html-to-markdown 2.26.1 → 2.26.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 48fbfb1438c5ad97a8984bbd111b90332abc08e7c2540709dfc6c3f02daea6c5
-  data.tar.gz: 7d579f3f883a4fb9d99ce6b8838e0b51c22ecb0c0a8759ca7bb3d01b723aa7d9
+  metadata.gz: 23d0242cd4fc575d8081e675fb8d16f09faa7fb1c6c0df9b18d21338c0391880
+  data.tar.gz: cf86724440a34a26e1f17c134a232b1b321edaacb95758e42f9eab59dc710f8b
 SHA512:
-  metadata.gz: 37ada8ad47408d91d98f065cf598ddfe04baa46900a92939f19f1290f54991cf59e4471b855d12889171d4e478ce2f9fc49bff938d814f8c3a3f4cd8ba009ebc
-  data.tar.gz: 3b3653ffc39d693111aa35fb8d7a255338801256856b55fe4e7933373b4cf06fd69b838d4177c046ee41ffc7654e36075dc6d18b1a58fa1e44b83ce6d225abb6
+  metadata.gz: 63afe8bdf9d36f4cc225859e3a7ebb62452e97feafccc5ea1e20564a47b7e037900b6af7300508eec4313d4306608e3b00bc5f6a3115001ee250d5c560880bb6
+  data.tar.gz: c8fcaa6e61fea4325b08ce39ebf0a2bd92ff4e6d58702497e38cc6a081c1eec3de095d72986ccdc828ee6e219697d53c12483087ee563d39f767fe989d72ffdb

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    html-to-markdown (2.26.1)
+    html-to-markdown (2.26.3)
       rb_sys (>= 0.9, < 1.0)
 
 GEM
@@ -20,6 +20,8 @@ GEM
       securerandom (>= 0.3)
       tzinfo (~> 2.0, >= 2.0.5)
       uri (>= 0.13.1)
+    addressable (2.8.9)
+      public_suffix (>= 2.0.2, < 8.0)
     ast (2.4.3)
     base64 (0.3.0)
     bigdecimal (4.0.1)
@@ -37,6 +39,9 @@ GEM
     i18n (1.14.8)
       concurrent-ruby (~> 1.0)
     json (2.18.1)
+    json-schema (6.1.0)
+      addressable (~> 2.8)
+      bigdecimal (>= 3.1, < 5)
     language_server-protocol (3.17.0.5)
     lint_roller (1.1.0)
     listen (3.10.0)
@@ -44,14 +49,18 @@ GEM
       rb-fsevent (~> 0.10, >= 0.10.3)
       rb-inotify (~> 0.9, >= 0.9.10)
     logger (1.7.0)
-    minitest (6.0.1)
+    mcp (0.7.1)
+      json-schema (>= 4.1)
+    minitest (6.0.2)
+      drb (~> 2.0)
       prism (~> 1.5)
     mutex_m (0.3.0)
     parallel (1.27.0)
-    parser (3.3.10.1)
+    parser (3.3.10.2)
       ast (~> 2.4.1)
       racc
     prism (1.9.0)
+    public_suffix (7.0.2)
     racc (1.8.1)
     rainbow (3.1.1)
     rake (13.3.1)
@@ -76,14 +85,15 @@ GEM
     rspec-expectations (3.13.5)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
-    rspec-mocks (3.13.7)
+    rspec-mocks (3.13.8)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
     rspec-support (3.13.7)
-    rubocop (1.84.2)
+    rubocop (1.85.0)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
       lint_roller (~> 1.1.0)
+      mcp (~> 0.6)
       parallel (~> 1.10)
       parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
@@ -147,6 +157,7 @@ DEPENDENCIES
 
