xberg 1.1.1 → 1.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d7be6ba0bd3df9bfdef3eb5586549a1fd63fafd8bd25d0910a842c850458ebb6
-  data.tar.gz: b4bad20fa6c0cd4058a85096ab475b4e5d9c4b1df1a53f3727b30ad54deae2b7
+  metadata.gz: 2fabfbdcbc215452a281c88a3475e71f593c63201e35d6aa12918dbc411f6e06
+  data.tar.gz: dcf0048b9f0a25c92f860d1b077c52440191864e184de6ebc623617d14ac584b
 SHA512:
-  metadata.gz: 36a2fe155923c2ee3928cd9dcd6d8fbd7a2f5361543df8f07ff6ab72cd9702eae131c6448176f56fd8a89ce8de34d2163779d37ff5cefd0a5640f298b3a1b353
-  data.tar.gz: 5f4a3ebdb856ff7702d0784be2153888c95ec3222101c5d83cda93e3a4ac0c75c830ccb2b7e999ae7c24d22d5a748c9407db99af0f260259cf8c4fe8c81c6cf2
+  metadata.gz: d594d585723df11fd18ab8c9fff9968ffb201f79b8f263d937219c664362ab2379d09cfaa733c952c4ec5954f04e9dcd84bb95f3a9c0d88f139fa67260051d9d
+  data.tar.gz: 9a927c4d766304d91f6c37a4fe36aacba53b5246ce3f28ad107cd5cf86a970f64cc8ab71afa95eee17d957f9c248dde38a475a43af506fbf60003ff7604a94b7

data/ext/xberg_rb/native/Cargo.lock

@@ -1501,6 +1501,12 @@ version = "0.8.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "773648b94d0e5d620f64f280777445740e61fe701025087ec8b57f45c791888b"
 
+[[package]]
+name = "core_detect"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7f8f80099a98041a3d1622845c271458a2d73e688351bf3cb999266764b81d48"
+
 [[package]]
 name = "core_maths"
 version = "0.1.1"
@@ -2027,11 +2033,17 @@ dependencies = [
 
 [[package]]
 name = "encoding_rs"
-version = "0.8.35"
+version = "0.8.40"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "75030f3c4f45dafd7586dd6780965a8c7e8e285a5ecb86713e63a79c5b2766f3"
+checksum = "2a7a45518d2863d18aa47f4a0cf9faec2aa4304cc09df5e41299f276b3ad135e"
 dependencies = [
  "cfg-if 1.0.4",
+ "core_detect",
+ "multiversion",
+ "multiversion_no_op",
+ "rustversion",
+ "scopeguard",
+ "simdutf8",
 ]
 
 [[package]]
@@ -4545,6 +4557,34 @@ dependencies = [
  "version_check",
 ]
 
+[[package]]
+name = "multiversion"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7edb7f0ff51249dfda9ab96b5823695e15a052dc15074c9dbf3d118afaf2c201"
+dependencies = [
+ "multiversion-macros",
+ "target-features",
+]
+
+[[package]]
+name = "multiversion-macros"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b093064383341eb3271f42e381cb8f10a01459478446953953c75d24bd339fc0"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.119",
+ "target-features",
+]
+
+[[package]]
+name = "multiversion_no_op"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "743fb55ba31b18fb1ecef6bdc9aa2743314978ac084044301a7eee33fb99a20d"
+
 [[package]]
 name = "nalgebra"
 version = "0.35.0"
@@ -4674,9 +4714,9 @@ dependencies = [
 
 [[package]]
 name = "nom-exif"
-version = "3.7.0"
+version = "3.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "47a28872847c2602f03d141629a212d7a8c3b6bd8983f518862c51aa474594f8"
+checksum = "3217a90c724aa7c99f53b2fc7eec37b63e3e9aa1801ebe031dee75234d9c8339"
 dependencies = [
  "bytes",
  "chrono",
@@ -7458,6 +7498,12 @@ dependencies = [
  "xattr",
 ]
 
+[[package]]
+name = "target-features"
+version = "0.1.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c1bbb9f3c5c463a01705937a24fdabc5047929ac764b2d5b9cf681c1f5041ed5"
+
 [[package]]
 name = "target-lexicon"
 version = "0.13.5"
@@ -9213,9 +9259,9 @@ dependencies = [
 
