords.json +54 -54
  284. data/vendor/kreuzberg/stopwords/th_stopwords.json +118 -118
  285. data/vendor/kreuzberg/stopwords/tl_stopwords.json +149 -149
  286. data/vendor/kreuzberg/stopwords/tr_stopwords.json +506 -506
  287. data/vendor/kreuzberg/stopwords/uk_stopwords.json +75 -75
  288. data/vendor/kreuzberg/stopwords/ur_stopwords.json +519 -519
  289. data/vendor/kreuzberg/stopwords/vi_stopwords.json +647 -647
  290. data/vendor/kreuzberg/stopwords/yo_stopwords.json +62 -62
  291. data/vendor/kreuzberg/stopwords/zh_stopwords.json +796 -796
  292. data/vendor/kreuzberg/stopwords/zu_stopwords.json +31 -31
  293. data/vendor/kreuzberg/tests/api_embed.rs +360 -0
  294. data/vendor/kreuzberg/tests/api_extract_multipart.rs +52 -52
  295. data/vendor/kreuzberg/tests/api_large_pdf_extraction.rs +471 -0
  296. data/vendor/kreuzberg/tests/api_large_pdf_extraction_diagnostics.rs +289 -0
  297. data/vendor/kreuzberg/tests/api_tests.rs +1472 -966
  298. data/vendor/kreuzberg/tests/archive_integration.rs +545 -543
  299. data/vendor/kreuzberg/tests/batch_orchestration.rs +587 -556
  300. data/vendor/kreuzberg/tests/batch_pooling_benchmark.rs +154 -0
  301. data/vendor/kreuzberg/tests/batch_processing.rs +328 -316
  302. data/vendor/kreuzberg/tests/bibtex_parity_test.rs +421 -421
  303. data/vendor/kreuzberg/tests/concurrency_stress.rs +541 -525
  304. data/vendor/kreuzberg/tests/config_features.rs +612 -598
  305. data/vendor/kreuzberg/tests/config_integration_test.rs +753 -0
  306. data/vendor/kreuzberg/tests/config_loading_tests.rs +416 -415
  307. data/vendor/kreuzberg/tests/core_integration.rs +519 -510
  308. data/vendor/kreuzberg/tests/csv_integration.rs +414 -414
  309. data/vendor/kreuzberg/tests/data/hierarchy_ground_truth.json +294 -0
  310. data/vendor/kreuzberg/tests/docbook_extractor_tests.rs +500 -498
  311. data/vendor/kreuzberg/tests/docx_metadata_extraction_test.rs +122 -122
  312. data/vendor/kreuzberg/tests/docx_vs_pandoc_comparison.rs +370 -370
  313. data/vendor/kreuzberg/tests/email_integration.rs +327 -325
  314. data/vendor/kreuzberg/tests/epub_native_extractor_tests.rs +275 -275
  315. data/vendor/kreuzberg/tests/error_handling.rs +402 -393
  316. data/vendor/kreuzberg/tests/fictionbook_extractor_tests.rs +228 -228
  317. data/vendor/kreuzberg/tests/format_integration.rs +165 -159
  318. data/vendor/kreuzberg/tests/helpers/mod.rs +202 -142
  319. data/vendor/kreuzberg/tests/html_table_test.rs +551 -551
  320. data/vendor/kreuzberg/tests/image_integration.rs +255 -253
  321. data/vendor/kreuzberg/tests/instrumentation_test.rs +139 -139
  322. data/vendor/kreuzberg/tests/jats_extractor_tests.rs +639 -639
  323. data/vendor/kreuzberg/tests/jupyter_extractor_tests.rs +704 -704
  324. data/vendor/kreuzberg/tests/keywords_integration.rs +479 -479
  325. data/vendor/kreuzberg/tests/keywords_quality.rs +509 -509
  326. data/vendor/kreuzberg/tests/latex_extractor_tests.rs +496 -496
  327. data/vendor/kreuzberg/tests/markdown_extractor_tests.rs +490 -490
  328. data/vendor/kreuzberg/tests/mime_detection.rs +429 -428
  329. data/vendor/kreuzberg/tests/ocr_configuration.rs +514 -510
  330. data/vendor/kreuzberg/tests/ocr_errors.rs +698 -676
  331. data/vendor/kreuzberg/tests/ocr_language_registry.rs +191 -0
  332. data/vendor/kreuzberg/tests/ocr_quality.rs +629 -627
  333. data/vendor/kreuzberg/tests/ocr_stress.rs +469 -469
  334. data/vendor/kreuzberg/tests/odt_extractor_tests.rs +674 -695
