html-to-markdown 2.23.4 → 2.24.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 204c58809e5e402d5fe7f4165cedf1bba22ac1b66866bd4ab7bdbd93109c5aaa
-  data.tar.gz: e9dabe4143e341129b2b08205fdd1cb47c70efdd4f6bf9848d3e4b82f7067b0e
+  metadata.gz: d52853c9ba91a063e69a508c1d75d9207d1bf1e86b1f1b116380ab016aa8c61a
+  data.tar.gz: 88b88a8730f8b775fc40fd3be833a26928caeefc48aaf7e90749b2b1942f3a65
 SHA512:
-  metadata.gz: 60e6c52864495463518ed6171b83b8efa734d1ce826462c81cc939112a67ea1ef6d03813e96ebd142cb4aeff1639f3b9072f9dfd07c81425c8861885ac6f7a95
-  data.tar.gz: '080bc5e1a9c0c83bcd5df465f836414edd52738861d13f9e749bc377189ff6cc8ddb4bdd8dc12209a303b679451bddbaed002c35b1c6212628e3627fd0c91d96'
+  metadata.gz: 15522f5476b3b204131bb0484a97f29d4aca42b433cafef79a89a4925c7e584c456900836188909086c94b0a4d2bd2716100703e7ebd04612307899d00c23cf6
+  data.tar.gz: f551468be46e15b4f173b58baeed51e6c039e6387a16a34654e9f6ce638bf213c3e79cfe80ea082445d7843e652c9fdc5f093fd96884198b8c75b6b94b626c7d

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    html-to-markdown (2.23.4)
+    html-to-markdown (2.24.1)
       rb_sys (>= 0.9, < 1.0)
 
 GEM
@@ -51,7 +51,7 @@ GEM
     parser (3.3.10.1)
       ast (~> 2.4.1)
       racc
-    prism (1.8.0)
+    prism (1.9.0)
     racc (1.8.1)
     rainbow (3.1.1)
     rake (13.3.1)
@@ -79,7 +79,7 @@ GEM
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
     rspec-support (3.13.6)
-    rubocop (1.82.1)
+    rubocop (1.84.0)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
       lint_roller (~> 1.1.0)
@@ -87,7 +87,7 @@ GEM
       parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 2.9.3, < 3.0)
-      rubocop-ast (>= 1.48.0, < 2.0)
+      rubocop-ast (>= 1.49.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 4.0)
     rubocop-ast (1.49.0)
@@ -159,7 +159,7 @@ CHECKSUMS
   ffi (1.17.3-x86_64-darwin) sha256=1f211811eb5cfaa25998322cdd92ab104bfbd26d1c4c08471599c511f2c00bb5
   ffi (1.17.3-x86_64-linux-gnu) sha256=3746b01f677aae7b16dc1acb7cb3cc17b3e35bdae7676a3f568153fb0e2c887f
   fileutils (1.8.0) sha256=8c6b1df54e2540bdb2f39258f08af78853aa70bad52b4d394bbc6424593c6e02
-  html-to-markdown (2.23.4)
+  html-to-markdown (2.24.1)
   i18n (1.14.8) sha256=285778639134865c5e0f6269e0b818256017e8cde89993fdfcbfb64d088824a5
   json (2.18.0) sha256=b10506aee4183f5cf49e0efc48073d7b75843ce3782c68dbeb763351c08fd505
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
@@ -170,7 +170,7 @@ CHECKSUMS
   mutex_m (0.3.0) sha256=cfcb04ac16b69c4813777022fdceda24e9f798e48092a2b817eb4c0a782b0751
   parallel (1.27.0) sha256=4ac151e1806b755fb4e2dc2332cbf0e54f2e24ba821ff2d3dcf86bf6dc4ae130
   parser (3.3.10.1) sha256=06f6a725d2cd91e5e7f2b7c32ba143631e1f7c8ae2fb918fc4cebec187e6a688
-  prism (1.8.0) sha256=84453a16ef5530ea62c5f03ec16b52a459575ad4e7b9c2b360fd8ce2c39c1254
+  prism (1.9.0) sha256=7b530c6a9f92c24300014919c9dcbc055bf4cdf51ec30aed099b06cd6674ef85
