xberg 1.0.0.pre.rc.32 → 1.0.0.pre.rc.33

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8a3acc55adf9c4de516a81017b6e02ae65953905e2debf75966da906066492e4
-  data.tar.gz: 1608076d7135673c19105818ac74f3b57c887c05dd8d20be9e503b2535768169
+  metadata.gz: e9c58dd2f7faf056ebc864d535ff983a8a4826c7999a6506ad82af5408c23961
+  data.tar.gz: effa2a8663dffe2f7136bb327464f88c06ffd58a5ad5f62a9493e883f46f52d8
 SHA512:
-  metadata.gz: 36a4980dfab6661a988a4b9b22cdcf7ee7337ce0aad71aa162a2f956e54e2864df9f55fe30d663222b4443289a17e2dc2c64732a818b637a26dcb7b48b3e02b9
-  data.tar.gz: ef5d6935da95424c1ce7287aadae1d3bce848504888efa31b024207c9642d74e2c1bee54a87cbe259fb279816cfc237d125c0965f191cf61287e6eab9df1e009
+  metadata.gz: eea8d788ea172a5dce2f192520bb55ae28af356bdee7f40c13f82b779bc92596a5499a0889708e0a1b6723522537e4cb1f906b0cbc44404d73e8e6a071c70f00
+  data.tar.gz: 2c1c5dbdb015e2abc5a487895ca995076b8650b0cce43f1184a6c5a61c68baaefe5abc31d0cc6861f19c40bc6eb1bea87bf04dbb0e7e8b912cd51c4e811d3e5f

data/ext/xberg_rb/native/Cargo.lock

@@ -281,7 +281,7 @@ checksum = "ae36dc4177970ef04fde5178d3e2429882def40e57a451f919c098f72baa6cec"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 3.0.2",
+ "syn 3.0.3",
 ]
 
 [[package]]
@@ -1020,9 +1020,9 @@ dependencies = [
 
 [[package]]
 name = "clap"
-version = "4.6.3"
+version = "4.6.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0fb99565819980999fb7b4a1796046a5c949e6d4ff132cf5fadf5a641e20d776"
+checksum = "d91e0c145792ef73a6ad36d27c75ac09f1832222a3c209689d90f534685ee5b7"
 dependencies = [
  "clap_builder",
  "clap_derive",
@@ -1042,14 +1042,14 @@ dependencies = [
 
 [[package]]
 name = "clap_derive"
-version = "4.6.3"
+version = "4.6.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32f2392eae7f16557a3d727ef3a12e57b2b2ca6f98566a5f4fb41ffe305df077"
+checksum = "d012d2b9d65aca7f18f4d9878a045bc17899bba951561ba5ec3c2ba1eed9a061"
 dependencies = [
  "heck",
  "proc-macro2",
  "quote",
- "syn 2.0.119",
+ "syn 3.0.3",
 ]
 
 [[package]]
@@ -1465,9 +1465,9 @@ dependencies = [
 
 [[package]]
 name = "ctor"
-version = "1.0.9"
+version = "1.0.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a394189d59f9befacce833f337f7b1eca5e9a91221bcdd4d28e0114d96e597b3"
+checksum = "e2e30e509674ef0ec91e21a7735766db37d163d46151b6a361d8b83dd79116bd"
 dependencies = [
  "link-section",
  "linktime-proc-macro",
@@ -2514,9 +2514,9 @@ checksum = "7f22fb22f065b308be0d8724e3706c7fa3fc2a6c7d6899df4cad7860e7a75436"
 
 [[package]]
 name = "glob"
-version = "0.3.3"
+version = "0.3.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0cc23270f6e1808e30a928bdc84dea0b9b4136a8bc82338574f23baf47bbd280"
+checksum = "e4eba85ea1d0a966a983acd07deee566e67395d2d96b6fb39e62b5a833f1eb0b"
 
 [[package]]
 name = "globset"
@@ -3519,9 +3519,9 @@ dependencies = [
 
 [[package]]
 name = "jsonschema"
-version = "0.48.2"
+version = "0.48.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "39d785c0683d1027839f1ac6a43b6d07c63d58ecd363d8146dda4531d53465a1"
+checksum = "f05a6cff806294a7699d48761d4ae2bdebab7fa2eef05e4a0b83320f0ff46901"
 dependencies = [
  "ahash 0.8.12",
  "bytecount",
@@ -3533,6 +3533,7 @@ dependencies = [
  "idna",
  "itoa",
  "jsonschema-regex",
+ "jsonschema-value",
  "num-cmp",
  "num-traits",
  "percent-encoding",
@@ -3547,13 +3548,27 @@ dependencies = [
 