  335. data/vendor/kreuzberg/tests/opml_extractor_tests.rs +616 -616
  336. data/vendor/kreuzberg/tests/orgmode_extractor_tests.rs +822 -822
  337. data/vendor/kreuzberg/tests/page_markers.rs +297 -0
  338. data/vendor/kreuzberg/tests/pdf_hierarchy_detection.rs +301 -0
  339. data/vendor/kreuzberg/tests/pdf_hierarchy_quality.rs +589 -0
  340. data/vendor/kreuzberg/tests/pdf_integration.rs +45 -43
  341. data/vendor/kreuzberg/tests/pdf_ocr_triggering.rs +301 -0
  342. data/vendor/kreuzberg/tests/pdf_text_merging.rs +475 -0
  343. data/vendor/kreuzberg/tests/pdfium_linking.rs +340 -0
  344. data/vendor/kreuzberg/tests/pipeline_integration.rs +1446 -1411
  345. data/vendor/kreuzberg/tests/plugin_ocr_backend_test.rs +776 -771
  346. data/vendor/kreuzberg/tests/plugin_postprocessor_test.rs +577 -560
  347. data/vendor/kreuzberg/tests/plugin_system.rs +927 -921
  348. data/vendor/kreuzberg/tests/plugin_validator_test.rs +783 -783
  349. data/vendor/kreuzberg/tests/registry_integration_tests.rs +587 -586
  350. data/vendor/kreuzberg/tests/rst_extractor_tests.rs +694 -692
  351. data/vendor/kreuzberg/tests/rtf_extractor_tests.rs +775 -776
  352. data/vendor/kreuzberg/tests/security_validation.rs +416 -415
  353. data/vendor/kreuzberg/tests/stopwords_integration_test.rs +888 -888
  354. data/vendor/kreuzberg/tests/test_fastembed.rs +631 -609
  355. data/vendor/kreuzberg/tests/typst_behavioral_tests.rs +1260 -1259
  356. data/vendor/kreuzberg/tests/typst_extractor_tests.rs +648 -647
  357. data/vendor/kreuzberg/tests/xlsx_metadata_extraction_test.rs +87 -87
  358. data/vendor/kreuzberg-ffi/Cargo.toml +67 -0
  359. data/vendor/kreuzberg-ffi/README.md +851 -0
  360. data/vendor/kreuzberg-ffi/benches/result_view_benchmark.rs +227 -0
  361. data/vendor/kreuzberg-ffi/build.rs +168 -0
  362. data/vendor/kreuzberg-ffi/cbindgen.toml +37 -0
  363. data/vendor/kreuzberg-ffi/kreuzberg-ffi.pc.in +12 -0
  364. data/vendor/kreuzberg-ffi/kreuzberg.h +3012 -0
  365. data/vendor/kreuzberg-ffi/src/batch_streaming.rs +588 -0
  366. data/vendor/kreuzberg-ffi/src/config.rs +1341 -0
  367. data/vendor/kreuzberg-ffi/src/error.rs +901 -0
  368. data/vendor/kreuzberg-ffi/src/extraction.rs +555 -0
  369. data/vendor/kreuzberg-ffi/src/helpers.rs +879 -0
  370. data/vendor/kreuzberg-ffi/src/lib.rs +977 -0
  371. data/vendor/kreuzberg-ffi/src/memory.rs +493 -0
  372. data/vendor/kreuzberg-ffi/src/mime.rs +329 -0
  373. data/vendor/kreuzberg-ffi/src/panic_shield.rs +265 -0
  374. data/vendor/kreuzberg-ffi/src/plugins/document_extractor.rs +442 -0
  375. data/vendor/kreuzberg-ffi/src/plugins/mod.rs +14 -0
  376. data/vendor/kreuzberg-ffi/src/plugins/ocr_backend.rs +628 -0
  377. data/vendor/kreuzberg-ffi/src/plugins/post_processor.rs +438 -0
  378. data/vendor/kreuzberg-ffi/src/plugins/validator.rs +329 -0
  379. data/vendor/kreuzberg-ffi/src/result.rs +510 -0
  380. data/vendor/kreuzberg-ffi/src/result_pool.rs +639 -0
  381. data/vendor/kreuzberg-ffi/src/result_view.rs +773 -0
  382. data/vendor/kreuzberg-ffi/src/string_intern.rs +568 -0
  383. data/vendor/kreuzberg-ffi/src/types.rs +363 -0
  384. data/vendor/kreuzberg-ffi/src/util.rs +210 -0
  385. data/vendor/kreuzberg-ffi/src/validation.rs +848 -0
  386. data/vendor/kreuzberg-ffi/tests.disabled/README.md +48 -0
  387. data/vendor/kreuzberg-ffi/tests.disabled/config_loading_tests.rs +299 -0
  388. data/vendor/kreuzberg-ffi/tests.disabled/config_tests.rs +346 -0