   racc (1.8.1) sha256=4a7f6929691dbec8b5209a0b373bc2614882b55fc5d2e447a21aaa691303d62f
   rainbow (3.1.1) sha256=039491aa3a89f42efa1d6dec2fc4e62ede96eb6acd95e52f1ad581182b79bc6a
   rake (13.3.1) sha256=8c9e89d09f66a26a01264e7e3480ec0607f0c497a861ef16063604b1b08eb19c
@@ -186,7 +186,7 @@ CHECKSUMS
   rspec-expectations (3.13.5) sha256=33a4d3a1d95060aea4c94e9f237030a8f9eae5615e9bd85718fe3a09e4b58836
   rspec-mocks (3.13.7) sha256=0979034e64b1d7a838aaaddf12bf065ea4dc40ef3d4c39f01f93ae2c66c62b1c
   rspec-support (3.13.6) sha256=2e8de3702427eab064c9352fe74488cc12a1bfae887ad8b91cba480ec9f8afb2
-  rubocop (1.82.1) sha256=09f1a6a654a960eda767aebea33e47603080f8e9c9a3f019bf9b94c9cab5e273
+  rubocop (1.84.0) sha256=88dec310153bb685a879f5a7cdb601f6287b8f0ee675d9dc63a17c7204c4190a
   rubocop-ast (1.49.0) sha256=49c3676d3123a0923d333e20c6c2dbaaae2d2287b475273fddee0c61da9f71fd
   rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33

data/README.md

@@ -18,13 +18,13 @@
     <img src="https://img.shields.io/maven-central/v/dev.kreuzberg/html-to-markdown?label=Java&color=007ec6" alt="Java">
   </a>
   <a href="https://pkg.go.dev/github.com/kreuzberg-dev/html-to-markdown/packages/go/v2/htmltomarkdown">
-    <img src="https://img.shields.io/badge/Go-v2.19.0-007ec6" alt="Go">
+    <img src="https://img.shields.io/badge/Go-v2.24.1-007ec6" alt="Go">
   </a>
   <a href="https://www.nuget.org/packages/KreuzbergDev.HtmlToMarkdown/">
     <img src="https://img.shields.io/nuget/v/KreuzbergDev.HtmlToMarkdown?label=C%23&color=007ec6" alt="C#">
   </a>
-  <a href="https://packagist.org/packages/goldziher/html-to-markdown">
-    <img src="https://img.shields.io/packagist/v/goldziher/html-to-markdown?label=PHP&color=007ec6" alt="PHP">
+  <a href="https://packagist.org/packages/kreuzberg-dev/html-to-markdown">
+    <img src="https://img.shields.io/packagist/v/kreuzberg-dev/html-to-markdown?label=PHP&color=007ec6" alt="PHP">
   </a>
   <a href="https://rubygems.org/gems/html-to-markdown">
     <img src="https://img.shields.io/gem/v/html-to-markdown?label=Ruby&color=007ec6" alt="Ruby">
@@ -169,20 +169,22 @@ The library supports converting HTML to [Djot](https://djot.net/), a lightweight
 
 ### Example Usage
 
-```python
-from html_to_markdown import convert, ConversionOptions
+
+```ruby
+require 'html_to_markdown'
 
 html = "<p>This is <strong>bold</strong> and <em>italic</em> text.</p>"
 
 # Default Markdown output
-markdown = convert(html)
+markdown = HtmlToMarkdown.convert(html)
 # Result: "This is **bold** and *italic* text."
 
 # Djot output
-djot = convert(html, ConversionOptions(output_format="djot"))
+djot = HtmlToMarkdown.convert(html, output_format: 'djot')
 # Result: "This is *bold* and _italic_ text."
 ```
 
+
 Djot's extended syntax allows you to express more semantic meaning in lightweight text, making it useful for documents that require strikethrough, insertion tracking, or mathematical notation.
