Initial commit: Go 1.23 release state

This commit is contained in:
Vorapol Rinsatitnon
2024-09-21 23:49:08 +10:00
commit 17cd57a668
13231 changed files with 3114330 additions and 0 deletions

2265
src/html/entity.go Normal file

File diff suppressed because it is too large Load Diff

37
src/html/entity_test.go Normal file
View File

@@ -0,0 +1,37 @@
// Copyright 2010 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package html
import (
"testing"
"unicode/utf8"
)
func init() {
UnescapeString("") // force load of entity maps
}
func TestEntityLength(t *testing.T) {
if len(entity) == 0 || len(entity2) == 0 {
t.Fatal("maps not loaded")
}
// We verify that the length of UTF-8 encoding of each value is <= 1 + len(key).
// The +1 comes from the leading "&". This property implies that the length of
// unescaped text is <= the length of escaped text.
for k, v := range entity {
if 1+len(k) < utf8.RuneLen(v) {
t.Error("escaped entity &" + k + " is shorter than its UTF-8 encoding " + string(v))
}
if len(k) > longestEntityWithoutSemicolon && k[len(k)-1] != ';' {
t.Errorf("entity name %s is %d characters, but longestEntityWithoutSemicolon=%d", k, len(k), longestEntityWithoutSemicolon)
}
}
for k, v := range entity2 {
if 1+len(k) < utf8.RuneLen(v[0])+utf8.RuneLen(v[1]) {
t.Error("escaped entity &" + k + " is shorter than its UTF-8 encoding " + string(v[0]) + string(v[1]))
}
}
}

214
src/html/escape.go Normal file
View File

@@ -0,0 +1,214 @@
// Copyright 2010 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Package html provides functions for escaping and unescaping HTML text.
package html
import (
"strings"
"unicode/utf8"
)
// These replacements permit compatibility with old numeric entities that
// assumed Windows-1252 encoding.
// https://html.spec.whatwg.org/multipage/parsing.html#numeric-character-reference-end-state
var replacementTable = [...]rune{
'\u20AC', // First entry is what 0x80 should be replaced with.
'\u0081',
'\u201A',
'\u0192',
'\u201E',
'\u2026',
'\u2020',
'\u2021',
'\u02C6',
'\u2030',
'\u0160',
'\u2039',
'\u0152',
'\u008D',
'\u017D',
'\u008F',
'\u0090',
'\u2018',
'\u2019',
'\u201C',
'\u201D',
'\u2022',
'\u2013',
'\u2014',
'\u02DC',
'\u2122',
'\u0161',
'\u203A',
'\u0153',
'\u009D',
'\u017E',
'\u0178', // Last entry is 0x9F.
// 0x00->'\uFFFD' is handled programmatically.
// 0x0D->'\u000D' is a no-op.
}
// unescapeEntity reads an entity like "&lt;" from b[src:] and writes the
// corresponding "<" to b[dst:], returning the incremented dst and src cursors.
// Precondition: b[src] == '&' && dst <= src.
func unescapeEntity(b []byte, dst, src int) (dst1, src1 int) {
const attribute = false
// http://www.whatwg.org/specs/web-apps/current-work/multipage/tokenization.html#consume-a-character-reference
// i starts at 1 because we already know that s[0] == '&'.
i, s := 1, b[src:]
if len(s) <= 1 {
b[dst] = b[src]
return dst + 1, src + 1
}
if s[i] == '#' {
if len(s) <= 3 { // We need to have at least "&#.".
b[dst] = b[src]
return dst + 1, src + 1
}
i++
c := s[i]
hex := false
if c == 'x' || c == 'X' {
hex = true
i++
}
x := '\x00'
for i < len(s) {
c = s[i]
i++
if hex {
if '0' <= c && c <= '9' {
x = 16*x + rune(c) - '0'
continue
} else if 'a' <= c && c <= 'f' {
x = 16*x + rune(c) - 'a' + 10
continue
} else if 'A' <= c && c <= 'F' {
x = 16*x + rune(c) - 'A' + 10
continue
}
} else if '0' <= c && c <= '9' {
x = 10*x + rune(c) - '0'
continue
}
if c != ';' {
i--
}
break
}
if i <= 3 { // No characters matched.
b[dst] = b[src]
return dst + 1, src + 1
}
if 0x80 <= x && x <= 0x9F {
// Replace characters from Windows-1252 with UTF-8 equivalents.
x = replacementTable[x-0x80]
} else if x == 0 || (0xD800 <= x && x <= 0xDFFF) || x > 0x10FFFF {
// Replace invalid characters with the replacement character.
x = '\uFFFD'
}
return dst + utf8.EncodeRune(b[dst:], x), src + i
}
// Consume the maximum number of characters possible, with the
// consumed characters matching one of the named references.
for i < len(s) {
c := s[i]
i++
// Lower-cased characters are more common in entities, so we check for them first.
if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' || '0' <= c && c <= '9' {
continue
}
if c != ';' {
i--
}
break
}
entityName := s[1:i]
if len(entityName) == 0 {
// No-op.
} else if attribute && entityName[len(entityName)-1] != ';' && len(s) > i && s[i] == '=' {
// No-op.
} else if x := entity[string(entityName)]; x != 0 {
return dst + utf8.EncodeRune(b[dst:], x), src + i
} else if x := entity2[string(entityName)]; x[0] != 0 {
dst1 := dst + utf8.EncodeRune(b[dst:], x[0])
return dst1 + utf8.EncodeRune(b[dst1:], x[1]), src + i
} else if !attribute {
maxLen := len(entityName) - 1
if maxLen > longestEntityWithoutSemicolon {
maxLen = longestEntityWithoutSemicolon
}
for j := maxLen; j > 1; j-- {
if x := entity[string(entityName[:j])]; x != 0 {
return dst + utf8.EncodeRune(b[dst:], x), src + j + 1
}
}
}
dst1, src1 = dst+i, src+i
copy(b[dst:dst1], b[src:src1])
return dst1, src1
}
var htmlEscaper = strings.NewReplacer(
`&`, "&amp;",
`'`, "&#39;", // "&#39;" is shorter than "&apos;" and apos was not in HTML until HTML5.
`<`, "&lt;",
`>`, "&gt;",
`"`, "&#34;", // "&#34;" is shorter than "&quot;".
)
// EscapeString escapes special characters like "<" to become "&lt;". It
// escapes only five such characters: <, >, &, ' and ".
// UnescapeString(EscapeString(s)) == s always holds, but the converse isn't
// always true.
func EscapeString(s string) string {
return htmlEscaper.Replace(s)
}
// UnescapeString unescapes entities like "&lt;" to become "<". It unescapes a
// larger range of entities than EscapeString escapes. For example, "&aacute;"
// unescapes to "á", as does "&#225;" and "&#xE1;".
// UnescapeString(EscapeString(s)) == s always holds, but the converse isn't
// always true.
func UnescapeString(s string) string {
populateMapsOnce.Do(populateMaps)
i := strings.IndexByte(s, '&')
if i < 0 {
return s
}
b := []byte(s)
dst, src := unescapeEntity(b, i, i)
for len(s[src:]) > 0 {
if s[src] == '&' {
i = 0
} else {
i = strings.IndexByte(s[src:], '&')
}
if i < 0 {
dst += copy(b[dst:], s[src:])
break
}
if i > 0 {
copy(b[dst:], s[src:src+i])
}
dst, src = unescapeEntity(b, dst+i, src+i)
}
return string(b[:dst])
}

169
src/html/escape_test.go Normal file
View File

@@ -0,0 +1,169 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package html
import (
"strings"
"testing"
)
type unescapeTest struct {
// A short description of the test case.
desc string
// The HTML text.
html string
// The unescaped text.
unescaped string
}
var unescapeTests = []unescapeTest{
// Handle no entities.
{
"copy",
"A\ttext\nstring",
"A\ttext\nstring",
},
// Handle simple named entities.
{
"simple",
"&amp; &gt; &lt;",
"& > <",
},
// Handle hitting the end of the string.
{
"stringEnd",
"&amp &amp",
"& &",
},
// Handle entities with two codepoints.
{
"multiCodepoint",
"text &gesl; blah",
"text \u22db\ufe00 blah",
},
// Handle decimal numeric entities.
{
"decimalEntity",
"Delta = &#916; ",
"Delta = Δ ",
},
// Handle hexadecimal numeric entities.
{
"hexadecimalEntity",
"Lambda = &#x3bb; = &#X3Bb ",
"Lambda = λ = λ ",
},
// Handle numeric early termination.
{
"numericEnds",
"&# &#x &#128;43 &copy = &#169f = &#xa9",
"&# &#x €43 © = ©f = ©",
},
// Handle numeric ISO-8859-1 entity replacements.
{
"numericReplacements",
"Footnote&#x87;",
"Footnote‡",
},
// Handle single ampersand.
{
"copySingleAmpersand",
"&",
"&",
},
// Handle ampersand followed by non-entity.
{
"copyAmpersandNonEntity",
"text &test",
"text &test",
},
// Handle "&#".
{
"copyAmpersandHash",
"text &#",
"text &#",
},
}
func TestUnescape(t *testing.T) {
for _, tt := range unescapeTests {
unescaped := UnescapeString(tt.html)
if unescaped != tt.unescaped {
t.Errorf("TestUnescape %s: want %q, got %q", tt.desc, tt.unescaped, unescaped)
}
}
}
func TestUnescapeEscape(t *testing.T) {
ss := []string{
``,
`abc def`,
`a & b`,
`a&amp;b`,
`a &amp b`,
`&quot;`,
`"`,
`"<&>"`,
`&quot;&lt;&amp;&gt;&quot;`,
`3&5==1 && 0<1, "0&lt;1", a+acute=&aacute;`,
`The special characters are: <, >, &, ' and "`,
}
for _, s := range ss {
if got := UnescapeString(EscapeString(s)); got != s {
t.Errorf("got %q want %q", got, s)
}
}
}
var (
benchEscapeData = strings.Repeat("AAAAA < BBBBB > CCCCC & DDDDD ' EEEEE \" ", 100)
benchEscapeNone = strings.Repeat("AAAAA x BBBBB x CCCCC x DDDDD x EEEEE x ", 100)
benchUnescapeSparse = strings.Repeat(strings.Repeat("AAAAA x BBBBB x CCCCC x DDDDD x EEEEE x ", 10)+"&amp;", 10)
benchUnescapeDense = strings.Repeat("&amp;&lt; &amp; &lt;", 100)
)
func BenchmarkEscape(b *testing.B) {
n := 0
for i := 0; i < b.N; i++ {
n += len(EscapeString(benchEscapeData))
}
}
func BenchmarkEscapeNone(b *testing.B) {
n := 0
for i := 0; i < b.N; i++ {
n += len(EscapeString(benchEscapeNone))
}
}
func BenchmarkUnescape(b *testing.B) {
s := EscapeString(benchEscapeData)
n := 0
for i := 0; i < b.N; i++ {
n += len(UnescapeString(s))
}
}
func BenchmarkUnescapeNone(b *testing.B) {
s := EscapeString(benchEscapeNone)
n := 0
for i := 0; i < b.N; i++ {
n += len(UnescapeString(s))
}
}
func BenchmarkUnescapeSparse(b *testing.B) {
n := 0
for i := 0; i < b.N; i++ {
n += len(UnescapeString(benchUnescapeSparse))
}
}
func BenchmarkUnescapeDense(b *testing.B) {
n := 0
for i := 0; i < b.N; i++ {
n += len(UnescapeString(benchUnescapeDense))
}
}

