kreuzberg 4.0.8 → 4.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/Gemfile.lock +2 -2
- data/README.md +1 -1
- data/ext/kreuzberg_rb/native/Cargo.lock +94 -98
- data/ext/kreuzberg_rb/native/Cargo.toml +4 -2
- data/ext/kreuzberg_rb/native/src/batch.rs +139 -0
- data/ext/kreuzberg_rb/native/src/config/mod.rs +10 -0
- data/ext/kreuzberg_rb/native/src/config/types.rs +1058 -0
- data/ext/kreuzberg_rb/native/src/error_handling.rs +125 -0
- data/ext/kreuzberg_rb/native/src/extraction.rs +79 -0
- data/ext/kreuzberg_rb/native/src/gc_guarded_value.rs +35 -0
- data/ext/kreuzberg_rb/native/src/helpers.rs +176 -0
- data/ext/kreuzberg_rb/native/src/lib.rs +342 -3622
- data/ext/kreuzberg_rb/native/src/metadata.rs +34 -0
- data/ext/kreuzberg_rb/native/src/plugins/mod.rs +92 -0
- data/ext/kreuzberg_rb/native/src/plugins/ocr_backend.rs +159 -0
- data/ext/kreuzberg_rb/native/src/plugins/post_processor.rs +126 -0
- data/ext/kreuzberg_rb/native/src/plugins/validator.rs +99 -0
- data/ext/kreuzberg_rb/native/src/result.rs +326 -0
- data/ext/kreuzberg_rb/native/src/validation.rs +4 -0
- data/lib/kreuzberg/config.rb +66 -0
- data/lib/kreuzberg/result.rb +107 -2
- data/lib/kreuzberg/types.rb +104 -0
- data/lib/kreuzberg/version.rb +1 -1
- data/lib/kreuzberg.rb +0 -4
- data/sig/kreuzberg.rbs +105 -1
- data/vendor/Cargo.toml +3 -3
- data/vendor/kreuzberg/Cargo.toml +4 -3
- data/vendor/kreuzberg/README.md +1 -1
- data/vendor/kreuzberg/src/api/config.rs +69 -0
- data/vendor/kreuzberg/src/api/handlers.rs +99 -2
- data/vendor/kreuzberg/src/api/mod.rs +14 -7
- data/vendor/kreuzberg/src/api/router.rs +214 -0
- data/vendor/kreuzberg/src/api/startup.rs +243 -0
- data/vendor/kreuzberg/src/api/types.rs +78 -0
- data/vendor/kreuzberg/src/cache/cleanup.rs +277 -0
- data/vendor/kreuzberg/src/cache/core.rs +428 -0
- data/vendor/kreuzberg/src/cache/mod.rs +21 -843
- data/vendor/kreuzberg/src/cache/utilities.rs +156 -0
- data/vendor/kreuzberg/src/chunking/boundaries.rs +301 -0
- data/vendor/kreuzberg/src/chunking/builder.rs +294 -0
- data/vendor/kreuzberg/src/chunking/config.rs +52 -0
- data/vendor/kreuzberg/src/chunking/core.rs +1017 -0
- data/vendor/kreuzberg/src/chunking/mod.rs +14 -2211
- data/vendor/kreuzberg/src/chunking/processor.rs +10 -0
- data/vendor/kreuzberg/src/chunking/validation.rs +686 -0
- data/vendor/kreuzberg/src/core/config/extraction/core.rs +169 -0
- data/vendor/kreuzberg/src/core/config/extraction/env.rs +179 -0
- data/vendor/kreuzberg/src/core/config/extraction/loaders.rs +204 -0
- data/vendor/kreuzberg/src/core/config/extraction/mod.rs +42 -0
- data/vendor/kreuzberg/src/core/config/extraction/types.rs +93 -0
- data/vendor/kreuzberg/src/core/config/formats.rs +135 -0
- data/vendor/kreuzberg/src/core/config/mod.rs +20 -0
- data/vendor/kreuzberg/src/core/config/ocr.rs +73 -0
- data/vendor/kreuzberg/src/core/config/page.rs +57 -0
- data/vendor/kreuzberg/src/core/config/pdf.rs +111 -0
- data/vendor/kreuzberg/src/core/config/processing.rs +312 -0
- data/vendor/kreuzberg/src/core/config_validation/dependencies.rs +187 -0
- data/vendor/kreuzberg/src/core/config_validation/mod.rs +386 -0
- data/vendor/kreuzberg/src/core/config_validation/sections.rs +401 -0
- data/vendor/kreuzberg/src/core/extractor/batch.rs +246 -0
- data/vendor/kreuzberg/src/core/extractor/bytes.rs +116 -0
