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 +1,1220 @@
1
+ //! Server configuration for the Kreuzberg API.
2
+ //!
3
+ //! This module provides the `ServerConfig` struct for managing API server settings
4
+ //! including host, port, CORS, and upload size limits. Configuration can be loaded
5
+ //! from TOML, YAML, or JSON files and can be overridden by environment variables.
6
+ //!
7
+ //! # Features
8
+ //!
9
+ //! - **Multi-format support**: Load configuration from TOML, YAML, or JSON files
10
+ //! - **Environment overrides**: All settings can be overridden via environment variables
11
+ //! - **Backward compatibility**: Supports legacy `max_upload_mb` field for smooth migrations
12
+ //! - **Sensible defaults**: All fields have reasonable defaults matching current behavior
13
+ //! - **Flexible CORS**: Support for all origins (default) or specific origin lists
14
+ //!
15
+ //! # Example
16
+ //!
17
+ //! ```rust,no_run
18
+ //! use kreuzberg::core::ServerConfig;
19
+ //!
20
+ //! # fn example() -> kreuzberg::Result<()> {
21
+ //! // Create with defaults
22
+ //! let mut config = ServerConfig::default();
23
+ //!
24
+ //! // Or load from file
25
+ //! let mut config = ServerConfig::from_file("kreuzberg.toml")?;
26
+ //!
27
+ //! // Apply environment variable overrides
28
+ //! config.apply_env_overrides()?;
29
+ //!
30
+ //! # Ok(())
31
+ //! # }
32
+ //! ```
33
+
34
+ use crate::{KreuzbergError, Result};
35
+ use serde::{Deserialize, Serialize};
36
+ use std::path::Path;
37
+
38
+ /// Default host address for API server
39
+ const DEFAULT_HOST: &str = "127.0.0.1";
40
+
41
+ /// Default port for API server
42
+ const DEFAULT_PORT: u16 = 8000;
43
+
44
+ /// Default maximum request body size: 100 MB
45
+ const DEFAULT_MAX_REQUEST_BODY_BYTES: usize = 104_857_600;
46
+
47
+ /// Default maximum multipart field size: 100 MB
48
+ const DEFAULT_MAX_MULTIPART_FIELD_BYTES: usize = 104_857_600;
49
+
50
+ /// API server configuration.
51
+ ///
52
+ /// This struct holds all configuration options for the Kreuzberg API server,
53
+ /// including host/port settings, CORS configuration, and upload limits.
54
+ ///
55
+ /// # Defaults
56
+ ///
57
+ /// - `host`: "127.0.0.1" (localhost only)
58
+ /// - `port`: 8000
59
+ /// - `cors_origins`: empty vector (allows all origins)
60
+ /// - `max_request_body_bytes`: 104_857_600 (100 MB)
61
+ /// - `max_multipart_field_bytes`: 104_857_600 (100 MB)
62
+ /// - `max_upload_mb`: None (legacy field, not used if other fields set)
63
+ #[derive(Debug, Clone, Serialize, Deserialize)]
64
+ #[serde(default)]
65
+ pub struct ServerConfig {
66
+ /// Server host address (e.g., "127.0.0.1", "0.0.0.0")
67
+ #[serde(default = "default_host")]
68
+ pub host: String,
69
+
70
+ /// Server port number
71
+ #[serde(default = "default_port")]
72
+ pub port: u16,
73
+
74
+ /// CORS allowed origins. Empty vector means allow all origins.
75
+ ///
76
+ /// If this is an empty vector, the server will accept requests from any origin.
77
+ /// If populated with specific origins (e.g., ["https://example.com"]), only
78
+ /// those origins will be allowed.
79
+ #[serde(default)]
80
+ pub cors_origins: Vec<String>,
81
+
82
+ /// Maximum size of request body in bytes (default: 100 MB)
83
+ #[serde(default = "default_max_request_body_bytes")]
84
+ pub max_request_body_bytes: usize,
85
+
86
+ /// Maximum size of multipart fields in bytes (default: 100 MB)
87
+ #[serde(default = "default_max_multipart_field_bytes")]
88
+ pub max_multipart_field_bytes: usize,
89
+
90
+ /// Legacy upload size limit in MB (for backward compatibility).
91
+ ///
92
+ /// This field is deprecated and only used for backward compatibility.
93
+ /// If set, it will override `max_multipart_field_bytes` during normalization.
94
+ /// New configurations should use `max_multipart_field_bytes` directly.
95
+ #[serde(skip_serializing_if = "Option::is_none")]
96
+ pub max_upload_mb: Option<usize>,
97
+ }
98
+
99
+ impl Default for ServerConfig {
100
+ fn default() -> Self {
101
+ Self {
102
+ host: default_host(),
103
+ port: default_port(),
104
+ cors_origins: Vec::new(),
105
+ max_request_body_bytes: default_max_request_body_bytes(),
106
+ max_multipart_field_bytes: default_max_multipart_field_bytes(),
107
+ max_upload_mb: None,
108
+ }
109
+ }
110
+ }
111
+
112
+ // Default value functions for serde
113
+ fn default_host() -> String {
114
+ DEFAULT_HOST.to_string()
115
+ }
116
+
117
+ fn default_port() -> u16 {
118
+ DEFAULT_PORT
119
+ }
120
+
121
+ fn default_max_request_body_bytes() -> usize {
122
+ DEFAULT_MAX_REQUEST_BODY_BYTES
123
+ }
124
+
125
+ fn default_max_multipart_field_bytes() -> usize {
126
+ DEFAULT_MAX_MULTIPART_FIELD_BYTES
127
+ }
128
+
129
+ impl ServerConfig {
130
+ /// Create a new `ServerConfig` with default values.
131
+ pub fn new() -> Self {
132
+ Self::default()
133
+ }
134
+
135
+ /// Get the server listen address (host:port).
136
+ ///
137
+ /// # Example
138
+ ///
139
+ /// ```rust
140
+ /// use kreuzberg::core::ServerConfig;
141
+ ///
142
+ /// let config = ServerConfig::default();
143
+ /// assert_eq!(config.listen_addr(), "127.0.0.1:8000");
144
+ /// ```
145
+ pub fn listen_addr(&self) -> String {
146
+ format!("{}:{}", self.host, self.port)
147
+ }
148
+
149
+ /// Check if CORS allows all origins.