 [[package]]
 name = "jsonschema-regex"
-version = "0.48.2"
+version = "0.48.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "917b3be16b2662538aafb55201790561e1f1ad1a0454baf42f33bd9de343778b"
+checksum = "84708dd3f6c5327478a96f37378adb27deed6a626cf84d19de780d9e8ae84b17"
 dependencies = [
  "regex-syntax",
 ]
 
+[[package]]
+name = "jsonschema-value"
+version = "0.48.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fd8c04ab4ce60e42c4e302d3bd9d184652bb56f212495d2eea23573bcbf72556"
+dependencies = [
+ "ahash 0.8.12",
+ "bytecount",
+ "fraction",
+ "num-cmp",
+ "num-traits",
+ "serde_json",
+]
+
 [[package]]
 name = "kdf"
 version = "0.1.0"
@@ -3693,9 +3708,9 @@ dependencies = [
 
 [[package]]
 name = "link-section"
-version = "0.19.0"
+version = "0.19.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e333fe507b738576d6da5bb3f1a7d7a1c80307ed9ef31624c057d844c19c93e9"
+checksum = "8dc98458dfe90986c5e2f6ddcf68360c7e5c4252600153e06aa4ee8176c0f8d1"
 
 [[package]]
 name = "linktime-proc-macro"
@@ -4520,9 +4535,9 @@ dependencies = [
 
 [[package]]
 name = "office_oxide"
-version = "0.1.7"
+version = "0.1.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7150eeae2b40a6926717058d0575ee2f48a36806725376a2ff6b2e69b61c88e9"
+checksum = "317c254cebe5b87c74a0fbd84e889d51ec0c4db885fa5aca98700ad85fd9ec5f"
 dependencies = [
  "atoi_simd",
  "encoding_rs",
@@ -4843,9 +4858,9 @@ checksum = "9b4f627cb1b25917193a259e49bdad08f671f8d9708acfd5fe0a8c1455d87220"
 
 [[package]]
 name = "pest"
-version = "2.8.7"
+version = "2.8.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "47627dd7305c6a2d6c8c6bcd24c5a4c17dbbf425f4f9c5313e724b38fc9782e9"
+checksum = "7df728be843c7070fab6ab7c328c4e9e9d78e23bf749c0669c86ee7ebfa050a2"
 dependencies = [
  "memchr",
  "ucd-trie",
@@ -4853,9 +4868,9 @@ dependencies = [
 
 [[package]]
 name = "pest_derive"
-version = "2.8.7"
+version = "2.8.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4b4254325ecad416ab689e27ba51da03ba01a9632bc6e108f5fe7c3c4ad29d58"
+checksum = "9e2dd6fc3b26b3462ee188aac870f5a41d398f1cd5e2408d16531bd71c9591fd"
 dependencies = [
  "pest",
  "pest_generator",
@@ -4863,9 +4878,9 @@ dependencies = [
 
 [[package]]
 name = "pest_generator"
-version = "2.8.7"
+version = "2.8.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6c4c0e91ead7a8f7acecbca6f003fc2e8282b1dbe2dd9c9d2f16aba42995e0a7"
+checksum = "6a7a9205cfb6f596a9e8b689c0a15f9ceb7a1aafae7aaf788150ac65b29975b6"
 dependencies = [
  "pest",
  "pest_meta",
@@ -4876,9 +4891,9 @@ dependencies = [
 
 [[package]]
 name = "pest_meta"
-version = "2.8.7"
+version = "2.8.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f9744bc48116fee06334924bb5f2bad41eed5e89bd26e29b0b799f9a3f82c210"
+checksum = "85abd351c0de1e8384fc791a0737111a350394937e92b956b743dac12429f57c"
 dependencies = [
  "pest",
 ]
@@ -5704,14 +5719,14 @@ checksum = "2c9283685feec7d69af75fb0e858d5e7378f33fe4fc699383b2916ab9273e03c"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 3.0.2",
+ "syn 3.0.3",
 ]
 
 [[package]]
 name = "referencing"
-version = "0.48.2"
+version = "0.48.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2d2b21a85c07fcf1cb95ff7ab8a687fcd37fa70db59975b5a032abdaadfdf42e"
+checksum = "980e77f5cfffc592c0e7dfb6d20770a0b96a6ab94d403de7454ed9799505e4ba"
 dependencies = [
  "ahash 0.8.12",
  "fluent-uri",
@@ -6358,7 +6373,7 @@ checksum = "e7a5d71263a5a7d47b41f6b3f06ba276f10cc18b0931f1799f710578e2309348"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 3.0.2",
+ "syn 3.0.3",
 ]
 