- data/vendor/kreuzberg/src/core/extractor/file.rs +240 -0
- data/vendor/kreuzberg/src/core/extractor/helpers.rs +71 -0
- data/vendor/kreuzberg/src/core/extractor/legacy.rs +62 -0
- data/vendor/kreuzberg/src/core/extractor/mod.rs +490 -0
- data/vendor/kreuzberg/src/core/extractor/sync.rs +208 -0
- data/vendor/kreuzberg/src/core/mod.rs +4 -1
- data/vendor/kreuzberg/src/core/pipeline/cache.rs +60 -0
- data/vendor/kreuzberg/src/core/pipeline/execution.rs +89 -0
- data/vendor/kreuzberg/src/core/pipeline/features.rs +108 -0
- data/vendor/kreuzberg/src/core/pipeline/format.rs +392 -0
- data/vendor/kreuzberg/src/core/pipeline/initialization.rs +67 -0
- data/vendor/kreuzberg/src/core/pipeline/mod.rs +135 -0
- data/vendor/kreuzberg/src/core/pipeline/tests.rs +975 -0
- data/vendor/kreuzberg/src/core/server_config/env.rs +90 -0
- data/vendor/kreuzberg/src/core/server_config/loader.rs +202 -0
- data/vendor/kreuzberg/src/core/server_config/mod.rs +380 -0
- data/vendor/kreuzberg/src/core/server_config/tests/basic_tests.rs +124 -0
- data/vendor/kreuzberg/src/core/server_config/tests/env_tests.rs +216 -0
- data/vendor/kreuzberg/src/core/server_config/tests/file_loading_tests.rs +341 -0
- data/vendor/kreuzberg/src/core/server_config/tests/mod.rs +5 -0
- data/vendor/kreuzberg/src/core/server_config/validation.rs +17 -0
- data/vendor/kreuzberg/src/embeddings.rs +136 -13
- data/vendor/kreuzberg/src/extraction/{archive.rs → archive/mod.rs} +45 -239
- data/vendor/kreuzberg/src/extraction/archive/sevenz.rs +98 -0
- data/vendor/kreuzberg/src/extraction/archive/tar.rs +118 -0
- data/vendor/kreuzberg/src/extraction/archive/zip.rs +101 -0
- data/vendor/kreuzberg/src/extraction/html/converter.rs +592 -0
- data/vendor/kreuzberg/src/extraction/html/image_handling.rs +95 -0
- data/vendor/kreuzberg/src/extraction/html/mod.rs +53 -0
- data/vendor/kreuzberg/src/extraction/html/processor.rs +659 -0
- data/vendor/kreuzberg/src/extraction/html/stack_management.rs +103 -0
- data/vendor/kreuzberg/src/extraction/html/types.rs +28 -0
- data/vendor/kreuzberg/src/extraction/mod.rs +6 -2
- data/vendor/kreuzberg/src/extraction/pptx/container.rs +159 -0
- data/vendor/kreuzberg/src/extraction/pptx/content_builder.rs +168 -0
- data/vendor/kreuzberg/src/extraction/pptx/elements.rs +132 -0
- data/vendor/kreuzberg/src/extraction/pptx/image_handling.rs +57 -0
- data/vendor/kreuzberg/src/extraction/pptx/metadata.rs +160 -0
- data/vendor/kreuzberg/src/extraction/pptx/mod.rs +558 -0
- data/vendor/kreuzberg/src/extraction/pptx/parser.rs +379 -0
- data/vendor/kreuzberg/src/extraction/transform/content.rs +205 -0
- data/vendor/kreuzberg/src/extraction/transform/elements.rs +211 -0
- data/vendor/kreuzberg/src/extraction/transform/mod.rs +480 -0
- data/vendor/kreuzberg/src/extraction/transform/types.rs +27 -0
- data/vendor/kreuzberg/src/extractors/archive.rs +2 -0
- data/vendor/kreuzberg/src/extractors/bibtex.rs +2 -0
- data/vendor/kreuzberg/src/extractors/djot_format/attributes.rs +134 -0
- data/vendor/kreuzberg/src/extractors/djot_format/conversion.rs +223 -0
- data/vendor/kreuzberg/src/extractors/djot_format/extractor.rs +172 -0