150
+ ///
151
+ /// Returns `true` if the `cors_origins` vector is empty, meaning all origins
152
+ /// are allowed. Returns `false` if specific origins are configured.
153
+ ///
154
+ /// # Example
155
+ ///
156
+ /// ```rust
157
+ /// use kreuzberg::core::ServerConfig;
158
+ ///
159
+ /// let mut config = ServerConfig::default();
160
+ /// assert!(config.cors_allows_all());
161
+ ///
162
+ /// config.cors_origins.push("https://example.com".to_string());
163
+ /// assert!(!config.cors_allows_all());
164
+ /// ```
165
+ pub fn cors_allows_all(&self) -> bool {
166
+ self.cors_origins.is_empty()
167
+ }
168
+
169
+ /// Check if a given origin is allowed by CORS configuration.
170
+ ///
171
+ /// Returns `true` if:
172
+ /// - CORS allows all origins (empty origins list), or
173
+ /// - The given origin is in the allowed origins list
174
+ ///
175
+ /// # Arguments
176
+ ///
177
+ /// * `origin` - The origin to check (e.g., "https://example.com")
178
+ ///
179
+ /// # Example
180
+ ///
181
+ /// ```rust
182
+ /// use kreuzberg::core::ServerConfig;
183
+ ///
184
+ /// let mut config = ServerConfig::default();
185
+ /// assert!(config.is_origin_allowed("https://example.com"));
186
+ ///
187
+ /// config.cors_origins.push("https://allowed.com".to_string());
188
+ /// assert!(config.is_origin_allowed("https://allowed.com"));
189
+ /// assert!(!config.is_origin_allowed("https://denied.com"));
190
+ /// ```
191
+ pub fn is_origin_allowed(&self, origin: &str) -> bool {
192
+ self.cors_origins.is_empty() || self.cors_origins.contains(&origin.to_string())
193
+ }
194
+
195
+ /// Get maximum request body size in megabytes (rounded up).
196
+ ///
197
+ /// # Example
198
+ ///
199
+ /// ```rust
200
+ /// use kreuzberg::core::ServerConfig;
201
+ ///
202
+ /// let mut config = ServerConfig::default();
203
+ /// assert_eq!(config.max_request_body_mb(), 100);
204
+ /// ```
205
+ pub fn max_request_body_mb(&self) -> usize {
206
+ self.max_request_body_bytes.div_ceil(1_048_576)
207
+ }
208
+
209
+ /// Get maximum multipart field size in megabytes (rounded up).
210
+ ///
211
+ /// # Example
212
+ ///
213
+ /// ```rust
214
+ /// use kreuzberg::core::ServerConfig;
215
+ ///
216
+ /// let mut config = ServerConfig::default();
217
+ /// assert_eq!(config.max_multipart_field_mb(), 100);
218
+ /// ```
219
+ pub fn max_multipart_field_mb(&self) -> usize {
220
+ self.max_multipart_field_bytes.div_ceil(1_048_576)
221
+ }
222
+
223
+ /// Normalize legacy field values for backward compatibility.
224
+ ///
225
+ /// If `max_upload_mb` is set, it will be converted to bytes and used to
226
+ /// override `max_multipart_field_bytes`. This allows old configurations
227
+ /// using the legacy field to continue working.
228
+ ///
229
+ /// This method is automatically called by `apply_env_overrides()`.
230
+ pub fn normalize_legacy_fields(&mut self) {
231
+ if let Some(max_upload_mb) = self.max_upload_mb {
232
+ // Convert MB to bytes
233
+ let max_bytes = max_upload_mb.saturating_mul(1_048_576);
234
+ self.max_multipart_field_bytes = max_bytes;
235
+ }
236
+ }
237
+
238
+ /// Apply environment variable overrides to the configuration.
239
+ ///
240
+ /// Reads the following environment variables and overrides config values if set:
241
+ ///
242
+ /// - `KREUZBERG_HOST` - Server host address
243
+ /// - `KREUZBERG_PORT` - Server port number (parsed as u16)
244
+ /// - `KREUZBERG_CORS_ORIGINS` - Comma-separated list of allowed origins
245
+ /// - `KREUZBERG_MAX_REQUEST_BODY_BYTES` - Max request body size in bytes
246
+ /// - `KREUZBERG_MAX_MULTIPART_FIELD_BYTES` - Max multipart field size in bytes
247
+ /// - `KREUZBERG_MAX_UPLOAD_SIZE_MB` - Max upload size in MB (legacy)
248
+ ///
249
+ /// # Errors
250
+ ///
251
+ /// Returns `KreuzbergError::Validation` if:
252
+ /// - `KREUZBERG_PORT` cannot be parsed as u16
253
+ /// - `KREUZBERG_MAX_REQUEST_BODY_BYTES` cannot be parsed as usize
254
+ /// - `KREUZBERG_MAX_MULTIPART_FIELD_BYTES` cannot be parsed as usize
255
+ /// - `KREUZBERG_MAX_UPLOAD_SIZE_MB` cannot be parsed as usize
256
+ ///
257
+ /// # Example
258
+ ///
259
+ /// ```rust,no_run
260
+ /// use kreuzberg::core::ServerConfig;
261
+ ///
262
+ /// # fn example() -> kreuzberg::Result<()> {
263
+ /// std::env::set_var("KREUZBERG_HOST", "0.0.0.0");
264
+ /// std::env::set_var("KREUZBERG_PORT", "3000");
265
+ ///
266
+ /// let mut config = ServerConfig::default();
267
+ /// config.apply_env_overrides()?;
268
+ ///
269
+ /// assert_eq!(config.host, "0.0.0.0");
270
+ /// assert_eq!(config.port, 3000);
271
+ /// # Ok(())
272
+ /// # }
273
+ /// ```
274
+ pub fn apply_env_overrides(&mut self) -> Result<()> {
275
+ // Host override
276
+ if let Ok(host) = std::env::var("KREUZBERG_HOST") {
277
+ self.host = host;
278
+ }
279
+
280
