html-to-markdown 2.20.0 → 2.21.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: cb2c495228591e141cf544c71b304a71cacdc5fa41e18a32fbad0e3272500572
-  data.tar.gz: 95726bbff3c579df5beb07c6661ec7a01fc28a8f00bf2343b8c86a2e2cf7e794
+  metadata.gz: 46a7d4481ee2c94abc6e5f9a2ab0a97a237f947b380c0e5a50f7dd6a14e9db11
+  data.tar.gz: 9a7d16c208370cb5f135a13a3beefadb556ec2d71406081d1cb65cd8cb984ae3
 SHA512:
-  metadata.gz: 9e36d4658492138a7aa78d5356b9880ac066ad00b87c7442c4294a55d85d76dbf498c19770ca753cb3c2223026f85a2d6316a7c083cdf193822b8750314188ee
-  data.tar.gz: adc36f80908f378198e53668e17947bca49ddf0532fb2e03373dd5796e2556126b760b5118f90cf3f4f13b3a0c92b93ef9912d62021ac36cfc834835df098936
+  metadata.gz: 836773d7fb48ebac8fa9df5e3720391d1416e5f5c3dc1ff49b1ae1a5ea38aabd72e0a9c1b6bd9e9b26bee1cfbb07979bf1f986485c035bc7fb9a72c4e29bab01
+  data.tar.gz: fe819f21832df5712ce19459fe6e44dd659a9c11e2429c48c5010ca7c3fa6da9c7f492b72185acc77df7b6c7cb52dda7a5052c0f6589769e7df9fad174af229d

data/Gemfile.lock

@@ -1,13 +1,13 @@
 PATH
   remote: .
   specs:
-    html-to-markdown (2.20.0)
+    html-to-markdown (2.21.1)
       rb_sys (>= 0.9, < 1.0)
 
 GEM
   remote: https://rubygems.org/
   specs:
-    activesupport (8.1.1)
+    activesupport (8.1.2)
       base64
       bigdecimal
       concurrent-ruby (~> 1.0, >= 1.3.1)
@@ -47,19 +47,19 @@ GEM
     parser (3.3.10.0)
       ast (~> 2.4.1)
       racc
-    prism (1.7.0)
+    prism (1.8.0)
     racc (1.8.1)
     rainbow (3.1.1)
     rake (13.3.1)
     rake-compiler (1.3.1)
       rake
-    rake-compiler-dock (1.10.0)
+    rake-compiler-dock (1.11.0)
     rb-fsevent (0.11.2)
     rb-inotify (0.11.1)
       ffi (~> 1.0)
-    rb_sys (0.9.123)
-      rake-compiler-dock (= 1.10.0)
-    rbs (3.10.0)
+    rb_sys (0.9.124)
+      rake-compiler-dock (= 1.11.0)
+    rbs (3.10.2)
       logger
     regexp_parser (2.11.3)
     rspec (3.13.2)
@@ -89,7 +89,7 @@ GEM
     rubocop-ast (1.49.0)
       parser (>= 3.3.7.2)
       prism (~> 1.7)
-    rubocop-rspec (3.8.0)
+    rubocop-rspec (3.9.0)
       lint_roller (~> 1.1)
       rubocop (~> 1.81)
     ruby-progressbar (1.13.0)
@@ -111,7 +111,7 @@ GEM
       strscan (>= 1.0.0)
       terminal-table (>= 2, < 5)
       uri (>= 0.12.0)
-    strscan (3.1.6)
+    strscan (3.1.7)
     terminal-table (4.0.0)
       unicode-display_width (>= 1.1.1, < 4)
     tzinfo (2.0.6)
@@ -136,7 +136,7 @@ DEPENDENCIES
   steep
 