- data/vendor/kreuzberg/src/extractors/djot_format/mod.rs +24 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/block_handlers.rs +271 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/content_extraction.rs +257 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/event_handlers.rs +101 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/inline_handlers.rs +201 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/mod.rs +16 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/state.rs +78 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/table_extraction.rs +68 -0
- data/vendor/kreuzberg/src/extractors/djot_format/parsing/text_extraction.rs +61 -0
- data/vendor/kreuzberg/src/extractors/djot_format/rendering.rs +452 -0
- data/vendor/kreuzberg/src/extractors/docbook.rs +2 -0
- data/vendor/kreuzberg/src/extractors/docx.rs +12 -1
- data/vendor/kreuzberg/src/extractors/email.rs +2 -0
- data/vendor/kreuzberg/src/extractors/epub/content.rs +333 -0
- data/vendor/kreuzberg/src/extractors/epub/metadata.rs +137 -0
- data/vendor/kreuzberg/src/extractors/epub/mod.rs +186 -0
- data/vendor/kreuzberg/src/extractors/epub/parsing.rs +86 -0
- data/vendor/kreuzberg/src/extractors/excel.rs +4 -0
- data/vendor/kreuzberg/src/extractors/fictionbook.rs +2 -0
- data/vendor/kreuzberg/src/extractors/frontmatter_utils.rs +466 -0
- data/vendor/kreuzberg/src/extractors/html.rs +80 -8
- data/vendor/kreuzberg/src/extractors/image.rs +8 -1
- data/vendor/kreuzberg/src/extractors/jats/elements.rs +350 -0
- data/vendor/kreuzberg/src/extractors/jats/metadata.rs +21 -0
- data/vendor/kreuzberg/src/extractors/{jats.rs → jats/mod.rs} +10 -412
- data/vendor/kreuzberg/src/extractors/jats/parser.rs +52 -0
- data/vendor/kreuzberg/src/extractors/jupyter.rs +2 -0
- data/vendor/kreuzberg/src/extractors/latex/commands.rs +93 -0
- data/vendor/kreuzberg/src/extractors/latex/environments.rs +157 -0
- data/vendor/kreuzberg/src/extractors/latex/metadata.rs +27 -0
- data/vendor/kreuzberg/src/extractors/latex/mod.rs +146 -0
- data/vendor/kreuzberg/src/extractors/latex/parser.rs +231 -0
- data/vendor/kreuzberg/src/extractors/latex/utilities.rs +126 -0
- data/vendor/kreuzberg/src/extractors/markdown.rs +39 -162
- data/vendor/kreuzberg/src/extractors/mod.rs +9 -1
- data/vendor/kreuzberg/src/extractors/odt.rs +2 -0
- data/vendor/kreuzberg/src/extractors/opml/core.rs +165 -0
- data/vendor/kreuzberg/src/extractors/opml/mod.rs +31 -0
- data/vendor/kreuzberg/src/extractors/opml/parser.rs +479 -0
- data/vendor/kreuzberg/src/extractors/orgmode.rs +2 -0
- data/vendor/kreuzberg/src/extractors/pdf/extraction.rs +106 -0
- data/vendor/kreuzberg/src/extractors/{pdf.rs → pdf/mod.rs} +25 -324
- data/vendor/kreuzberg/src/extractors/pdf/ocr.rs +214 -0
- data/vendor/kreuzberg/src/extractors/pdf/pages.rs +51 -0
- data/vendor/kreuzberg/src/extractors/pptx.rs +9 -2
- data/vendor/kreuzberg/src/extractors/rst.rs +2 -0
- data/vendor/kreuzberg/src/extractors/rtf/encoding.rs +116 -0
- data/vendor/kreuzberg/src/extractors/rtf/formatting.rs +24 -0
- data/vendor/kreuzberg/src/extractors/rtf/images.rs +72 -0
- data/vendor/kreuzberg/src/extractors/rtf/metadata.rs +216 -0
- data/vendor/kreuzberg/src/extractors/rtf/mod.rs +142 -0
- data/vendor/kreuzberg/src/extractors/rtf/parser.rs +259 -0
- data/vendor/kreuzberg/src/extractors/rtf/tables.rs +83 -0
- data/vendor/kreuzberg/src/extractors/structured.rs +2 -0
- data/vendor/kreuzberg/src/extractors/text.rs +4 -0
- data/vendor/kreuzberg/src/extractors/typst.rs +2 -0