+ // Port override
281
+ if let Ok(port_str) = std::env::var("KREUZBERG_PORT") {
282
+ self.port = port_str.parse::<u16>().map_err(|e| {
283
+ KreuzbergError::validation(format!(
284
+ "KREUZBERG_PORT must be a valid u16 number, got '{}': {}",
285
+ port_str, e
286
+ ))
287
+ })?;
288
+ }
289
+
290
+ // CORS origins override (comma-separated)
291
+ if let Ok(origins_str) = std::env::var("KREUZBERG_CORS_ORIGINS") {
292
+ self.cors_origins = origins_str
293
+ .split(',')
294
+ .map(|s| s.trim().to_string())
295
+ .filter(|s| !s.is_empty())
296
+ .collect();
297
+ }
298
+
299
+ // Max request body bytes override
300
+ if let Ok(bytes_str) = std::env::var("KREUZBERG_MAX_REQUEST_BODY_BYTES") {
301
+ self.max_request_body_bytes = bytes_str.parse::<usize>().map_err(|e| {
302
+ KreuzbergError::validation(format!(
303
+ "KREUZBERG_MAX_REQUEST_BODY_BYTES must be a valid usize, got '{}': {}",
304
+ bytes_str, e
305
+ ))
306
+ })?;
307
+ }
308
+
309
+ // Max multipart field bytes override
310
+ if let Ok(bytes_str) = std::env::var("KREUZBERG_MAX_MULTIPART_FIELD_BYTES") {
311
+ self.max_multipart_field_bytes = bytes_str.parse::<usize>().map_err(|e| {
312
+ KreuzbergError::validation(format!(
313
+ "KREUZBERG_MAX_MULTIPART_FIELD_BYTES must be a valid usize, got '{}': {}",
314
+ bytes_str, e
315
+ ))
316
+ })?;
317
+ }
318
+
319
+ // Legacy max upload size override (in MB)
320
+ if let Ok(mb_str) = std::env::var("KREUZBERG_MAX_UPLOAD_SIZE_MB") {
321
+ let mb = mb_str.parse::<usize>().map_err(|e| {
322
+ KreuzbergError::validation(format!(
323
+ "KREUZBERG_MAX_UPLOAD_SIZE_MB must be a valid usize, got '{}': {}",
324
+ mb_str, e
325
+ ))
326
+ })?;
327
+ self.max_upload_mb = Some(mb);
328
+ }
329
+
330
+ // Apply legacy field normalization
331
+ self.normalize_legacy_fields();
332
+
333
+ Ok(())
334
+ }
335
+
336
+ /// Load server configuration from a file.
337
+ ///
338
+ /// Automatically detects the file format based on extension:
339
+ /// - `.toml` - TOML format
340
+ /// - `.yaml` or `.yml` - YAML format
341
+ /// - `.json` - JSON format
342
+ ///
343
+ /// This function handles two config file formats:
344
+ /// 1. Flat format: Server config at root level
345
+ /// 2. Nested format: Server config under `[server]` section (combined with ExtractionConfig)
346
+ ///
347
+ /// # Arguments
348
+ ///
349
+ /// * `path` - Path to the configuration file
350
+ ///
351
+ /// # Errors
352
+ ///
353
+ /// Returns `KreuzbergError::Validation` if:
354
+ /// - File doesn't exist or cannot be read
355
+ /// - File extension is not recognized
356
+ /// - File content is invalid for the detected format
357
+ ///
358
+ /// # Example
359
+ ///
360
+ /// ```rust,no_run
361
+ /// use kreuzberg::core::ServerConfig;
362
+ ///
363
+ /// # fn example() -> kreuzberg::Result<()> {
364
+ /// let config = ServerConfig::from_file("kreuzberg.toml")?;
365
+ /// # Ok(())
366
+ /// # }
367
+ /// ```
368
+ pub fn from_file(path: impl AsRef<Path>) -> Result<Self> {
369
+ let path = path.as_ref();
370
+
371
+ let content = std::fs::read_to_string(path)
372
+ .map_err(|e| KreuzbergError::validation(format!("Failed to read config file {}: {}", path.display(), e)))?;
373
+
374
+ let extension = path.extension().and_then(|ext| ext.to_str()).ok_or_else(|| {
375
+ KreuzbergError::validation(format!(
376
+ "Cannot determine file format: no extension found in {}",
377
+ path.display()
378
+ ))
379
+ })?;
380
+
381
+ let mut config = match extension.to_lowercase().as_str() {
382
+ "toml" => {
383
+ // Try nested format first (with [server] section)
384
+ #[derive(Deserialize)]
385
+ struct RootConfig {
386
+ #[serde(default)]
387
+ server: Option<ServerConfig>,
388
+ }
389
+
390
+ if let Ok(root) = toml::from_str::<RootConfig>(&content) {
391
+ if let Some(server) = root.server {
392
+ server
393
+ } else {
394
+ // No [server] section, try flat format
395
+ toml::from_str::<Self>(&content).map_err(|e| {
396
+ KreuzbergError::validation(format!("Invalid TOML in {}: {}", path.display(), e))
397
+ })?
398
+ }
399
+ } else {
400
+ // Fall back to flat format
401
+ toml::from_str::<Self>(&content)
402
+ .map_err(|e| KreuzbergError::validation(format!("Invalid TOML in {}: {}", path.display(), e)))?
403
+ }
404
+ }
405
+ "yaml" | "yml" => {
406
+ // Try nested format first (with server: section)
407
+ #[derive(Deserialize)]
408
+ struct RootConfig {
409
+ #[serde(default)]
410
+ server: Option<ServerConfig>,
411
+ }
412
+
413
+ if let Ok(root) = serde_yaml_ng::from_str::<RootConfig>(&content) {
414
+ if let Some(server) = root.server {
415
+ server
416
+ } else {
417
+ // No server section, try flat format
418
+ serde_yaml_ng::from_str::<Self>(&content).map_err(|e| {
419
+ KreuzbergError::validation(format!("Invalid YAML in {}: {}", path.display(), e))
420
+ })?