 
 

data/ext/html-to-markdown-rb/native/Cargo.lock

@@ -148,9 +148,9 @@ checksum = "8f1fe948ff07f4bd06c30984e69f5b4899c516a3ef74f34df92a2df2ab535495"
 
 [[package]]
 name = "cc"
-version = "1.2.53"
+version = "1.2.54"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "755d2fce177175ffca841e9a06afdb2c4ab0f593d53b4dee48147dfaade85932"
+checksum = "6354c81bbfd62d9cfa9cb3c773c2b7b2a3a482d569de977fd0e961f6e7c00583"
 dependencies = [
  "find-msvc-tools",
  "shlex",
@@ -384,7 +384,7 @@ dependencies = [
 
 [[package]]
 name = "html-to-markdown-rb"
-version = "2.23.4"
+version = "2.24.1"
 dependencies = [
  "html-to-markdown-rs",
  "magnus",
@@ -837,9 +837,9 @@ dependencies = [
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.105"
+version = "1.0.106"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "535d180e0ecab6268a3e718bb9fd44db66bbbc256257165fc699dadf70d16fe7"
+checksum = "8fd00f0bb2e90d81d1044c2b32617f68fcb9fa3bb7640c23e9c748e53fb30934"
 dependencies = [
  "unicode-ident",
 ]
@@ -870,9 +870,9 @@ dependencies = [
 
 [[package]]
 name = "quote"
-version = "1.0.43"
+version = "1.0.44"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dc74d9a594b72ae6656596548f56f667211f8a97b3d4c3d467150794690dc40a"
+checksum = "21b2ebcf727b7760c461f091f9f0f539b77b8e87f2fd88131e7f1b433b3cece4"
 dependencies = [
  "proc-macro2",
 ]
@@ -1066,9 +1066,9 @@ checksum = "e320a6c5ad31d271ad523dcf3ad13e2767ad8b1cb8f047f75a8aeaf8da139da2"
 
 [[package]]
 name = "siphasher"
-version = "1.0.1"
+version = "1.0.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "56199f7ddabf13fe5074ce809e7d3f42b42ae711800501b5b16ea82ad029c39d"
+checksum = "b2aa850e253778c88a04c3d7323b043aeda9d3e30d5971937c1855769763678e"
 
 [[package]]
 name = "smallvec"
@@ -1219,9 +1219,9 @@ checksum = "1292c0d970b54115d14f2492fe0170adf21d68a1de108eebc51c1df4f346a091"
 
 [[package]]
 name = "uuid"
-version = "1.19.0"
+version = "1.20.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e2e054861b4bd027cd373e18e8d8d8e6548085000e41290d95ce0c373a654b4a"
+checksum = "ee48d38b119b0cd71fe4141b30f5ba9c7c5d9f4e7a3a8b4a674e4b6ef789976f"
 dependencies = [
  "js-sys",
  "wasm-bindgen",
@@ -1289,9 +1289,9 @@ dependencies = [
 
 [[package]]
 name = "web_atoms"
-version = "0.2.1"
+version = "0.2.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "30e588f10c7bc3465f5fc1ab087fc97877ec1064a7ec89fb685ac4ee998dac4a"
+checksum = "57a9779e9f04d2ac1ce317aee707aa2f6b773afba7b931222bff6983843b1576"
 dependencies = [
  "phf",
  "phf_codegen",
@@ -1366,18 +1366,18 @@ checksum = "cfe53a6657fd280eaa890a3bc59152892ffa3e30101319d168b781ed6529b049"
 
 [[package]]
 name = "zerocopy"
-version = "0.8.33"
+version = "0.8.35"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "668f5168d10b9ee831de31933dc111a459c97ec93225beb307aed970d1372dfd"
+checksum = "fdea86ddd5568519879b8187e1cf04e24fce28f7fe046ceecbce472ff19a2572"
 dependencies = [
  "zerocopy-derive",
 ]
 
 [[package]]
 name = "zerocopy-derive"
-version = "0.8.33"
+version = "0.8.35"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2c7962b26b0a8685668b671ee4b54d007a67d4eaf05fda79ac0ecf41e32270f1"
+checksum = "0c15e1b46eff7c6c91195752e0eeed8ef040e391cdece7c25376957d5f15df22"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -1386,9 +1386,9 @@ dependencies = [
 
 [[package]]
 name = "zmij"
-version = "1.0.16"
+version = "1.0.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dfcd145825aace48cff44a8844de64bf75feec3080e0aa5cdbde72961ae51a65"