- data/vendor/kreuzberg/src/extractors/xml.rs +2 -0
- data/vendor/kreuzberg/src/keywords/processor.rs +14 -0
- data/vendor/kreuzberg/src/language_detection/processor.rs +10 -0
- data/vendor/kreuzberg/src/lib.rs +2 -2
- data/vendor/kreuzberg/src/mcp/errors.rs +312 -0
- data/vendor/kreuzberg/src/mcp/format.rs +211 -0
- data/vendor/kreuzberg/src/mcp/mod.rs +9 -3
- data/vendor/kreuzberg/src/mcp/params.rs +196 -0
- data/vendor/kreuzberg/src/mcp/server.rs +39 -1438
- data/vendor/kreuzberg/src/mcp/tools/cache.rs +179 -0
- data/vendor/kreuzberg/src/mcp/tools/extraction.rs +403 -0
- data/vendor/kreuzberg/src/mcp/tools/mime.rs +150 -0
- data/vendor/kreuzberg/src/mcp/tools/mod.rs +11 -0
- data/vendor/kreuzberg/src/ocr/backends/easyocr.rs +96 -0
- data/vendor/kreuzberg/src/ocr/backends/mod.rs +7 -0
- data/vendor/kreuzberg/src/ocr/backends/paddleocr.rs +27 -0
- data/vendor/kreuzberg/src/ocr/backends/tesseract.rs +134 -0
- data/vendor/kreuzberg/src/ocr/hocr.rs +60 -16
- data/vendor/kreuzberg/src/ocr/language_registry.rs +11 -235
- data/vendor/kreuzberg/src/ocr/mod.rs +1 -0
- data/vendor/kreuzberg/src/ocr/processor/config.rs +203 -0
- data/vendor/kreuzberg/src/ocr/processor/execution.rs +494 -0
- data/vendor/kreuzberg/src/ocr/processor/mod.rs +265 -0
- data/vendor/kreuzberg/src/ocr/processor/validation.rs +145 -0
- data/vendor/kreuzberg/src/ocr/tesseract_backend.rs +41 -24
- data/vendor/kreuzberg/src/pdf/bindings.rs +21 -8
- data/vendor/kreuzberg/src/pdf/hierarchy/bounding_box.rs +289 -0
- data/vendor/kreuzberg/src/pdf/hierarchy/clustering.rs +199 -0
- data/vendor/kreuzberg/src/pdf/{hierarchy.rs → hierarchy/extraction.rs} +6 -346
- data/vendor/kreuzberg/src/pdf/hierarchy/mod.rs +18 -0
- data/vendor/kreuzberg/src/plugins/extractor/mod.rs +319 -0
- data/vendor/kreuzberg/src/plugins/extractor/registry.rs +434 -0
- data/vendor/kreuzberg/src/plugins/extractor/trait.rs +391 -0
- data/vendor/kreuzberg/src/plugins/mod.rs +13 -0
- data/vendor/kreuzberg/src/plugins/ocr.rs +11 -0
- data/vendor/kreuzberg/src/plugins/processor/mod.rs +365 -0
- data/vendor/kreuzberg/src/plugins/processor/registry.rs +37 -0
- data/vendor/kreuzberg/src/plugins/processor/trait.rs +284 -0
- data/vendor/kreuzberg/src/plugins/registry/extractor.rs +416 -0
- data/vendor/kreuzberg/src/plugins/registry/mod.rs +116 -0
- data/vendor/kreuzberg/src/plugins/registry/ocr.rs +293 -0
- data/vendor/kreuzberg/src/plugins/registry/processor.rs +304 -0
- data/vendor/kreuzberg/src/plugins/registry/validator.rs +238 -0
- data/vendor/kreuzberg/src/plugins/validator/mod.rs +424 -0
- data/vendor/kreuzberg/src/plugins/validator/registry.rs +355 -0
- data/vendor/kreuzberg/src/plugins/validator/trait.rs +276 -0
- data/vendor/kreuzberg/src/stopwords/languages/asian.rs +40 -0
- data/vendor/kreuzberg/src/stopwords/languages/germanic.rs +36 -0
- data/vendor/kreuzberg/src/stopwords/languages/mod.rs +10 -0
- data/vendor/kreuzberg/src/stopwords/languages/other.rs +44 -0
- data/vendor/kreuzberg/src/stopwords/languages/romance.rs +36 -0
- data/vendor/kreuzberg/src/stopwords/languages/slavic.rs +36 -0
- data/vendor/kreuzberg/src/stopwords/mod.rs +7 -33
- data/vendor/kreuzberg/src/text/quality.rs +1 -1