421
+ }
422
+ } else {
423
+ // Fall back to flat format
424
+ serde_yaml_ng::from_str::<Self>(&content)
425
+ .map_err(|e| KreuzbergError::validation(format!("Invalid YAML in {}: {}", path.display(), e)))?
426
+ }
427
+ }
428
+ "json" => {
429
+ // Try nested format first (with "server" key)
430
+ #[derive(Deserialize)]
431
+ struct RootConfig {
432
+ #[serde(default)]
433
+ server: Option<ServerConfig>,
434
+ }
435
+
436
+ if let Ok(root) = serde_json::from_str::<RootConfig>(&content) {
437
+ if let Some(server) = root.server {
438
+ server
439
+ } else {
440
+ // No server key, try flat format
441
+ serde_json::from_str::<Self>(&content).map_err(|e| {
442
+ KreuzbergError::validation(format!("Invalid JSON in {}: {}", path.display(), e))
443
+ })?
444
+ }
445
+ } else {
446
+ // Fall back to flat format
447
+ serde_json::from_str::<Self>(&content)
448
+ .map_err(|e| KreuzbergError::validation(format!("Invalid JSON in {}: {}", path.display(), e)))?
449
+ }
450
+ }
451
+ _ => {
452
+ return Err(KreuzbergError::validation(format!(
453
+ "Unsupported config file format: .{}. Supported formats: .toml, .yaml, .yml, .json",
454
+ extension
455
+ )));
456
+ }
457
+ };
458
+
459
+ // Normalize legacy fields
460
+ config.normalize_legacy_fields();
461
+
462
+ Ok(config)
463
+ }
464
+
465
+ /// Load server configuration from a TOML file.
466
+ ///
467
+ /// # Arguments
468
+ ///
469
+ /// * `path` - Path to the TOML file
470
+ ///
471
+ /// # Errors
472
+ ///
473
+ /// Returns `KreuzbergError::Validation` if the file doesn't exist or is invalid TOML.
474
+ ///
475
+ /// # Example
476
+ ///
477
+ /// ```rust,no_run
478
+ /// use kreuzberg::core::ServerConfig;
479
+ ///
480
+ /// # fn example() -> kreuzberg::Result<()> {
481
+ /// let config = ServerConfig::from_toml_file("kreuzberg.toml")?;
482
+ /// # Ok(())
483
+ /// # }
484
+ /// ```
485
+ pub fn from_toml_file(path: impl AsRef<Path>) -> Result<Self> {
486
+ let path = path.as_ref();
487
+
488
+ let content = std::fs::read_to_string(path)
489
+ .map_err(|e| KreuzbergError::validation(format!("Failed to read config file {}: {}", path.display(), e)))?;
490
+
491
+ let mut config: Self = toml::from_str(&content)
492
+ .map_err(|e| KreuzbergError::validation(format!("Invalid TOML in {}: {}", path.display(), e)))?;
493
+
494
+ config.normalize_legacy_fields();
495
+
496
+ Ok(config)
497
+ }
498
+
499
+ /// Load server configuration from a YAML file.
500
+ ///
501
+ /// # Arguments
502
+ ///
503
+ /// * `path` - Path to the YAML file
504
+ ///
505
+ /// # Errors
506
+ ///
507
+ /// Returns `KreuzbergError::Validation` if the file doesn't exist or is invalid YAML.
508
+ pub fn from_yaml_file(path: impl AsRef<Path>) -> Result<Self> {
509
+ let path = path.as_ref();
510
+
511
+ let content = std::fs::read_to_string(path)
512
+ .map_err(|e| KreuzbergError::validation(format!("Failed to read config file {}: {}", path.display(), e)))?;
513
+
514
+ let mut config: Self = serde_yaml_ng::from_str(&content)
515
+ .map_err(|e| KreuzbergError::validation(format!("Invalid YAML in {}: {}", path.display(), e)))?;
516
+
517
+ config.normalize_legacy_fields();
518
+
519
+ Ok(config)
520
+ }
521
+
522
+ /// Load server configuration from a JSON file.
523
+ ///
524
+ /// # Arguments
525
+ ///
526
+ /// * `path` - Path to the JSON file
527
+ ///
528
+ /// # Errors
529
+ ///
530
+ /// Returns `KreuzbergError::Validation` if the file doesn't exist or is invalid JSON.
531
+ pub fn from_json_file(path: impl AsRef<Path>) -> Result<Self> {
532
+ let path = path.as_ref();
533
+
534
+ let content = std::fs::read_to_string(path)
535
+ .map_err(|e| KreuzbergError::validation(format!("Failed to read config file {}: {}", path.display(), e)))?;
536
+
537
+ let mut config: Self = serde_json::from_str(&content)
538
+ .map_err(|e| KreuzbergError::validation(format!("Invalid JSON in {}: {}", path.display(), e)))?;
539
+
540
+ config.normalize_legacy_fields();
541
+
542
+ Ok(config)
543
+ }
544
+ }
545
+
546
+ #[cfg(test)]
547
+ #[allow(unsafe_code)]
548
+ mod tests {
549
+ use super::*;
550
+ use std::fs;
551
+ use tempfile::tempdir;
552
+
553
+ #[test]
554
+ fn test_default_config() {
555
+ let config = ServerConfig::default();
556
+ assert_eq!(config.host, "127.0.0.1");
557
+ assert_eq!(config.port, 8000);
558
+ assert!(config.cors_origins.is_empty());
559
+ assert_eq!(config.max_request_body_bytes, 104_857_600);
560
+ assert_eq!(config.max_multipart_field_bytes, 104_857_600);
561
+ assert!(config.max_upload_mb.is_none());
562
+ }
563
+
564
+ #[test]
565
+ fn test_listen_addr() {
566
+ let config = ServerConfig::default();
567
+ assert_eq!(config.listen_addr(), "127.0.0.1:8000");
568
+ }
569
+
570
