Document SanitizedAnchorName algorithm, copy implementation. (#509)

The goal of this change is to reduce number of non-standard library
packages (repositories) that blackfriday imports (not counting imports
used only for tests) 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.

This change is a port of PR #352 from v1 into v2.

Updates #348.
Updates #350.

README.md

@@ -104,6 +104,8 @@ html := bluemonday.UGCPolicy().SanitizeBytes(unsafe)
 If you want to customize the set of options, use `blackfriday.WithExtensions`,
 `blackfriday.WithRenderer` and `blackfriday.WithRefOverride`.
 
+### `blackfriday-tool`
+
 You can also check out `blackfriday-tool` for a more complete example
 of how to use it. Download and install it using:
 
@@ -123,6 +125,22 @@ 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 `AutoHeadingIDs` extension 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/gopkg.in/russross/blackfriday.v2#hdr-Sanitized_Anchor_Names.
+
+[`SanitizedAnchorName`](https://godoc.org/gopkg.in/russross/blackfriday.v2#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
 --------

block.go

@@ -18,8 +18,7 @@ import (
 	"html"
 	"regexp"
 	"strings"
-
-	"github.com/shurcooL/sanitized_anchor_name"
+	"unicode"
 )
 
 const (
@@ -259,7 +258,7 @@ func (p *Markdown) prefixHeading(data []byte) int {
 	}
 	if end > i {
 		if id == "" && p.extensions&AutoHeadingIDs != 0 {
-			id = sanitized_anchor_name.Create(string(data[i:end]))
+			id = SanitizedAnchorName(string(data[i:end]))
 		}
 		block := p.addBlock(Heading, data[i:end])
 		block.HeadingID = id
@@ -1505,7 +1504,7 @@ func (p *Markdown) paragraph(data []byte) int {
 
 				id := ""
 				if p.extensions&AutoHeadingIDs != 0 {
-					id = sanitized_anchor_name.Create(string(data[prev:eol]))
+					id = SanitizedAnchorName(string(data[prev:eol]))
 				}
 
 				block := p.addBlock(Heading, data[prev:eol])
@@ -1590,3 +1589,24 @@ func skipUntilChar(text []byte, start int, char byte) int {
 	}
 	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)
+}

block_test.go

@@ -1875,3 +1875,44 @@ func TestIsFenceLine(t *testing.T) {
 		}
 	}
 }
+
+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)
+		}
+	}
+}

doc.go

@@ -15,4 +15,32 @@
 //
 // If you're interested in calling Blackfriday from command line, see
 // https://github.com/russross/blackfriday-tool.
+//
+// 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 AutoHeadingIDs extension 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 precede 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.

go.mod

@@ -1,6 +1,3 @@
 module github.com/russross/blackfriday/v2
 
-require (
-	github.com/pmezard/go-difflib v1.0.0
-	github.com/shurcooL/sanitized_anchor_name v0.0.0-20170918181015-86672fcb3f95
-)
+require github.com/pmezard/go-difflib v1.0.0

go.sum

@@ -1,4 +1,2 @@
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
-github.com/shurcooL/sanitized_anchor_name v0.0.0-20170918181015-86672fcb3f95 h1:/vdW8Cb7EXrkqWGufVMES1OH2sU9gKVb2n9/1y5NMBY=
-github.com/shurcooL/sanitized_anchor_name v0.0.0-20170918181015-86672fcb3f95/go.mod h1:1NzhyTcUVG4SuEtjjoZeVRXNmyL/1OwPU0+IJeTBvfc=