- data/vendor/kreuzberg/src/text/quality_processor.rs +10 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/analysis.rs +238 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/mod.rs +8 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/punctuation.rs +54 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/reducer.rs +384 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/sentence_selection.rs +68 -0
- data/vendor/kreuzberg/src/text/token_reduction/core/word_filtering.rs +156 -0
- data/vendor/kreuzberg/src/text/token_reduction/filters/general.rs +377 -0
- data/vendor/kreuzberg/src/text/token_reduction/filters/html.rs +51 -0
- data/vendor/kreuzberg/src/text/token_reduction/filters/markdown.rs +285 -0
- data/vendor/kreuzberg/src/text/token_reduction/filters.rs +131 -246
- data/vendor/kreuzberg/src/types/djot.rs +209 -0
- data/vendor/kreuzberg/src/types/extraction.rs +301 -0
- data/vendor/kreuzberg/src/types/formats.rs +443 -0
- data/vendor/kreuzberg/src/types/metadata.rs +560 -0
- data/vendor/kreuzberg/src/types/mod.rs +281 -0
- data/vendor/kreuzberg/src/types/page.rs +182 -0
- data/vendor/kreuzberg/src/types/serde_helpers.rs +132 -0
- data/vendor/kreuzberg/src/types/tables.rs +39 -0
- data/vendor/kreuzberg/src/utils/quality/heuristics.rs +58 -0
- data/vendor/kreuzberg/src/utils/{quality.rs → quality/mod.rs} +168 -489
- data/vendor/kreuzberg/src/utils/quality/patterns.rs +117 -0
- data/vendor/kreuzberg/src/utils/quality/scoring.rs +178 -0
- data/vendor/kreuzberg/src/utils/string_pool/buffer_pool.rs +325 -0
- data/vendor/kreuzberg/src/utils/string_pool/interned.rs +102 -0
- data/vendor/kreuzberg/src/utils/string_pool/language_pool.rs +119 -0
- data/vendor/kreuzberg/src/utils/string_pool/mime_pool.rs +235 -0
- data/vendor/kreuzberg/src/utils/string_pool/mod.rs +41 -0
- data/vendor/kreuzberg/tests/api_chunk.rs +313 -0
- data/vendor/kreuzberg/tests/api_embed.rs +6 -9
- data/vendor/kreuzberg/tests/batch_orchestration.rs +1 -0
- data/vendor/kreuzberg/tests/concurrency_stress.rs +7 -0
- data/vendor/kreuzberg/tests/core_integration.rs +1 -0
- data/vendor/kreuzberg/tests/docx_metadata_extraction_test.rs +130 -0
- data/vendor/kreuzberg/tests/epub_native_extractor_tests.rs +5 -14
- data/vendor/kreuzberg/tests/format_integration.rs +2 -0
- data/vendor/kreuzberg/tests/helpers/mod.rs +1 -0
- data/vendor/kreuzberg/tests/html_table_test.rs +11 -11
- data/vendor/kreuzberg/tests/ocr_configuration.rs +16 -0
- data/vendor/kreuzberg/tests/ocr_errors.rs +18 -0
- data/vendor/kreuzberg/tests/ocr_quality.rs +9 -0
- data/vendor/kreuzberg/tests/ocr_stress.rs +1 -0
- data/vendor/kreuzberg/tests/pipeline_integration.rs +50 -0
- data/vendor/kreuzberg/tests/plugin_ocr_backend_test.rs +13 -0
- data/vendor/kreuzberg/tests/plugin_system.rs +12 -0
- data/vendor/kreuzberg/tests/registry_integration_tests.rs +2 -0
- data/vendor/kreuzberg-ffi/Cargo.toml +2 -1
- data/vendor/kreuzberg-ffi/benches/result_view_benchmark.rs +2 -0
- data/vendor/kreuzberg-ffi/kreuzberg.h +347 -178
- data/vendor/kreuzberg-ffi/src/config/html.rs +318 -0
- data/vendor/kreuzberg-ffi/src/config/loader.rs +154 -0
- data/vendor/kreuzberg-ffi/src/config/merge.rs +104 -0
- data/vendor/kreuzberg-ffi/src/config/mod.rs +385 -0
- data/vendor/kreuzberg-ffi/src/config/parse.rs +91 -0
- data/vendor/kreuzberg-ffi/src/config/serialize.rs +118 -0
- data/vendor/kreuzberg-ffi/src/config_builder.rs +598 -0
- data/vendor/kreuzberg-ffi/src/error.rs +46 -14
- data/vendor/kreuzberg-ffi/src/helpers.rs +10 -0