 CHECKSUMS
   activesupport (8.1.2) sha256=88842578ccd0d40f658289b0e8c842acfe9af751afee2e0744a7873f50b6fdae
+  addressable (2.8.9) sha256=cc154fcbe689711808a43601dee7b980238ce54368d23e127421753e46895485
   ast (2.4.3) sha256=954615157c1d6a382bc27d690d973195e79db7f55e9765ac7c481c60bdb4d383
   base64 (0.3.0) sha256=27337aeabad6ffae05c265c450490628ef3ebd4b67be58257393227588f5a97b
   bigdecimal (4.0.1) sha256=8b07d3d065a9f921c80ceaea7c9d4ae596697295b584c296fe599dd0ad01c4a7
@@ -161,18 +172,21 @@ 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 (2.26.1)
+  html-to-markdown (2.26.3)
   i18n (1.14.8) sha256=285778639134865c5e0f6269e0b818256017e8cde89993fdfcbfb64d088824a5
   json (2.18.1) sha256=fe112755501b8d0466b5ada6cf50c8c3f41e897fa128ac5d263ec09eedc9f986
+  json-schema (6.1.0) sha256=6bf70a2cfb6dfd5a06da28093fa8190f324c88eabd36a7f47097f227321dc702
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   listen (3.10.0) sha256=c6e182db62143aeccc2e1960033bebe7445309c7272061979bb098d03760c9d2
   logger (1.7.0) sha256=196edec7cc44b66cfb40f9755ce11b392f21f7967696af15d274dde7edff0203
-  minitest (6.0.1) sha256=7854c74f48e2e975969062833adc4013f249a4b212f5e7b9d5c040bf838d54bb
+  mcp (0.7.1) sha256=fa967895d6952bad0d981ea907731d8528d2c246d2079d56a9c8bae83d14f1c7
+  minitest (6.0.2) sha256=db6e57956f6ecc6134683b4c87467d6dd792323c7f0eea7b93f66bd284adbc3d
   mutex_m (0.3.0) sha256=cfcb04ac16b69c4813777022fdceda24e9f798e48092a2b817eb4c0a782b0751
   parallel (1.27.0) sha256=4ac151e1806b755fb4e2dc2332cbf0e54f2e24ba821ff2d3dcf86bf6dc4ae130
-  parser (3.3.10.1) sha256=06f6a725d2cd91e5e7f2b7c32ba143631e1f7c8ae2fb918fc4cebec187e6a688
+  parser (3.3.10.2) sha256=6f60c84aa4bdcedb6d1a2434b738fe8a8136807b6adc8f7f53b97da9bc4e9357
   prism (1.9.0) sha256=7b530c6a9f92c24300014919c9dcbc055bf4cdf51ec30aed099b06cd6674ef85
+  public_suffix (7.0.2) sha256=9114090c8e4e7135c1fd0e7acfea33afaab38101884320c65aaa0ffb8e26a857
   racc (1.8.1) sha256=4a7f6929691dbec8b5209a0b373bc2614882b55fc5d2e447a21aaa691303d62f
   rainbow (3.1.1) sha256=039491aa3a89f42efa1d6dec2fc4e62ede96eb6acd95e52f1ad581182b79bc6a
   rake (13.3.1) sha256=8c9e89d09f66a26a01264e7e3480ec0607f0c497a861ef16063604b1b08eb19c
@@ -186,9 +200,9 @@ CHECKSUMS
   rspec (3.13.2) sha256=206284a08ad798e61f86d7ca3e376718d52c0bc944626b2349266f239f820587
   rspec-core (3.13.6) sha256=a8823c6411667b60a8bca135364351dda34cd55e44ff94c4be4633b37d828b2d
   rspec-expectations (3.13.5) sha256=33a4d3a1d95060aea4c94e9f237030a8f9eae5615e9bd85718fe3a09e4b58836
-  rspec-mocks (3.13.7) sha256=0979034e64b1d7a838aaaddf12bf065ea4dc40ef3d4c39f01f93ae2c66c62b1c
+  rspec-mocks (3.13.8) sha256=086ad3d3d17533f4237643de0b5c42f04b66348c28bf6b9c2d3f4a3b01af1d47
   rspec-support (3.13.7) sha256=0640e5570872aafefd79867901deeeeb40b0c9875a36b983d85f54fb7381c47c
-  rubocop (1.84.2) sha256=5692cea54168f3dc8cb79a6fe95c5424b7ea893c707ad7a4307b0585e88dbf5f
+  rubocop (1.85.0) sha256=317407feb681a07d54f64d2f9e1d6b6af1ce7678e51cd658e3ad8bd66da48c01
   rubocop-ast (1.49.0) sha256=49c3676d3123a0923d333e20c6c2dbaaae2d2287b475273fddee0c61da9f71fd
   rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33

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

@@ -1,6 +1,6 @@
 [package]
 name = "html-to-markdown-rb"
-version ="2.26.1"
+version ="2.26.3"
 edition = "2024"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
 license = "MIT"

data/lib/html_to_markdown/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HtmlToMarkdown
-  VERSION = '2.26.1'
+  VERSION = '2.26.3'
 end

data/rust-vendor/html-to-markdown-rs/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "html-to-markdown-rs"
-version = "2.26.1"
+version = "2.26.3"
 edition = "2024"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
 license = "MIT"

