From 5fa05e53f6d8c73d282cc3182927c8a8c8dee8b4 Mon Sep 17 00:00:00 2001
From: Gleb Mazovetskiy <glex.spb@gmail.com>
Date: Thu, 17 May 2018 22:22:23 +0900
Subject: [PATCH] Optimize header parsing

Fixes #505

Benchmarks:

"setext":

```bash
ruby -rbenchmark -Ilib -rkramdown -e 'p Benchmark.measure{Kramdown::Document.new("1#{" "*20000}2\n==\n")}'
```

"atx":

```bash
ruby -rbenchmark -Ilib -rkramdown -e 'p Benchmark.measure{Kramdown::Document.new("## 1#{" "*20000}2")}'
```
---
 .gitignore                                    |  5 +++
 lib/kramdown/parser/gfm.rb                    | 19 ++++++---
 lib/kramdown/parser/kramdown/header.rb        | 39 +++++++++++++------
 test/testcases/block/04_header/atx_header.hcd |  0
 4 files changed, 47 insertions(+), 16 deletions(-)
 create mode 100644 test/testcases/block/04_header/atx_header.hcd

diff --git a/.gitignore b/.gitignore
index a2990bfb..7e3ec420 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,8 @@
 htmldoc
 pkg
 webgen-tmp
+.bundle/
+CONTRIBUTERS
+VERSION
+kramdown.gemspec
+man/man1/kramdown.1
diff --git a/lib/kramdown/parser/gfm.rb b/lib/kramdown/parser/gfm.rb
index 3ae5396a..756179d1 100644
--- a/lib/kramdown/parser/gfm.rb
+++ b/lib/kramdown/parser/gfm.rb
@@ -112,19 +112,28 @@ def generate_gfm_header_id(text)
         @options[:auto_id_prefix] + result
       end
 
-      ATX_HEADER_START = /^\#{1,6}\s/
+      ATX_HEADER_START = /^(?<level>\#{1,6})[\t ]+(?<contents>.*)\n/
       define_parser(:atx_header_gfm, ATX_HEADER_START, nil, 'parse_atx_header')
       define_parser(:atx_header_gfm_quirk, ATX_HEADER_START)
 
       # Copied from kramdown/parser/kramdown/header.rb, removed the first line
       def parse_atx_header_gfm_quirk
-        start_line_number = @src.current_line_number
-        @src.check(ATX_HEADER_MATCH)
-        level, text, id = @src[1], @src[2].to_s.strip, @src[3]
+        text = @src["contents"]
+        text.rstrip!
+        id_match = HEADER_ID.match(text)
+        if id_match
+          id = id_match["id"]
+          text = text[0...-id_match[0].length]
+          text.rstrip!
+        end
+        text.sub!(/[\t ]#+\z/, '')
+        text.rstrip!
         return false if text.empty?
+        level = @src["level"].length
 
+        start_line_number = @src.current_line_number
         @src.pos += @src.matched_size
-        el = new_block_el(:header, nil, nil, :level => level.length, :raw_text => text, :location => start_line_number)
+        el = new_block_el(:header, nil, nil, :level => level, :raw_text => text, :location => start_line_number)
         add_text(text, el)
         el.attr['id'] = id if id
         @tree.children << el
diff --git a/lib/kramdown/parser/kramdown/header.rb b/lib/kramdown/parser/kramdown/header.rb
index aa882dd9..fc2af227 100644
--- a/lib/kramdown/parser/kramdown/header.rb
+++ b/lib/kramdown/parser/kramdown/header.rb
@@ -13,18 +13,27 @@ module Kramdown
   module Parser
     class Kramdown
 
-      HEADER_ID=/(?:[ \t]+\{#([A-Za-z][\w:-]*)\})?/
-      SETEXT_HEADER_START = /^(#{OPT_SPACE}[^ \t].*?)#{HEADER_ID}[ \t]*?\n(-|=)+\s*?\n/
+      HEADER_ID = /[\t ]{#(?<id>[A-Za-z][\w:-]*)}\z/
+      SETEXT_HEADER_START = /^#{OPT_SPACE}(?<contents>.*)\n(?<level>[-=])[-=]*[ \t\r\f\v]*\n/
 
       # Parse the Setext header at the current location.
       def parse_setext_header
         return false if !after_block_boundary?
 
+        text = @src["contents"]
+        text.strip!
+        id_match = HEADER_ID.match(text)
+        if id_match
+          id = id_match["id"]
+          text = text[0...-id_match[0].length]
+          text.rstrip!
+        end
+        return false if text.empty?
+        level = @src["level"] == '-' ? 2 : 1
+
         start_line_number = @src.current_line_number
         @src.pos += @src.matched_size
-        text, id, level = @src[1], @src[2], @src[3]
-        text.strip!
-        el = new_block_el(:header, nil, nil, :level => (level == '-' ? 2 : 1), :raw_text => text, :location => start_line_number)
+        el = new_block_el(:header, nil, nil, :level => level, :raw_text => text, :location => start_line_number)
         add_text(text, el)
         el.attr['id'] = id if id
         @tree.children << el
@@ -33,20 +42,28 @@ def parse_setext_header
       define_parser(:setext_header, SETEXT_HEADER_START)
 
 
-      ATX_HEADER_START = /^\#{1,6}/
-      ATX_HEADER_MATCH = /^(\#{1,6})(.+?(?:\\#)?)\s*?#*#{HEADER_ID}\s*?\n/
+      ATX_HEADER_START = /^(?<level>\#{1,6})[\t ]*(?<contents>.*)\n/
 
       # Parse the Atx header at the current location.
       def parse_atx_header
         return false if !after_block_boundary?
 
-        start_line_number = @src.current_line_number
-        @src.check(ATX_HEADER_MATCH)
-        level, text, id = @src[1], @src[2].to_s.strip, @src[3]
+        text = @src["contents"]
+        text.rstrip!
+        id_match = HEADER_ID.match(text)
+        if id_match
+          id = id_match["id"]
+          text = text[0...-id_match[0].length]
+          text.rstrip!
+        end
+        text.sub!(/[\t ]#+\z/, '')
+        text.rstrip!
         return false if text.empty?
+        level = @src["level"].length
 
+        start_line_number = @src.current_line_number
         @src.pos += @src.matched_size
-        el = new_block_el(:header, nil, nil, :level => level.length, :raw_text => text, :location => start_line_number)
+        el = new_block_el(:header, nil, nil, :level => level, :raw_text => text, :location => start_line_number)
         add_text(text, el)
         el.attr['id'] = id if id
         @tree.children << el
diff --git a/test/testcases/block/04_header/atx_header.hcd b/test/testcases/block/04_header/atx_header.hcd
new file mode 100644
index 00000000..e69de29b