- data/vendor/kreuzberg-ffi/src/html_options.rs +421 -0
- data/vendor/kreuzberg-ffi/src/lib.rs +16 -0
- data/vendor/kreuzberg-ffi/src/panic_shield.rs +11 -0
- data/vendor/kreuzberg-ffi/src/plugins/ocr_backend.rs +2 -0
- data/vendor/kreuzberg-ffi/src/result.rs +148 -122
- data/vendor/kreuzberg-ffi/src/result_view.rs +4 -0
- data/vendor/kreuzberg-tesseract/Cargo.toml +2 -2
- metadata +200 -28
- data/vendor/kreuzberg/src/api/server.rs +0 -518
- data/vendor/kreuzberg/src/core/config.rs +0 -1914
- data/vendor/kreuzberg/src/core/config_validation.rs +0 -949
- data/vendor/kreuzberg/src/core/extractor.rs +0 -1200
- data/vendor/kreuzberg/src/core/pipeline.rs +0 -1223
- data/vendor/kreuzberg/src/core/server_config.rs +0 -1220
- data/vendor/kreuzberg/src/extraction/html.rs +0 -1830
- data/vendor/kreuzberg/src/extraction/pptx.rs +0 -3102
- data/vendor/kreuzberg/src/extractors/epub.rs +0 -696
- data/vendor/kreuzberg/src/extractors/latex.rs +0 -653
- data/vendor/kreuzberg/src/extractors/opml.rs +0 -635
- data/vendor/kreuzberg/src/extractors/rtf.rs +0 -809
- data/vendor/kreuzberg/src/ocr/processor.rs +0 -858
- data/vendor/kreuzberg/src/plugins/extractor.rs +0 -1042
- data/vendor/kreuzberg/src/plugins/processor.rs +0 -650
- data/vendor/kreuzberg/src/plugins/registry.rs +0 -1339
- data/vendor/kreuzberg/src/plugins/validator.rs +0 -967
- data/vendor/kreuzberg/src/text/token_reduction/core.rs +0 -832
- data/vendor/kreuzberg/src/types.rs +0 -1713
- data/vendor/kreuzberg/src/utils/string_pool.rs +0 -762
- data/vendor/kreuzberg-ffi/src/config.rs +0 -1341
|
@@ -0,0 +1,67 @@
|
|
|
1
|
+
//! Pipeline initialization and setup logic.
|
|
2
|
+
//!
|
|
3
|
+
//! This module handles the initialization of features and processor cache
|
|
4
|
+
//! required for pipeline execution.
|
|
5
|
+
|
|
6
|
+
use crate::Result;
|
|
7
|
+
|
|
8
|
+
use super::cache::{PROCESSOR_CACHE, ProcessorCache};
|
|
9
|
+
|
|
10
|
+
/// Type alias for processor stages tuple (Early, Middle, Late).
|
|
11
|
+
type ProcessorStages = (
|
|
12
|
+
std::sync::Arc<Vec<std::sync::Arc<dyn crate::plugins::PostProcessor>>>,
|
|
13
|
+
std::sync::Arc<Vec<std::sync::Arc<dyn crate::plugins::PostProcessor>>>,
|
|
14
|
+
std::sync::Arc<Vec<std::sync::Arc<dyn crate::plugins::PostProcessor>>>,
|
|
15
|
+
);
|
|
16
|
+
|
|
17
|
+
/// Initialize feature-specific systems that may be needed during pipeline execution.
|
|
18
|
+
pub(super) fn initialize_features() {
|
|
19
|
+
#[cfg(any(feature = "keywords-yake", feature = "keywords-rake"))]
|
|
20
|
+
{
|
|
21
|
+
let _ = crate::keywords::ensure_initialized();
|
|
22
|
+
}
|
|
23
|
+
|
|
24
|
+
#[cfg(feature = "language-detection")]
|
|
25
|
+
{
|
|
26
|
+
let _ = crate::language_detection::ensure_initialized();
|
|
27
|
+
}
|
|
28
|
+
|
|
29
|
+
#[cfg(feature = "chunking")]
|
|
30
|
+
{
|
|
31
|
+
let _ = crate::chunking::ensure_initialized();
|
|
32
|
+
}
|
|
33
|
+
|
|
34
|
+
#[cfg(feature = "quality")]
|
|
35
|
+
{
|
|
36
|
+
let registry = crate::plugins::registry::get_post_processor_registry();
|
|
37
|
+
if let Ok(mut reg) = registry.write() {
|
|
38
|
+
let _ = reg.register(std::sync::Arc::new(crate::text::QualityProcessor), 30);
|
|
39
|
+
}
|
|
40
|
+
}
|
|
41
|
+
}
|
|
42
|
+
|
|
43
|
+
/// Initialize the processor cache if not already initialized.
