html-to-markdown 2.5.2 → 2.5.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: dbdd98e748fec609d1b033648e4e78aef576a320a127daaa83202fff9a662f51
-  data.tar.gz: 6a9ea3fa2b0eaaabc6d9618369da96899ea5295585b9354bbf9348915b04bac6
+  metadata.gz: 0f6d03456f76e7f7b1157f052c51dc333e1bd68e72a6bb7664f4701f9a592ed2
+  data.tar.gz: e4a09afb52aba7580f9805b91981d96007f04c37fa7ba6059867d0e818e260e0
 SHA512:
-  metadata.gz: 141b341e6b3729cb804864a1af4e7a7411dd46efcd37645a401169bc57385ae623726a7b136049e5a78010695f1750e56635e670887319bdfb22a62493d7b7e1
-  data.tar.gz: 3d9b17aa842772c81ca3f212983d614b4318be7635adaa8581005aca3909829d332975275d40dc33d79aac2bd7c3d72d8c05ca24e3d7ecf44f0b1c2e07518758
+  metadata.gz: d37723f521d0765ff47578457ba24d6636fe5102af2672cc3f8485bef323770cdb8155512378db4025ea47aa4c31146291da960a69ccdacb93a461e4c8e64593
+  data.tar.gz: 31523b5389627bfb1476133f1daa39ae2273d32dfdd72dac0f7927cc45267e41def028ff07bec8a833c567d07d02babda7e8e4e53c7dc63f7490c8ad7dc00481

data/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "html-to-markdown-rb"
-version = "2.5.2"
+version = "2.5.4"
 edition = "2021"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
 license = "MIT"
@@ -21,7 +21,7 @@ crate-type = ["cdylib", "rlib"]
 default = []
 
 [dependencies]
-html-to-markdown-rs = { version = "2.5.2", features = ["inline-images"] }
+html-to-markdown-rs = { version = "2.5.4", features = ["inline-images"] }
 magnus = { git = "https://github.com/matsadler/magnus", rev = "f6db11769efb517427bf7f121f9c32e18b059b38", features = ["rb-sys"] }
 
 [dev-dependencies]

data/README.md

@@ -1,10 +1,22 @@
 # html-to-markdown-rb
 
-Ruby bindings for the `html-to-markdown` Rust engine – the same core that powers the Python wheels, Node.js NAPI bindings, WebAssembly package, and CLI. The gem exposes fast HTML → Markdown conversion with identical rendering behaviour across every supported language.
+Blazing-fast HTML → Markdown conversion for Ruby, powered by the same Rust engine used by our Python, Node.js, and WebAssembly packages. Ship identical Markdown across every runtime while enjoying native extension performance.
 
-[![RubyGems](https://badge.fury.io/rb/html-to-markdown.svg)](https://rubygems.org/gems/html-to-markdown)
+[![Crates.io](https://img.shields.io/crates/v/html-to-markdown-rs.svg)](https://crates.io/crates/html-to-markdown-rs)
+[![npm version](https://badge.fury.io/js/html-to-markdown-node.svg)](https://www.npmjs.com/package/html-to-markdown-node)
+[![PyPI version](https://badge.fury.io/py/html-to-markdown.svg)](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)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://github.com/Goldziher/html-to-markdown/blob/main/LICENSE)
 
+## Features
+
+- ⚡ **Rust-fast**: Ruby bindings around a highly optimised Rust core (60‑80× faster than BeautifulSoup-based converters).
+- 🔁 **Identical output**: Shares logic with the Python wheels, npm bindings, WASM package, and CLI — consistent Markdown everywhere.
+- ⚙️ **Rich configuration**: Control heading styles, list indentation, whitespace handling, HTML preprocessing, and more.
+- 🖼️ **Inline image extraction**: Pull out embedded images (PNG/JPEG/SVG/data URIs) alongside Markdown.
+- 🧰 **Bundled CLI proxy**: Call the Rust CLI straight from Ruby or shell scripts.
+- 🛠️ **First-class Rails support**: Works with `Gem.win_platform?` builds, supports Trusted Publishing, and compiles on install if no native gem matches.
+
 ## Installation
 
 ```bash
@@ -29,6 +41,18 @@ ridk exec pacman -S --needed --noconfirm base-devel mingw-w64-ucrt-x86_64-toolch
 
 This provides the standard headers (including `strings.h`) required for the bindgen step.
 
+## Performance Snapshot
+
+Apple M4 • Real Wikipedia documents • `HtmlToMarkdown.convert` (Ruby)
+
+| Document            | Size  | Latency | Throughput | Docs/sec |
+| ------------------- | ----- | ------- | ---------- | -------- |
+| Lists (Timeline)    | 129KB | 0.69ms  | 187 MB/s   | 1,450    |
+| Tables (Countries)  | 360KB | 2.19ms  | 164 MB/s   | 456      |
+| Mixed (Python wiki) | 656KB | 4.88ms  | 134 MB/s   | 205      |
+
+> Same core, same benchmarks: the Ruby extension stays within single-digit % of the Rust CLI and mirrors the Python/Node numbers.
+
 ## Quick Start
 