+checksum = "02aae0f83f69aafc94776e879363e9771d7ecbffe2c7fbb6c14c5e00dfe88439"
 
 [[package]]
 name = "zune-core"
@@ -1398,9 +1398,9 @@ checksum = "cb8a0807f7c01457d0379ba880ba6322660448ddebc890ce29bb64da71fb40f9"
 
 [[package]]
 name = "zune-jpeg"
-version = "0.5.11"
+version = "0.5.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2959ca473aae96a14ecedf501d20b3608d2825ba280d5adb57d651721885b0c2"
+checksum = "410e9ecef634c709e3831c2cfdb8d9c32164fae1c67496d5b68fff728eec37fe"
 dependencies = [
  "zune-core",
 ]

data/ext/html-to-markdown-rb/native/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "html-to-markdown-rb"
-version ="2.23.4"
+version ="2.24.1"
 edition = "2024"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
 license = "MIT"
@@ -20,8 +20,6 @@ crate-type = ["cdylib", "rlib"]
 [dependencies]
 html-to-markdown-rs = { path = "../../../rust-vendor/html-to-markdown-rs", features = ["inline-images", "visitor", "metadata"] }
 magnus = { git = "https://github.com/matsadler/magnus", rev = "f6db11769efb517427bf7f121f9c32e18b059b38", features = ["rb-sys"] }
-
-[target.'cfg(not(target_os = "windows"))'.dependencies]
 pprof = { version = "0.15", features = ["flamegraph"], optional = true }
 
 [dev-dependencies]

data/lib/html_to_markdown/cli_proxy.rb

@@ -5,8 +5,11 @@ require 'pathname'
 
 module HtmlToMarkdown
   module CLIProxy
-    Error = Class.new(StandardError)
-    MissingBinaryError = Class.new(Error)
+    class Error < StandardError
+    end
+
+    class MissingBinaryError < Error
+    end
 
     class CLIExecutionError < Error
       attr_reader :stderr, :status

data/lib/html_to_markdown/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HtmlToMarkdown
-  VERSION = '2.23.4'
+  VERSION = '2.24.1'
 end

data/rust-vendor/cc/.cargo-checksum.json

@@ -1 +1 @@
-{"files":{".cargo_vcs_info.json":"180fb62a60d18d0593b0fc92d2d5cce9cb1a0d648839a317c2cf3bba57c744a6","CHANGELOG.md":"f7c860164ad789e1c4e35aafaff755530d873d35036a819e841ee20b9cb4a817","Cargo.lock":"497c499e308a7b0473bfd575197740bb1fd72483a7926953793abe18e4aa6fbb","Cargo.toml":"ea7867647b43346615f986e1952d00b2fae84f660e6ddf1a921d12fe66f9842d","Cargo.toml.orig":"83f378d0731eee4f911ae9e4a177263bb0471198475c7f907ae706a73c616c7a","LICENSE-APACHE":"a60eea817514531668d7e00765731449fe14d059d3249e0bc93b36de45f759f2","LICENSE-MIT":"378f5840b258e2779c39418f3f2d7b2ba96f1c7917dd6be0713f88305dbda397","README.md":"cc952a506dc726eed3e128d47d62defbac4470bbb0e465f6a043ae8bf0c96d19","clippy.toml":"aa7850db4350883c8f373bd0d6b4d19bf3b75f13c1c238e24368c109cb52fb1d","src/command_helpers.rs":"2e53ac6e632845157829c3cb8845c3301eb00d57567e3607884a624715024dab","src/detect_compiler_family.c":"97ca4b021495611e828becea6187add37414186a16dfedd26c2947cbce6e8b2f","src/flags.rs":"11f22aa9741b0d445b19ab29ea1076cc32a4143316a74855b6bd682004217543","src/lib.rs":"a8fa06bbb9573f5a657433e437c3f9e1470d2447d8723ddc719f16dd55073dcc","src/parallel/async_executor.rs":"4ce24435fff6b6555b43fee042c16bd65d4150d0346567f246b9190d85b45983","src/parallel/command_runner.rs":"1d17d9a037de93b31cc7ee6a914a1e1f6c1ac156cf8619f0014dcbaca6a0b215","src/parallel/job_token.rs":"6cceffcf0f7e7d82d775287e88a4f9a62dc690c321933204f9b531cf40072bf1","src/parallel/mod.rs":"bf156170790d40a76015a8c78cc2a391cbe1b60bb96c1fac1d047d26e342386a","src/parallel/stderr.rs":"840fe8d0ba613c98fa92cbebc374a028a3212ba7dcea58285483223a31acc7eb","src/target.rs":"a9d2d347a3a08015531f00b32bc8a