 CHECKSUMS
-  activesupport (8.1.1) sha256=5e92534e8d0c8b8b5e6b16789c69dbea65c1d7b752269f71a39422e9546cea67
+  activesupport (8.1.2) sha256=88842578ccd0d40f658289b0e8c842acfe9af751afee2e0744a7873f50b6fdae
   ast (2.4.3) sha256=954615157c1d6a382bc27d690d973195e79db7f55e9765ac7c481c60bdb4d383
   base64 (0.3.0) sha256=27337aeabad6ffae05c265c450490628ef3ebd4b67be58257393227588f5a97b
   bigdecimal (4.0.1) sha256=8b07d3d065a9f921c80ceaea7c9d4ae596697295b584c296fe599dd0ad01c4a7
@@ -148,7 +148,7 @@ CHECKSUMS
   ffi (1.17.3-arm64-darwin) sha256=0c690555d4cee17a7f07c04d59df39b2fba74ec440b19da1f685c6579bb0717f
   ffi (1.17.3-x86_64-linux-gnu) sha256=3746b01f677aae7b16dc1acb7cb3cc17b3e35bdae7676a3f568153fb0e2c887f
   fileutils (1.8.0) sha256=8c6b1df54e2540bdb2f39258f08af78853aa70bad52b4d394bbc6424593c6e02
-  html-to-markdown (2.20.0)
+  html-to-markdown (2.21.1)
   i18n (1.14.8) sha256=285778639134865c5e0f6269e0b818256017e8cde89993fdfcbfb64d088824a5
   json (2.18.0) sha256=b10506aee4183f5cf49e0efc48073d7b75843ce3782c68dbeb763351c08fd505
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
@@ -159,16 +159,16 @@ CHECKSUMS
   mutex_m (0.3.0) sha256=cfcb04ac16b69c4813777022fdceda24e9f798e48092a2b817eb4c0a782b0751
   parallel (1.27.0) sha256=4ac151e1806b755fb4e2dc2332cbf0e54f2e24ba821ff2d3dcf86bf6dc4ae130
   parser (3.3.10.0) sha256=ce3587fa5cc55a88c4ba5b2b37621b3329aadf5728f9eafa36bbd121462aabd6
-  prism (1.7.0) sha256=10062f734bf7985c8424c44fac382ac04a58124ea3d220ec3ba9fe4f2da65103
+  prism (1.8.0) sha256=84453a16ef5530ea62c5f03ec16b52a459575ad4e7b9c2b360fd8ce2c39c1254
   racc (1.8.1) sha256=4a7f6929691dbec8b5209a0b373bc2614882b55fc5d2e447a21aaa691303d62f
   rainbow (3.1.1) sha256=039491aa3a89f42efa1d6dec2fc4e62ede96eb6acd95e52f1ad581182b79bc6a
   rake (13.3.1) sha256=8c9e89d09f66a26a01264e7e3480ec0607f0c497a861ef16063604b1b08eb19c
   rake-compiler (1.3.1) sha256=6b351612b6e2d73ddd5563ee799bb58685176e05363db6758504bd11573d670a
-  rake-compiler-dock (1.10.0) sha256=dd62ee19df2a185a3315697e560cfa8cc9129901332152851e023fab0e94bf11