  389. data/vendor/kreuzberg-ffi/tests.disabled/extractor_tests.rs +232 -0
  390. data/vendor/kreuzberg-ffi/tests.disabled/plugin_registration_tests.rs +470 -0
  391. data/vendor/kreuzberg-tesseract/.commitlintrc.json +13 -0
  392. data/vendor/kreuzberg-tesseract/.crate-ignore +2 -0
  393. data/vendor/kreuzberg-tesseract/Cargo.lock +2933 -0
  394. data/vendor/kreuzberg-tesseract/Cargo.toml +57 -0
  395. data/vendor/{rb-sys/LICENSE-MIT → kreuzberg-tesseract/LICENSE} +22 -21
  396. data/vendor/kreuzberg-tesseract/README.md +399 -0
  397. data/vendor/kreuzberg-tesseract/build.rs +1127 -0
  398. data/vendor/kreuzberg-tesseract/patches/README.md +71 -0
  399. data/vendor/kreuzberg-tesseract/patches/tesseract.diff +199 -0
  400. data/vendor/kreuzberg-tesseract/src/api.rs +1371 -0
  401. data/vendor/kreuzberg-tesseract/src/choice_iterator.rs +77 -0
  402. data/vendor/kreuzberg-tesseract/src/enums.rs +297 -0
  403. data/vendor/kreuzberg-tesseract/src/error.rs +81 -0
  404. data/vendor/kreuzberg-tesseract/src/lib.rs +145 -0
  405. data/vendor/kreuzberg-tesseract/src/monitor.rs +57 -0
  406. data/vendor/kreuzberg-tesseract/src/mutable_iterator.rs +197 -0
  407. data/vendor/kreuzberg-tesseract/src/page_iterator.rs +253 -0
  408. data/vendor/kreuzberg-tesseract/src/result_iterator.rs +286 -0
  409. data/vendor/kreuzberg-tesseract/src/result_renderer.rs +183 -0
  410. data/vendor/kreuzberg-tesseract/tests/integration_test.rs +211 -0
  411. metadata +196 -45
  412. data/vendor/kreuzberg/benches/otel_overhead.rs +0 -48
  413. data/vendor/kreuzberg/src/extractors/fictionbook.rs.backup2 +0 -738
  414. data/vendor/rb-sys/.cargo-ok +0 -1
  415. data/vendor/rb-sys/.cargo_vcs_info.json +0 -6
  416. data/vendor/rb-sys/Cargo.lock +0 -393
  417. data/vendor/rb-sys/Cargo.toml +0 -70
  418. data/vendor/rb-sys/Cargo.toml.orig +0 -57
  419. data/vendor/rb-sys/LICENSE-APACHE +0 -190
  420. data/vendor/rb-sys/bin/release.sh +0 -21
  421. data/vendor/rb-sys/build/features.rs +0 -108
  422. data/vendor/rb-sys/build/main.rs +0 -246
  423. data/vendor/rb-sys/build/stable_api_config.rs +0 -153
  424. data/vendor/rb-sys/build/version.rs +0 -48
  425. data/vendor/rb-sys/readme.md +0 -36
  426. data/vendor/rb-sys/src/bindings.rs +0 -21
  427. data/vendor/rb-sys/src/hidden.rs +0 -11
  428. data/vendor/rb-sys/src/lib.rs +0 -34
  429. data/vendor/rb-sys/src/macros.rs +0 -371
  430. data/vendor/rb-sys/src/memory.rs +0 -53
  431. data/vendor/rb-sys/src/ruby_abi_version.rs +0 -38
  432. data/vendor/rb-sys/src/special_consts.rs +0 -31
  433. data/vendor/rb-sys/src/stable_api/compiled.c +0 -179
  434. data/vendor/rb-sys/src/stable_api/compiled.rs +0 -257
  435. data/vendor/rb-sys/src/stable_api/ruby_2_6.rs +0 -316
  436. data/vendor/rb-sys/src/stable_api/ruby_2_7.rs +0 -316
  437. data/vendor/rb-sys/src/stable_api/ruby_3_0.rs +0 -324
  438. data/vendor/rb-sys/src/stable_api/ruby_3_1.rs +0 -317
  439. data/vendor/rb-sys/src/stable_api/ruby_3_2.rs +0 -315
  440. data/vendor/rb-sys/src/stable_api/ruby_3_3.rs +0 -326
  441. data/vendor/rb-sys/src/stable_api/ruby_3_4.rs +0 -327
  442. data/vendor/rb-sys/src/stable_api.rs +0 -261
  443. data/vendor/rb-sys/src/symbol.rs +0 -31
  444. data/vendor/rb-sys/src/tracking_allocator.rs +0 -332
  445. data/vendor/rb-sys/src/utils.rs +0 -89
  446. data/vendor/rb-sys/src/value_type.rs +0 -7
@@ -0,0 +1,475 @@
1
+ //! Tests for PDF text extraction with font information and bounding box operations.