7478f5636822cc8e0237ea9efa3f07d4cd4","src/target/apple.rs":"da9411b2c4db419e0fa39f765ee53b8665b3837fb39827679a53238734a4a1c1","src/target/generated.rs":"7eb87ba6f8c60901cfd09221afed60bb2fe639ac2471d7ad6b92ce85b3b556b0","src/target/llvm.rs":"190fe8d2b204cd4a6e68f2a1aada17ecd6390799564ed25792fcc08ab34710ba","src/target/parser.rs":"fb18f00f92c6a5160b208317d8739bacab687da18716fa8a50cea0d4271add2e","src/tempfile.rs":"3d9a4bd894862a345aa230a61ec266f0c68f4ef9713d1d9c727482e61f1ea7c3","src/tool.rs":"1d279a6f0738f9164ca794c85bd55951aa95e2adc788e75a6bb8133393d6cce8","src/utilities.rs":"52b30b24a1c31cdefb105309ee5220cfc9fca76eaf4e6d6509c3e19f431448fe"},"package":"755d2fce177175ffca841e9a06afdb2c4ab0f593d53b4dee48147dfaade85932"}
+{"files":{".cargo_vcs_info.json":"1076e91b3aa51afd96889c343752acf6d3bed25dced0d88b0e2ad265e4f2aa3a","CHANGELOG.md":"ed73228f983d352b0c567120f9d200c82bbf096e1ba08fb8ef7ffd2b0761bb2b","Cargo.lock":"8b0dfc5676cb4ce8af8c9f55c352949131c165bf1430a89b73076384fabc6924","Cargo.toml":"e6ffb9cb188f8b83e937bb0d9773b36749624a68aae8d4e176ea84e1c3a0a5bd","Cargo.toml.orig":"a9e316d9ea93dc4931fd29b641a37d0395645f579361d0de5cd36c8de90c00fa","LICENSE-APACHE":"a60eea817514531668d7e00765731449fe14d059d3249e0bc93b36de45f759f2","LICENSE-MIT":"378f5840b258e2779c39418f3f2d7b2ba96f1c7917dd6be0713f88305dbda397","README.md":"cc952a506dc726eed3e128d47d62defbac4470bbb0e465f6a043ae8bf0c96d19","clippy.toml":"aa7850db4350883c8f373bd0d6b4d19bf3b75f13c1c238e24368c109cb52fb1d","src/command_helpers.rs":"2e53ac6e632845157829c3cb8845c3301eb00d57567e3607884a624715024dab","src/detect_compiler_family.c":"97ca4b021495611e828becea6187add37414186a16dfedd26c2947cbce6e8b2f","src/flags.rs":"11f22aa9741b0d445b19ab29ea1076cc32a4143316a74855b6bd682004217543","src/lib.rs":"a8fa06bbb9573f5a657433e437c3f9e1470d2447d8723ddc719f16dd55073dcc","src/parallel/async_executor.rs":"4ce24435fff6b6555b43fee042c16bd65d4150d0346567f246b9190d85b45983","src/parallel/command_runner.rs":"1d17d9a037de93b31cc7ee6a914a1e1f6c1ac156cf8619f0014dcbaca6a0b215","src/parallel/job_token.rs":"6cceffcf0f7e7d82d775287e88a4f9a62dc690c321933204f9b531cf40072bf1","src/parallel/mod.rs":"bf156170790d40a76015a8c78cc2a391cbe1b60bb96c1fac1d047d26e342386a","src/parallel/stderr.rs":"840fe8d0ba613c98fa92cbebc374a028a3212ba7dcea58285483223a31acc7eb","src/target.rs":"a9d2d347a3a08015531f00b32bc8a7478f5636822cc8e0237ea9efa3f07d4cd4","src/target/apple.rs":"da9411b2c4db419e0fa39f765ee53b8665b3837fb39827679a53238734a4a1c1","src/target/generated.rs":"1098456df6be02eea133785236080efc04bc4318e8581cf4199b8f1ab2a3fab9","src/target/llvm.rs":"190fe8d2b204cd4a6e68f2a1aada17ecd6390799564ed25792fcc08ab34710ba","src/target/parser.rs":"0b901fbc3d18f663cb00741fcac6dc390b28c5286c99c1c8ec1727b44563e529","src/tempfile.rs":"3d9a4bd894862a345aa230a61ec266f0c68f4ef9713d1d9c727482e61f1ea7c3","src/tool.rs":"1d279a6f0738f9164ca794c85bd55951aa95e2adc788e75a6bb8133393d6cce8","src/utilities.rs":"52b30b24a1c31cdefb105309ee5220cfc9fca76eaf4e6d6509c3e19f431448fe"},"package":"6354c81bbfd62d9cfa9cb3c773c2b7b2a3a482d569de977fd0e961f6e7c00583"}

data/rust-vendor/cc/.cargo_vcs_info.json

@@ -1,6 +1,6 @@
 {
   "git": {
-    "sha1": "5be292b6bbaddda74e2196b5679d9e7ba5834604"
+    "sha1": "85466d853fbdb288e1e2c627539bcfc035fa2ed3"
   },
   "path_in_vcs": ""
 }

data/rust-vendor/cc/CHANGELOG.md

@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.2.54](https://github.com/rust-lang/cc-rs/compare/cc-v1.2.53...cc-v1.2.54) - 2026-01-23