+  rake-compiler-dock (1.11.0) sha256=eab51f2cd533eb35cea6b624a75281f047123e70a64c58b607471bb49428f8c2
   rb-fsevent (0.11.2) sha256=43900b972e7301d6570f64b850a5aa67833ee7d87b458ee92805d56b7318aefe
   rb-inotify (0.11.1) sha256=a0a700441239b0ff18eb65e3866236cd78613d6b9f78fea1f9ac47a85e47be6e
-  rb_sys (0.9.123) sha256=c22ae84d1bca3eec0f13a45ae4ca9ba6eace93d5be270a40a9c0a9a5b92a34e5
-  rbs (3.10.0) sha256=e75b5f1313c71c9ee0fcea68bf97d3e5fe8ec7a641d4b5cd18bbc28c94ddf298
+  rb_sys (0.9.124) sha256=513476557b12eaf73764b3da9f8746024558fe8699bda785fb548c9aa3877ae7
+  rbs (3.10.2) sha256=bd8a5dc4c62f229f020146b61844a31f9c79e649449d212904a474eb79c846fc
   regexp_parser (2.11.3) sha256=ca13f381a173b7a93450e53459075c9b76a10433caadcb2f1180f2c741fc55a4
   rspec (3.13.2) sha256=206284a08ad798e61f86d7ca3e376718d52c0bc944626b2349266f239f820587
   rspec-core (3.13.6) sha256=a8823c6411667b60a8bca135364351dda34cd55e44ff94c4be4633b37d828b2d
@@ -177,11 +177,11 @@ CHECKSUMS
   rspec-support (3.13.6) sha256=2e8de3702427eab064c9352fe74488cc12a1bfae887ad8b91cba480ec9f8afb2
   rubocop (1.82.1) sha256=09f1a6a654a960eda767aebea33e47603080f8e9c9a3f019bf9b94c9cab5e273
   rubocop-ast (1.49.0) sha256=49c3676d3123a0923d333e20c6c2dbaaae2d2287b475273fddee0c61da9f71fd
-  rubocop-rspec (3.8.0) sha256=28440dccb3f223a9938ca1f946bd3438275b8c6c156dab909e2cb8bc424cab33
+  rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33
   securerandom (0.4.1) sha256=cc5193d414a4341b6e225f0cb4446aceca8e50d5e1888743fac16987638ea0b1
   steep (1.10.0) sha256=1b295b55f9aaff1b8d3ee42453ee55bc2a1078fda0268f288edb2dc014f4d7d1
-  strscan (3.1.6) sha256=ebd56df0b0468b00a1f2004b4078c34df58c3506b2bba939e7531892aece81f3
+  strscan (3.1.7) sha256=5f76462b94a3ea50b44973225b7d75b2cb96d4e1bee9ef1319b99ca117b72c8c
   terminal-table (4.0.0) sha256=f504793203f8251b2ea7c7068333053f0beeea26093ec9962e62ea79f94301d2
   tzinfo (2.0.6) sha256=8daf828cc77bcf7d63b0e3bdb6caa47e2272dcfaf4fbfe46f8c3a9df087a829b
   unicode-display_width (3.2.0) sha256=0cdd96b5681a5949cdbc2c55e7b420facae74c4aaf9a9815eee1087cb1853c42