2
+ //!
3
+ //! This module tests character extraction functionality following TDD principles,
4
+ //! verifying that characters are extracted with positions and font sizes,
5
+ //! and that bounding box operations work correctly.
6
+
7
+ #![cfg(feature = "pdf")]
8
+
9
+ mod helpers;
10
+
11
+ use helpers::get_test_file_path;
12
+ use kreuzberg::pdf::hierarchy::{BoundingBox, extract_chars_with_fonts};
13
+ use pdfium_render::prelude::*;
14
+
15
+ // ============================================================================
16
+ // Character Extraction Tests (Following TDD)
17
+ // ============================================================================
18
+
19
+ /// Test basic character extraction with positions and font sizes.
20
+ ///
21
+ /// Verifies that:
22
+ /// - Characters are extracted with their positions
23
+ /// - Font sizes are captured
24
+ /// - Position order is preserved
25
+ #[test]
26
+ fn test_extract_chars_basic() {
27
+ let pdf_path = get_test_file_path("pdfs_with_tables/tiny.pdf");
28
+
29
+ // Load PDF
30
+ let pdfium = Pdfium;
31
+ let document = pdfium
32
+ .load_pdf_from_file(pdf_path.to_str().unwrap(), None)
33
+ .expect("Failed to load test PDF");
34
+
35
+ // Get first page
36
+ let page = document.pages().get(0).expect("Failed to get first page");
37
+
38
+ // Extract characters with fonts
39
+ let chars = extract_chars_with_fonts(&page).expect("Failed to extract characters with fonts");
40
+
41
+ // Verify we got some characters
42
+ assert!(!chars.is_empty(), "Should extract at least one character from test PDF");
43
+
44
+ // Verify each character has required fields
45
+ for char_data in chars.iter() {
46
+ assert!(!char_data.text.is_empty(), "Character text should not be empty");
47
+ assert!(char_data.font_size > 0.0, "Font size should be positive");
48
+ assert!(char_data.x >= 0.0, "X position should be non-negative");
49
+ assert!(char_data.y >= 0.0, "Y position should be non-negative");
50
+ }
51
+ }
52
+
53
+ /// Test that character extraction preserves reading order.