 [[package]]
 name = "xberg"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8d88afb006b949d6de97555cb528437726bc4aa2c641c618d3b1c536828a9804"
+checksum = "771b6187d6d275a97243bb452e909d9a736db1aa1ad0391b388f27d43c2364d1"
 dependencies = [
  "aes-gcm",
  "ahash",
@@ -9350,9 +9396,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-candle-ocr"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "73993303311e4441b037b0a1aed64ed40b4f230daf1b0990e68abfc54e625101"
+checksum = "06533bb7ef81505e00fb366ef399e119014ca2b0377b75468364fe1537ad6b87"
 dependencies = [
  "base64 0.23.1",
  "candle-core",
@@ -9372,9 +9418,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-gliner"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "53b8fb46063d23a6609ff2f189289458f756fa8f2e59575e3464753143b845d5"
+checksum = "c475644a8ef045d071c9e59032feb9f73a5191610c00240200a08acac4dbc57f"
 dependencies = [
  "getrandom 0.2.17",
  "getrandom 0.3.4",
@@ -9388,9 +9434,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-libheif"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1b7da27f2e68e5e66c568cedc24892c220ba7b8362fc4b074e32d0f016cca937"
+checksum = "2e91e8d98372702dc5836f645b9ef020a7ebae824a00ab12d9182181a63067c9"
 dependencies = [
  "cfg-if 1.0.4",
  "enumn",
@@ -9401,9 +9447,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-libwpd"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "972715f175bc4180c255d76ebdfbbd935adca2a818f818151995feab9c0952e4"
+checksum = "d09f47b32f5b9066c9619437ccad09e278a73b53e50adeb211dcb8db5b72ead9"
 dependencies = [
  "cc",
  "flate2",
@@ -9416,9 +9462,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-native-pdf"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b768494ffc05dcd59a5003ead16e493affd36a741aa34f73f4ba3ef1b9045a7a"
+checksum = "70739a2e890c3f30c3b64c8f0911ba48064cf68008b463d268babc2985d6b55f"
 dependencies = [
  "aes 0.9.3",
  "base64 0.23.1",
@@ -9463,9 +9509,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-paddle-ocr"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "18ce27e08abe961f04239ed4dbaa789792d9f49bae755336242dedaaff97c6ff"
+checksum = "8bdc84f871be63230c84861f6bdf911e99eb21153663a89be25882f4ec2d706a"
 dependencies = [
  "geo-clipper",
  "geo-types",
@@ -9480,7 +9526,7 @@ dependencies = [
 
 [[package]]
 name = "xberg-rb"
-version = "1.1.1"
+version = "1.1.2"
 dependencies = [
  "async-trait",
  "crawlberg",
@@ -9496,9 +9542,9 @@ dependencies = [
 
 [[package]]
 name = "xberg-tesseract"
-version = "1.1.1"
+version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2ca506451bb0ef5c3636fb091ca02ee31eec420c42154b3dfebfafdc0986a5c4"
+checksum = "362ecf382c207857401e64184513656a42de37345711cf2e1554caa6d9752519"
 dependencies = [
  "cc",
  "cmake",

data/ext/xberg_rb/native/Cargo.toml

@@ -1,11 +1,11 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:1efb91aaf98fa801f534983b9e6216c6d04d660986f75509d467311c3a719c8e
+# alef:hash:07aa5bc5a192a77758e42f213f201383ef913b7d438265bf93de7863c4bfbfaf
 # To regenerate: alef generate
 # To verify freshness: alef verify
 
 [package]
 name = "xberg-rb"
-version = "1.1.1"
+version = "1.1.2"
 edition = "2024"
 license = "MIT"
 description = "High-performance document intelligence library"
@@ -70,13 +70,13 @@ tokio = { version = "1", features = ["rt-multi-thread"] }
 tracing = "0.1"
 
 [target.'cfg(all(target_os = "macos", target_arch = "x86_64"))'.dependencies]
-xberg = { version = "1.1.1", default-features = false, features = ["macos-intel-target", "simd-utf8"] }
+xberg = { version = "1.1.2", default-features = false, features = ["macos-intel-target", "simd-utf8"] }
 
 [target.'cfg(not(any(target_os = "windows", all(target_os = "macos", target_arch = "x86_64"))))'.dependencies]
-xberg = { version = "1.1.1", 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 = "1.1.2", 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 = "1.1.1", default-features = false, features = ["windows-gnu-target", "simd-utf8"] }
+xberg = { version = "1.1.2", default-features = false, features = ["windows-gnu-target", "simd-utf8"] }
 
 # This crate deliberately does not use `[lints]` / `workspace = true`: its C-ABI /
 # PyO3 / napi / ext-php-rs / NIF boundary requires `unsafe` code, and the workspace's

