From 0ba0f2b6ed7c475a92e4df8641825cb7a11d1fa3 Mon Sep 17 00:00:00 2001
From: Dmitri Shuralyov <shurcooL@gmail.com>
Date: Tue, 9 May 2017 02:07:14 -0400
Subject: [PATCH] Document SanitizedAnchorName algorithm, copy implementation.
 (#352)

The goal of this change is to reduce number of non-standard library
packages (repositories) that blackfriday imports from 1 to 0, and in
turn, reduce the cost of importing blackfriday into other projects.

Do so by documenting the algorithm of SanitizedAnchorName, and include
a copy of the small function inside blackfriday itself. The same
functionality continues to be available in the original location,
github.com/shurcooL/sanitized_anchor_name.Create. It can be used by
existing users and those that look for a small package, and don't need
all of blackfriday functionality. Existing users of blackfriday can use
the new SanitizedAnchorName function directly and avoid an extra
package import.

Resolves #350.
---
 README.md     | 16 ++++++++++++++++
 block.go      | 28 ++++++++++++++++++++++++----
 block_test.go | 41 +++++++++++++++++++++++++++++++++++++++++
 doc.go        | 32 ++++++++++++++++++++++++++++++++
 markdown.go   |  3 ---
 5 files changed, 113 insertions(+), 7 deletions(-)
 create mode 100644 doc.go

diff --git a/README.md b/README.md
index 3c113cb..0d1ac9a 100644
--- a/README.md
+++ b/README.md
@@ -93,6 +93,22 @@ installed in `$GOPATH/bin`.  This is a statically-linked binary that
 can be copied to wherever you need it without worrying about
 dependencies and library versions.
 
+### Sanitized anchor names
+
+Blackfriday includes an algorithm for creating sanitized anchor names
+corresponding to a given input text. This algorithm is used to create
+anchors for headings when `EXTENSION_AUTO_HEADER_IDS` is enabled. The
+algorithm has a specification, so that other packages can create
+compatible anchor names and links to those anchors.
+
+The specification is located at https://godoc.org/github.com/russross/blackfriday#hdr-Sanitized_Anchor_Names.
+
+[`SanitizedAnchorName`](https://godoc.org/github.com/russross/blackfriday#SanitizedAnchorName) exposes this functionality, and can be used to
+create compatible links to the anchor names generated by blackfriday.
+This algorithm is also implemented in a small standalone package at
+[`github.com/shurcooL/sanitized_anchor_name`](https://godoc.org/github.com/shurcooL/sanitized_anchor_name). It can be useful for clients
+that want a small package and don't need full functionality of blackfriday.
+
 
 Features
 --------
diff --git a/block.go b/block.go
index 9cf451f..7fc731d 100644
--- a/block.go
+++ b/block.go
@@ -15,8 +15,7 @@ package blackfriday
 
 import (
 	"bytes"
-
-	"github.com/shurcooL/sanitized_anchor_name"
+	"unicode"
 )
 
 // Parse block-level data.
@@ -243,7 +242,7 @@ func (p *parser) prefixHeader(out *bytes.Buffer, data []byte) int {
 	}
 	if end > i {
 		if id == "" && p.flags&EXTENSION_AUTO_HEADER_IDS != 0 {
-			id = sanitized_anchor_name.Create(string(data[i:end]))
+			id = SanitizedAnchorName(string(data[i:end]))
 		}
 		work := func() bool {
 			p.inline(out, data[i:end])
@@ -1364,7 +1363,7 @@ func (p *parser) paragraph(out *bytes.Buffer, data []byte) int {
 
 				id := ""
 				if p.flags&EXTENSION_AUTO_HEADER_IDS != 0 {
-					id = sanitized_anchor_name.Create(string(data[prev:eol]))
+					id = SanitizedAnchorName(string(data[prev:eol]))
 				}
 