54
+ ///
55
+ /// Verifies that:
56
+ /// - Characters appear in left-to-right order
57
+ /// - Y-coordinates generally decrease as we move down the page
58
+ #[test]
59
+ fn test_extract_chars_preserves_order() {
60
+ let pdf_path = get_test_file_path("pdfs_with_tables/tiny.pdf");
61
+
62
+ // Load PDF
63
+ let pdfium = Pdfium;
64
+ let document = pdfium
65
+ .load_pdf_from_file(pdf_path.to_str().unwrap(), None)
66
+ .expect("Failed to load test PDF");
67
+
68
+ // Get first page
69
+ let page = document.pages().get(0).expect("Failed to get first page");
70
+
71
+ // Extract characters with fonts
72
+ let chars = extract_chars_with_fonts(&page).expect("Failed to extract characters with fonts");
73
+
74
+ assert!(!chars.is_empty(), "Should extract at least one character");
75
+
76
+ // Within each line (similar y-coordinate), characters should be left-to-right
77
+ let mut last_y = f32::NEG_INFINITY;
78
+ let mut last_x = f32::NEG_INFINITY;
79
+ let y_line_threshold = 5.0; // Consider within 5 units as same line
80
+
81
+ for char_data in chars.iter() {
82
+ // If we're on a new line
83
+ if (char_data.y - last_y).abs() > y_line_threshold {
84
+ last_x = f32::NEG_INFINITY;
85
+ last_y = char_data.y;
86
+ }
87
+
88
+ // On same line, x should generally increase (allowing for small variations)
89
+ // We use a threshold to allow for measurement precision issues
90
+ if (char_data.y - last_y).abs() <= y_line_threshold && char_data.x < last_x - 1.0 {
91
+ // This is acceptable if it's a new line or small variation
92
+ if last_x != f32::NEG_INFINITY && (char_data.y - last_y).abs() <= y_line_threshold {
93
+ panic!(
94
+ "Characters should be left-to-right on same line: {} < {} at y={}",
95
+ char_data.x, last_x, char_data.y
96
+ );
97
+ }
98
+ }
99
+
100
+ last_x = char_data.x;
101
+ last_y = char_data.y;
102
+ }
103
+ }
104
+
105
+ // ============================================================================
106
+ // Bounding Box Tests
107
+ // ============================================================================
108
+
109
+ /// Helper function to create a BoundingBox from x, y, width, height
110
+ fn create_bbox(x: f32, y: f32, width: f32, height: f32) -> BoundingBox {
111
+ BoundingBox {
112
+ left: x,
113
+ top: y,
114
+ right: x + width,
115
+ bottom: y + height,
116
+ }
117
+ }
118
+
119
+ #[test]
120
+ fn test_iou_calculation() {
121
+ // Two overlapping boxes
122
+ let bbox1 = create_bbox(0.0, 0.0, 10.0, 10.0);
123
+ let bbox2 = create_bbox(5.0, 5.0, 10.0, 10.0);
124
+
125
+ // Expected intersection: 5x5 = 25
126
+ // Expected union: 100 + 100 - 25 = 175
127
+ // Expected IOU: 25/175 ≈ 0.1429
128
+ let iou = bbox1.iou(&bbox2);
129
+ assert!((iou - 0.1429).abs() < 0.001, "IOU calculation failed");
130
+ }
131
+
132
+ #[test]
133
+ fn test_weighted_distance_calculation() {
134
+ // Two boxes with different X and Y distances
135
+ let bbox1 = create_bbox(0.0, 0.0, 10.0, 10.0);
136
+ let bbox2 = create_bbox(20.0, 5.0, 10.0, 10.0);
137
+
138
+ // Distance: X=20, Y=5
139
+ // Weighted: X*5.0 + Y*1.0 = 20*5.0 + 5*1.0 = 100 + 5 = 105
140
+ let weighted_dist = bbox1.weighted_distance(&bbox2);
141
+ assert!(
142
+ (weighted_dist - 105.0).abs() < 0.001,
143
+ "Weighted distance calculation failed"
144
+ );
145
+
146
+ // Verify X weight (5.0) > Y weight (1.0) by checking ratio
147
+ let bbox3 = create_bbox(0.0, 0.0, 10.0, 10.0);
148
+ let bbox4 = create_bbox(10.0, 0.0, 10.0, 10.0);
149
+ let only_x_dist = bbox3.weighted_distance(&bbox4);
150
+
151
+ let bbox5 = create_bbox(0.0, 0.0, 10.0, 10.0);
152
+ let bbox6 = create_bbox(0.0, 10.0, 10.0, 10.0);
153
+ let only_y_dist = bbox5.weighted_distance(&bbox6);
154
+
155
+ // X distance of 10 with weight 5.0 = 50
156
+ // Y distance of 10 with weight 1.0 = 10
157
+ // X weight should be 5x larger than Y weight
158
+ assert!(only_x_dist > only_y_dist, "X weight should be greater than Y weight");
159
+ assert!((only_x_dist - 50.0).abs() < 0.001, "X-only weighted distance failed");
160