22
src/html/example_test.go Normal file
View File

@@ -0,0 +1,22 @@
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package html_test
import (
"fmt"
"html"
)
func ExampleEscapeString() {
const s = `"Fran & Freddie's Diner" <tasty@example.com>`
fmt.Println(html.EscapeString(s))
// Output: &#34;Fran &amp; Freddie&#39;s Diner&#34; &lt;tasty@example.com&gt;
}
func ExampleUnescapeString() {
const s = `&quot;Fran &amp; Freddie&#39;s Diner&quot; &lt;tasty@example.com&gt;`
fmt.Println(html.UnescapeString(s))
// Output: "Fran & Freddie's Diner" <tasty@example.com>
}

22
src/html/fuzz_test.go Normal file
View File

@@ -0,0 +1,22 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package html
import "testing"
func FuzzEscapeUnescape(f *testing.F) {
f.Fuzz(func(t *testing.T, v string) {
e := EscapeString(v)
u := UnescapeString(e)
if u != v {
t.Errorf("EscapeString(%q) = %q, UnescapeString(%q) = %q, want %q", v, e, e, u, v)
}
// As per the documentation, this isn't always equal to v, so it makes
// no sense to check for equality. It can still be interesting to find
// panics in it though.
EscapeString(UnescapeString(v))
})
}

175
src/html/template/attr.go Normal file
View File

@@ -0,0 +1,175 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"strings"
)
// attrTypeMap[n] describes the value of the given attribute.
// If an attribute affects (or can mask) the encoding or interpretation of
// other content, or affects the contents, idempotency, or credentials of a
// network message, then the value in this map is contentTypeUnsafe.
// This map is derived from HTML5, specifically
// https://www.w3.org/TR/html5/Overview.html#attributes-1
// as well as "%URI"-typed attributes from
// https://www.w3.org/TR/html4/index/attributes.html
var attrTypeMap = map[string]contentType{
"accept": contentTypePlain,
"accept-charset": contentTypeUnsafe,
"action": contentTypeURL,
"alt": contentTypePlain,
"archive": contentTypeURL,
"async": contentTypeUnsafe,
"autocomplete": contentTypePlain,
"autofocus": contentTypePlain,
"autoplay": contentTypePlain,
"background": contentTypeURL,
"border": contentTypePlain,
"checked": contentTypePlain,
"cite": contentTypeURL,
"challenge": contentTypeUnsafe,
"charset": contentTypeUnsafe,
"class": contentTypePlain,
"classid": contentTypeURL,
"codebase": contentTypeURL,
"cols": contentTypePlain,
"colspan": contentTypePlain,
"content": contentTypeUnsafe,
"contenteditable": contentTypePlain,
"contextmenu": contentTypePlain,
"controls": contentTypePlain,
"coords": contentTypePlain,
"crossorigin": contentTypeUnsafe,
"data": contentTypeURL,
"datetime": contentTypePlain,
"default": contentTypePlain,
"defer": contentTypeUnsafe,
"dir": contentTypePlain,
"dirname": contentTypePlain,
"disabled": contentTypePlain,
"draggable": contentTypePlain,
"dropzone": contentTypePlain,
"enctype": contentTypeUnsafe,
"for": contentTypePlain,
"form": contentTypeUnsafe,
"formaction": contentTypeURL,
"formenctype": contentTypeUnsafe,
"formmethod": contentTypeUnsafe,
"formnovalidate": contentTypeUnsafe,
"formtarget": contentTypePlain,
"headers": contentTypePlain,
"height": contentTypePlain,
"hidden": contentTypePlain,
"high": contentTypePlain,
"href": contentTypeURL,
"hreflang": contentTypePlain,
"http-equiv": contentTypeUnsafe,
"icon": contentTypeURL,
"id": contentTypePlain,
"ismap": contentTypePlain,
"keytype": contentTypeUnsafe,
"kind": contentTypePlain,
"label": contentTypePlain,
"lang": contentTypePlain,
"language": contentTypeUnsafe,
"list": contentTypePlain,
"longdesc": contentTypeURL,
"loop": contentTypePlain,
"low": contentTypePlain,
"manifest": contentTypeURL,
"max": contentTypePlain,
"maxlength": contentTypePlain,
"media": contentTypePlain,
"mediagroup": contentTypePlain,
"method": contentTypeUnsafe,
"min": contentTypePlain,
"multiple": contentTypePlain,
"name": contentTypePlain,
"novalidate": contentTypeUnsafe,
// Skip handler names from
// https://www.w3.org/TR/html5/webappapis.html#event-handlers-on-elements,-document-objects,-and-window-objects
// since we have special handling in attrType.
"open": contentTypePlain,
"optimum": contentTypePlain,
"pattern": contentTypeUnsafe,
"placeholder": contentTypePlain,
"poster": contentTypeURL,
"profile": contentTypeURL,
"preload": contentTypePlain,
"pubdate": contentTypePlain,
"radiogroup": contentTypePlain,
"readonly": contentTypePlain,
"rel": contentTypeUnsafe,
"required": contentTypePlain,
"reversed": contentTypePlain,
"rows": contentTypePlain,
"rowspan": contentTypePlain,
"sandbox": contentTypeUnsafe,
"spellcheck": contentTypePlain,
"scope": contentTypePlain,
"scoped": contentTypePlain,
"seamless": contentTypePlain,
"selected": contentTypePlain,
"shape": contentTypePlain,
"size": contentTypePlain,
"sizes": contentTypePlain,
"span": contentTypePlain,
"src": contentTypeURL,
"srcdoc": contentTypeHTML,
"srclang": contentTypePlain,
"srcset": contentTypeSrcset,
"start": contentTypePlain,
"step": contentTypePlain,
"style": contentTypeCSS,
"tabindex": contentTypePlain,
"target": contentTypePlain,
"title": contentTypePlain,
"type": contentTypeUnsafe,
"usemap": contentTypeURL,
"value": contentTypeUnsafe,
"width": contentTypePlain,
"wrap": contentTypePlain,
"xmlns": contentTypeURL,
}
// attrType returns a conservative (upper-bound on authority) guess at the
// type of the lowercase named attribute.
func attrType(name string) contentType {
if strings.HasPrefix(name, "data-") {
// Strip data- so that custom attribute heuristics below are
// widely applied.
// Treat data-action as URL below.
name = name[5:]
} else if prefix, short, ok := strings.Cut(name, ":"); ok {
if prefix == "xmlns" {
return contentTypeURL
}
// Treat svg:href and xlink:href as href below.
name = short
}
if t, ok := attrTypeMap[name]; ok {
return t
}
// Treat partial event handler names as script.
if strings.HasPrefix(name, "on") {
return contentTypeJS
}
// Heuristics to prevent "javascript:..." injection in custom
// data attributes and custom attributes like g:tweetUrl.
// https://www.w3.org/TR/html5/dom.html#embedding-custom-non-visible-data-with-the-data-*-attributes
// "Custom data attributes are intended to store custom data
// private to the page or application, for which there are no
// more appropriate attributes or elements."
// Developers seem to store URL content in data URLs that start
// or end with "URI" or "URL".
if strings.Contains(name, "src") ||
strings.Contains(name, "uri") ||
strings.Contains(name, "url") {
return contentTypeURL
}
return contentTypePlain
}

View File

@@ -0,0 +1,28 @@
// Code generated by "stringer -type attr"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[attrNone-0]
_ = x[attrScript-1]
_ = x[attrScriptType-2]
_ = x[attrStyle-3]
_ = x[attrURL-4]
_ = x[attrSrcset-5]
}
const _attr_name = "attrNoneattrScriptattrScriptTypeattrStyleattrURLattrSrcset"
var _attr_index = [...]uint8{0, 8, 18, 32, 41, 48, 58}
func (i attr) String() string {
if i >= attr(len(_attr_index)-1) {
return "attr(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _attr_name[_attr_index[i]:_attr_index[i+1]]
}