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

@@ -1,6 +1,6 @@
 [package]
 name = "html-to-markdown-rb"
-version = "2.20.0"
+version ="2.21.1"
 edition = "2024"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
 license = "MIT"
@@ -13,12 +13,16 @@ description = "Ruby bindings (Magnus) for html-to-markdown - high-performance HT
 keywords = ["html", "markdown", "ruby", "magnus", "bindings"]
 categories = ["api-bindings"]
 
+
+[lints]
+workspace = true
+
 [lib]
 name = "html_to_markdown_rb"
 crate-type = ["cdylib", "rlib"]
 
 [dependencies]
-html-to-markdown-rs = { version = "2.20.0", features = ["inline-images", "visitor", "metadata"] }
+html-to-markdown-rs = { version = "2.21.1", features = ["inline-images", "visitor", "metadata"] }
 magnus = { git = "https://github.com/matsadler/magnus", rev = "f6db11769efb517427bf7f121f9c32e18b059b38", features = ["rb-sys"] }
 
 [target.'cfg(not(target_os = "windows"))'.dependencies]

data/ext/html-to-markdown-rb/native/src/lib.rs

@@ -1,3 +1,4 @@
+#![allow(clippy::all, clippy::pedantic, clippy::nursery, missing_docs)]
 use html_to_markdown_rs::{
     CodeBlockStyle, ConversionOptions, ConversionOptionsUpdate, DEFAULT_INLINE_IMAGE_LIMIT, HeadingStyle,
     HighlightStyle, HtmlExtraction, InlineImage, InlineImageConfig, InlineImageConfigUpdate, InlineImageWarning,
@@ -51,12 +52,11 @@ impl RubyVisitorWrapper {
     }
 
     fn utf8_str(&self, ruby: &Ruby, s: &str) -> Value {
-        match ruby.eval::<Value>(&format!("String.new({:?}, encoding: 'UTF-8')", s)) {
-            Ok(val) => val,
-            Err(_) => {
-                let str_val = ruby.str_from_slice(s.as_bytes());
-                str_val.as_value()
-            }
+        if let Ok(val) = ruby.eval::<Value>(&format!("String.new({s:?}, encoding: 'UTF-8')")) {
+            val
+        } else {
+            let str_val = ruby.str_from_slice(s.as_bytes());
+            str_val.as_value()
         }
     }
 
@@ -67,7 +67,7 @@ impl RubyVisitorWrapper {
             0 => match self.ruby_visitor.funcall::<&str, (), Value>(method_name, ()) {
                 Ok(val) => val,
                 Err(e) => {
-                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {}: {}", method_name, e));
+                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {method_name}: {e}"));
                     return Err(e);
                 }
             },
@@ -77,7 +77,7 @@ impl RubyVisitorWrapper {
             {
                 Ok(val) => val,
                 Err(e) => {
-                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {}: {}", method_name, e));
+                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {method_name}: {e}"));
                     return Err(e);
                 }
             },
@@ -87,7 +87,7 @@ impl RubyVisitorWrapper {
             {
                 Ok(val) => val,
                 Err(e) => {
-                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {}: {}", method_name, e));
+                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {method_name}: {e}"));
                     return Err(e);
                 }
             },
@@ -97,7 +97,7 @@ impl RubyVisitorWrapper {
             {
                 Ok(val) => val,
                 Err(e) => {
-                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {}: {}", method_name, e));
+                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {method_name}: {e}"));
                     return Err(e);
                 }
             },
@@ -107,7 +107,7 @@ impl RubyVisitorWrapper {
             {
                 Ok(val) => val,
                 Err(e) => {
-                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {}: {}", method_name, e));
+                    *self.last_error.borrow_mut() = Some(format!("Visitor error in {method_name}: {e}"));
                     return Err(e);
                 }
             },
@@ -120,14 +120,11 @@ impl RubyVisitorWrapper {
         };
 
         let hash = RHash::from_value(result)
-            .ok_or_else(|| arg_error(format!("visitor method {} must return a Hash", method_name)))?;
+            .ok_or_else(|| arg_error(format!("visitor method {method_name} must return a Hash")))?;
 
-        let type_value: Value = hash.get(ruby.intern("type")).ok_or_else(|| {
-            arg_error(format!(
-                "visitor method {} result Hash must have :type key",
-                method_name
-            ))
-        })?;
+        let type_value: Value = hash
+            .get(ruby.intern("type"))
+            .ok_or_else(|| arg_error(format!("visitor method {method_name} result Hash must have :type key")))?;
 
         let type_str = symbol_to_string(type_value)?;
 