+ assert!((only_y_dist - 10.0).abs() < 0.001, "Y-only weighted distance failed");
161
+ }
162
+
163
+ #[test]
164
+ fn test_intersection_ratio() {
165
+ // Two overlapping boxes
166
+ let bbox1 = create_bbox(0.0, 0.0, 10.0, 10.0);
167
+ let bbox2 = create_bbox(5.0, 5.0, 10.0, 10.0);
168
+
169
+ // Expected intersection: 5x5 = 25
170
+ // bbox1 area: 100
171
+ // Expected ratio: 25/100 = 0.25
172
+ let ratio = bbox1.intersection_ratio(&bbox2);
173
+ assert!((ratio - 0.25).abs() < 0.001, "Intersection ratio calculation failed");
174
+ }
175
+
176
+ #[test]
177
+ fn test_edge_case_no_overlap() {
178
+ // Two non-overlapping boxes
179
+ let bbox1 = create_bbox(0.0, 0.0, 10.0, 10.0);
180
+ let bbox2 = create_bbox(20.0, 20.0, 10.0, 10.0);
181
+
182
+ // IOU should be 0
183
+ let iou = bbox1.iou(&bbox2);
184
+ assert!((iou - 0.0).abs() < 0.001, "Non-overlapping boxes should have IOU of 0");
185
+
186
+ // Intersection ratio should be 0
187
+ let ratio = bbox1.intersection_ratio(&bbox2);
188
+ assert!(
189
+ (ratio - 0.0).abs() < 0.001,
190
+ "Non-overlapping boxes should have intersection ratio of 0"
191
+ );
192
+ }
193
+
194
+ #[test]
195
+ fn test_edge_case_fully_contained() {
196
+ // Smaller box fully contained in larger box
197
+ let bbox_large = create_bbox(0.0, 0.0, 20.0, 20.0);
198
+ let bbox_small = create_bbox(5.0, 5.0, 10.0, 10.0);
199
+
200
+ // Intersection: 10x10 = 100
201
+ // Union: 400 + 100 - 100 = 400
202
+ // IOU: 100/400 = 0.25
203
+ let iou = bbox_large.iou(&bbox_small);
204
+ assert!((iou - 0.25).abs() < 0.001, "Fully contained box IOU calculation failed");
205
+
206
+ // Intersection ratio: 100/400 = 0.25
207
+ let ratio = bbox_large.intersection_ratio(&bbox_small);
208
+ assert!(
209
+ (ratio - 0.25).abs() < 0.001,
210
+ "Fully contained box intersection ratio failed"
211
+ );
212
+ }
213
+
214
+ // ============================================================================
215
+ // Character Merging Tests (Following TDD)
216
+ // ============================================================================
217
+
218
+ use kreuzberg::pdf::hierarchy::{CharData, merge_chars_into_blocks};
219
+
220
+ /// Factory helper to create a CharData with minimal parameters.
221
+ fn create_char(text: &str, x: f32, y: f32, font_size: f32) -> CharData {
222
+ CharData {
223
+ text: text.to_string(),
224
+ font_size,
225
+ x,
226
+ y,
227
+ width: font_size * 0.6,
228
+ height: font_size,
229
+ }
230
+ }
231
+
232
+ /// Test horizontal text merging: characters at (0,10), (10,10), (20,10) should merge into 1 block.
233
+ ///
234
+ /// This test validates that characters on the same horizontal line and within the merge
235
+ /// distance threshold are grouped into a single text block.
236
+ #[test]
237
+ fn test_merge_horizontal_text_merging() {
238
+ let chars = vec![
239
+ create_char("H", 0.0, 10.0, 12.0),
240
+ create_char("e", 10.0, 10.0, 12.0),
241
+ create_char("y", 20.0, 10.0, 12.0),
242
+ ];
243
+
244
+ let blocks = merge_chars_into_blocks(chars);
245
+
246
+ assert_eq!(blocks.len(), 1, "Expected 1 block for horizontal text");
247
+ assert_eq!(blocks[0].text, "Hey", "Expected merged text 'Hey'");
248
+ }
249
+
250
+ /// Test vertical text separation: characters at (0,10) and (0,50) should be 2 blocks.
251
+ ///
252
+ /// This test validates that characters with large vertical separation are kept in separate blocks.
253
+ #[test]
254
+ fn test_merge_vertical_text_separation() {
255
+ let chars = vec![create_char("A", 0.0, 10.0, 12.0), create_char("B", 0.0, 50.0, 12.0)];
256
+
257
+ let blocks = merge_chars_into_blocks(chars);
258
+
259
+ assert_eq!(blocks.len(), 2, "Expected 2 blocks for vertically separated text");
260
+ assert_eq!(blocks[0].text, "A", "Expected first block to contain 'A'");
261
+ assert_eq!(blocks[1].text, "B", "Expected second block to contain 'B'");
262
+ }
263
+
264
+ /// Test edge case with negative coordinates.
265
+ ///
266
+ /// This test ensures the merging algorithm handles negative coordinates correctly.