View File

@@ -0,0 +1,278 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"errors"
"fmt"
"io"
"strings"
"sync"
"testing"
"text/template/parse"
)
func TestAddParseTreeHTML(t *testing.T) {
root := Must(New("root").Parse(`{{define "a"}} {{.}} {{template "b"}} {{.}} "></a>{{end}}`))
tree, err := parse.Parse("t", `{{define "b"}}<a href="{{end}}`, "", "", nil, nil)
if err != nil {
t.Fatal(err)
}
added := Must(root.AddParseTree("b", tree["b"]))
b := new(strings.Builder)
err = added.ExecuteTemplate(b, "a", "1>0")
if err != nil {
t.Fatal(err)
}
if got, want := b.String(), ` 1&gt;0 <a href=" 1%3e0 "></a>`; got != want {
t.Errorf("got %q want %q", got, want)
}
}
func TestClone(t *testing.T) {
// The {{.}} will be executed with data "<i>*/" in different contexts.
// In the t0 template, it will be in a text context.
// In the t1 template, it will be in a URL context.
// In the t2 template, it will be in a JavaScript context.
// In the t3 template, it will be in a CSS context.
const tmpl = `{{define "a"}}{{template "lhs"}}{{.}}{{template "rhs"}}{{end}}`
b := new(strings.Builder)
// Create an incomplete template t0.
t0 := Must(New("t0").Parse(tmpl))
// Clone t0 as t1.
t1 := Must(t0.Clone())
Must(t1.Parse(`{{define "lhs"}} <a href=" {{end}}`))
Must(t1.Parse(`{{define "rhs"}} "></a> {{end}}`))
// Execute t1.
b.Reset()
if err := t1.ExecuteTemplate(b, "a", "<i>*/"); err != nil {
t.Fatal(err)
}
if got, want := b.String(), ` <a href=" %3ci%3e*/ "></a> `; got != want {
t.Errorf("t1: got %q want %q", got, want)
}
// Clone t0 as t2.
t2 := Must(t0.Clone())
Must(t2.Parse(`{{define "lhs"}} <p onclick="javascript: {{end}}`))
Must(t2.Parse(`{{define "rhs"}} "></p> {{end}}`))
// Execute t2.
b.Reset()
if err := t2.ExecuteTemplate(b, "a", "<i>*/"); err != nil {
t.Fatal(err)
}
if got, want := b.String(), ` <p onclick="javascript: &#34;\u003ci\u003e*/&#34; "></p> `; got != want {
t.Errorf("t2: got %q want %q", got, want)
}
// Clone t0 as t3, but do not execute t3 yet.
t3 := Must(t0.Clone())
Must(t3.Parse(`{{define "lhs"}} <style> {{end}}`))
Must(t3.Parse(`{{define "rhs"}} </style> {{end}}`))
// Complete t0.
Must(t0.Parse(`{{define "lhs"}} ( {{end}}`))
Must(t0.Parse(`{{define "rhs"}} ) {{end}}`))
// Clone t0 as t4. Redefining the "lhs" template should not fail.
t4 := Must(t0.Clone())
if _, err := t4.Parse(`{{define "lhs"}} OK {{end}}`); err != nil {
t.Errorf(`redefine "lhs": got err %v want nil`, err)
}
// Cloning t1 should fail as it has been executed.
if _, err := t1.Clone(); err == nil {
t.Error("cloning t1: got nil err want non-nil")
}
// Redefining the "lhs" template in t1 should fail as it has been executed.
if _, err := t1.Parse(`{{define "lhs"}} OK {{end}}`); err == nil {
t.Error(`redefine "lhs": got nil err want non-nil`)
}
// Execute t0.
b.Reset()
if err := t0.ExecuteTemplate(b, "a", "<i>*/"); err != nil {
t.Fatal(err)
}
if got, want := b.String(), ` ( &lt;i&gt;*/ ) `; got != want {
t.Errorf("t0: got %q want %q", got, want)
}
// Clone t0. This should fail, as t0 has already executed.
if _, err := t0.Clone(); err == nil {
t.Error(`t0.Clone(): got nil err want non-nil`)
}
// Similarly, cloning sub-templates should fail.
if _, err := t0.Lookup("a").Clone(); err == nil {
t.Error(`t0.Lookup("a").Clone(): got nil err want non-nil`)
}
if _, err := t0.Lookup("lhs").Clone(); err == nil {
t.Error(`t0.Lookup("lhs").Clone(): got nil err want non-nil`)
}
// Execute t3.
b.Reset()
if err := t3.ExecuteTemplate(b, "a", "<i>*/"); err != nil {
t.Fatal(err)
}
if got, want := b.String(), ` <style> ZgotmplZ </style> `; got != want {
t.Errorf("t3: got %q want %q", got, want)
}
}
func TestTemplates(t *testing.T) {
names := []string{"t0", "a", "lhs", "rhs"}
// Some template definitions borrowed from TestClone.
const tmpl = `
{{define "a"}}{{template "lhs"}}{{.}}{{template "rhs"}}{{end}}
{{define "lhs"}} <a href=" {{end}}
{{define "rhs"}} "></a> {{end}}`
t0 := Must(New("t0").Parse(tmpl))
templates := t0.Templates()
if len(templates) != len(names) {
t.Errorf("expected %d templates; got %d", len(names), len(templates))
}
for _, name := range names {
found := false
for _, tmpl := range templates {
if name == tmpl.text.Name() {
found = true
break
}
}
if !found {
t.Error("could not find template", name)
}
}
}
// This used to crash; https://golang.org/issue/3281
func TestCloneCrash(t *testing.T) {
t1 := New("all")
Must(t1.New("t1").Parse(`{{define "foo"}}foo{{end}}`))
t1.Clone()
}
// Ensure that this guarantee from the docs is upheld:
// "Further calls to Parse in the copy will add templates
// to the copy but not to the original."
func TestCloneThenParse(t *testing.T) {
t0 := Must(New("t0").Parse(`{{define "a"}}{{template "embedded"}}{{end}}`))
t1 := Must(t0.Clone())
Must(t1.Parse(`{{define "embedded"}}t1{{end}}`))
if len(t0.Templates())+1 != len(t1.Templates()) {
t.Error("adding a template to a clone added it to the original")
}
// double check that the embedded template isn't available in the original
err := t0.ExecuteTemplate(io.Discard, "a", nil)
if err == nil {
t.Error("expected 'no such template' error")
}
}
// https://golang.org/issue/5980
func TestFuncMapWorksAfterClone(t *testing.T) {
funcs := FuncMap{"customFunc": func() (string, error) {
return "", errors.New("issue5980")
}}
// get the expected error output (no clone)
uncloned := Must(New("").Funcs(funcs).Parse("{{customFunc}}"))
wantErr := uncloned.Execute(io.Discard, nil)
// toClone must be the same as uncloned. It has to be recreated from scratch,
// since cloning cannot occur after execution.
toClone := Must(New("").Funcs(funcs).Parse("{{customFunc}}"))
cloned := Must(toClone.Clone())
gotErr := cloned.Execute(io.Discard, nil)
if wantErr.Error() != gotErr.Error() {
t.Errorf("clone error message mismatch want %q got %q", wantErr, gotErr)
}
}
// https://golang.org/issue/16101
func TestTemplateCloneExecuteRace(t *testing.T) {
const (
input = `<title>{{block "a" .}}a{{end}}</title><body>{{block "b" .}}b{{end}}<body>`
overlay = `{{define "b"}}A{{end}}`
)
outer := Must(New("outer").Parse(input))
tmpl := Must(Must(outer.Clone()).Parse(overlay))
var wg sync.WaitGroup
for i := 0; i < 10; i++ {
wg.Add(1)
go func() {
defer wg.Done()
for i := 0; i < 100; i++ {
if err := tmpl.Execute(io.Discard, "data"); err != nil {
panic(err)
}
}
}()
}
wg.Wait()
}
func TestTemplateCloneLookup(t *testing.T) {
// Template.escape makes an assumption that the template associated
// with t.Name() is t. Check that this holds.
tmpl := Must(New("x").Parse("a"))
tmpl = Must(tmpl.Clone())
if tmpl.Lookup(tmpl.Name()) != tmpl {
t.Error("after Clone, tmpl.Lookup(tmpl.Name()) != tmpl")
}
}
func TestCloneGrowth(t *testing.T) {
tmpl := Must(New("root").Parse(`<title>{{block "B". }}Arg{{end}}</title>`))
tmpl = Must(tmpl.Clone())
Must(tmpl.Parse(`{{define "B"}}Text{{end}}`))
for i := 0; i < 10; i++ {
tmpl.Execute(io.Discard, nil)
}
if len(tmpl.DefinedTemplates()) > 200 {
t.Fatalf("too many templates: %v", len(tmpl.DefinedTemplates()))
}
}
// https://golang.org/issue/17735
func TestCloneRedefinedName(t *testing.T) {
const base = `
{{ define "a" -}}<title>{{ template "b" . -}}</title>{{ end -}}
{{ define "b" }}{{ end -}}
`
const page = `{{ template "a" . }}`
t1 := Must(New("a").Parse(base))
for i := 0; i < 2; i++ {
t2 := Must(t1.Clone())
t2 = Must(t2.New(fmt.Sprintf("%d", i)).Parse(page))
err := t2.Execute(io.Discard, nil)
if err != nil {
t.Fatal(err)
}
}
}
// Issue 24791.
func TestClonePipe(t *testing.T) {
a := Must(New("a").Parse(`{{define "a"}}{{range $v := .A}}{{$v}}{{end}}{{end}}`))
data := struct{ A []string }{A: []string{"hi"}}
b := Must(a.Clone())
var buf strings.Builder
if err := b.Execute(&buf, &data); err != nil {
t.Fatal(err)
}
if got, want := buf.String(), "hi"; got != want {
t.Errorf("got %q want %q", got, want)
}
}

View File

@@ -0,0 +1,185 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"fmt"
"reflect"
)
// Strings of content from a trusted source.
type (
// CSS encapsulates known safe content that matches any of:
// 1. The CSS3 stylesheet production, such as `p { color: purple }`.
// 2. The CSS3 rule production, such as `a[href=~"https:"].foo#bar`.
// 3. CSS3 declaration productions, such as `color: red; margin: 2px`.
// 4. The CSS3 value production, such as `rgba(0, 0, 255, 127)`.
// See https://www.w3.org/TR/css3-syntax/#parsing and
// https://web.archive.org/web/20090211114933/http://w3.org/TR/css3-syntax#style
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
CSS string
// HTML encapsulates a known safe HTML document fragment.
// It should not be used for HTML from a third-party, or HTML with
// unclosed tags or comments. The outputs of a sound HTML sanitizer
// and a template escaped by this package are fine for use with HTML.
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
HTML string
// HTMLAttr encapsulates an HTML attribute from a trusted source,
// for example, ` dir="ltr"`.
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
HTMLAttr string
// JS encapsulates a known safe EcmaScript5 Expression, for example,
// `(x + y * z())`.
// Template authors are responsible for ensuring that typed expressions
// do not break the intended precedence and that there is no
// statement/expression ambiguity as when passing an expression like
// "{ foo: bar() }\n['foo']()", which is both a valid Expression and a
// valid Program with a very different meaning.
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
//
// Using JS to include valid but untrusted JSON is not safe.
// A safe alternative is to parse the JSON with json.Unmarshal and then
// pass the resultant object into the template, where it will be
// converted to sanitized JSON when presented in a JavaScript context.
JS string
// JSStr encapsulates a sequence of characters meant to be embedded
// between quotes in a JavaScript expression.
// The string must match a series of StringCharacters:
// StringCharacter :: SourceCharacter but not `\` or LineTerminator
// | EscapeSequence
// Note that LineContinuations are not allowed.
// JSStr("foo\\nbar") is fine, but JSStr("foo\\\nbar") is not.
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
JSStr string
// URL encapsulates a known safe URL or URL substring (see RFC 3986).
// A URL like `javascript:checkThatFormNotEditedBeforeLeavingPage()`
// from a trusted source should go in the page, but by default dynamic
// `javascript:` URLs are filtered out since they are a frequently
// exploited injection vector.
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
URL string
// Srcset encapsulates a known safe srcset attribute
// (see https://w3c.github.io/html/semantics-embedded-content.html#element-attrdef-img-srcset).
//
// Use of this type presents a security risk:
// the encapsulated content should come from a trusted source,
// as it will be included verbatim in the template output.
Srcset string
)
type contentType uint8
const (
contentTypePlain contentType = iota
contentTypeCSS
contentTypeHTML
contentTypeHTMLAttr
contentTypeJS
contentTypeJSStr
contentTypeURL
contentTypeSrcset
// contentTypeUnsafe is used in attr.go for values that affect how
// embedded content and network messages are formed, vetted,
// or interpreted; or which credentials network messages carry.
contentTypeUnsafe
)
// indirect returns the value, after dereferencing as many times
// as necessary to reach the base type (or nil).
func indirect(a any) any {
if a == nil {
return nil
}
if t := reflect.TypeOf(a); t.Kind() != reflect.Pointer {
// Avoid creating a reflect.Value if it's not a pointer.
return a
}
v := reflect.ValueOf(a)
for v.Kind() == reflect.Pointer && !v.IsNil() {
v = v.Elem()
}
return v.Interface()
}
var (
errorType = reflect.TypeFor[error]()
fmtStringerType = reflect.TypeFor[fmt.Stringer]()
)
// indirectToStringerOrError returns the value, after dereferencing as many times
// as necessary to reach the base type (or nil) or an implementation of fmt.Stringer
// or error.
func indirectToStringerOrError(a any) any {
if a == nil {
return nil
}
v := reflect.ValueOf(a)
for !v.Type().Implements(fmtStringerType) && !v.Type().Implements(errorType) && v.Kind() == reflect.Pointer && !v.IsNil() {
v = v.Elem()
}
return v.Interface()
}
// stringify converts its arguments to a string and the type of the content.
// All pointers are dereferenced, as in the text/template package.
func stringify(args ...any) (string, contentType) {
if len(args) == 1 {
switch s := indirect(args[0]).(type) {
case string:
return s, contentTypePlain
case CSS:
return string(s), contentTypeCSS
case HTML:
return string(s), contentTypeHTML
case HTMLAttr:
return string(s), contentTypeHTMLAttr
case JS:
return string(s), contentTypeJS
case JSStr:
return string(s), contentTypeJSStr
case URL:
return string(s), contentTypeURL
case Srcset:
return string(s), contentTypeSrcset
}
}
i := 0
for _, arg := range args {
// We skip untyped nil arguments for backward compatibility.
// Without this they would be output as <nil>, escaped.
// See issue 25875.
if arg == nil {
continue
}
args[i] = indirectToStringerOrError(arg)
i++
}
return fmt.Sprint(args[:i]...), contentTypePlain
}