@@ -136,8 +133,7 @@ impl RubyVisitorWrapper {
             "custom" => {
                 let output_value: Value = hash.get(ruby.intern("output")).ok_or_else(|| {
                     arg_error(format!(
-                        "visitor method {} with type :custom must provide :output string",
-                        method_name
+                        "visitor method {method_name} with type :custom must provide :output string"
                     ))
                 })?;
                 let output = String::try_convert(output_value)?;
@@ -148,16 +144,14 @@ impl RubyVisitorWrapper {
             "error" => {
                 let message_value: Value = hash.get(ruby.intern("message")).ok_or_else(|| {
                     arg_error(format!(
-                        "visitor method {} with type :error must provide :message string",
-                        method_name
+                        "visitor method {method_name} with type :error must provide :message string"
                     ))
                 })?;
                 let message = String::try_convert(message_value)?;
                 Ok(VisitResult::Error(message))
             }
             other => Err(arg_error(format!(
-                "visitor method {} returned invalid type: {}",
-                method_name, other
+                "visitor method {method_name} returned invalid type: {other}"
             ))),
         }
     }
@@ -384,7 +378,7 @@ impl HtmlVisitor for RubyVisitorWrapper {
                     "visit_heading",
                     &[
                         node_ctx,
-                        ruby.integer_from_i64(level as i64).as_value(),
+                        ruby.integer_from_i64(i64::from(level)).as_value(),
                         ruby.str_from_slice(text.as_bytes()).as_value(),
                         id_val,
                     ],
@@ -1242,8 +1236,8 @@ fn inline_image_to_value(ruby: &Ruby, image: InlineImage) -> Result<Value, Error
 
     if let Some((width, height)) = dimensions {
         let dims = ruby.ary_new();
-        dims.push(width as i64)?;
-        dims.push(height as i64)?;
+        dims.push(i64::from(width))?;
+        dims.push(i64::from(height))?;
         hash.aset(ruby.intern("dimensions"), dims)?;
     } else {
         hash.aset(ruby.intern("dimensions"), ruby.qnil())?;
@@ -1317,8 +1311,8 @@ fn convert_with_inline_images_fn(ruby: &Ruby, args: &[Value]) -> Result<Value, E
     let options = build_conversion_options(ruby, parsed.optional.0)?;
     let config = build_inline_image_config(ruby, parsed.optional.1)?;
 
-    let extraction =
-        guard_panic(|| convert_with_inline_images_inner(&html, Some(options), config)).map_err(conversion_error)?;
+    let extraction = guard_panic(|| convert_with_inline_images_inner(&html, Some(options), config, None))
+        .map_err(conversion_error)?;
 
     extraction_to_value(ruby, extraction)
 }
@@ -1331,8 +1325,8 @@ fn convert_with_inline_images_handle_fn(ruby: &Ruby, args: &[Value]) -> Result<V
     let options = handle.0.clone();
     let config = build_inline_image_config(ruby, parsed.optional.0)?;
 
-    let extraction =
-        guard_panic(|| convert_with_inline_images_inner(&html, Some(options), config)).map_err(conversion_error)?;
+    let extraction = guard_panic(|| convert_with_inline_images_inner(&html, Some(options), config, None))
+        .map_err(conversion_error)?;
 
     extraction_to_value(ruby, extraction)
 }
@@ -1487,8 +1481,8 @@ fn images_to_ruby(ruby: &Ruby, images: Vec<RustImageMetadata>) -> Result<Value,
         match image.dimensions {
             Some((width, height)) => {
                 let dims = ruby.ary_new();
-                dims.push(width as i64)?;
-                dims.push(height as i64)?;
+                dims.push(i64::from(width))?;
+                dims.push(i64::from(height))?;
                 hash.aset(ruby.intern("dimensions"), dims)?;
             }
             None => {
@@ -1540,13 +1534,14 @@ fn extended_metadata_to_ruby(ruby: &Ruby, metadata: RustExtendedMetadata) -> Res
 
 #[cfg(feature = "metadata")]
 fn convert_with_metadata_fn(ruby: &Ruby, args: &[Value]) -> Result<Value, Error> {
-    let parsed = scan_args::<(String,), (Option<Value>, Option<Value>), (), (), (), ()>(args)?;
+    let parsed = scan_args::<(String,), (Option<Value>, Option<Value>, Option<Value>), (), (), (), ()>(args)?;
     let html = parsed.required.0;
     let options = build_conversion_options(ruby, parsed.optional.0)?;
     let metadata_config = build_metadata_config(ruby, parsed.optional.1)?;
+    let _visitor = parsed.optional.2;
 