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

@@ -61,6 +61,7 @@ pub(crate) fn handle(
 
     let p_ctx = Context {
         in_paragraph: true,
+        block_content_start: output.len(),
         ..ctx.clone()
     };
 

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

@@ -63,6 +63,10 @@ pub struct Context {
     pub(crate) heading_allow_inline_images: bool,
     /// Are we inside a paragraph element?
     pub(crate) in_paragraph: bool,
+    /// Output buffer position where the current block's content starts.
+    /// Used to distinguish paragraph-break newlines from a previous block
+    /// vs. newlines generated within the current block.
+    pub(crate) block_content_start: usize,
     /// Are we inside a ruby element?
     pub(crate) in_ruby: bool,
     /// Are we inside a `<strong>` / `<b>` element?
@@ -152,6 +156,7 @@ impl Context {
             in_heading: false,
             heading_allow_inline_images: false,
             in_paragraph: false,
+            block_content_start: 0,
             in_ruby: false,
             in_strong: false,
             in_link: false,

data/rust-vendor/html-to-markdown-rs/src/converter/inline/semantic/typography.rs

@@ -87,6 +87,8 @@ pub fn handle_subscript(
             } else {
                 output.push_str(&options.sub_symbol);
             }
+        } else {
+            output.push_str(trimmed);
         }
         append_inline_suffix(output, suffix, !trimmed.is_empty(), node_handle, parser, dom_ctx);
     }
@@ -139,6 +141,8 @@ pub fn handle_superscript(
             } else {
                 output.push_str(&options.sup_symbol);
             }
+        } else {
+            output.push_str(trimmed);
         }
         append_inline_suffix(output, suffix, !trimmed.is_empty(), node_handle, parser, dom_ctx);
     }

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

@@ -82,6 +82,12 @@ pub fn process_text_node(
             if !output.ends_with("\n\n") {
                 if let Some(next_tag) = get_next_sibling_tag(node_handle, parser, dom_ctx) {
                     if is_inline_element(next_tag) {
+                        // Newlines between inline elements collapse to a single space
+                        // in HTML rendering (per CSS white-space: normal). Preserve
+                        // this word boundary so adjacent inline content doesn't merge.
+                        if !output.ends_with(' ') && !output.ends_with('\n') {
+                            output.push(' ');
+                        }
                         return;
                     }
                 }
@@ -178,7 +184,12 @@ pub fn process_text_node(
         if !suffix.is_empty() {
             final_text.push_str(suffix);
         } else if has_trailing_single_newline {
-            let at_paragraph_break = output.ends_with("\n\n");
+            // Check if the "\n\n" at the end of the output buffer came from within
+            // the current block's content, not from a previous block's closing.
+            // Without this distinction, the second paragraph after a "\n\n" boundary
+            // would incorrectly suppress the trailing space before inline elements.
+            let current_block_output = &output[ctx.block_content_start..];
+            let at_paragraph_break = current_block_output.ends_with("\n\n");
             if !at_paragraph_break {
                 if has_double_newline {
                     final_text.push('\n');

data/rust-vendor/html-to-markdown-rs/tests/integration_test.rs

@@ -373,6 +373,66 @@ fn test_superscript_leading_whitespace() {
     assert_eq!(result, "hello ^world^\n");
 }
 
+#[test]
+fn test_subscript_default_passthrough() {
+    let html = "<p>H<sub>2</sub>O</p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "H2O\n");
+}
+
+#[test]
+fn test_superscript_default_passthrough() {
+    let html = "<p>x<sup>2</sup> + y<sup>3</sup></p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "x2 + y3\n");
+}
+
+#[test]
+fn test_subscript_superscript_combined_default() {
+    let html = "<p>CO<sub>2</sub><sup>*</sup></p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "CO2*\n");
+}
+
+#[test]
+fn test_subscript_html_tag_symbol() {
+    let html = "<p>H<sub>2</sub>O</p>";
+    let opts = ConversionOptions {
+        sub_symbol: "<sub>".to_string(),
+        ..Default::default()
+    };
+    let result = convert(html, Some(opts)).unwrap();
+    assert_eq!(result, "H<sub>2</sub>O\n");
+}
+
+#[test]
+fn test_adjacent_links_with_newline_separator() {
+    let html = "<p>\n<a href=\"/page1\">Link 1</a>\n<a href=\"/page2\">Link 2</a>\n</p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "[Link 1](/page1) [Link 2](/page2)\n");
+}
+
+#[test]
+fn test_adjacent_links_no_whitespace() {
+    let html = "<p><a href=\"/page1\">Link 1</a><a href=\"/page2\">Link 2</a></p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "[Link 1](/page1)[Link 2](/page2)\n");
+}
+
+#[test]
+fn test_adjacent_links_with_space() {
+    let html = "<p><a href=\"/page1\">Link 1</a> <a href=\"/page2\">Link 2</a></p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "[Link 1](/page1) [Link 2](/page2)\n");
+}
+
+#[test]
+fn test_adjacent_inline_elements_with_newline() {
+    let html = "<p><strong>bold</strong>\n<em>italic</em></p>";
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "**bold** *italic*\n");
+}
+
 #[test]
 fn test_autolink() {
     let html = "<p><a href=\"https://example.com\">https://example.com</a></p>";