 				p.r.Header(out, work, level, id)
@@ -1428,3 +1427,24 @@ func (p *parser) paragraph(out *bytes.Buffer, data []byte) int {
 	p.renderParagraph(out, data[:i])
 	return i
 }
+
+// SanitizedAnchorName returns a sanitized anchor name for the given text.
+//
+// It implements the algorithm specified in the package comment.
+func SanitizedAnchorName(text string) string {
+	var anchorName []rune
+	futureDash := false
+	for _, r := range text {
+		switch {
+		case unicode.IsLetter(r) || unicode.IsNumber(r):
+			if futureDash && len(anchorName) > 0 {
+				anchorName = append(anchorName, '-')
+			}
+			futureDash = false
+			anchorName = append(anchorName, unicode.ToLower(r))
+		default:
+			futureDash = true
+		}
+	}
+	return string(anchorName)
+}
diff --git a/block_test.go b/block_test.go
index 83ed2e1..89d5775 100644
--- a/block_test.go
+++ b/block_test.go
@@ -1738,3 +1738,44 @@ func TestJoinLines(t *testing.T) {
 		t.Error("output dose not match.")
 	}
 }
+
+func TestSanitizedAnchorName(t *testing.T) {
+	tests := []struct {
+		text string
+		want string
+	}{
+		{
+			text: "This is a header",
+			want: "this-is-a-header",
+		},
+		{
+			text: "This is also          a header",
+			want: "this-is-also-a-header",
+		},
+		{
+			text: "main.go",
+			want: "main-go",
+		},
+		{
+			text: "Article 123",
+			want: "article-123",
+		},
+		{
+			text: "<- Let's try this, shall we?",
+			want: "let-s-try-this-shall-we",
+		},
+		{
+			text: "        ",
+			want: "",
+		},
+		{
+			text: "Hello, 世界",
+			want: "hello-世界",
+		},
+	}
+	for _, test := range tests {
+		if got := SanitizedAnchorName(test.text); got != test.want {
+			t.Errorf("SanitizedAnchorName(%q):\ngot %q\nwant %q", test.text, got, test.want)
+		}
+	}
+}
diff --git a/doc.go b/doc.go
new file mode 100644
index 0000000..9656c42
--- /dev/null
+++ b/doc.go
@@ -0,0 +1,32 @@
+// Package blackfriday is a Markdown processor.
+//
+// It translates plain text with simple formatting rules into HTML or LaTeX.
+//
+// Sanitized Anchor Names
+//
+// Blackfriday includes an algorithm for creating sanitized anchor names
+// corresponding to a given input text. This algorithm is used to create
+// anchors for headings when EXTENSION_AUTO_HEADER_IDS is enabled. The
+// algorithm is specified below, so that other packages can create
+// compatible anchor names and links to those anchors.
+//
+// The algorithm iterates over the input text, interpreted as UTF-8,
+// one Unicode code point (rune) at a time. All runes that are letters (category L)
+// or numbers (category N) are considered valid characters. They are mapped to
+// lower case, and included in the output. All other runes are considered
+// invalid characters. Invalid characters that preceed the first valid character,
+// as well as invalid character that follow the last valid character
+// are dropped completely. All other sequences of invalid characters
+// between two valid characters are replaced with a single dash character '-'.
+//
+// SanitizedAnchorName exposes this functionality, and can be used to
+// create compatible links to the anchor names generated by blackfriday.
+// This algorithm is also implemented in a small standalone package at
+// github.com/shurcooL/sanitized_anchor_name. It can be useful for clients
+// that want a small package and don't need full functionality of blackfriday.
+package blackfriday
+
+// NOTE: Keep Sanitized Anchor Name algorithm in sync with package
+//       github.com/shurcooL/sanitized_anchor_name.
+//       Otherwise, users of sanitized_anchor_name will get anchor names
+//       that are incompatible with those generated by blackfriday.
diff --git a/markdown.go b/markdown.go
index 09fc2d7..6d842d3 100644
--- a/markdown.go
+++ b/markdown.go
@@ -13,9 +13,6 @@
 //
 //
 
-// Blackfriday markdown processor.
-//
-// Translates plain text with simple formatting rules into HTML or LaTeX.
 package blackfriday
 
 import (