|
|
44
|
+
pub(super) fn initialize_processor_cache() -> Result<()> {
|
|
45
|
+
let mut cache_lock = PROCESSOR_CACHE
|
|
46
|
+
.write()
|
|
47
|
+
.map_err(|e| crate::KreuzbergError::Other(format!("Processor cache lock poisoned: {}", e)))?;
|
|
48
|
+
if cache_lock.is_none() {
|
|
49
|
+
*cache_lock = Some(ProcessorCache::new()?);
|
|
50
|
+
}
|
|
51
|
+
Ok(())
|
|
52
|
+
}
|
|
53
|
+
|
|
54
|
+
/// Get processors from the cache, organized by stage.
|
|
55
|
+
pub(super) fn get_processors_from_cache() -> Result<ProcessorStages> {
|
|
56
|
+
let cache_lock = PROCESSOR_CACHE
|
|
57
|
+
.read()
|
|
58
|
+
.map_err(|e| crate::KreuzbergError::Other(format!("Processor cache lock poisoned: {}", e)))?;
|
|
59
|
+
let cache = cache_lock
|
|
60
|
+
.as_ref()
|
|
61
|
+
.ok_or_else(|| crate::KreuzbergError::Other("Processor cache not initialized".to_string()))?;
|
|
62
|
+
Ok((
|
|
63
|
+
std::sync::Arc::clone(&cache.early),
|
|
64
|
+
std::sync::Arc::clone(&cache.middle),
|
|
65
|
+
std::sync::Arc::clone(&cache.late),
|
|
66
|
+
))
|
|
67
|
+
}
|
|
@@ -0,0 +1,135 @@
|
|
|
1
|
+
//! Post-processing pipeline orchestration.
|
|
2
|
+
//!
|
|
3
|
+
//! This module orchestrates the post-processing pipeline, executing validators,
|
|
4
|
+
//! quality processing, chunking, and custom hooks in the correct order.
|
|
5
|
+
|
|
6
|
+
mod cache;
|
|
7
|
+
mod execution;
|
|
8
|
+
mod features;
|
|
9
|
+
mod format;
|
|
10
|
+
mod initialization;
|
|
11
|
+
|
|
12
|
+
#[cfg(test)]
|
|
13
|
+
mod tests;
|
|
14
|
+
|
|
15
|
+
pub use cache::clear_processor_cache;
|
|
16
|
+
pub use format::apply_output_format;
|
|
17
|
+
|
|
18
|
+
use crate::Result;
|
|
19
|
+
use crate::core::config::ExtractionConfig;
|
|
20
|
+
use crate::types::ExtractionResult;
|
|
21
|
+
|
|
22
|
+
use execution::{execute_processors, execute_validators};
|
|
23
|
+
use features::{execute_chunking, execute_language_detection};
|
|
24
|
+
use initialization::{get_processors_from_cache, initialize_features, initialize_processor_cache};
|
|
25
|
+
|
|
26
|
+
/// Run the post-processing pipeline on an extraction result.
|
|
27
|
+
///
|
|
28
|
+
/// Executes post-processing in the following order:
|
|
29
|
+
/// 1. Post-Processors - Execute by stage (Early, Middle, Late) to modify/enhance the result
|
|
30
|
+
/// 2. Quality Processing - Text cleaning and quality scoring
|
|
31
|
+
/// 3. Chunking - Text splitting if enabled
|
|
32
|
+
/// 4. Validators - Run validation hooks on the processed result (can fail fast)
|
|
33
|
+
///
|
|
34
|
+
/// # Arguments
|
|
35
|
+
///
|
|
36
|
+
/// * `result` - The extraction result to process
|
|
37
|
+
/// * `config` - Extraction configuration
|
|
38
|
+
///
|
|
39
|
+
/// # Returns
|
|
40
|
+
///
|
|
41
|
+
/// The processed extraction result.