 [[package]]
@@ -6405,7 +6420,7 @@ checksum = "8d3b1629de253c70a0508c3899572da79ca359fdab27c7920ff00406df418906"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 3.0.2",
+ "syn 3.0.3",
 ]
 
 [[package]]
@@ -6712,9 +6727,9 @@ dependencies = [
 
 [[package]]
 name = "sse-stream"
-version = "0.2.4"
+version = "0.2.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "39f24a9b78c40b90817bbcd1821c74ddfd74916aadd29403d001532a9195532d"
+checksum = "c123f296ade4ec4b8b0f6162116e6629f5146922ca5ab40ca9d3c2e73ab4761e"
 dependencies = [
  "bytes",
  "futures-util",
@@ -7042,9 +7057,9 @@ dependencies = [
 
 [[package]]
 name = "syn"
-version = "3.0.2"
+version = "3.0.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a207d6d6a2b7fc470b80443726053f18a2481b7e1eee970597051596567987a3"
+checksum = "53e9bae58849f64dfa4f5d5ae372c8341f7305f82a3868709269343628b659a3"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -7219,7 +7234,7 @@ checksum = "43cbfe0cf76104d42a574802844187e84a305e531ed54455f11fbde0f10541cd"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 3.0.2",
+ "syn 3.0.3",
 ]
 
 [[package]]
@@ -7475,9 +7490,9 @@ dependencies = [
 
 [[package]]
 name = "tokio-stream"
-version = "0.1.18"
+version = "0.1.19"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32da49809aab5c3bc678af03902d4ccddea2a87d028d86392a4b1560c6906c70"
+checksum = "a3d06f0b082ba57c26b79407372e57cf2a1e28124f78e9479fe80322cf53420b"
 dependencies = [
  "futures-core",
  "pin-project-lite",
@@ -7870,9 +7885,9 @@ checksum = "2896d95c02a80c6d6a5d6e953d479f5ddf2dfdb6a244441010e373ac0fb88971"
 
 [[package]]
 name = "unhwp"
-version = "0.5.3"
+version = "0.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "54ecdf25ce4d83badd909fca9eeaa121d57c342c500147ea19173122efcc39fb"
+checksum = "4a9cdee1f69dfbc8f20bfd4028c5653c8769183f73be170341bd9a1568a74899"
 dependencies = [
  "cfb 0.10.0",
  "flate2",
@@ -8852,9 +8867,9 @@ dependencies = [
 
 [[package]]
 name = "xberg"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "da88ef5c30844ee5d145658a1fc9f30a2f19366a0416fc0754f86c45cfd4c0aa"
+checksum = "0bf1060d70870a43018bf4c9a44d8582fa8d5f3be3ecea967f03e86d93ebb2ca"
 dependencies = [
  "aes-gcm",
  "ahash 0.8.12",
@@ -8976,9 +8991,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-gliner"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "134fbe490d200ed2d4b9f54c5195ecd29511f63c3ef6ae5ff21e64852c3dd09b"
+checksum = "65f21f1c7a0f96f502d1ec5aa35405c3c1a1ec168fb77f7401ab14450264f395"
 dependencies = [
  "getrandom 0.2.17",
  "getrandom 0.3.4",
@@ -8992,9 +9007,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-libheif"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bdf81ff26a309442055e126ccf1bb507dbb23815b4bb234de8a35367474ce0cb"
+checksum = "2147eac7d54ba62b5f6dd9cd170fdafb32b29924532bafb1e161cd8e5b9c51fa"
 dependencies = [
  "cfg-if 1.0.4",
  "enumn",
@@ -9005,9 +9020,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-paddle-ocr"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ab63cdc6606e5b52a112f6b97e65416907113110dd2f60d535af07dba8fb9703"
+checksum = "0d574aa7d5752564411a8ee1d78fc98efdfa69ae2e385b3fdab919cb29247c8e"
 dependencies = [
  "geo-clipper",
  "geo-types",
@@ -9022,7 +9037,7 @@ dependencies = [
 