View File

@@ -0,0 +1,458 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"fmt"
"strings"
"testing"
)
func TestTypedContent(t *testing.T) {
data := []any{
`<b> "foo%" O'Reilly &bar;`,
CSS(`a[href =~ "//example.com"]#foo`),
HTML(`Hello, <b>World</b> &amp;tc!`),
HTMLAttr(` dir="ltr"`),
JS(`c && alert("Hello, World!");`),
JSStr(`Hello, World & O'Reilly\u0021`),
URL(`greeting=H%69,&addressee=(World)`),
Srcset(`greeting=H%69,&addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`),
URL(`,foo/,`),
}
// For each content sensitive escaper, see how it does on
// each of the typed strings above.
tests := []struct {
// A template containing a single {{.}}.
input string
want []string
}{
{
`<style>{{.}} { color: blue }</style>`,
[]string{
`ZgotmplZ`,
// Allowed but not escaped.
`a[href =~ "//example.com"]#foo`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
},
},
{
`<div style="{{.}}">`,
[]string{
`ZgotmplZ`,
// Allowed and HTML escaped.
`a[href =~ &#34;//example.com&#34;]#foo`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
},
},
{
`{{.}}`,
[]string{
`&lt;b&gt; &#34;foo%&#34; O&#39;Reilly &amp;bar;`,
`a[href =~ &#34;//example.com&#34;]#foo`,
// Not escaped.
`Hello, <b>World</b> &amp;tc!`,
` dir=&#34;ltr&#34;`,
`c &amp;&amp; alert(&#34;Hello, World!&#34;);`,
`Hello, World &amp; O&#39;Reilly\u0021`,
`greeting=H%69,&amp;addressee=(World)`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`,foo/,`,
},
},
{
`<a{{.}}>`,
[]string{
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
// Allowed and HTML escaped.
` dir="ltr"`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
`ZgotmplZ`,
},
},
{
`<a title={{.}}>`,
[]string{
`&lt;b&gt;&#32;&#34;foo%&#34;&#32;O&#39;Reilly&#32;&amp;bar;`,
`a[href&#32;&#61;~&#32;&#34;//example.com&#34;]#foo`,
// Tags stripped, spaces escaped, entity not re-escaped.
`Hello,&#32;World&#32;&amp;tc!`,
`&#32;dir&#61;&#34;ltr&#34;`,
`c&#32;&amp;&amp;&#32;alert(&#34;Hello,&#32;World!&#34;);`,
`Hello,&#32;World&#32;&amp;&#32;O&#39;Reilly\u0021`,
`greeting&#61;H%69,&amp;addressee&#61;(World)`,
`greeting&#61;H%69,&amp;addressee&#61;(World)&#32;2x,&#32;https://golang.org/favicon.ico&#32;500.5w`,
`,foo/,`,
},
},
{
`<a title='{{.}}'>`,
[]string{
`&lt;b&gt; &#34;foo%&#34; O&#39;Reilly &amp;bar;`,
`a[href =~ &#34;//example.com&#34;]#foo`,
// Tags stripped, entity not re-escaped.
`Hello, World &amp;tc!`,
` dir=&#34;ltr&#34;`,
`c &amp;&amp; alert(&#34;Hello, World!&#34;);`,
`Hello, World &amp; O&#39;Reilly\u0021`,
`greeting=H%69,&amp;addressee=(World)`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`,foo/,`,
},
},
{
`<textarea>{{.}}</textarea>`,
[]string{
`&lt;b&gt; &#34;foo%&#34; O&#39;Reilly &amp;bar;`,
`a[href =~ &#34;//example.com&#34;]#foo`,
// Angle brackets escaped to prevent injection of close tags, entity not re-escaped.
`Hello, &lt;b&gt;World&lt;/b&gt; &amp;tc!`,
` dir=&#34;ltr&#34;`,
`c &amp;&amp; alert(&#34;Hello, World!&#34;);`,
`Hello, World &amp; O&#39;Reilly\u0021`,
`greeting=H%69,&amp;addressee=(World)`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`,foo/,`,
},
},
{
`<script>alert({{.}})</script>`,
[]string{
`"\u003cb\u003e \"foo%\" O'Reilly \u0026bar;"`,
`"a[href =~ \"//example.com\"]#foo"`,
`"Hello, \u003cb\u003eWorld\u003c/b\u003e \u0026amp;tc!"`,
`" dir=\"ltr\""`,
// Not escaped.
`c && alert("Hello, World!");`,
// Escape sequence not over-escaped.
`"Hello, World & O'Reilly\u0021"`,
`"greeting=H%69,\u0026addressee=(World)"`,
`"greeting=H%69,\u0026addressee=(World) 2x, https://golang.org/favicon.ico 500.5w"`,
`",foo/,"`,
},
},
{
`<button onclick="alert({{.}})">`,
[]string{
`&#34;\u003cb\u003e \&#34;foo%\&#34; O&#39;Reilly \u0026bar;&#34;`,
`&#34;a[href =~ \&#34;//example.com\&#34;]#foo&#34;`,
`&#34;Hello, \u003cb\u003eWorld\u003c/b\u003e \u0026amp;tc!&#34;`,
`&#34; dir=\&#34;ltr\&#34;&#34;`,
// Not JS escaped but HTML escaped.
`c &amp;&amp; alert(&#34;Hello, World!&#34;);`,
// Escape sequence not over-escaped.
`&#34;Hello, World &amp; O&#39;Reilly\u0021&#34;`,
`&#34;greeting=H%69,\u0026addressee=(World)&#34;`,
`&#34;greeting=H%69,\u0026addressee=(World) 2x, https://golang.org/favicon.ico 500.5w&#34;`,
`&#34;,foo/,&#34;`,
},
},
{
`<script>alert("{{.}}")</script>`,
[]string{
`\u003cb\u003e \u0022foo%\u0022 O\u0027Reilly \u0026bar;`,
`a[href =~ \u0022\/\/example.com\u0022]#foo`,
`Hello, \u003cb\u003eWorld\u003c\/b\u003e \u0026amp;tc!`,
` dir=\u0022ltr\u0022`,
`c \u0026\u0026 alert(\u0022Hello, World!\u0022);`,
// Escape sequence not over-escaped.
`Hello, World \u0026 O\u0027Reilly\u0021`,
`greeting=H%69,\u0026addressee=(World)`,
`greeting=H%69,\u0026addressee=(World) 2x, https:\/\/golang.org\/favicon.ico 500.5w`,
`,foo\/,`,
},
},
{
`<script type="text/javascript">alert("{{.}}")</script>`,
[]string{
`\u003cb\u003e \u0022foo%\u0022 O\u0027Reilly \u0026bar;`,
`a[href =~ \u0022\/\/example.com\u0022]#foo`,
`Hello, \u003cb\u003eWorld\u003c\/b\u003e \u0026amp;tc!`,
` dir=\u0022ltr\u0022`,
`c \u0026\u0026 alert(\u0022Hello, World!\u0022);`,
// Escape sequence not over-escaped.
`Hello, World \u0026 O\u0027Reilly\u0021`,
`greeting=H%69,\u0026addressee=(World)`,
`greeting=H%69,\u0026addressee=(World) 2x, https:\/\/golang.org\/favicon.ico 500.5w`,
`,foo\/,`,
},
},
{
`<script type="text/javascript">alert({{.}})</script>`,
[]string{
`"\u003cb\u003e \"foo%\" O'Reilly \u0026bar;"`,
`"a[href =~ \"//example.com\"]#foo"`,
`"Hello, \u003cb\u003eWorld\u003c/b\u003e \u0026amp;tc!"`,
`" dir=\"ltr\""`,
// Not escaped.
`c && alert("Hello, World!");`,
// Escape sequence not over-escaped.
`"Hello, World & O'Reilly\u0021"`,
`"greeting=H%69,\u0026addressee=(World)"`,
`"greeting=H%69,\u0026addressee=(World) 2x, https://golang.org/favicon.ico 500.5w"`,
`",foo/,"`,
},
},
{
// Not treated as JS. The output is same as for <div>{{.}}</div>
`<script type="text/template">{{.}}</script>`,
[]string{
`&lt;b&gt; &#34;foo%&#34; O&#39;Reilly &amp;bar;`,
`a[href =~ &#34;//example.com&#34;]#foo`,
// Not escaped.
`Hello, <b>World</b> &amp;tc!`,
` dir=&#34;ltr&#34;`,
`c &amp;&amp; alert(&#34;Hello, World!&#34;);`,
`Hello, World &amp; O&#39;Reilly\u0021`,
`greeting=H%69,&amp;addressee=(World)`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`,foo/,`,
},
},
{
`<button onclick='alert("{{.}}")'>`,
[]string{
`\u003cb\u003e \u0022foo%\u0022 O\u0027Reilly \u0026bar;`,
`a[href =~ \u0022\/\/example.com\u0022]#foo`,
`Hello, \u003cb\u003eWorld\u003c\/b\u003e \u0026amp;tc!`,
` dir=\u0022ltr\u0022`,
`c \u0026\u0026 alert(\u0022Hello, World!\u0022);`,
// Escape sequence not over-escaped.
`Hello, World \u0026 O\u0027Reilly\u0021`,
`greeting=H%69,\u0026addressee=(World)`,
`greeting=H%69,\u0026addressee=(World) 2x, https:\/\/golang.org\/favicon.ico 500.5w`,
`,foo\/,`,
},
},
{
`<a href="?q={{.}}">`,
[]string{
`%3cb%3e%20%22foo%25%22%20O%27Reilly%20%26bar%3b`,
`a%5bhref%20%3d~%20%22%2f%2fexample.com%22%5d%23foo`,
`Hello%2c%20%3cb%3eWorld%3c%2fb%3e%20%26amp%3btc%21`,
`%20dir%3d%22ltr%22`,
`c%20%26%26%20alert%28%22Hello%2c%20World%21%22%29%3b`,
`Hello%2c%20World%20%26%20O%27Reilly%5cu0021`,
// Quotes and parens are escaped but %69 is not over-escaped. HTML escaping is done.
`greeting=H%69,&amp;addressee=%28World%29`,
`greeting%3dH%2569%2c%26addressee%3d%28World%29%202x%2c%20https%3a%2f%2fgolang.org%2ffavicon.ico%20500.5w`,
`,foo/,`,
},
},
{
`<style>body { background: url('?img={{.}}') }</style>`,
[]string{
`%3cb%3e%20%22foo%25%22%20O%27Reilly%20%26bar%3b`,
`a%5bhref%20%3d~%20%22%2f%2fexample.com%22%5d%23foo`,
`Hello%2c%20%3cb%3eWorld%3c%2fb%3e%20%26amp%3btc%21`,
`%20dir%3d%22ltr%22`,
`c%20%26%26%20alert%28%22Hello%2c%20World%21%22%29%3b`,
`Hello%2c%20World%20%26%20O%27Reilly%5cu0021`,
// Quotes and parens are escaped but %69 is not over-escaped. HTML escaping is not done.
`greeting=H%69,&addressee=%28World%29`,
`greeting%3dH%2569%2c%26addressee%3d%28World%29%202x%2c%20https%3a%2f%2fgolang.org%2ffavicon.ico%20500.5w`,
`,foo/,`,
},
},
{
`<img srcset="{{.}}">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
// Commas are not escaped.
`Hello,#ZgotmplZ`,
// Leading spaces are not percent escapes.
` dir=%22ltr%22`,
// Spaces after commas are not percent escaped.
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
// Metadata is not escaped.
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
{
`<img srcset={{.}}>`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
// Spaces are HTML escaped not %-escaped
`&#32;dir&#61;%22ltr%22`,
`#ZgotmplZ,&#32;World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting&#61;H%69%2c&amp;addressee&#61;%28World%29`,
`greeting&#61;H%69,&amp;addressee&#61;(World)&#32;2x,&#32;https://golang.org/favicon.ico&#32;500.5w`,
// Commas are escaped.
`%2cfoo/%2c`,
},
},
{
`<img srcset="{{.}} 2x, https://golang.org/ 500.5w">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
` dir=%22ltr%22`,
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
{
`<img srcset="http://godoc.org/ {{.}}, https://golang.org/ 500.5w">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
` dir=%22ltr%22`,
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
{
`<img srcset="http://godoc.org/?q={{.}} 2x, https://golang.org/ 500.5w">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
` dir=%22ltr%22`,
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
{
`<img srcset="http://godoc.org/ 2x, {{.}} 500.5w">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
` dir=%22ltr%22`,
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
{
`<img srcset="http://godoc.org/ 2x, https://golang.org/ {{.}}">`,
[]string{
`#ZgotmplZ`,
`#ZgotmplZ`,
`Hello,#ZgotmplZ`,
` dir=%22ltr%22`,
`#ZgotmplZ, World!%22%29;`,
`Hello,#ZgotmplZ`,
`greeting=H%69%2c&amp;addressee=%28World%29`,
`greeting=H%69,&amp;addressee=(World) 2x, https://golang.org/favicon.ico 500.5w`,
`%2cfoo/%2c`,
},
},
}
for _, test := range tests {
tmpl := Must(New("x").Parse(test.input))
pre := strings.Index(test.input, "{{.}}")
post := len(test.input) - (pre + 5)
var b strings.Builder
for i, x := range data {
b.Reset()
if err := tmpl.Execute(&b, x); err != nil {
t.Errorf("%q with %v: %s", test.input, x, err)
continue
}
if want, got := test.want[i], b.String()[pre:b.Len()-post]; want != got {
t.Errorf("%q with %v:\nwant\n\t%q,\ngot\n\t%q\n", test.input, x, want, got)
continue
}
}
}
}
// Test that we print using the String method. Was issue 3073.
type myStringer struct {
v int
}
func (s *myStringer) String() string {
return fmt.Sprintf("string=%d", s.v)
}
type errorer struct {
v int
}
func (s *errorer) Error() string {
return fmt.Sprintf("error=%d", s.v)
}
func TestStringer(t *testing.T) {
s := &myStringer{3}
b := new(strings.Builder)
tmpl := Must(New("x").Parse("{{.}}"))
if err := tmpl.Execute(b, s); err != nil {
t.Fatal(err)
}
var expect = "string=3"
if b.String() != expect {
t.Errorf("expected %q got %q", expect, b.String())
}
e := &errorer{7}
b.Reset()
if err := tmpl.Execute(b, e); err != nil {
t.Fatal(err)
}
expect = "error=7"
if b.String() != expect {
t.Errorf("expected %q got %q", expect, b.String())
}
}
// https://golang.org/issue/5982
func TestEscapingNilNonemptyInterfaces(t *testing.T) {
tmpl := Must(New("x").Parse("{{.E}}"))
got := new(bytes.Buffer)
testData := struct{ E error }{} // any non-empty interface here will do; error is just ready at hand
tmpl.Execute(got, testData)
// A non-empty interface should print like an empty interface.
want := new(bytes.Buffer)
data := struct{ E any }{}
tmpl.Execute(want, data)
if !bytes.Equal(want.Bytes(), got.Bytes()) {
t.Errorf("expected %q got %q", string(want.Bytes()), string(got.Bytes()))
}
}

View File

@@ -0,0 +1,291 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"fmt"
"text/template/parse"
)
// context describes the state an HTML parser must be in when it reaches the
// portion of HTML produced by evaluating a particular template node.
//
// The zero value of type context is the start context for a template that
// produces an HTML fragment as defined at
// https://www.w3.org/TR/html5/syntax.html#the-end
// where the context element is null.
type context struct {
state state
delim delim
urlPart urlPart
jsCtx jsCtx
// jsBraceDepth contains the current depth, for each JS template literal
// string interpolation expression, of braces we've seen. This is used to
// determine if the next } will close a JS template literal string
// interpolation expression or not.
jsBraceDepth []int
attr attr
element element
n parse.Node // for range break/continue
err *Error
}
func (c context) String() string {
var err error
if c.err != nil {
err = c.err
}
return fmt.Sprintf("{%v %v %v %v %v %v %v}", c.state, c.delim, c.urlPart, c.jsCtx, c.attr, c.element, err)
}
// eq reports whether two contexts are equal.
func (c context) eq(d context) bool {
return c.state == d.state &&
c.delim == d.delim &&
c.urlPart == d.urlPart &&
c.jsCtx == d.jsCtx &&
c.attr == d.attr &&
c.element == d.element &&
c.err == d.err
}
// mangle produces an identifier that includes a suffix that distinguishes it
// from template names mangled with different contexts.
func (c context) mangle(templateName string) string {
// The mangled name for the default context is the input templateName.
if c.state == stateText {
return templateName
}
s := templateName + "$htmltemplate_" + c.state.String()
if c.delim != delimNone {
s += "_" + c.delim.String()
}
if c.urlPart != urlPartNone {
s += "_" + c.urlPart.String()
}
if c.jsCtx != jsCtxRegexp {
s += "_" + c.jsCtx.String()
}
if c.attr != attrNone {
s += "_" + c.attr.String()
}
if c.element != elementNone {
s += "_" + c.element.String()
}
return s
}
// state describes a high-level HTML parser state.
//
// It bounds the top of the element stack, and by extension the HTML insertion
// mode, but also contains state that does not correspond to anything in the
// HTML5 parsing algorithm because a single token production in the HTML
// grammar may contain embedded actions in a template. For instance, the quoted
// HTML attribute produced by
//
// <div title="Hello {{.World}}">
//
// is a single token in HTML's grammar but in a template spans several nodes.
type state uint8
//go:generate stringer -type state
const (
// stateText is parsed character data. An HTML parser is in
// this state when its parse position is outside an HTML tag,
// directive, comment, and special element body.
stateText state = iota
// stateTag occurs before an HTML attribute or the end of a tag.
stateTag
// stateAttrName occurs inside an attribute name.
// It occurs between the ^'s in ` ^name^ = value`.
stateAttrName
// stateAfterName occurs after an attr name has ended but before any
// equals sign. It occurs between the ^'s in ` name^ ^= value`.
stateAfterName
// stateBeforeValue occurs after the equals sign but before the value.
// It occurs between the ^'s in ` name =^ ^value`.
stateBeforeValue
// stateHTMLCmt occurs inside an <!-- HTML comment -->.
stateHTMLCmt
// stateRCDATA occurs inside an RCDATA element (<textarea> or <title>)
// as described at https://www.w3.org/TR/html5/syntax.html#elements-0
stateRCDATA
// stateAttr occurs inside an HTML attribute whose content is text.
stateAttr
// stateURL occurs inside an HTML attribute whose content is a URL.
stateURL
// stateSrcset occurs inside an HTML srcset attribute.
stateSrcset
// stateJS occurs inside an event handler or script element.
stateJS
// stateJSDqStr occurs inside a JavaScript double quoted string.
stateJSDqStr
// stateJSSqStr occurs inside a JavaScript single quoted string.
stateJSSqStr
// stateJSTmplLit occurs inside a JavaScript back quoted string.
stateJSTmplLit
// stateJSRegexp occurs inside a JavaScript regexp literal.
stateJSRegexp
// stateJSBlockCmt occurs inside a JavaScript /* block comment */.
stateJSBlockCmt
// stateJSLineCmt occurs inside a JavaScript // line comment.
stateJSLineCmt
// stateJSHTMLOpenCmt occurs inside a JavaScript <!-- HTML-like comment.
stateJSHTMLOpenCmt
// stateJSHTMLCloseCmt occurs inside a JavaScript --> HTML-like comment.
stateJSHTMLCloseCmt
// stateCSS occurs inside a <style> element or style attribute.
stateCSS
// stateCSSDqStr occurs inside a CSS double quoted string.
stateCSSDqStr
// stateCSSSqStr occurs inside a CSS single quoted string.
stateCSSSqStr
// stateCSSDqURL occurs inside a CSS double quoted url("...").
stateCSSDqURL
// stateCSSSqURL occurs inside a CSS single quoted url('...').
stateCSSSqURL
// stateCSSURL occurs inside a CSS unquoted url(...).
stateCSSURL
// stateCSSBlockCmt occurs inside a CSS /* block comment */.
stateCSSBlockCmt
// stateCSSLineCmt occurs inside a CSS // line comment.
stateCSSLineCmt
// stateError is an infectious error state outside any valid
// HTML/CSS/JS construct.
stateError
// stateDead marks unreachable code after a {{break}} or {{continue}}.
stateDead
)
// isComment is true for any state that contains content meant for template
// authors & maintainers, not for end-users or machines.
func isComment(s state) bool {
switch s {
case stateHTMLCmt, stateJSBlockCmt, stateJSLineCmt, stateJSHTMLOpenCmt, stateJSHTMLCloseCmt, stateCSSBlockCmt, stateCSSLineCmt:
return true
}
return false
}
// isInTag return whether s occurs solely inside an HTML tag.
func isInTag(s state) bool {
switch s {
case stateTag, stateAttrName, stateAfterName, stateBeforeValue, stateAttr:
return true
}
return false
}
// isInScriptLiteral returns true if s is one of the literal states within a
// <script> tag, and as such occurrences of "<!--", "<script", and "</script"
// need to be treated specially.
func isInScriptLiteral(s state) bool {
// Ignore the comment states (stateJSBlockCmt, stateJSLineCmt,
// stateJSHTMLOpenCmt, stateJSHTMLCloseCmt) because their content is already
// omitted from the output.
switch s {
case stateJSDqStr, stateJSSqStr, stateJSTmplLit, stateJSRegexp:
return true
}
return false
}
// delim is the delimiter that will end the current HTML attribute.
type delim uint8
//go:generate stringer -type delim
const (
// delimNone occurs outside any attribute.
delimNone delim = iota
// delimDoubleQuote occurs when a double quote (") closes the attribute.
delimDoubleQuote
// delimSingleQuote occurs when a single quote (') closes the attribute.
delimSingleQuote
// delimSpaceOrTagEnd occurs when a space or right angle bracket (>)
// closes the attribute.
delimSpaceOrTagEnd
)
// urlPart identifies a part in an RFC 3986 hierarchical URL to allow different
// encoding strategies.
type urlPart uint8
//go:generate stringer -type urlPart
const (
// urlPartNone occurs when not in a URL, or possibly at the start:
// ^ in "^http://auth/path?k=v#frag".
urlPartNone urlPart = iota
// urlPartPreQuery occurs in the scheme, authority, or path; between the
// ^s in "h^ttp://auth/path^?k=v#frag".
urlPartPreQuery
// urlPartQueryOrFrag occurs in the query portion between the ^s in
// "http://auth/path?^k=v#frag^".
urlPartQueryOrFrag
// urlPartUnknown occurs due to joining of contexts both before and
// after the query separator.
urlPartUnknown
)
// jsCtx determines whether a '/' starts a regular expression literal or a
// division operator.
type jsCtx uint8
//go:generate stringer -type jsCtx
const (
// jsCtxRegexp occurs where a '/' would start a regexp literal.
jsCtxRegexp jsCtx = iota
// jsCtxDivOp occurs where a '/' would start a division operator.
jsCtxDivOp
// jsCtxUnknown occurs where a '/' is ambiguous due to context joining.
jsCtxUnknown
)
// element identifies the HTML element when inside a start tag or special body.
// Certain HTML element (for example <script> and <style>) have bodies that are
// treated differently from stateText so the element type is necessary to
// transition into the correct context at the end of a tag and to identify the
// end delimiter for the body.
type element uint8
//go:generate stringer -type element
const (
// elementNone occurs outside a special tag or special element body.
elementNone element = iota
// elementScript corresponds to the raw text <script> element
// with JS MIME type or no type attribute.
elementScript
// elementStyle corresponds to the raw text <style> element.
elementStyle
// elementTextarea corresponds to the RCDATA <textarea> element.
elementTextarea
// elementTitle corresponds to the RCDATA <title> element.
elementTitle
)
//go:generate stringer -type attr
// attr identifies the current HTML attribute when inside the attribute,
// that is, starting from stateAttrName until stateTag/stateText (exclusive).
type attr uint8
const (
// attrNone corresponds to a normal attribute or no attribute.
attrNone attr = iota
// attrScript corresponds to an event handler attribute.
attrScript
// attrScriptType corresponds to the type attribute in script HTML element
attrScriptType
// attrStyle corresponds to the style attribute whose value is CSS.
attrStyle
// attrURL corresponds to an attribute whose value is a URL.
attrURL
// attrSrcset corresponds to a srcset attribute.
attrSrcset
)

260
src/html/template/css.go Normal file
View File

@@ -0,0 +1,260 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"fmt"
"strings"
"unicode"
"unicode/utf8"
)
// endsWithCSSKeyword reports whether b ends with an ident that
// case-insensitively matches the lower-case kw.
func endsWithCSSKeyword(b []byte, kw string) bool {
i := len(b) - len(kw)
if i < 0 {
// Too short.
return false
}
if i != 0 {
r, _ := utf8.DecodeLastRune(b[:i])
if isCSSNmchar(r) {
// Too long.
return false
}
}
// Many CSS keywords, such as "!important" can have characters encoded,
// but the URI production does not allow that according to
// https://www.w3.org/TR/css3-syntax/#TOK-URI
// This does not attempt to recognize encoded keywords. For example,
// given "\75\72\6c" and "url" this return false.
return string(bytes.ToLower(b[i:])) == kw
}
// isCSSNmchar reports whether rune is allowed anywhere in a CSS identifier.
func isCSSNmchar(r rune) bool {
// Based on the CSS3 nmchar production but ignores multi-rune escape
// sequences.
// https://www.w3.org/TR/css3-syntax/#SUBTOK-nmchar
return 'a' <= r && r <= 'z' ||
'A' <= r && r <= 'Z' ||
'0' <= r && r <= '9' ||
r == '-' ||
r == '_' ||
// Non-ASCII cases below.
0x80 <= r && r <= 0xd7ff ||
0xe000 <= r && r <= 0xfffd ||
0x10000 <= r && r <= 0x10ffff
}
// decodeCSS decodes CSS3 escapes given a sequence of stringchars.
// If there is no change, it returns the input, otherwise it returns a slice
// backed by a new array.
// https://www.w3.org/TR/css3-syntax/#SUBTOK-stringchar defines stringchar.
func decodeCSS(s []byte) []byte {
i := bytes.IndexByte(s, '\\')
if i == -1 {
return s
}
// The UTF-8 sequence for a codepoint is never longer than 1 + the
// number hex digits need to represent that codepoint, so len(s) is an
// upper bound on the output length.
b := make([]byte, 0, len(s))
for len(s) != 0 {
i := bytes.IndexByte(s, '\\')
if i == -1 {
i = len(s)
}
b, s = append(b, s[:i]...), s[i:]
if len(s) < 2 {
break
}
// https://www.w3.org/TR/css3-syntax/#SUBTOK-escape
// escape ::= unicode | '\' [#x20-#x7E#x80-#xD7FF#xE000-#xFFFD#x10000-#x10FFFF]
if isHex(s[1]) {
// https://www.w3.org/TR/css3-syntax/#SUBTOK-unicode
// unicode ::= '\' [0-9a-fA-F]{1,6} wc?
j := 2
for j < len(s) && j < 7 && isHex(s[j]) {
j++
}
r := hexDecode(s[1:j])
if r > unicode.MaxRune {
r, j = r/16, j-1
}
n := utf8.EncodeRune(b[len(b):cap(b)], r)
// The optional space at the end allows a hex
// sequence to be followed by a literal hex.
// string(decodeCSS([]byte(`\A B`))) == "\nB"
b, s = b[:len(b)+n], skipCSSSpace(s[j:])
} else {
// `\\` decodes to `\` and `\"` to `"`.
_, n := utf8.DecodeRune(s[1:])
b, s = append(b, s[1:1+n]...), s[1+n:]
}
}
return b
}
// isHex reports whether the given character is a hex digit.
func isHex(c byte) bool {
return '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F'
}
// hexDecode decodes a short hex digit sequence: "10" -> 16.
func hexDecode(s []byte) rune {
n := '\x00'
for _, c := range s {
n <<= 4
switch {
case '0' <= c && c <= '9':
n |= rune(c - '0')
case 'a' <= c && c <= 'f':
n |= rune(c-'a') + 10
case 'A' <= c && c <= 'F':
n |= rune(c-'A') + 10
default:
panic(fmt.Sprintf("Bad hex digit in %q", s))
}
}
return n
}
// skipCSSSpace returns a suffix of c, skipping over a single space.
func skipCSSSpace(c []byte) []byte {
if len(c) == 0 {
return c
}
// wc ::= #x9 | #xA | #xC | #xD | #x20
switch c[0] {
case '\t', '\n', '\f', ' ':
return c[1:]
case '\r':
// This differs from CSS3's wc production because it contains a
// probable spec error whereby wc contains all the single byte
// sequences in nl (newline) but not CRLF.
if len(c) >= 2 && c[1] == '\n' {
return c[2:]
}
return c[1:]
}
return c
}
// isCSSSpace reports whether b is a CSS space char as defined in wc.
func isCSSSpace(b byte) bool {
switch b {
case '\t', '\n', '\f', '\r', ' ':
return true
}
return false
}
// cssEscaper escapes HTML and CSS special characters using \<hex>+ escapes.
func cssEscaper(args ...any) string {
s, _ := stringify(args...)
var b strings.Builder
r, w, written := rune(0), 0, 0
for i := 0; i < len(s); i += w {
// See comment in htmlEscaper.
r, w = utf8.DecodeRuneInString(s[i:])
var repl string
switch {
case int(r) < len(cssReplacementTable) && cssReplacementTable[r] != "":
repl = cssReplacementTable[r]
default:
continue
}
if written == 0 {
b.Grow(len(s))
}
b.WriteString(s[written:i])
b.WriteString(repl)
written = i + w
if repl != `\\` && (written == len(s) || isHex(s[written]) || isCSSSpace(s[written])) {
b.WriteByte(' ')
}
}
if written == 0 {
return s
}
b.WriteString(s[written:])
return b.String()
}
var cssReplacementTable = []string{
0: `\0`,
'\t': `\9`,
'\n': `\a`,
'\f': `\c`,
'\r': `\d`,
// Encode HTML specials as hex so the output can be embedded
// in HTML attributes without further encoding.
'"': `\22`,
'&': `\26`,
'\'': `\27`,
'(': `\28`,
')': `\29`,
'+': `\2b`,
'/': `\2f`,
':': `\3a`,
';': `\3b`,
'<': `\3c`,
'>': `\3e`,
'\\': `\\`,
'{': `\7b`,
'}': `\7d`,
}
var expressionBytes = []byte("expression")
var mozBindingBytes = []byte("mozbinding")
// cssValueFilter allows innocuous CSS values in the output including CSS
// quantities (10px or 25%), ID or class literals (#foo, .bar), keyword values
// (inherit, blue), and colors (#888).
// It filters out unsafe values, such as those that affect token boundaries,
// and anything that might execute scripts.
func cssValueFilter(args ...any) string {
s, t := stringify(args...)
if t == contentTypeCSS {
return s
}
b, id := decodeCSS([]byte(s)), make([]byte, 0, 64)
// CSS3 error handling is specified as honoring string boundaries per
// https://www.w3.org/TR/css3-syntax/#error-handling :
// Malformed declarations. User agents must handle unexpected
// tokens encountered while parsing a declaration by reading until
// the end of the declaration, while observing the rules for
// matching pairs of (), [], {}, "", and '', and correctly handling
// escapes. For example, a malformed declaration may be missing a
// property, colon (:) or value.
// So we need to make sure that values do not have mismatched bracket
// or quote characters to prevent the browser from restarting parsing
// inside a string that might embed JavaScript source.
for i, c := range b {
switch c {
case 0, '"', '\'', '(', ')', '/', ';', '@', '[', '\\', ']', '`', '{', '}', '<', '>':
return filterFailsafe
case '-':
// Disallow <!-- or -->.
// -- should not appear in valid identifiers.
if i != 0 && b[i-1] == '-' {
return filterFailsafe
}
default:
if c < utf8.RuneSelf && isCSSNmchar(rune(c)) {
id = append(id, c)
}
}
}
id = bytes.ToLower(id)
if bytes.Contains(id, expressionBytes) || bytes.Contains(id, mozBindingBytes) {
return filterFailsafe
}
return string(b)
}

View File

@@ -0,0 +1,283 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"strconv"
"strings"
"testing"
)
func TestEndsWithCSSKeyword(t *testing.T) {
tests := []struct {
css, kw string
want bool
}{
{"", "url", false},
{"url", "url", true},
{"URL", "url", true},
{"Url", "url", true},
{"url", "important", false},
{"important", "important", true},
{"image-url", "url", false},
{"imageurl", "url", false},
{"image url", "url", true},
}
for _, test := range tests {
got := endsWithCSSKeyword([]byte(test.css), test.kw)
if got != test.want {
t.Errorf("want %t but got %t for css=%v, kw=%v", test.want, got, test.css, test.kw)
}
}
}
func TestIsCSSNmchar(t *testing.T) {
tests := []struct {
rune rune
want bool
}{
{0, false},
{'0', true},
{'9', true},
{'A', true},
{'Z', true},
{'a', true},
{'z', true},
{'_', true},
{'-', true},
{':', false},
{';', false},
{' ', false},
{0x7f, false},
{0x80, true},
{0x1234, true},
{0xd800, false},
{0xdc00, false},
{0xfffe, false},
{0x10000, true},
{0x110000, false},
}
for _, test := range tests {
got := isCSSNmchar(test.rune)
if got != test.want {
t.Errorf("%q: want %t but got %t", string(test.rune), test.want, got)
}
}
}
func TestDecodeCSS(t *testing.T) {
tests := []struct {
css, want string
}{
{``, ``},
{`foo`, `foo`},
{`foo\`, `foo`},
{`foo\\`, `foo\`},
{`\`, ``},
{`\A`, "\n"},
{`\a`, "\n"},
{`\0a`, "\n"},
{`\00000a`, "\n"},
{`\000000a`, "\u0000a"},
{`\1234 5`, "\u1234" + "5"},
{`\1234\20 5`, "\u1234" + " 5"},
{`\1234\A 5`, "\u1234" + "\n5"},
{"\\1234\t5", "\u1234" + "5"},
{"\\1234\n5", "\u1234" + "5"},
{"\\1234\r\n5", "\u1234" + "5"},
{`\12345`, "\U00012345"},
{`\\`, `\`},
{`\\ `, `\ `},
{`\"`, `"`},
{`\'`, `'`},
{`\.`, `.`},
{`\. .`, `. .`},
{
`The \3c i\3equick\3c/i\3e,\d\A\3cspan style=\27 color:brown\27\3e brown\3c/span\3e fox jumps\2028over the \3c canine class=\22lazy\22 \3e dog\3c/canine\3e`,
"The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>",
},
}
for _, test := range tests {
got1 := string(decodeCSS([]byte(test.css)))
if got1 != test.want {
t.Errorf("%q: want\n\t%q\nbut got\n\t%q", test.css, test.want, got1)
}
recoded := cssEscaper(got1)
if got2 := string(decodeCSS([]byte(recoded))); got2 != test.want {
t.Errorf("%q: escape & decode not dual for %q", test.css, recoded)
}
}
}
func TestHexDecode(t *testing.T) {
for i := 0; i < 0x200000; i += 101 /* coprime with 16 */ {
s := strconv.FormatInt(int64(i), 16)
if got := int(hexDecode([]byte(s))); got != i {
t.Errorf("%s: want %d but got %d", s, i, got)
}
s = strings.ToUpper(s)
if got := int(hexDecode([]byte(s))); got != i {
t.Errorf("%s: want %d but got %d", s, i, got)
}
}
}
func TestSkipCSSSpace(t *testing.T) {
tests := []struct {
css, want string
}{
{"", ""},
{"foo", "foo"},
{"\n", ""},
{"\r\n", ""},
{"\r", ""},
{"\t", ""},
{" ", ""},
{"\f", ""},
{" foo", "foo"},
{" foo", " foo"},
{`\20`, `\20`},
}
for _, test := range tests {
got := string(skipCSSSpace([]byte(test.css)))
if got != test.want {
t.Errorf("%q: want %q but got %q", test.css, test.want, got)
}
}
}
func TestCSSEscaper(t *testing.T) {
input := ("\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f" +
"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !"#$%&'()*+,-./` +
`0123456789:;<=>?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\]^_` +
"`abcdefghijklmno" +
"pqrstuvwxyz{|}~\x7f" +
"\u00A0\u0100\u2028\u2029\ufeff\U0001D11E")
want := ("\\0\x01\x02\x03\x04\x05\x06\x07" +
"\x08\\9 \\a\x0b\\c \\d\x0E\x0F" +
"\x10\x11\x12\x13\x14\x15\x16\x17" +
"\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !\22#$%\26\27\28\29*\2b,-.\2f ` +
`0123456789\3a\3b\3c=\3e?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\\]^_` +
"`abcdefghijklmno" +
`pqrstuvwxyz\7b|\7d~` + "\u007f" +
"\u00A0\u0100\u2028\u2029\ufeff\U0001D11E")
got := cssEscaper(input)
if got != want {
t.Errorf("encode: want\n\t%q\nbut got\n\t%q", want, got)
}
got = string(decodeCSS([]byte(got)))
if input != got {
t.Errorf("decode: want\n\t%q\nbut got\n\t%q", input, got)
}
}
func TestCSSValueFilter(t *testing.T) {
tests := []struct {
css, want string
}{
{"", ""},
{"foo", "foo"},
{"0", "0"},
{"0px", "0px"},
{"-5px", "-5px"},
{"1.25in", "1.25in"},
{"+.33em", "+.33em"},
{"100%", "100%"},
{"12.5%", "12.5%"},
{".foo", ".foo"},
{"#bar", "#bar"},
{"corner-radius", "corner-radius"},
{"-moz-corner-radius", "-moz-corner-radius"},
{"#000", "#000"},
{"#48f", "#48f"},
{"#123456", "#123456"},
{"U+00-FF, U+980-9FF", "U+00-FF, U+980-9FF"},
{"color: red", "color: red"},
{"<!--", "ZgotmplZ"},
{"-->", "ZgotmplZ"},
{"<![CDATA[", "ZgotmplZ"},
{"]]>", "ZgotmplZ"},
{"</style", "ZgotmplZ"},
{`"`, "ZgotmplZ"},
{`'`, "ZgotmplZ"},
{"`", "ZgotmplZ"},
{"\x00", "ZgotmplZ"},
{"/* foo */", "ZgotmplZ"},
{"//", "ZgotmplZ"},
{"[href=~", "ZgotmplZ"},
{"expression(alert(1337))", "ZgotmplZ"},
{"-expression(alert(1337))", "ZgotmplZ"},
{"expression", "ZgotmplZ"},
{"Expression", "ZgotmplZ"},
{"EXPRESSION", "ZgotmplZ"},
{"-moz-binding", "ZgotmplZ"},
{"-expr\x00ession(alert(1337))", "ZgotmplZ"},
{`-expr\0ession(alert(1337))`, "ZgotmplZ"},
{`-express\69on(alert(1337))`, "ZgotmplZ"},
{`-express\69 on(alert(1337))`, "ZgotmplZ"},
{`-exp\72 ession(alert(1337))`, "ZgotmplZ"},
{`-exp\52 ession(alert(1337))`, "ZgotmplZ"},
{`-exp\000052 ession(alert(1337))`, "ZgotmplZ"},
{`-expre\0000073sion`, "-expre\x073sion"},
{`@import url evil.css`, "ZgotmplZ"},
{"<", "ZgotmplZ"},
{">", "ZgotmplZ"},
}
for _, test := range tests {
got := cssValueFilter(test.css)
if got != test.want {
t.Errorf("%q: want %q but got %q", test.css, test.want, got)
}
}
}
func BenchmarkCSSEscaper(b *testing.B) {
for i := 0; i < b.N; i++ {
cssEscaper("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}
func BenchmarkCSSEscaperNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
cssEscaper("The quick, brown fox jumps over the lazy dog.")
}
}
func BenchmarkDecodeCSS(b *testing.B) {
s := []byte(`The \3c i\3equick\3c/i\3e,\d\A\3cspan style=\27 color:brown\27\3e brown\3c/span\3e fox jumps\2028over the \3c canine class=\22lazy\22 \3edog\3c/canine\3e`)
b.ResetTimer()
for i := 0; i < b.N; i++ {
decodeCSS(s)
}
}
func BenchmarkDecodeCSSNoSpecials(b *testing.B) {
s := []byte("The quick, brown fox jumps over the lazy dog.")
b.ResetTimer()
for i := 0; i < b.N; i++ {
decodeCSS(s)
}
}
func BenchmarkCSSValueFilter(b *testing.B) {
for i := 0; i < b.N; i++ {
cssValueFilter(` e\78preS\0Sio/**/n(alert(1337))`)
}
}
func BenchmarkCSSValueFilterOk(b *testing.B) {
for i := 0; i < b.N; i++ {
cssValueFilter(`Times New Roman`)
}
}

View File

@@ -0,0 +1,26 @@
// Code generated by "stringer -type delim"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[delimNone-0]
_ = x[delimDoubleQuote-1]
_ = x[delimSingleQuote-2]
_ = x[delimSpaceOrTagEnd-3]
}
const _delim_name = "delimNonedelimDoubleQuotedelimSingleQuotedelimSpaceOrTagEnd"
var _delim_index = [...]uint8{0, 9, 25, 41, 59}
func (i delim) String() string {
if i >= delim(len(_delim_index)-1) {
return "delim(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _delim_name[_delim_index[i]:_delim_index[i+1]]
}