+ #[test]
571
+ fn test_listen_addr_custom() {
572
+ let config = ServerConfig {
573
+ host: "0.0.0.0".to_string(),
574
+ port: 3000,
575
+ ..Default::default()
576
+ };
577
+ assert_eq!(config.listen_addr(), "0.0.0.0:3000");
578
+ }
579
+
580
+ #[test]
581
+ fn test_cors_allows_all() {
582
+ let mut config = ServerConfig::default();
583
+ assert!(config.cors_allows_all());
584
+
585
+ config.cors_origins.push("https://example.com".to_string());
586
+ assert!(!config.cors_allows_all());
587
+ }
588
+
589
+ #[test]
590
+ fn test_is_origin_allowed_all() {
591
+ let config = ServerConfig::default();
592
+ assert!(config.is_origin_allowed("https://example.com"));
593
+ assert!(config.is_origin_allowed("https://other.com"));
594
+ }
595
+
596
+ #[test]
597
+ fn test_is_origin_allowed_specific() {
598
+ let mut config = ServerConfig::default();
599
+ config.cors_origins.push("https://allowed.com".to_string());
600
+
601
+ assert!(config.is_origin_allowed("https://allowed.com"));
602
+ assert!(!config.is_origin_allowed("https://denied.com"));
603
+ }
604
+
605
+ #[test]
606
+ fn test_max_request_body_mb() {
607
+ let config = ServerConfig::default();
608
+ assert_eq!(config.max_request_body_mb(), 100);
609
+ }
610
+
611
+ #[test]
612
+ fn test_max_multipart_field_mb() {
613
+ let config = ServerConfig::default();
614
+ assert_eq!(config.max_multipart_field_mb(), 100);
615
+ }
616
+
617
+ #[test]
618
+ fn test_max_bytes_to_mb_rounding() {
619
+ let mut config = ServerConfig {
620
+ max_request_body_bytes: 1_048_576, // 1 MB
621
+ ..Default::default()
622
+ };
623
+ assert_eq!(config.max_request_body_mb(), 1);
624
+
625
+ config.max_request_body_bytes = 1_048_577; // 1 MB + 1 byte
626
+ assert_eq!(config.max_request_body_mb(), 2); // Rounds up
627
+ }
628
+
629
+ #[test]
630
+ fn test_normalize_legacy_max_upload_mb() {
631
+ let mut config = ServerConfig {
632
+ max_upload_mb: Some(50),
633
+ ..Default::default()
634
+ };
635
+
636
+ config.normalize_legacy_fields();
637
+
638
+ assert_eq!(config.max_multipart_field_bytes, 50 * 1_048_576);
639
+ }
640
+
641
+ #[test]
642
+ fn test_normalize_legacy_max_upload_mb_zero() {
643
+ let mut config = ServerConfig {
644
+ max_upload_mb: Some(0),
645
+ ..Default::default()
646
+ };
647
+
648
+ config.normalize_legacy_fields();
649
+
650
+ assert_eq!(config.max_multipart_field_bytes, 0);
651
+ }
652
+
653
+ #[test]
654
+ fn test_from_toml_file() {
655
+ let dir = tempdir().unwrap();
656
+ let config_path = dir.path().join("server.toml");
657
+
658
+ fs::write(
659
+ &config_path,
660
+ r#"
661
+ host = "0.0.0.0"
662
+ port = 3000
663
+ cors_origins = ["https://example.com", "https://other.com"]
664
+ max_request_body_bytes = 50000000
665
+ max_multipart_field_bytes = 75000000
666
+ "#,
667
+ )
668
+ .unwrap();
669
+
670
+ let config = ServerConfig::from_toml_file(&config_path).unwrap();
671
+ assert_eq!(config.host, "0.0.0.0");
672
+ assert_eq!(config.port, 3000);
673
+ assert_eq!(config.cors_origins.len(), 2);
674
+ assert_eq!(config.max_request_body_bytes, 50_000_000);
675
+ assert_eq!(config.max_multipart_field_bytes, 75_000_000);
676
+ }
677
+
678
+ #[test]
679
+ fn test_from_yaml_file() {
680
+ let dir = tempdir().unwrap();
681
+ let config_path = dir.path().join("server.yaml");
682
+
683
+ fs::write(
684
+ &config_path,
685
+ r#"
686
+ host: 0.0.0.0
687
+ port: 3000
688
+ cors_origins:
689
+ - https://example.com
690
+ - https://other.com
691
+ max_request_body_bytes: 50000000
692
+ max_multipart_field_bytes: 75000000
693
+ "#,
694
+ )
695
+ .unwrap();
696
+
697
+ let config = ServerConfig::from_yaml_file(&config_path).unwrap();
698
+ assert_eq!(config.host, "0.0.0.0");
699
+ assert_eq!(config.port, 3000);
700
+ assert_eq!(config.cors_origins.len(), 2);
701
+ assert_eq!(config.max_request_body_bytes, 50_000_000);
702
+ assert_eq!(config.max_multipart_field_bytes, 75_000_000);
703
+ }
704
+
705
+ #[test]
706
+ fn test_from_json_file() {
707
+ let dir = tempdir().unwrap();
708
+ let config_path = dir.path().join("server.json");
709
+
710
+ fs::write(
711
+ &config_path,
712
+ r#"{
713
+ "host": "0.0.0.0",
714
+ "port": 3000,
715
+ "cors_origins": ["https://example.com", "https://other.com"],
716
+ "max_request_body_bytes": 50000000,
717
+ "max_multipart_field_bytes": 75000000
718
+ }
719
+ "#,
720
+ )
721
+ .unwrap();
722
+
723
+ let config = ServerConfig::from_json_file(&config_path).unwrap();
724
+ assert_eq!(config.host, "0.0.0.0");
725
+ assert_eq!(config.port, 3000);
726
+ assert_eq!(config.cors_origins.len(), 2);
727
+ assert_eq!(config.max_request_body_bytes, 50_000_000);
728
+ assert_eq!(config.max_multipart_field_bytes, 75_000_000);
729
+ }
730
+
731
+ #[test]