data/rust-vendor/html-to-markdown-rs/tests/issue_212_regressions.rs

@@ -0,0 +1,63 @@
+#![allow(missing_docs)]
+
+use html_to_markdown_rs::convert;
+
+/// Regression test for https://github.com/kreuzberg-dev/html-to-markdown/issues/212
+///
+/// When `\n` precedes an `<a>` tag inside a `<p>`, the whitespace was
+/// inconsistently handled between the first and subsequent paragraphs.
+/// The bug was stateful: identical HTML structures produced different
+/// results depending on their position in the document.
+#[test]
+fn consistent_whitespace_before_link_across_paragraphs_issue_212() {
+    let html = r#"<p>text before
+<a href="https://example.com">the link</a>
+after</p>
+<p>text before
+<a href="https://example.com">the link</a>
+after</p>"#;
+
+    let result = convert(html, None).unwrap();
+    assert_eq!(
+        result,
+        "text before [the link](https://example.com) after\n\ntext before [the link](https://example.com) after\n"
+    );
+}
+
+/// Same bug but with three paragraphs to ensure no accumulation effects.
+#[test]
+fn consistent_whitespace_three_paragraphs_issue_212() {
+    let html = r#"<p>click
+<a href="/a">here</a></p>
+<p>click
+<a href="/b">here</a></p>
+<p>click
+<a href="/c">here</a></p>"#;
+
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "click [here](/a)\n\nclick [here](/b)\n\nclick [here](/c)\n");
+}
+
+/// Verify the fix doesn't break whitespace between text nodes without links.
+#[test]
+fn newline_before_inline_elements_consistent_issue_212() {
+    let html = r#"<p>before
+<strong>bold</strong> after</p>
+<p>before
+<strong>bold</strong> after</p>"#;
+
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "before **bold** after\n\nbefore **bold** after\n");
+}
+
+/// Verify with `<em>` tags across multiple paragraphs.
+#[test]
+fn newline_before_em_across_paragraphs_issue_212() {
+    let html = r#"<p>some text
+<em>emphasized</em> end</p>
+<p>some text
+<em>emphasized</em> end</p>"#;
+
+    let result = convert(html, None).unwrap();
+    assert_eq!(result, "some text *emphasized* end\n\nsome text *emphasized* end\n");
+}

data/spec/visitor_spec.rb

@@ -35,7 +35,7 @@ RSpec.describe HtmlToMarkdown do
         visit_definition_list_end visit_form visit_input visit_button visit_audio visit_video
         visit_iframe visit_details visit_summary visit_figure_start visit_figcaption
         visit_figure_end
-      ].each_with_object({}) { |name, hash| hash[name.to_sym] = { type: :continue } }
+      ].to_h { |name| [name.to_sym, { type: :continue }] }
     end
 
     def create_visitor(**overrides)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: html-to-markdown
 version: !ruby/object:Gem::Version
-  version: 2.26.1
+  version: 2.26.3
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-02-27 00:00:00.000000000 Z
+date: 2026-02-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys
@@ -1946,6 +1946,7 @@ files:
 - rust-vendor/html-to-markdown-rs/tests/issue_190_regressions.rs
 - rust-vendor/html-to-markdown-rs/tests/issue_199_regressions.rs
 - rust-vendor/html-to-markdown-rs/tests/issue_200_regressions.rs
+- rust-vendor/html-to-markdown-rs/tests/issue_212_regressions.rs
 - rust-vendor/html-to-markdown-rs/tests/json_ld_script_extraction.rs
 - rust-vendor/html-to-markdown-rs/tests/lists_test.rs
 - rust-vendor/html-to-markdown-rs/tests/preprocessing_tests.rs