 ```ruby
@@ -53,9 +77,11 @@ puts markdown
 # - Identical output across languages
 ```
 
-### Conversion with Options
+## API
 
-All configuration mirrors the Rust API. Options accept symbols or strings and match the same defaults as the other bindings.
+### Conversion Options
+
+Pass a Ruby hash (string or symbol keys) to tweak rendering. Every option maps one-for-one with the Rust/Python/Node APIs.
 
 ```ruby
 require 'html_to_markdown'
@@ -64,10 +90,31 @@ markdown = HtmlToMarkdown.convert(
   '<pre><code class="language-ruby">puts "hi"</code></pre>',
   heading_style: :atx,
   code_block_style: :fenced,
-  bullets: ['*', '-', '+'],
-  wrap: true,
-  wrap_width: 80,
-  preserve_tags: %w[table figure]
+  bullets: '*+-',
+  list_indent_type: :spaces,
+  list_indent_width: 2,
+  whitespace_mode: :normalized,
+  highlight_style: :double_equal
+)
+
+puts markdown
+```
+
+### HTML Preprocessing
+
+Clean up scraped HTML (navigation, forms, malformed markup) before conversion:
+
+```ruby
+require 'html_to_markdown'
+
+markdown = HtmlToMarkdown.convert(
+  html,
+  preprocessing: {
+    enabled: true,
+    preset: :aggressive, # :minimal, :standard, :aggressive
+    remove_navigation: true,
+    remove_forms: true
+  }
 )
 ```
 
@@ -94,7 +141,7 @@ result.inline_images.each do |img|
 end
 ```
 
-### CLI Proxy
+## CLI
 
 The gem bundles a small proxy for the Rust CLI binary. Use it when you need parity with the standalone `html-to-markdown` executable.
 
@@ -109,10 +156,17 @@ You can also call the CLI binary directly for scripting:
 
 ```ruby
 HtmlToMarkdown::CLIProxy.call(['--version'])
-# => "html-to-markdown 2.5.2"
+# => "html-to-markdown 2.5.4"
+```
+
+Rebuild the CLI locally if you see `CLI binary not built` during tests:
+
+```bash
+bundle exec rake compile          # builds the extension
+bundle exec ruby scripts/prepare_ruby_gem.rb  # copies the CLI into lib/bin/
 ```
 
-### Error Handling
+## Error Handling
 
 Conversion errors raise `HtmlToMarkdown::Error` (wrapping the Rust error context). CLI invocations use specialised subclasses:
 

data/lib/html_to_markdown/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module HtmlToMarkdown
-  VERSION = '2.5.2'
+  VERSION = '2.5.4'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: html-to-markdown
 version: !ruby/object:Gem::Version
-  version: 2.5.2
+  version: 2.5.4
 platform: ruby
 authors:
 - Na'aman Hirschfeld
@@ -30,8 +30,9 @@ dependencies:
     - - "<"
       - !ruby/object:Gem::Version
         version: '1.0'
-description: High-performance HTML to Markdown conversion from Ruby using Magnus and
-  rb-sys.
+description: |-
+  html-to-markdown wraps our ultra-fast Rust converter with a Ruby-native API via Magnus and rb-sys.
+  Enjoy identical output to the Python, Node, and WASM bindings, a bundled CLI proxy, and seamless cross-platform installs.
 email:
 - nhirschfeld@gmail.com
 executables:
@@ -44,7 +45,6 @@ files:
 - README.md
 - exe/html-to-markdown
 - extconf.rb
-- lib/bin/html-to-markdown
 - lib/html_to_markdown.rb
 - lib/html_to_markdown/cli.rb
 - lib/html_to_markdown/cli_proxy.rb
@@ -81,5 +81,5 @@ requirements: []
 rubygems_version: 3.5.22
 signing_key:
 specification_version: 4
-summary: Ruby bindings for the html-to-markdown Rust library
+summary: Blazing-fast HTML to Markdown conversion for Ruby, powered by Rust.
 test_files: []