732
+ fn test_from_file_auto_detects_toml() {
733
+ let dir = tempdir().unwrap();
734
+ let config_path = dir.path().join("server.toml");
735
+
736
+ fs::write(
737
+ &config_path,
738
+ r#"
739
+ host = "0.0.0.0"
740
+ port = 3000
741
+ "#,
742
+ )
743
+ .unwrap();
744
+
745
+ let config = ServerConfig::from_file(&config_path).unwrap();
746
+ assert_eq!(config.host, "0.0.0.0");
747
+ assert_eq!(config.port, 3000);
748
+ }
749
+
750
+ #[test]
751
+ fn test_from_file_auto_detects_yaml() {
752
+ let dir = tempdir().unwrap();
753
+ let config_path = dir.path().join("server.yaml");
754
+
755
+ fs::write(
756
+ &config_path,
757
+ r#"
758
+ host: 0.0.0.0
759
+ port: 3000
760
+ "#,
761
+ )
762
+ .unwrap();
763
+
764
+ let config = ServerConfig::from_file(&config_path).unwrap();
765
+ assert_eq!(config.host, "0.0.0.0");
766
+ assert_eq!(config.port, 3000);
767
+ }
768
+
769
+ #[test]
770
+ fn test_from_file_auto_detects_json() {
771
+ let dir = tempdir().unwrap();
772
+ let config_path = dir.path().join("server.json");
773
+
774
+ fs::write(&config_path, r#"{"host": "0.0.0.0", "port": 3000}"#).unwrap();
775
+
776
+ let config = ServerConfig::from_file(&config_path).unwrap();
777
+ assert_eq!(config.host, "0.0.0.0");
778
+ assert_eq!(config.port, 3000);
779
+ }
780
+
781
+ #[test]
782
+ fn test_from_file_unsupported_extension() {
783
+ let dir = tempdir().unwrap();
784
+ let config_path = dir.path().join("server.txt");
785
+
786
+ fs::write(&config_path, "host = 0.0.0.0").unwrap();
787
+
788
+ let result = ServerConfig::from_file(&config_path);
789
+ assert!(result.is_err());
790
+ assert!(
791
+ result
792
+ .unwrap_err()
793
+ .to_string()
794
+ .contains("Unsupported config file format")
795
+ );
796
+ }
797
+
798
+ #[test]
799
+ fn test_from_file_no_extension() {
800
+ let dir = tempdir().unwrap();
801
+ let config_path = dir.path().join("server");
802
+
803
+ fs::write(&config_path, "host = 0.0.0.0").unwrap();
804
+
805
+ let result = ServerConfig::from_file(&config_path);
806
+ assert!(result.is_err());
807
+ assert!(result.unwrap_err().to_string().contains("no extension found"));
808
+ }
809
+
810
+ #[test]
811
+ fn test_legacy_max_upload_mb_in_file() {
812
+ let dir = tempdir().unwrap();
813
+ let config_path = dir.path().join("server.toml");
814
+
815
+ fs::write(
816
+ &config_path,
817
+ r#"
818
+ host = "127.0.0.1"
819
+ port = 8000
820
+ max_upload_mb = 50
821
+ "#,
822
+ )
823
+ .unwrap();
824
+
825
+ let config = ServerConfig::from_toml_file(&config_path).unwrap();
826
+ assert_eq!(config.max_upload_mb, Some(50));
827
+ assert_eq!(config.max_multipart_field_bytes, 50 * 1_048_576);
828
+ }
829
+
830
+ #[serial_test::serial]
831
+ #[test]
832
+ fn test_apply_env_host_override() {
833
+ let original = std::env::var("KREUZBERG_HOST").ok();
834
+ unsafe {
835
+ std::env::set_var("KREUZBERG_HOST", "192.168.1.1");
836
+ }
837
+
838
+ let mut config = ServerConfig::default();
839
+ config.apply_env_overrides().unwrap();
840
+
841
+ assert_eq!(config.host, "192.168.1.1");
842
+
843
+ // Cleanup
844
+ unsafe {
845
+ if let Some(orig) = original {
846
+ std::env::set_var("KREUZBERG_HOST", orig);
847
+ } else {
848
+ std::env::remove_var("KREUZBERG_HOST");
849
+ }
850
+ }
851
+ }
852
+
853
+ #[serial_test::serial]
854
+ #[test]
855
+ fn test_apply_env_port_override() {
856
+ let original = std::env::var("KREUZBERG_PORT").ok();
857
+ unsafe {
858
+ std::env::set_var("KREUZBERG_PORT", "5000");
859
+ }
860
+
861
+ let mut config = ServerConfig::default();
862
+ config.apply_env_overrides().unwrap();
863
+
864
+ assert_eq!(config.port, 5000);
865
+
866
+ // Cleanup
867
+ unsafe {
868
+ if let Some(orig) = original {
869
+ std::env::set_var("KREUZBERG_PORT", orig);
870
+ } else {
871
+ std::env::remove_var("KREUZBERG_PORT");
872
+ }
873
+ }
874
+ }
875
+
876
+ #[serial_test::serial]
877
+ #[test]
878
+ fn test_apply_env_port_invalid() {
879
+ let original = std::env::var("KREUZBERG_PORT").ok();
880
+ unsafe {
881
+ std::env::set_var("KREUZBERG_PORT", "not_a_number");
882
+ }
883
+
884
+ let mut config = ServerConfig::default();
885
+ let result = config.apply_env_overrides();
886
+
887
+ assert!(result.is_err());
888
+ assert!(
889
+ result
890
+ .unwrap_err()
891
+ .to_string()
892
+ .contains("KREUZBERG_PORT must be a valid u16")
893
+ );
894
+
895
+ // Cleanup
896
+ unsafe {
897
+ if let Some(orig) = original {
898
+ std::env::set_var("KREUZBERG_PORT", orig);
899
+ } else {
900
+ std::env::remove_var("KREUZBERG_PORT");
901
+ }
902
+ }
903
+ }
904
+
905
+ #[serial_test::serial]
906
+ #[test]
907
+ fn test_apply_env_cors_origins_override() {
908
+ let original = std::env::var("KREUZBERG_CORS_ORIGINS").ok();
909