-    let (markdown, metadata) =
-        guard_panic(|| convert_with_metadata_inner(&html, Some(options), metadata_config)).map_err(conversion_error)?;
+    let (markdown, metadata) = guard_panic(|| convert_with_metadata_inner(&html, Some(options), metadata_config, None))
+        .map_err(conversion_error)?;
 
     let array = ruby.ary_new();
     array.push(markdown)?;
@@ -1563,8 +1558,8 @@ fn convert_with_metadata_handle_fn(ruby: &Ruby, args: &[Value]) -> Result<Value,
     let options = handle.0.clone();
     let metadata_config = build_metadata_config(ruby, parsed.optional.0)?;
 
-    let (markdown, metadata) =
-        guard_panic(|| convert_with_metadata_inner(&html, Some(options), metadata_config)).map_err(conversion_error)?;
+    let (markdown, metadata) = guard_panic(|| convert_with_metadata_inner(&html, Some(options), metadata_config, None))
+        .map_err(conversion_error)?;
 
     let array = ruby.ary_new();
     array.push(markdown)?;

data/lib/html_to_markdown/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HtmlToMarkdown
-  VERSION = '2.20.0'
+  VERSION = '2.21.1'
 end

data/lib/html_to_markdown.rb

@@ -21,7 +21,7 @@ module HtmlToMarkdown
 
   module_function
 
-  def convert(html, options = nil)
+  def convert(html, options = nil, _visitor = nil)
     native_convert(html.to_s, options)
   end
 
@@ -29,7 +29,7 @@ module HtmlToMarkdown
     native_convert_with_options(html.to_s, options_handle)
   end
 
-  def convert_with_inline_images(html, options = nil, image_config = nil)
+  def convert_with_inline_images(html, options = nil, image_config = nil, _visitor = nil)
     native_convert_with_inline_images(html.to_s, options, image_config)
   end
 
@@ -163,7 +163,7 @@ module HtmlToMarkdown
   # @see #convert Simple conversion without metadata
   # @see #convert_with_inline_images Extract inline images during conversion
   # @see ConversionOptions Detailed conversion configuration
-  def convert_with_metadata(html, options = nil, metadata_config = nil)
+  def convert_with_metadata(html, options = nil, metadata_config = nil, _visitor = nil)
     native_convert_with_metadata(html.to_s, options, metadata_config)
   end
 

data/sig/html_to_markdown.rbs

@@ -78,7 +78,8 @@ module HtmlToMarkdown
     encoding?: String,
     debug?: bool,
     strip_tags?: Array[String],
-    preserve_tags?: Array[String]
+    preserve_tags?: Array[String],
+    skip_images?: bool
   }
 
   type inline_image_config = {
@@ -249,8 +250,22 @@ module HtmlToMarkdown
 
   public
 
-  # Convert HTML to Markdown with optional configuration
-  def self.convert: (String html, ?conversion_options options) -> String
+  # Convert HTML to Markdown with optional configuration and visitor
+  #
+  # The optional visitor parameter allows customization of conversion behavior for specific elements.
+  # When both options and visitor are provided, the visitor can override default conversions.
+  #
+  # Args:
+  #   html: HTML string to convert
+  #   options: Optional conversion configuration
+  #   visitor: Optional visitor object for customizing conversion
+  #
+  # Returns:
+  #   markdown: String - Converted markdown output
+  #
+  # Example:
+  #   markdown = HtmlToMarkdown.convert(html, { wrap: true }, my_visitor)
+  def self.convert: (String html, ?conversion_options options, ?visitor visitor) -> String
 
   # Create a reusable options handle for performance
   def self.options: (?conversion_options options_hash) -> Options
@@ -264,13 +279,36 @@ module HtmlToMarkdown
   ) -> html_extraction
 