data/ext/xberg_rb/src/lib.rs

@@ -1,5 +1,5 @@
 // This file is auto-generated by alef. DO NOT EDIT.
-// alef:hash:2633d345d440ea6f70d180e16fe231f2acf5cd4edb2ac17c3d97ac84ddd59763
+// alef:hash:4ce5add6ef19794221c2b201292d5f8a63d94b5413e0adb88c4b0379154e256f
 // Re-generate with: alef generate
 #![allow(dead_code, unused_imports, unused_variables)]
 #![allow(
@@ -32141,7 +32141,7 @@ impl TableGrid {
 #[magnus::wrap(class = "Xberg::TesseractConfig")]
 pub struct TesseractConfig {
     language: Vec<String>,
-    psm: i32,
+    psm: Option<i32>,
     output_format: String,
     oem: i32,
     min_confidence: f64,
@@ -32222,13 +32222,13 @@ impl TesseractConfig {
                 None => vec!["eng".to_string()],
             },
             psm: match kwargs.get(ruby.to_symbol("psm")) {
-                Some(v) => i32::try_convert(v).map_err(|e| {
+                Some(v) => Some(i32::try_convert(v).map_err(|e| {
                     magnus::Error::new(
                         unsafe { magnus::Ruby::get_unchecked() }.exception_type_error(),
                         format!("invalid value for `psm`: {}", e),
                     )
-                })?,
-                None => 3,
+                })?),
+                None => None,
             },
             output_format: match kwargs.get(ruby.to_symbol("output_format")) {
                 Some(v) => String::try_convert(v).map_err(|e| {
@@ -32408,7 +32408,7 @@ impl TesseractConfig {
         self.language.clone()
     }
 
-    fn psm(&self) -> i32 {
+    fn psm(&self) -> Option<i32> {
         self.psm
     }
 

data/lib/xberg/native.rb

@@ -1,5 +1,5 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:95dd25c8cd2698d5ee247c92e046162237595aeaacf861c48f18911ce66ec4cf
+# alef:hash:2ed8d28561d80117ab2d77bb163218ecfac8cd3a943b8656907f3fbc2a23425c
 # To regenerate: alef generate
 # To verify freshness: alef verify
 # frozen_string_literal: true
@@ -5295,15 +5295,23 @@ module Xberg
   ## Try the classical OCR backend first. If the quality score is below
   ## `quality_threshold`, send the page to the VLM.
   ##
-  ## `quality_threshold` is in the `[0.0, 1.0]` range produced by
-  ## `text::quality::calculate_quality_score`. A value of `0.5` is a
-  ## reasonable starting point; calibrate with the Stage 0 benchmark harness.
+  ## `quality_threshold` is in the `[0.0, 1.0]` range, but it is **not** the same
+  ## quantity reported on `score` (GH#1584).
+  ## The accept decision blends text-shape quality with confidence, weighted 0.7/0.3
+  ## (`extractors::pdf::ocr::pipeline_stage_score`) -- when the backend's confidence is on
+  ## a known scale it contributes only 30% of the compared score, so a page can clear this
+  ## threshold on clean-looking text even while its own `PageOcrConfidence.score` reads
+  ## below it. Do not calibrate this value by reading `PageOcrConfidence.score` off a
+  ## sample page and expecting an equal `quality_threshold` to reproduce the same
+  ## accept/reject outcome. A value of `0.5` is a reasonable starting point; calibrate with
+  ## the Stage 0 benchmark harness.
   VlmFallbackPolicyOnLowQuality = Data.define(:quality_threshold) do
     include VlmFallbackPolicy
     extend T::Sig
 
-    # Minimum acceptable quality score from the classical backend.
-    # Pages scoring below this are retried with VLM.
+    # Minimum acceptable quality score from the classical backend. Pages scoring below
+    # this are retried with VLM -- see this variant's doc comment for what "scoring"
+    # means here.
     sig { returns(Float) }
     # rubocop:disable Lint/UselessMethodDefinition
     def quality_threshold = super

data/lib/xberg/version.rb

@@ -1,10 +1,10 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:71dc9990e02ee56c24327e6bf2b5874c3d7c8a177543612ea54d00c7891b0a16
+# alef:hash:b27bcf116f7d5dec9c7278cd662f93e13ba51460490348bade51c096c52c1798
 # To regenerate: alef generate
 # To verify freshness: alef verify
 # frozen_string_literal: true
 
 module Xberg
   ## The version string for this package.
-  VERSION = "1.1.1"
+  VERSION = "1.1.2"
 end

data/sig/types.rbs

@@ -1,5 +1,5 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:f3e7d9702356d5d894258ff5f2a1b2558cabf46397da7c92dbf36f6d9a11f43e
+# alef:hash:8935255e8d7580a6b3409f10f303f1c20025de46e73a29d529837410d3a2acda
 # To regenerate: alef generate
 # To verify freshness: alef verify
 
@@ -2484,7 +2484,7 @@ def initialize: (?rows: Integer, ?cols: Integer, ?cells: Array[GridCell]) -> voi
 
     class TesseractConfig
     attr_reader language: Array[String]
-        attr_reader psm: Integer
+        attr_reader psm: Integer?
         attr_reader output_format: String
         attr_reader oem: Integer
         attr_reader min_confidence: Float

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: xberg
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.1.2
 platform: ruby
 authors:
 - Na'aman Hirschfeld <naaman@xberg.io>