+
+### Other
+
+- Fix x86_64-unknown-linux-gnuasan parsing ([#1661](https://github.com/rust-lang/cc-rs/pull/1661))
+- Regenerate target info ([#1660](https://github.com/rust-lang/cc-rs/pull/1660))
+
 ## [1.2.53](https://github.com/rust-lang/cc-rs/compare/cc-v1.2.52...cc-v1.2.53) - 2026-01-16
 
 ### Other

data/rust-vendor/cc/Cargo.lock

@@ -10,7 +10,7 @@ checksum = "812e12b5285cc515a9c72a5c1d3b6d46a19dac5acfef5265968c166106e31dd3"
 
 [[package]]
 name = "cc"
-version = "1.2.53"
+version = "1.2.54"
 dependencies = [
  "find-msvc-tools",
  "jobserver",

data/rust-vendor/cc/Cargo.toml

@@ -13,7 +13,7 @@
 edition = "2018"
 rust-version = "1.63"
 name = "cc"
-version = "1.2.53"
+version = "1.2.54"
 authors = ["Alex Crichton <alex@alexcrichton.com>"]
 build = false
 exclude = [

data/rust-vendor/cc/Cargo.toml.orig

@@ -1,6 +1,6 @@
 [package]
 name = "cc"
-version = "1.2.53"
+version = "1.2.54"
 authors = ["Alex Crichton <alex@alexcrichton.com>"]
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/rust-lang/cc-rs"

data/rust-vendor/cc/src/target/generated.rs

@@ -296,6 +296,7 @@ pub(crate) const LLVM_TARGETS: &[(&str, &str)] = &[
     ("x86_64-unknown-illumos", "x86_64-pc-solaris"),
     ("x86_64-unknown-l4re-uclibc", "x86_64-unknown-l4re-gnu"),
     ("x86_64-unknown-linux-gnu", "x86_64-unknown-linux-gnu"),
+    ("x86_64-unknown-linux-gnuasan", "x86_64-unknown-linux-gnu"),
     ("x86_64-unknown-linux-gnux32", "x86_64-unknown-linux-gnux32"),
     ("x86_64-unknown-linux-musl", "x86_64-unknown-linux-musl"),
     ("x86_64-unknown-linux-none", "x86_64-unknown-linux-none"),

data/rust-vendor/cc/src/target/parser.rs

@@ -445,6 +445,10 @@ impl<'a> TargetInfo<'a> {
             abi = "elfv2";
         }
 
+        if abi == "asan" {
+            abi = "";
+        }
+
         Ok(Self {
             full_arch,
             arch,

data/rust-vendor/html-to-markdown-rs/README.md

@@ -8,7 +8,7 @@ This crate is the core engine compiled into the Python wheels, Ruby gem, Node.js
 [![npm version](https://img.shields.io/npm/v/html-to-markdown-node.svg?logo=npm)](https://www.npmjs.com/package/html-to-markdown-node)
 [![PyPI version](https://img.shields.io/pypi/v/html-to-markdown.svg?logo=pypi)](https://pypi.org/project/html-to-markdown/)
 [![Gem Version](https://badge.fury.io/rb/html-to-markdown.svg)](https://rubygems.org/gems/html-to-markdown)
-[![Packagist](https://img.shields.io/packagist/v/goldziher/html-to-markdown.svg)](https://packagist.org/packages/goldziher/html-to-markdown)
+[![Packagist](https://img.shields.io/packagist/v/kreuzberg-dev/html-to-markdown.svg)](https://packagist.org/packages/kreuzberg-dev/html-to-markdown)
 [![docs.rs](https://docs.rs/html-to-markdown-rs/badge.svg)](https://docs.rs/html-to-markdown-rs)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://github.com/kreuzberg-dev/html-to-markdown/blob/main/LICENSE)
 

data/rust-vendor/html-to-markdown-rs/src/convert_api.rs

@@ -8,7 +8,7 @@ use std::borrow::Cow;
 use crate::error::Result;
 use crate::options::{ConversionOptions, WhitespaceMode};
 use crate::text;
-use crate::validation::validate_input;
+use crate::validation::{Utf16Encoding, detect_utf16_encoding, validate_input};
 use crate::{ConversionError, ConversionOptionsUpdate};
 
 #[cfg(feature = "visitor")]
@@ -40,10 +40,9 @@ use crate::{HtmlExtraction, InlineImageConfig};
 ///
 /// Returns an error if HTML parsing fails or if the input contains invalid UTF-8.