267
+ #[test]
268
+ fn test_merge_edge_case_negative_coordinates() {
269
+ let chars = vec![
270
+ create_char("X", -10.0, -5.0, 12.0),
271
+ create_char("Y", 0.0, -5.0, 12.0),
272
+ create_char("Z", 10.0, -5.0, 12.0),
273
+ ];
274
+
275
+ let blocks = merge_chars_into_blocks(chars);
276
+
277
+ assert_eq!(blocks.len(), 1, "Expected 1 block for negative coordinates");
278
+ assert_eq!(blocks[0].text, "XYZ", "Expected merged text 'XYZ'");
279
+ }
280
+
281
+ /// Test edge case with overlapping blocks.
282
+ ///
283
+ /// This test validates behavior when characters have overlapping or very close bounding boxes.
284
+ #[test]
285
+ fn test_merge_edge_case_overlapping_blocks() {
286
+ let chars = vec![
287
+ create_char("O", 0.0, 0.0, 12.0),
288
+ create_char("V", 1.0, 0.0, 12.0),
289
+ create_char("E", 2.0, 0.0, 12.0),
290
+ ];
291
+
292
+ let blocks = merge_chars_into_blocks(chars);
293
+
294
+ assert_eq!(blocks.len(), 1, "Expected 1 block for overlapping characters");
295
+ assert_eq!(blocks[0].text, "OVE", "Expected merged text 'OVE'");
296
+ }
297
+
298
+ /// Test max merge distance threshold.
299
+ ///
300
+ /// This test validates that characters beyond the maximum merge distance are kept separate.
301
+ #[test]
302
+ fn test_merge_max_merge_distance_threshold() {
303
+ let chars = vec![
304
+ create_char("T", 0.0, 10.0, 12.0),
305
+ create_char("e", 50.0, 10.0, 12.0), // Large gap, should be separate
306
+ create_char("s", 100.0, 10.0, 12.0), // Even larger gap
307
+ ];
308
+
309
+ let blocks = merge_chars_into_blocks(chars);
310
+
311
+ // With reasonable merge distance (should be ~2.5x font size for distance),
312
+ // characters at 50 and 100 units apart should create separate blocks
313
+ assert!(blocks.len() > 1, "Expected multiple blocks due to large gaps");
314
+ assert_eq!(
315
+ blocks.iter().map(|b| b.text.len()).sum::<usize>(),
316
+ 3,
317
+ "Expected all 3 characters to be preserved"
318
+ );
319
+ }
320
+
321
+ /// Test edge case with zero font size.
322
+ ///
323
+ /// This test ensures the algorithm handles zero or near-zero font sizes gracefully
324
+ /// without panicking or causing division by zero.
325
+ #[test]
326
+ fn test_merge_zero_font_size() {
327
+ // Create characters with zero font size (edge case)
328
+ let chars = vec![
329
+ CharData {
330
+ text: "A".to_string(),
331
+ x: 0.0,
332
+ y: 10.0,
333
+ font_size: 0.0,
334
+ width: 0.0,
335
+ height: 0.0,
336
+ },
337
+ CharData {
338
+ text: "B".to_string(),
339
+ x: 1.0,
340
+ y: 10.0,
341
+ font_size: 0.0,
342
+ width: 0.0,
343
+ height: 0.0,
344
+ },
345
+ ];
346
+
347
+ // Should not panic
348
+ let blocks = merge_chars_into_blocks(chars);
349
+
350
+ // Should still produce some output
351
+ assert!(!blocks.is_empty(), "Should produce blocks even with zero font size");
352
+ }
353
+
354
+ /// Test IOU calculation with zero area boxes.
355
+ ///
356
+ /// This test validates that IOU calculations handle degenerate bounding boxes
357
+ /// with zero area without panicking.
358
+ #[test]
359
+ fn test_iou_zero_area_boxes() {
360
+ // Two boxes with zero area (point-like)
361
+ let bbox1 = create_bbox(0.0, 0.0, 0.0, 0.0);
362
+ let bbox2 = create_bbox(5.0, 5.0, 0.0, 0.0);
363
+
364
+ // Should not panic and should return 0 or similar
365
+ let iou = bbox1.iou(&bbox2);
366
+ assert!(
367
+ (0.0..=1.0).contains(&iou),
368
+ "IOU should be in valid range for zero-area boxes"
369
+ );
370
+
371
+ // Intersection ratio should also be safe
372
+ let ratio = bbox1.intersection_ratio(&bbox2);
373
+ assert!(
374
+ (0.0..=1.0).contains(&ratio),
375
+ "Intersection ratio should be in valid range"
376
+ );
377
+ }
378
+
379
+ /// Test IOU calculation with identical boxes.