 [[package]]
 name = "xberg-rb"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 dependencies = [
  "async-trait",
  "crawlberg",
@@ -9037,9 +9052,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-tesseract"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "60e18e77a45d62cc3b07da430e6ad59f935e4c8a376810d80c1afb13a012f62a"
+checksum = "955a17a1528e0b1f3260f5ec3cc0bacc31645e4db59300d1db6f79782329253d"
 dependencies = [
  "cc",
  "cmake",

data/ext/xberg_rb/native/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "xberg-rb"
-version = "1.0.0-rc.32"
+version = "1.0.0-rc.33"
 edition = "2024"
 license = "MIT"
 description = "High-performance document intelligence library"
@@ -17,11 +17,12 @@ path = "../src/lib.rs"
 crate-type = ["cdylib"]
 
 [features]
-default = ["api", "api-types", "auto-rotate-types", "chunking-tokenizers", "diff", "embedding-presets", "heuristics", "html", "keywords-rake", "keywords-yake", "late-interaction", "late-interaction-presets", "layout-types", "liter-llm", "markdown-footnotes", "ocr", "ocr-wasm", "office", "paddle-ocr-types", "pdf", "presets", "quality", "redaction", "reranker", "reranker-presets", "sparse-embedding-presets", "sparse-embeddings", "svg", "tokio-runtime", "transcription-types", "tree-sitter", "url-config-types", "url-ingestion", "xml"]
+default = ["api", "api-types", "auto-rotate-types", "chunking-tokenizers", "classification", "diff", "embedding-presets", "heuristics", "html", "keywords-rake", "keywords-yake", "late-interaction", "late-interaction-presets", "layout-types", "liter-llm", "markdown-footnotes", "ocr", "ocr-wasm", "office", "paddle-ocr-types", "pdf", "presets", "quality", "redaction", "reranker", "reranker-presets", "sparse-embedding-presets", "sparse-embeddings", "svg", "tokio-runtime", "transcription-types", "tree-sitter", "url-config-types", "url-ingestion", "xml"]
 api = ["xberg/api"]
 api-types = ["xberg/api-types"]
 auto-rotate-types = ["xberg/auto-rotate-types"]
 chunking-tokenizers = ["xberg/chunking-tokenizers"]
+classification = ["xberg/classification"]
 diff = ["xberg/diff"]
 embedding-presets = ["xberg/embedding-presets"]
 heuristics = ["xberg/heuristics"]
@@ -63,10 +64,10 @@ serde = { version = "1", features = ["derive"] }
 serde_json = "1"
 tokio = { version = "1", features = ["rt-multi-thread"] }
 [target.'cfg(not(any(target_os = "windows", all(target_os = "macos", target_arch = "x86_64"))))'.dependencies]
-xberg = { version = "1.0.0-rc.32", features = ["full", "pdf", "ocr", "paddle-ocr", "paddle-ocr-types", "layout-detection", "layout-types", "embeddings", "embedding-presets", "reranker", "reranker-presets", "sparse-embeddings", "sparse-embedding-presets", "late-interaction", "late-interaction-presets", "chunking", "keywords-yake", "keywords-rake", "language-detection", "html", "tree-sitter", "office", "email", "archives", "stopwords", "auto-rotate", "auto-rotate-types", "tokio-runtime", "api", "api-types", "mcp", "liter-llm", "quality", "svg", "transcription", "transcription-types", "classification", "captioning", "ner-onnx", "ner-llm", "diff", "markdown-footnotes", "redaction", "enrichment", "heuristics", "presets", "structured", "url-ingestion"] }
+xberg = { version = "1.0.0-rc.33", features = ["full", "pdf", "ocr", "paddle-ocr", "paddle-ocr-types", "layout-detection", "layout-types", "embeddings", "embedding-presets", "reranker", "reranker-presets", "sparse-embeddings", "sparse-embedding-presets", "late-interaction", "late-interaction-presets", "chunking", "keywords-yake", "keywords-rake", "language-detection", "html", "tree-sitter", "office", "email", "archives", "stopwords", "auto-rotate", "auto-rotate-types", "tokio-runtime", "api", "api-types", "mcp", "liter-llm", "quality", "svg", "transcription", "transcription-types", "classification", "captioning", "ner-onnx", "ner-llm", "diff", "markdown-footnotes", "redaction", "enrichment", "heuristics", "presets", "structured", "url-ingestion"] }
 
 [target.'cfg(target_os = "windows")'.dependencies]
-xberg = { version = "1.0.0-rc.32", features = ["windows-target"] }
+xberg = { version = "1.0.0-rc.33", features = ["windows-target"] }
 
 [target.'cfg(all(target_os = "macos", target_arch = "x86_64"))'.dependencies]
-xberg = { version = "1.0.0-rc.32", features = ["macos-intel-target"] }
+xberg = { version = "1.0.0-rc.33", features = ["macos-intel-target"] }