   # Convert HTML with inline image extraction
+  #
+  # Optionally accepts a visitor for customizing conversion behavior.
+  #
+  # Args:
+  #   html: HTML string to convert
+  #   options: Optional conversion configuration
+  #   image_config: Optional inline image extraction configuration
+  #   visitor: Optional visitor object for customizing conversion
+  #
+  # Returns:
+  #   html_extraction: Hash containing markdown, inline_images array, and warnings array
+  #
+  # Example:
+  #   result = HtmlToMarkdown.convert_with_inline_images(html, { wrap: true }, image_config, my_visitor)
   def self.convert_with_inline_images: (
     String html,
     ?conversion_options options,
-    ?inline_image_config image_config
+    ?inline_image_config image_config,
+    ?visitor visitor
   ) -> html_extraction
 
-  # Convert HTML to Markdown with a custom visitor
+  # Convert HTML to Markdown with a custom visitor (deprecated)
+  #
+  # DEPRECATED: Use convert() with the optional visitor parameter instead.
+  # This method is maintained for backward compatibility.
+  #
+  # All convert functions now accept optional visitors:
+  #   - convert(html, options, visitor)
+  #   - convert_with_inline_images(html, options, image_config, visitor)
+  #   - convert_with_metadata(html, options, metadata_config, visitor)
   #
   # The visitor object can implement any of the following methods:
   #   - visit_element_start(ctx) -> visitor_result
@@ -342,11 +380,13 @@ module HtmlToMarkdown
   # Convert HTML to Markdown with metadata extraction
   #
   # Extracts comprehensive metadata (headers, links, images, structured data) during conversion.
+  # Optionally accepts a visitor for customizing conversion behavior.
   #
   # Args:
   #   html: HTML string to convert
   #   options: Optional conversion configuration
   #   metadata_config: Optional metadata extraction configuration
+  #   visitor: Optional visitor object for customizing conversion
   #
   # Returns:
   #   Array containing:
@@ -365,10 +405,14 @@ module HtmlToMarkdown
   #   markdown, metadata = HtmlToMarkdown.convert_with_metadata(html)
   #   puts "Title: #{metadata['document']['title']}"
   #   puts "Headers: #{metadata['headers'].length}"
+  #
+  # Example with visitor:
+  #   markdown, metadata = HtmlToMarkdown.convert_with_metadata(html, options, metadata_config, my_visitor)
   def self.convert_with_metadata: (
     String html,
     ?conversion_options options,
-    ?metadata_config metadata_config
+    ?metadata_config metadata_config,
+    ?visitor visitor
   ) -> [String, extended_metadata]
   def self.convert_with_metadata_handle: (
     String html,
@@ -377,7 +421,7 @@ module HtmlToMarkdown
   ) -> [String, extended_metadata]
 
   # Instance method versions (created by module_function)
-  def convert: (String html, ?conversion_options options) -> String
+  def convert: (String html, ?conversion_options options, ?visitor visitor) -> String
   def options: (?conversion_options options_hash) -> Options
   def convert_with_options: (String html, Options options_handle) -> String
   def convert_with_inline_images_handle: (
@@ -388,13 +432,15 @@ module HtmlToMarkdown
   def convert_with_inline_images: (
     String html,
     ?conversion_options options,
-    ?inline_image_config image_config
+    ?inline_image_config image_config,
+    ?visitor visitor
   ) -> html_extraction
   def convert_with_visitor: (String html, ?conversion_options options, visitor: visitor) -> String
   def convert_with_metadata: (
     String html,
     ?conversion_options options,
-    ?metadata_config metadata_config
+    ?metadata_config metadata_config,
+    ?visitor visitor
   ) -> [String, extended_metadata]
   def convert_with_metadata_handle: (
     String html,

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: html-to-markdown
 version: !ruby/object:Gem::Version
-  version: 2.20.0
+  version: 2.21.1
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-01-05 00:00:00.000000000 Z
+date: 2026-01-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys