shithub: hugo

Download patch

ref: 366c557251c73288987a700c63d148a69e9f0ec0
parent: ba53799fdb7c59af8733df2027d245584ca6749f
author: Jeffrey Tolar <[email protected]>
date: Wed Jan 28 16:11:41 EST 2015 

Use a regular expression in replaceShortcodeTokens

This fixes a bug where a shortcode needs to be expanded multiple times,
which can arise in practice when using reference links.


--- a/hugolib/handler_page.go
+++ b/hugolib/handler_page.go
@@ -60,7 +60,7 @@
 	tmpContent, tmpTableOfContents := helpers.ExtractTOC(p.renderContent(helpers.RemoveSummaryDivider(p.rawContent)))
 
 	if len(p.contentShortCodes) > 0 {
-		tmpContentWithTokensReplaced, err := replaceShortcodeTokens(tmpContent, shortcodePlaceholderPrefix, -1, true, p.contentShortCodes)
+		tmpContentWithTokensReplaced, err := replaceShortcodeTokens(tmpContent, shortcodePlaceholderPrefix, true, p.contentShortCodes)
 
 		if err != nil {
 			jww.FATAL.Printf("Fail to replace short code tokens in %s:\n%s", p.BaseFileName(), err.Error())
@@ -113,7 +113,7 @@
 	tmpContent, tmpTableOfContents := helpers.ExtractTOC(p.renderContent(helpers.RemoveSummaryDivider(p.rawContent)))
 
 	if len(p.contentShortCodes) > 0 {
-		tmpContentWithTokensReplaced, err := replaceShortcodeTokens(tmpContent, shortcodePlaceholderPrefix, -1, true, p.contentShortCodes)
+		tmpContentWithTokensReplaced, err := replaceShortcodeTokens(tmpContent, shortcodePlaceholderPrefix, true, p.contentShortCodes)
 
 		if err != nil {
 			jww.FATAL.Printf("Fail to replace short code tokens in %s:\n%s", p.BaseFileName(), err.Error())
--- a/hugolib/page.go
+++ b/hugolib/page.go
@@ -162,10 +162,9 @@
 		p.Truncated = true // by definition
 		header := bytes.Split(p.rawContent, helpers.SummaryDivider)[0]
 		renderedHeader := p.renderBytes(header)
-		numShortcodesInHeader := bytes.Count(header, []byte(shortcodePlaceholderPrefix))
 		if len(p.contentShortCodes) > 0 {
 			tmpContentWithTokensReplaced, err :=
-				replaceShortcodeTokens(renderedHeader, shortcodePlaceholderPrefix, numShortcodesInHeader, true, p.contentShortCodes)
+				replaceShortcodeTokens(renderedHeader, shortcodePlaceholderPrefix, true, p.contentShortCodes)
 			if err != nil {
 				jww.FATAL.Printf("Failed to replace short code tokens in Summary for %s:\n%s", p.BaseFileName(), err.Error())
 			} else {
--- a/hugolib/shortcode.go
+++ b/hugolib/shortcode.go
@@ -20,7 +20,6 @@
 	"reflect"
 	"regexp"
 	"sort"
-	"strconv"
 	"strings"
 	"sync"
 
@@ -132,7 +131,7 @@
 	tmpContent, tmpShortcodes := extractAndRenderShortcodes(stringToParse, page, t)
 
 	if len(tmpShortcodes) > 0 {
-		tmpContentWithTokensReplaced, err := replaceShortcodeTokens([]byte(tmpContent), shortcodePlaceholderPrefix, -1, true, tmpShortcodes)
+		tmpContentWithTokensReplaced, err := replaceShortcodeTokens([]byte(tmpContent), shortcodePlaceholderPrefix, true, tmpShortcodes)
 
 		if err != nil {
 			jww.ERROR.Printf("Fail to replace short code tokens in %s:\n%s", page.BaseFileName(), err.Error())
@@ -428,60 +427,44 @@
 }
 
 // Replace prefixed shortcode tokens (HUGOSHORTCODE-1, HUGOSHORTCODE-2) with the real content.
-// This assumes that all tokens exist in the input string and that they are in order.
-// numReplacements = -1 will do len(replacements), and it will always start from the beginning (1)
 // wrapped = true means that the token has been wrapped in {@{@/@}@}
-func replaceShortcodeTokens(source []byte, prefix string, numReplacements int, wrapped bool, replacements map[string]string) ([]byte, error) {
+func replaceShortcodeTokens(source []byte, prefix string, wrapped bool, replacements map[string]string) (b []byte, err error) {
+	var re *regexp.Regexp
 
-	if numReplacements < 0 {
-		numReplacements = len(replacements)
-	}
-
-	if numReplacements == 0 {
-		return source, nil
-	}
-
-	newLen := len(source)
-
-	for i := 1; i <= numReplacements; i++ {
-		key := prefix + "-" + strconv.Itoa(i)
-
-		if wrapped {
-			key = "{@{@" + key + "@}@}"
+	if wrapped {
+		re, err = regexp.Compile(`\{@\{@` + regexp.QuoteMeta(prefix) + `-\d+@\}@\}`)
+		if err != nil {
+			return nil, err
 		}
-		val := []byte(replacements[key])
-
-		newLen += (len(val) - len(key))
+	} else {
+		re, err = regexp.Compile(regexp.QuoteMeta(prefix) + `-(\d+)`)
+		if err != nil {
+			return nil, err
+		}
 	}
 
-	buff := make([]byte, newLen)
-
-	width := 0
-	start := 0
-
-	for i := 0; i < numReplacements; i++ {
-		tokenNum := i + 1
-		oldVal := prefix + "-" + strconv.Itoa(tokenNum)
-		if wrapped {
-			oldVal = "{@{@" + oldVal + "@}@}"
+	// use panic/recover for reporting if an unknown
+	defer func() {
+		if r := recover(); r != nil {
+			var ok bool
+			b = nil
+			err, ok = r.(error)
+			if !ok {
+				err = fmt.Errorf("unexpected panic during replaceShortcodeTokens: %v", r)
+			}
 		}
-		newVal := []byte(replacements[oldVal])
-		j := start
+	}()
+	b = re.ReplaceAllFunc(source, func(m []byte) []byte {
+		key := string(m)
 
-		k := bytes.Index(source[start:], []byte(oldVal))
-
-		if k < 0 {
-			// this should never happen, but let the caller decide to panic or not
-			return nil, fmt.Errorf("illegal state in content; shortcode token #%d is missing or out of order (%q)", tokenNum, source)
+		if val, ok := replacements[key]; ok {
+			return []byte(val)
+		} else {
+			panic(fmt.Errorf("unknown shortcode token %q", key))
 		}
-		j += k
+	})
 
-		width += copy(buff[width:], source[start:j])
-		width += copy(buff[width:], newVal)
-		start = j + len(oldVal)
-	}
-	width += copy(buff[width:], source[start:])
-	return buff[0:width], nil
+	return b, err
 }
 
 func GetTemplate(name string, t tpl.Template) *template.Template {
--- a/hugolib/shortcode_test.go
+++ b/hugolib/shortcode_test.go
@@ -281,23 +281,24 @@
 
 func TestReplaceShortcodeTokens(t *testing.T) {
 	for i, this := range []struct {
-		input           []byte
-		prefix          string
-		replacements    map[string]string
-		numReplacements int
-		wrappedInDiv    bool
-		expect          interface{}
+		input        []byte
+		prefix       string
+		replacements map[string]string
+		wrappedInDiv bool
+		expect       interface{}
 	}{
-		{[]byte("Hello PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "World"}, -1, false, []byte("Hello World.")},
-		{[]byte("A {@{@A-1@}@} asdf {@{@A-2@}@}."), "A", map[string]string{"{@{@A-1@}@}": "v1", "{@{@A-2@}@}": "v2"}, -1, true, []byte("A v1 asdf v2.")},
-		{[]byte("Hello PREFIX2-1. Go PREFIX2-2, Go, Go PREFIX2-3 Go Go!."), "PREFIX2", map[string]string{"PREFIX2-1": "Europe", "PREFIX2-2": "Jonny", "PREFIX2-3": "Johnny"}, -1, false, []byte("Hello Europe. Go Jonny, Go, Go Johnny Go Go!.")},
-		{[]byte("A PREFIX-2 PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, -1, false, false},
-		{[]byte("A PREFIX-1 PREFIX-2"), "PREFIX", map[string]string{"PREFIX-1": "A"}, -1, false, []byte("A A PREFIX-2")},
-		{[]byte("A PREFIX-1 but not the second."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, -1, false, false},
-		{[]byte("An PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, 1, false, []byte("An A.")},
-		{[]byte("An PREFIX-1 PREFIX-2."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, 1, false, []byte("An A PREFIX-2.")},
+		{[]byte("Hello PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "World"}, false, []byte("Hello World.")},
+		{[]byte("A {@{@A-1@}@} asdf {@{@A-2@}@}."), "A", map[string]string{"{@{@A-1@}@}": "v1", "{@{@A-2@}@}": "v2"}, true, []byte("A v1 asdf v2.")},
+		{[]byte("Hello PREFIX2-1. Go PREFIX2-2, Go, Go PREFIX2-3 Go Go!."), "PREFIX2", map[string]string{"PREFIX2-1": "Europe", "PREFIX2-2": "Jonny", "PREFIX2-3": "Johnny"}, false, []byte("Hello Europe. Go Jonny, Go, Go Johnny Go Go!.")},
+		{[]byte("A PREFIX-2 PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, false, []byte("A B A.")},
+		{[]byte("A PREFIX-1 PREFIX-2"), "PREFIX", map[string]string{"PREFIX-1": "A"}, false, false},
+		{[]byte("A PREFIX-1 but not the second."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, false, []byte("A A but not the second.")},
+		{[]byte("An PREFIX-1."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, false, []byte("An A.")},
+		{[]byte("An PREFIX-1 PREFIX-2."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B"}, false, []byte("An A B.")},
+		{[]byte("A PREFIX-1 PREFIX-2 PREFIX-3 PREFIX-1 PREFIX-3."), "PREFIX", map[string]string{"PREFIX-1": "A", "PREFIX-2": "B", "PREFIX-3": "C"}, false, []byte("A A B C A C.")},
+		{[]byte("A {@{@PREFIX-1@}@} {@{@PREFIX-2@}@} {@{@PREFIX-3@}@} {@{@PREFIX-1@}@} {@{@PREFIX-3@}@}."), "PREFIX", map[string]string{"{@{@PREFIX-1@}@}": "A", "{@{@PREFIX-2@}@}": "B", "{@{@PREFIX-3@}@}": "C"}, true, []byte("A A B C A C.")},
 	} {
-		results, err := replaceShortcodeTokens(this.input, this.prefix, this.numReplacements, this.wrappedInDiv, this.replacements)
+		results, err := replaceShortcodeTokens(this.input, this.prefix, this.wrappedInDiv, this.replacements)
 
 		if b, ok := this.expect.(bool); ok && !b {
 			if err == nil {