380
+ ///
381
+ /// This test validates that identical boxes have IOU of 1.0 (perfect overlap).
382
+ #[test]
383
+ fn test_iou_identical_boxes() {
384
+ // Two identical boxes
385
+ let bbox1 = create_bbox(10.0, 20.0, 30.0, 40.0);
386
+ let bbox2 = create_bbox(10.0, 20.0, 30.0, 40.0);
387
+
388
+ // IOU should be 1.0 for identical boxes
389
+ let iou = bbox1.iou(&bbox2);
390
+ assert!(
391
+ (iou - 1.0).abs() < 0.001,
392
+ "Identical boxes should have IOU of 1.0, got {}",
393
+ iou
394
+ );
395
+
396
+ // Intersection ratio should also be 1.0
397
+ let ratio = bbox1.intersection_ratio(&bbox2);
398
+ assert!(
399
+ (ratio - 1.0).abs() < 0.001,
400
+ "Identical boxes should have intersection ratio of 1.0, got {}",
401
+ ratio
402
+ );
403
+ }
404
+
405
+ /// Test contains() method with various box configurations.
406
+ #[test]
407
+ fn test_contains_method() {
408
+ let large_box = create_bbox(0.0, 0.0, 100.0, 100.0);
409
+ let small_box = create_bbox(10.0, 10.0, 50.0, 50.0);
410
+ let outside_box = create_bbox(110.0, 110.0, 150.0, 150.0);
411
+
412
+ // Small box should be contained in large box
413
+ assert!(large_box.contains(&small_box), "Large box should contain small box");
414
+
415
+ // Large box should not be contained in small box
416
+ assert!(
417
+ !small_box.contains(&large_box),
418
+ "Small box should not contain large box"
419
+ );
420
+
421
+ // Outside box should not be contained
422
+ assert!(
423
+ !large_box.contains(&outside_box),
424
+ "Large box should not contain outside box"
425
+ );
426
+ }
427
+
428
+ /// Test center() method for bounding box centerpoint calculation.
429
+ #[test]
430
+ fn test_center_method() {
431
+ let bbox = create_bbox(0.0, 0.0, 100.0, 100.0);
432
+ let center = bbox.center();
433
+
434
+ assert_eq!(center.0, 50.0, "Center X should be 50.0");
435
+ assert_eq!(center.1, 50.0, "Center Y should be 50.0");
436
+
437
+ // Test with offset box
438
+ let offset_bbox = create_bbox(20.0, 30.0, 80.0, 70.0);
439
+ let offset_center = offset_bbox.center();
440
+
441
+ assert_eq!(offset_center.0, 60.0, "Offset center X should be 60.0");
442
+ assert_eq!(offset_center.1, 65.0, "Offset center Y should be 65.0");
443
+ }
444
+
445
+ /// Test merge() method for combining bounding boxes.
446
+ #[test]
447
+ fn test_merge_method() {
448
+ let bbox1 = create_bbox(0.0, 0.0, 50.0, 50.0);
449
+ let bbox2 = create_bbox(30.0, 30.0, 100.0, 100.0);
450
+
451
+ let merged = bbox1.merge(&bbox2);
452
+
453
+ assert_eq!(merged.left, 0.0, "Merged left should be 0.0");
454
+ assert_eq!(merged.top, 0.0, "Merged top should be 0.0");
455
+ assert_eq!(merged.right, 130.0, "Merged right should be 130.0");
456
+ assert_eq!(merged.bottom, 130.0, "Merged bottom should be 130.0");
457
+ }
458
+
459
+ /// Test relaxed_iou() method with expansion factor.
460
+ #[test]
461
+ fn test_relaxed_iou_method() {
462
+ let bbox1 = create_bbox(0.0, 0.0, 10.0, 10.0);
463
+ let bbox2 = create_bbox(15.0, 15.0, 25.0, 25.0);
464
+
465
+ // Without relaxation, IOU should be 0
466
+ let normal_iou = bbox1.iou(&bbox2);
467
+ assert!(normal_iou < 0.01, "Non-overlapping boxes should have near-zero IOU");
468
+
469
+ // With relaxation, IOU should increase
470
+ let relaxed_iou = bbox1.relaxed_iou(&bbox2, 0.5);
471
+ assert!(
472
+ relaxed_iou > normal_iou,
473
+ "Relaxed IOU should be greater than normal IOU"
474
+ );
475
+ }