|
|
42
|
+
///
|
|
43
|
+
/// # Errors
|
|
44
|
+
///
|
|
45
|
+
/// - Validator errors bubble up immediately
|
|
46
|
+
/// - Post-processor errors are caught and recorded in metadata
|
|
47
|
+
/// - System errors (IO, RuntimeError equivalents) always bubble up
|
|
48
|
+
#[cfg_attr(feature = "otel", tracing::instrument(
|
|
49
|
+
skip(result, config),
|
|
50
|
+
fields(
|
|
51
|
+
pipeline.stage = "post_processing",
|
|
52
|
+
content.length = result.content.len(),
|
|
53
|
+
)
|
|
54
|
+
))]
|
|
55
|
+
pub async fn run_pipeline(mut result: ExtractionResult, config: &ExtractionConfig) -> Result<ExtractionResult> {
|
|
56
|
+
let pp_config = config.postprocessor.as_ref();
|
|
57
|
+
let postprocessing_enabled = pp_config.is_none_or(|c| c.enabled);
|
|
58
|
+
|
|
59
|
+
if postprocessing_enabled {
|
|
60
|
+
initialize_features();
|
|
61
|
+
initialize_processor_cache()?;
|
|
62
|
+
|
|
63
|
+
let (early_processors, middle_processors, late_processors) = get_processors_from_cache()?;
|
|
64
|
+
|
|
65
|
+
execute_processors(
|
|
66
|
+
&mut result,
|
|
67
|
+
config,
|
|
68
|
+
&pp_config,
|
|
69
|
+
early_processors,
|
|
70
|
+
middle_processors,
|
|
71
|
+
late_processors,
|
|
72
|
+
)
|
|
73
|
+
.await?;
|
|
74
|
+
}
|
|
75
|
+
|
|
76
|
+
execute_chunking(&mut result, config)?;
|
|
77
|
+
execute_language_detection(&mut result, config)?;
|
|
78
|
+
execute_validators(&result, config).await?;
|
|
79
|
+
|
|
80
|
+
// Transform to element-based output if requested
|
|
81
|
+
if config.result_format == crate::types::OutputFormat::ElementBased {
|
|
82
|
+
result.elements = Some(crate::extraction::transform::transform_extraction_result_to_elements(
|
|
83
|
+
&result,
|
|
84
|
+
));
|
|
85
|
+
}
|
|
86
|
+
|
|
87
|
+
// Apply output format conversion as the final step
|
|
88
|
+
apply_output_format(&mut result, config.output_format);
|
|
89
|
+
|
|
90
|
+
Ok(result)
|
|
91
|
+
}
|
|
92
|
+
|
|
93
|
+
/// Run the post-processing pipeline synchronously (WASM-compatible version).
|
|
94
|
+
///
|
|
95
|
+
/// This is a synchronous implementation for WASM and non-async contexts.
|
|
96
|
+
/// It performs a subset of the full async pipeline, excluding async post-processors
|
|
97
|
+
/// and validators.
|
|
98
|
+
///
|
|
99
|
+
/// # Arguments
|
|
100
|
+
///
|
|
101
|
+
/// * `result` - The extraction result to process
|
|
102
|
+
/// * `config` - Extraction configuration
|
|
103
|
+
///
|
|
104
|
+
/// # Returns
|
|
105
|
+
///
|
|
106
|
+
/// The processed extraction result.
|
|
107
|
+
///
|
|
108
|
+
/// # Notes
|
|
109
|
+
///
|
|
110
|
+
/// This function is only available when the `tokio-runtime` feature is disabled.
|
|
111
|
+
/// It handles:
|
|
112
|
+
/// - Quality processing (if enabled)
|
|
113
|
+
/// - Chunking (if enabled)
|
|
114
|
+
/// - Language detection (if enabled)
|
|
115
|
+
///
|
|
116
|
+
/// It does NOT handle:
|
|
117
|
+
/// - Async post-processors
|
|
118
|
+
/// - Async validators
|
|
119
|
+
#[cfg(not(feature = "tokio-runtime"))]
|
|
120
|
+
pub fn run_pipeline_sync(mut result: ExtractionResult, config: &ExtractionConfig) -> Result<ExtractionResult> {
|
|
121
|
+
execute_chunking(&mut result, config)?;
|
|
122
|
+
execute_language_detection(&mut result, config)?;
|
|
123
|
+
|
|
124
|
+
// Transform to element-based output if requested
|
|
125
|
+
if config.result_format == crate::types::OutputFormat::ElementBased {
|
|
126
|
+
result.elements = Some(crate::extraction::transform::transform_extraction_result_to_elements(
|
|
127
|
+
&result,
|
|
128
|
+
));
|
|
129
|
+
}
|
|
130
|
+
|
|
131
|
+
// Apply output format conversion as the final step
|
|
132
|
+
apply_output_format(&mut result, config.output_format);
|
|
133
|
+
|
|
134
|
+
Ok(result)
|
|
135
|
+
}
|