+ unsafe {
910
+ std::env::set_var("KREUZBERG_CORS_ORIGINS", "https://example.com, https://other.com");
911
+ }
912
+
913
+ let mut config = ServerConfig::default();
914
+ config.apply_env_overrides().unwrap();
915
+
916
+ assert_eq!(config.cors_origins.len(), 2);
917
+ assert!(config.cors_origins.contains(&"https://example.com".to_string()));
918
+ assert!(config.cors_origins.contains(&"https://other.com".to_string()));
919
+
920
+ // Cleanup
921
+ unsafe {
922
+ if let Some(orig) = original {
923
+ std::env::set_var("KREUZBERG_CORS_ORIGINS", orig);
924
+ } else {
925
+ std::env::remove_var("KREUZBERG_CORS_ORIGINS");
926
+ }
927
+ }
928
+ }
929
+
930
+ #[serial_test::serial]
931
+ #[test]
932
+ fn test_apply_env_max_request_body_bytes_override() {
933
+ let original = std::env::var("KREUZBERG_MAX_REQUEST_BODY_BYTES").ok();
934
+ unsafe {
935
+ std::env::set_var("KREUZBERG_MAX_REQUEST_BODY_BYTES", "52428800");
936
+ }
937
+
938
+ let mut config = ServerConfig::default();
939
+ config.apply_env_overrides().unwrap();
940
+
941
+ assert_eq!(config.max_request_body_bytes, 52_428_800);
942
+
943
+ // Cleanup
944
+ unsafe {
945
+ if let Some(orig) = original {
946
+ std::env::set_var("KREUZBERG_MAX_REQUEST_BODY_BYTES", orig);
947
+ } else {
948
+ std::env::remove_var("KREUZBERG_MAX_REQUEST_BODY_BYTES");
949
+ }
950
+ }
951
+ }
952
+
953
+ #[serial_test::serial]
954
+ #[test]
955
+ fn test_apply_env_max_multipart_field_bytes_override() {
956
+ let original = std::env::var("KREUZBERG_MAX_MULTIPART_FIELD_BYTES").ok();
957
+ unsafe {
958
+ std::env::set_var("KREUZBERG_MAX_MULTIPART_FIELD_BYTES", "78643200");
959
+ }
960
+
961
+ let mut config = ServerConfig::default();
962
+ config.apply_env_overrides().unwrap();
963
+
964
+ assert_eq!(config.max_multipart_field_bytes, 78_643_200);
965
+
966
+ // Cleanup
967
+ unsafe {
968
+ if let Some(orig) = original {
969
+ std::env::set_var("KREUZBERG_MAX_MULTIPART_FIELD_BYTES", orig);
970
+ } else {
971
+ std::env::remove_var("KREUZBERG_MAX_MULTIPART_FIELD_BYTES");
972
+ }
973
+ }
974
+ }
975
+
976
+ #[serial_test::serial]
977
+ #[test]
978
+ fn test_apply_env_legacy_max_upload_size_mb_override() {
979
+ let original = std::env::var("KREUZBERG_MAX_UPLOAD_SIZE_MB").ok();
980
+ unsafe {
981
+ std::env::set_var("KREUZBERG_MAX_UPLOAD_SIZE_MB", "75");
982
+ }
983
+
984
+ let mut config = ServerConfig::default();
985
+ config.apply_env_overrides().unwrap();
986
+
987
+ assert_eq!(config.max_upload_mb, Some(75));
988
+ assert_eq!(config.max_multipart_field_bytes, 75 * 1_048_576);
989
+
990
+ // Cleanup
991
+ unsafe {
992
+ if let Some(orig) = original {
993
+ std::env::set_var("KREUZBERG_MAX_UPLOAD_SIZE_MB", orig);
994
+ } else {
995
+ std::env::remove_var("KREUZBERG_MAX_UPLOAD_SIZE_MB");
996
+ }
997
+ }
998
+ }
999
+
1000
+ #[serial_test::serial]
1001
+ #[test]
1002
+ fn test_apply_env_multiple_overrides() {
1003
+ let host_orig = std::env::var("KREUZBERG_HOST").ok();
1004
+ let port_orig = std::env::var("KREUZBERG_PORT").ok();
1005
+ let cors_orig = std::env::var("KREUZBERG_CORS_ORIGINS").ok();
1006
+
1007
+ unsafe {
1008
+ std::env::set_var("KREUZBERG_HOST", "0.0.0.0");
1009
+ std::env::set_var("KREUZBERG_PORT", "4000");
1010
+ std::env::set_var("KREUZBERG_CORS_ORIGINS", "https://api.example.com");
1011
+ }
1012
+
1013
+ let mut config = ServerConfig::default();
1014
+ config.apply_env_overrides().unwrap();
1015
+
1016
+ assert_eq!(config.host, "0.0.0.0");
1017
+ assert_eq!(config.port, 4000);
1018
+ assert_eq!(config.cors_origins.len(), 1);
1019
+ assert_eq!(config.cors_origins[0], "https://api.example.com");
1020
+
1021
+ // Cleanup
1022
+ unsafe {
1023
+ if let Some(orig) = host_orig {
1024
+ std::env::set_var("KREUZBERG_HOST", orig);
1025
+ } else {
1026
+ std::env::remove_var("KREUZBERG_HOST");
1027
+ }
1028
+ if let Some(orig) = port_orig {
1029
+ std::env::set_var("KREUZBERG_PORT", orig);
1030
+ } else {
1031
+ std::env::remove_var("KREUZBERG_PORT");
1032
+ }
1033
+ if let Some(orig) = cors_orig {
1034
+ std::env::set_var("KREUZBERG_CORS_ORIGINS", orig);
1035
+ } else {
1036
+ std::env::remove_var("KREUZBERG_CORS_ORIGINS");
1037
+ }
1038
+ }
1039
+ }
1040
+
1041
+ #[test]
1042
+ fn test_serde_default_serialization() {
1043
+ let config = ServerConfig::default();
1044
+ let json = serde_json::to_string(&config).unwrap();
1045
+
1046
+ // Should serialize without the max_upload_mb field when None
1047
+ assert!(!json.contains("max_upload_mb"));
1048
+ }
1049
+
1050
+ #[test]
1051
+ fn test_serde_with_max_upload_mb_serialization() {
1052
+ let config = ServerConfig {
1053
+ max_upload_mb: Some(50),
1054
+ ..Default::default()
1055
+ };
1056
+ let json = serde_json::to_string(&config).unwrap();