 pub fn convert(html: &str, options: Option<ConversionOptions>) -> Result<String> {
-    validate_input(html)?;
     let options = options.unwrap_or_default();
 
-    let normalized_html = normalize_line_endings(html);
+    let normalized_html = normalize_input(html)?;
 
     if !options.wrap {
         if let Some(markdown) = fast_text_only(normalized_html.as_ref(), &options) {
@@ -84,10 +83,9 @@ pub fn convert_with_inline_images(
     use std::cell::RefCell;
     use std::rc::Rc;
 
-    validate_input(html)?;
     let options = options.unwrap_or_default();
 
-    let normalized_html = normalize_line_endings(html);
+    let normalized_html = normalize_input(html)?;
 
     let collector = Rc::new(RefCell::new(crate::inline_images::InlineImageCollector::new(
         image_cfg,
@@ -257,10 +255,9 @@ pub fn convert_with_metadata(
     use std::cell::RefCell;
     use std::rc::Rc;
 
-    validate_input(html)?;
     let options = options.unwrap_or_default();
+    let normalized_html = normalize_input(html)?;
     if !metadata_cfg.any_enabled() {
-        let normalized_html = normalize_line_endings(html);
         #[cfg(feature = "visitor")]
         let markdown = crate::converter::convert_html_impl(normalized_html.as_ref(), &options, None, None, visitor)?;
         #[cfg(not(feature = "visitor"))]
@@ -273,8 +270,6 @@ pub fn convert_with_metadata(
         return Ok((markdown, ExtendedMetadata::default()));
     }
 
-    let normalized_html = normalize_line_endings(html);
-
     let metadata_collector = Rc::new(RefCell::new(crate::metadata::MetadataCollector::new(metadata_cfg)));
 
     #[cfg(feature = "visitor")]
@@ -332,8 +327,8 @@ pub fn convert_with_metadata(
 ///     fn visit_code_block(
 ///         &mut self,
 ///         _ctx: &NodeContext,
-///         code: &str,
 ///         language: Option<&str>,
+///         code: &str,
 ///     ) -> VisitResult {
 ///         VisitResult::Custom(format!("```{}\n{}\n```", language.unwrap_or(""), code))
 ///     }
@@ -352,10 +347,9 @@ pub fn convert_with_visitor(
     options: Option<ConversionOptions>,
     visitor: Option<visitor::VisitorHandle>,
 ) -> Result<String> {
-    validate_input(html)?;
     let options = options.unwrap_or_default();
 
-    let normalized_html = normalize_line_endings(html);
+    let normalized_html = normalize_input(html)?;
 
     let markdown = crate::converter::convert_html_with_visitor(normalized_html.as_ref(), &options, visitor)?;
 
@@ -412,8 +406,8 @@ pub fn convert_with_visitor(
 ///     async fn visit_code_block(
 ///         &mut self,
 ///         _ctx: &NodeContext,
-///         code: &str,
 ///         language: Option<&str>,
+///         code: &str,
 ///     ) -> VisitResult {
 ///         // Can perform async operations here (e.g., syntax highlighting via service)
 ///         VisitResult::Custom(format!("```{}\n{}\n```", language.unwrap_or(""), code))
@@ -433,10 +427,9 @@ pub async fn convert_with_async_visitor(
     options: Option<ConversionOptions>,
     visitor: Option<visitor_helpers::AsyncVisitorHandle>,
 ) -> Result<String> {
-    validate_input(html)?;
     let options = options.unwrap_or_default();
 
-    let normalized_html = normalize_line_endings(html);
+    let normalized_html = normalize_input(html)?;
 
     // Use the async implementation that properly awaits visitor callbacks
     let markdown =
@@ -449,6 +442,88 @@ pub async fn convert_with_async_visitor(
     }
 }
 
+/// Validate and normalize HTML input for conversion.
+fn normalize_input(html: &str) -> Result<Cow<'_, str>> {
+    let decoded = decode_utf16_if_needed(html);
+    match decoded {
+        Cow::Borrowed(borrowed) => {
+            validate_input(borrowed)?;
+            let sanitized = strip_nul_bytes(borrowed);
+            match sanitized {
+                Cow::Borrowed(b) => Ok(normalize_line_endings(b)),
+                Cow::Owned(o) => Ok(Cow::Owned(normalize_line_endings(&o).into_owned())),
+            }
+        }
+        Cow::Owned(mut owned) => {
+            validate_input(&owned)?;
+            if owned.contains('\0') {
+                owned = owned.replace('\0', "");
+            }
+            if owned.contains('\r') {
+                owned = owned.replace("\r\n", "\n").replace('\r', "\n");
+            }
+            Ok(Cow::Owned(owned))
+        }
+    }
+}
+
+/// Attempt to decode UTF-16 HTML that was provided as a lossy UTF-8 string.
+///
+/// Some callers read raw bytes and convert with `from_utf8_lossy`, which preserves
+/// the NUL-byte pattern of UTF-16 input. When we detect that pattern, we can
+/// recover the original HTML instead of rejecting it as binary data.