1057
+
1058
+ // Should serialize with max_upload_mb when Some
1059
+ assert!(json.contains("max_upload_mb"));
1060
+ }
1061
+
1062
+ #[test]
1063
+ fn test_cors_origins_empty_in_toml() {
1064
+ let dir = tempdir().unwrap();
1065
+ let config_path = dir.path().join("server.toml");
1066
+
1067
+ fs::write(
1068
+ &config_path,
1069
+ r#"
1070
+ host = "127.0.0.1"
1071
+ port = 8000
1072
+ "#,
1073
+ )
1074
+ .unwrap();
1075
+
1076
+ let config = ServerConfig::from_toml_file(&config_path).unwrap();
1077
+ assert!(config.cors_origins.is_empty());
1078
+ assert!(config.cors_allows_all());
1079
+ }
1080
+
1081
+ #[test]
1082
+ fn test_full_configuration_toml() {
1083
+ let dir = tempdir().unwrap();
1084
+ let config_path = dir.path().join("server.toml");
1085
+
1086
+ fs::write(
1087
+ &config_path,
1088
+ r#"
1089
+ host = "192.168.1.100"
1090
+ port = 9000
1091
+ cors_origins = ["https://app1.com", "https://app2.com", "https://app3.com"]
1092
+ max_request_body_bytes = 200000000
1093
+ max_multipart_field_bytes = 150000000
1094
+ "#,
1095
+ )
1096
+ .unwrap();
1097
+
1098
+ let config = ServerConfig::from_toml_file(&config_path).unwrap();
1099
+ assert_eq!(config.host, "192.168.1.100");
1100
+ assert_eq!(config.port, 9000);
1101
+ assert_eq!(config.listen_addr(), "192.168.1.100:9000");
1102
+ assert_eq!(config.cors_origins.len(), 3);
1103
+ assert!(!config.cors_allows_all());
1104
+ assert!(config.is_origin_allowed("https://app1.com"));
1105
+ assert!(!config.is_origin_allowed("https://app4.com"));
1106
+ assert_eq!(config.max_request_body_bytes, 200_000_000);
1107
+ assert_eq!(config.max_multipart_field_bytes, 150_000_000);
1108
+ assert_eq!(config.max_request_body_mb(), 191);
1109
+ assert_eq!(config.max_multipart_field_mb(), 144);
1110
+ }
1111
+
1112
+ #[test]
1113
+ fn test_from_file_with_nested_server_section_toml() {
1114
+ let dir = tempdir().unwrap();
1115
+ let config_path = dir.path().join("kreuzberg.toml");
1116
+
1117
+ // Config file with [server] section and other sections (like ExtractionConfig)
1118
+ fs::write(
1119
+ &config_path,
1120
+ r#"
1121
+ [server]
1122
+ host = "0.0.0.0"
1123
+ port = 3000
1124
+ cors_origins = ["https://example.com"]
1125
+
1126
+ [ocr]
1127
+ backend = "tesseract"
1128
+ language = "eng"
1129
+
1130
+ [extraction]
1131
+ enabled = true
1132
+ "#,
1133
+ )
1134
+ .unwrap();
1135
+
1136
+ let config = ServerConfig::from_file(&config_path).unwrap();
1137
+ assert_eq!(config.host, "0.0.0.0");
1138
+ assert_eq!(config.port, 3000);
1139
+ assert_eq!(config.cors_origins.len(), 1);
1140
+ assert_eq!(config.cors_origins[0], "https://example.com");
1141
+ }
1142
+
1143
+ #[test]
1144
+ fn test_from_file_with_nested_server_section_yaml() {
1145
+ let dir = tempdir().unwrap();
1146
+ let config_path = dir.path().join("kreuzberg.yaml");
1147
+
1148
+ // Config file with server: section and other sections
1149
+ fs::write(
1150
+ &config_path,
1151
+ r#"
1152
+ server:
1153
+ host: 0.0.0.0
1154
+ port: 4000
1155
+ cors_origins:
1156
+ - https://example.com
1157
+
1158
+ ocr:
1159
+ backend: tesseract
1160
+ language: eng
1161
+ "#,
1162
+ )
1163
+ .unwrap();
1164
+
1165
+ let config = ServerConfig::from_file(&config_path).unwrap();
1166
+ assert_eq!(config.host, "0.0.0.0");
1167
+ assert_eq!(config.port, 4000);
1168
+ assert_eq!(config.cors_origins.len(), 1);
1169
+ }
1170
+
1171
+ #[test]
1172
+ fn test_from_file_with_nested_server_section_json() {
1173
+ let dir = tempdir().unwrap();
1174
+ let config_path = dir.path().join("kreuzberg.json");
1175
+
1176
+ // Config file with "server" key and other sections
1177
+ fs::write(
1178
+ &config_path,
1179
+ r#"
1180
+ {
1181
+ "server": {
1182
+ "host": "0.0.0.0",
1183
+ "port": 5000,
1184
+ "cors_origins": ["https://example.com"]
1185
+ },
1186
+ "ocr": {
1187
+ "backend": "tesseract",
1188
+ "language": "eng"
1189
+ }
1190
+ }
1191
+ "#,
1192
+ )
1193
+ .unwrap();
1194
+
1195
+ let config = ServerConfig::from_file(&config_path).unwrap();
1196
+ assert_eq!(config.host, "0.0.0.0");
1197
+ assert_eq!(config.port, 5000);
1198
+ assert_eq!(config.cors_origins.len(), 1);
1199
+ }
1200
+
1201
+ #[test]
1202
+ fn test_from_file_flat_format_still_works() {
1203
+ let dir = tempdir().unwrap();
1204
+ let config_path = dir.path().join("server.toml");
1205
+
1206
+ // Old flat format without [server] section
1207
+ fs::write(
1208
+ &config_path,
1209
+ r#"
1210
+ host = "192.168.1.1"
1211
+ port = 6000
1212
+ "#,
1213
+ )
1214
+ .unwrap();
1215
+
1216
+ let config = ServerConfig::from_file(&config_path).unwrap();
1217
+ assert_eq!(config.host, "192.168.1.1");
1218
+ assert_eq!(config.port, 6000);
1219
+ }
1220
+ }