+fn decode_utf16_if_needed(html: &str) -> Cow<'_, str> {
+    let bytes = html.as_bytes();
+    if !bytes.contains(&0) {
+        return Cow::Borrowed(html);
+    }
+
+    let Some(encoding) = detect_utf16_encoding(bytes) else {
+        return Cow::Borrowed(html);
+    };
+
+    let decoded = decode_utf16_bytes(bytes, encoding);
+    if decoded.is_empty() {
+        Cow::Borrowed(html)
+    } else {
+        Cow::Owned(decoded)
+    }
+}
+
+fn decode_utf16_bytes(bytes: &[u8], encoding: Utf16Encoding) -> String {
+    let (is_little_endian, skip_bom) = match encoding {
+        Utf16Encoding::BomLe => (true, true),
+        Utf16Encoding::BomBe => (false, true),
+        Utf16Encoding::NoBomLe => (true, false),
+        Utf16Encoding::NoBomBe => (false, false),
+    };
+
+    let mut units = Vec::with_capacity(bytes.len() / 2);
+    for chunk in bytes.chunks_exact(2) {
+        let unit = if is_little_endian {
+            u16::from_le_bytes([chunk[0], chunk[1]])
+        } else {
+            u16::from_be_bytes([chunk[0], chunk[1]])
+        };
+        units.push(unit);
+    }
+
+    let mut decoded = String::from_utf16_lossy(&units);
+    if skip_bom {
+        decoded = decoded.trim_start_matches('\u{FEFF}').to_string();
+    }
+    decoded
+}
+
+/// Strip NUL bytes that can appear in malformed HTML inputs.
+fn strip_nul_bytes(html: &str) -> Cow<'_, str> {
+    if html.contains('\0') {
+        Cow::Owned(html.replace('\0', ""))
+    } else {
+        Cow::Borrowed(html)
+    }
+}
+
 /// Normalize line endings in HTML input.
 ///
 /// Converts CRLF and CR line endings to LF for consistent processing.

data/rust-vendor/html-to-markdown-rs/src/converter/block/preformatted.rs

@@ -268,7 +268,7 @@ fn dedent_code_block(content: &str) -> String {
     let min_indent = lines
         .iter()
         .filter(|line| !line.trim().is_empty())
-        .map(|line| line.len() - line.trim_start().len())
+        .map(|line| line.chars().take_while(|c| c.is_whitespace()).count())
         .min()
         .unwrap_or(0);
 
@@ -277,10 +277,21 @@ fn dedent_code_block(content: &str) -> String {
         .map(|line| {
             if line.trim().is_empty() {
                 String::new()
-            } else if line.len() >= min_indent {
-                line[min_indent..].to_string()
             } else {
-                line.to_string()
+                let mut remaining = min_indent;
+                let mut cut = 0;
+                for (idx, ch) in line.char_indices() {
+                    if remaining == 0 {
+                        break;
+                    }
+                    if ch.is_whitespace() {
+                        remaining -= 1;
+                        cut = idx + ch.len_utf8();
+                    } else {
+                        break;
+                    }
+                }
+                line[cut..].to_string()
             }
         })
         .collect();

data/rust-vendor/html-to-markdown-rs/src/converter/handlers/graphic.rs

@@ -35,7 +35,7 @@ pub fn handle_graphic(
     tag: &tl::HTMLTag,
     parser: &tl::Parser,
     output: &mut String,
-    _options: &ConversionOptions,
+    options: &ConversionOptions,
     ctx: &Context,
     depth: usize,
     dom_ctx: &DomContext,
@@ -158,8 +158,10 @@ pub fn handle_graphic(
         should_use_alt_text,
     ));
 
-    if let Some(graphic_text) = graphic_output {
-        output.push_str(&graphic_text);
+    if !options.skip_images {
+        if let Some(graphic_text) = graphic_output {
+            output.push_str(&graphic_text);
+        }
     }
 
     // Add graphic to metadata collector

data/rust-vendor/html-to-markdown-rs/src/converter/media/svg.rs

@@ -174,7 +174,7 @@ pub(crate) fn handle_svg(
     tag: &tl::HTMLTag,
     parser: &Parser,
     output: &mut String,
-    _options: &crate::options::ConversionOptions,
+    options: &crate::options::ConversionOptions,
     ctx: &super::Context,
     _depth: usize,
     dom_ctx: &super::DomContext,
@@ -216,6 +216,10 @@ pub(crate) fn handle_svg(
         handle_inline_svg(collector_ref, node_handle, parser, title_opt, attributes_map);
     }
 
+    if options.skip_images {
+        return;
+    }
+
     if ctx.convert_as_inline {
         output.push_str(&title);
     } else {