240
src/html/template/doc.go Normal file
View File

@@ -0,0 +1,240 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
/*
Package template (html/template) implements data-driven templates for
generating HTML output safe against code injection. It provides the
same interface as [text/template] and should be used instead of
[text/template] whenever the output is HTML.
The documentation here focuses on the security features of the package.
For information about how to program the templates themselves, see the
documentation for [text/template].
# Introduction
This package wraps [text/template] so you can share its template API
to parse and execute HTML templates safely.
tmpl, err := template.New("name").Parse(...)
// Error checking elided
err = tmpl.Execute(out, data)
If successful, tmpl will now be injection-safe. Otherwise, err is an error
defined in the docs for ErrorCode.
HTML templates treat data values as plain text which should be encoded so they
can be safely embedded in an HTML document. The escaping is contextual, so
actions can appear within JavaScript, CSS, and URI contexts.
The security model used by this package assumes that template authors are
trusted, while Execute's data parameter is not. More details are
provided below.
Example
import "text/template"
...
t, err := template.New("foo").Parse(`{{define "T"}}Hello, {{.}}!{{end}}`)
err = t.ExecuteTemplate(out, "T", "<script>alert('you have been pwned')</script>")
produces
Hello, <script>alert('you have been pwned')</script>!
but the contextual autoescaping in html/template
import "html/template"
...
t, err := template.New("foo").Parse(`{{define "T"}}Hello, {{.}}!{{end}}`)
err = t.ExecuteTemplate(out, "T", "<script>alert('you have been pwned')</script>")
produces safe, escaped HTML output
Hello, &lt;script&gt;alert(&#39;you have been pwned&#39;)&lt;/script&gt;!
# Contexts
This package understands HTML, CSS, JavaScript, and URIs. It adds sanitizing
functions to each simple action pipeline, so given the excerpt
<a href="/search?q={{.}}">{{.}}</a>
At parse time each {{.}} is overwritten to add escaping functions as necessary.
In this case it becomes
<a href="/search?q={{. | urlescaper | attrescaper}}">{{. | htmlescaper}}</a>
where urlescaper, attrescaper, and htmlescaper are aliases for internal escaping
functions.
For these internal escaping functions, if an action pipeline evaluates to
a nil interface value, it is treated as though it were an empty string.
# Namespaced and data- attributes
Attributes with a namespace are treated as if they had no namespace.
Given the excerpt
<a my:href="{{.}}"></a>
At parse time the attribute will be treated as if it were just "href".
So at parse time the template becomes:
<a my:href="{{. | urlescaper | attrescaper}}"></a>
Similarly to attributes with namespaces, attributes with a "data-" prefix are
treated as if they had no "data-" prefix. So given
<a data-href="{{.}}"></a>
At parse time this becomes
<a data-href="{{. | urlescaper | attrescaper}}"></a>
If an attribute has both a namespace and a "data-" prefix, only the namespace
will be removed when determining the context. For example
<a my:data-href="{{.}}"></a>
This is handled as if "my:data-href" was just "data-href" and not "href" as
it would be if the "data-" prefix were to be ignored too. Thus at parse
time this becomes just
<a my:data-href="{{. | attrescaper}}"></a>
As a special case, attributes with the namespace "xmlns" are always treated
as containing URLs. Given the excerpts
<a xmlns:title="{{.}}"></a>
<a xmlns:href="{{.}}"></a>
<a xmlns:onclick="{{.}}"></a>
At parse time they become:
<a xmlns:title="{{. | urlescaper | attrescaper}}"></a>
<a xmlns:href="{{. | urlescaper | attrescaper}}"></a>
<a xmlns:onclick="{{. | urlescaper | attrescaper}}"></a>
# Errors
See the documentation of ErrorCode for details.
# A fuller picture
The rest of this package comment may be skipped on first reading; it includes
details necessary to understand escaping contexts and error messages. Most users
will not need to understand these details.
# Contexts
Assuming {{.}} is `O'Reilly: How are <i>you</i>?`, the table below shows
how {{.}} appears when used in the context to the left.
Context {{.}} After
{{.}} O'Reilly: How are &lt;i&gt;you&lt;/i&gt;?
<a title='{{.}}'> O&#39;Reilly: How are you?
<a href="/{{.}}"> O&#39;Reilly: How are %3ci%3eyou%3c/i%3e?
<a href="?q={{.}}"> O&#39;Reilly%3a%20How%20are%3ci%3e...%3f
<a onx='f("{{.}}")'> O\x27Reilly: How are \x3ci\x3eyou...?
<a onx='f({{.}})'> "O\x27Reilly: How are \x3ci\x3eyou...?"
<a onx='pattern = /{{.}}/;'> O\x27Reilly: How are \x3ci\x3eyou...\x3f
If used in an unsafe context, then the value might be filtered out:
Context {{.}} After
<a href="{{.}}"> #ZgotmplZ
since "O'Reilly:" is not an allowed protocol like "http:".
If {{.}} is the innocuous word, `left`, then it can appear more widely,
Context {{.}} After
{{.}} left
<a title='{{.}}'> left
<a href='{{.}}'> left
<a href='/{{.}}'> left
<a href='?dir={{.}}'> left
<a style="border-{{.}}: 4px"> left
<a style="align: {{.}}"> left
<a style="background: '{{.}}'> left
<a style="background: url('{{.}}')> left
<style>p.{{.}} {color:red}</style> left
Non-string values can be used in JavaScript contexts.
If {{.}} is
struct{A,B string}{ "foo", "bar" }
in the escaped template
<script>var pair = {{.}};</script>
then the template output is
<script>var pair = {"A": "foo", "B": "bar"};</script>
See package json to understand how non-string content is marshaled for
embedding in JavaScript contexts.
# Typed Strings
By default, this package assumes that all pipelines produce a plain text string.
It adds escaping pipeline stages necessary to correctly and safely embed that
plain text string in the appropriate context.
When a data value is not plain text, you can make sure it is not over-escaped
by marking it with its type.
Types HTML, JS, URL, and others from content.go can carry safe content that is
exempted from escaping.
The template
Hello, {{.}}!
can be invoked with
tmpl.Execute(out, template.HTML(`<b>World</b>`))
to produce
Hello, <b>World</b>!
instead of the
Hello, &lt;b&gt;World&lt;b&gt;!
that would have been produced if {{.}} was a regular string.
# Security Model
https://rawgit.com/mikesamuel/sanitized-jquery-templates/trunk/safetemplate.html#problem_definition defines "safe" as used by this package.
This package assumes that template authors are trusted, that Execute's data
parameter is not, and seeks to preserve the properties below in the face
of untrusted data:
Structure Preservation Property:
"... when a template author writes an HTML tag in a safe templating language,
the browser will interpret the corresponding portion of the output as a tag
regardless of the values of untrusted data, and similarly for other structures
such as attribute boundaries and JS and CSS string boundaries."
Code Effect Property:
"... only code specified by the template author should run as a result of
injecting the template output into a page and all code specified by the
template author should run as a result of the same."
Least Surprise Property:
"A developer (or code reviewer) familiar with HTML, CSS, and JavaScript, who
knows that contextual autoescaping happens should be able to look at a {{.}}
and correctly infer what sanitization happens."
Previously, ECMAScript 6 template literal were disabled by default, and could be
enabled with the GODEBUG=jstmpllitinterp=1 environment variable. Template
literals are now supported by default, and setting jstmpllitinterp has no
effect.
*/
package template

View File

@@ -0,0 +1,27 @@
// Code generated by "stringer -type element"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[elementNone-0]
_ = x[elementScript-1]
_ = x[elementStyle-2]
_ = x[elementTextarea-3]
_ = x[elementTitle-4]
}
const _element_name = "elementNoneelementScriptelementStyleelementTextareaelementTitle"
var _element_index = [...]uint8{0, 11, 24, 36, 51, 63}
func (i element) String() string {
if i >= element(len(_element_index)-1) {
return "element(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _element_name[_element_index[i]:_element_index[i+1]]
}

248
src/html/template/error.go Normal file
View File

@@ -0,0 +1,248 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"fmt"
"text/template/parse"
)
// Error describes a problem encountered during template Escaping.
type Error struct {
// ErrorCode describes the kind of error.
ErrorCode ErrorCode
// Node is the node that caused the problem, if known.
// If not nil, it overrides Name and Line.
Node parse.Node
// Name is the name of the template in which the error was encountered.
Name string
// Line is the line number of the error in the template source or 0.
Line int
// Description is a human-readable description of the problem.
Description string
}
// ErrorCode is a code for a kind of error.
type ErrorCode int
// We define codes for each error that manifests while escaping templates, but
// escaped templates may also fail at runtime.
//
// Output: "ZgotmplZ"
// Example:
//
// <img src="{{.X}}">
// where {{.X}} evaluates to `javascript:...`
//
// Discussion:
//
// "ZgotmplZ" is a special value that indicates that unsafe content reached a
// CSS or URL context at runtime. The output of the example will be
// <img src="#ZgotmplZ">
// If the data comes from a trusted source, use content types to exempt it
// from filtering: URL(`javascript:...`).
const (
// OK indicates the lack of an error.
OK ErrorCode = iota
// ErrAmbigContext: "... appears in an ambiguous context within a URL"
// Example:
// <a href="
// {{if .C}}
// /path/
// {{else}}
// /search?q=
// {{end}}
// {{.X}}
// ">
// Discussion:
// {{.X}} is in an ambiguous URL context since, depending on {{.C}},
// it may be either a URL suffix or a query parameter.
// Moving {{.X}} into the condition removes the ambiguity:
// <a href="{{if .C}}/path/{{.X}}{{else}}/search?q={{.X}}">
ErrAmbigContext
// ErrBadHTML: "expected space, attr name, or end of tag, but got ...",
// "... in unquoted attr", "... in attribute name"
// Example:
// <a href = /search?q=foo>
// <href=foo>
// <form na<e=...>
// <option selected<
// Discussion:
// This is often due to a typo in an HTML element, but some runes
// are banned in tag names, attribute names, and unquoted attribute
// values because they can tickle parser ambiguities.
// Quoting all attributes is the best policy.
ErrBadHTML
// ErrBranchEnd: "{{if}} branches end in different contexts"
// Example:
// {{if .C}}<a href="{{end}}{{.X}}
// Discussion:
// Package html/template statically examines each path through an
// {{if}}, {{range}}, or {{with}} to escape any following pipelines.
// The example is ambiguous since {{.X}} might be an HTML text node,
// or a URL prefix in an HTML attribute. The context of {{.X}} is
// used to figure out how to escape it, but that context depends on
// the run-time value of {{.C}} which is not statically known.
//
// The problem is usually something like missing quotes or angle
// brackets, or can be avoided by refactoring to put the two contexts
// into different branches of an if, range or with. If the problem
// is in a {{range}} over a collection that should never be empty,
// adding a dummy {{else}} can help.
ErrBranchEnd
// ErrEndContext: "... ends in a non-text context: ..."
// Examples:
// <div
// <div title="no close quote>
// <script>f()
// Discussion:
// Executed templates should produce a DocumentFragment of HTML.
// Templates that end without closing tags will trigger this error.
// Templates that should not be used in an HTML context or that
// produce incomplete Fragments should not be executed directly.
//
// {{define "main"}} <script>{{template "helper"}}</script> {{end}}
// {{define "helper"}} document.write(' <div title=" ') {{end}}
//
// "helper" does not produce a valid document fragment, so should
// not be Executed directly.
ErrEndContext
// ErrNoSuchTemplate: "no such template ..."
// Examples:
// {{define "main"}}<div {{template "attrs"}}>{{end}}
// {{define "attrs"}}href="{{.URL}}"{{end}}
// Discussion:
// Package html/template looks through template calls to compute the
// context.
// Here the {{.URL}} in "attrs" must be treated as a URL when called
// from "main", but you will get this error if "attrs" is not defined
// when "main" is parsed.
ErrNoSuchTemplate
// ErrOutputContext: "cannot compute output context for template ..."
// Examples:
// {{define "t"}}{{if .T}}{{template "t" .T}}{{end}}{{.H}}",{{end}}
// Discussion:
// A recursive template does not end in the same context in which it
// starts, and a reliable output context cannot be computed.
// Look for typos in the named template.
// If the template should not be called in the named start context,
// look for calls to that template in unexpected contexts.
// Maybe refactor recursive templates to not be recursive.
ErrOutputContext
// ErrPartialCharset: "unfinished JS regexp charset in ..."
// Example:
// <script>var pattern = /foo[{{.Chars}}]/</script>
// Discussion:
// Package html/template does not support interpolation into regular
// expression literal character sets.
ErrPartialCharset
// ErrPartialEscape: "unfinished escape sequence in ..."
// Example:
// <script>alert("\{{.X}}")</script>
// Discussion:
// Package html/template does not support actions following a
// backslash.
// This is usually an error and there are better solutions; for
// example
// <script>alert("{{.X}}")</script>
// should work, and if {{.X}} is a partial escape sequence such as
// "xA0", mark the whole sequence as safe content: JSStr(`\xA0`)
ErrPartialEscape
// ErrRangeLoopReentry: "on range loop re-entry: ..."
// Example:
// <script>var x = [{{range .}}'{{.}},{{end}}]</script>
// Discussion:
// If an iteration through a range would cause it to end in a
// different context than an earlier pass, there is no single context.
// In the example, there is missing a quote, so it is not clear
// whether {{.}} is meant to be inside a JS string or in a JS value
// context. The second iteration would produce something like
//
// <script>var x = ['firstValue,'secondValue]</script>
ErrRangeLoopReentry
// ErrSlashAmbig: '/' could start a division or regexp.
// Example:
// <script>
// {{if .C}}var x = 1{{end}}
// /-{{.N}}/i.test(x) ? doThis : doThat();
// </script>
// Discussion:
// The example above could produce `var x = 1/-2/i.test(s)...`
// in which the first '/' is a mathematical division operator or it
// could produce `/-2/i.test(s)` in which the first '/' starts a
// regexp literal.
// Look for missing semicolons inside branches, and maybe add
// parentheses to make it clear which interpretation you intend.
ErrSlashAmbig
// ErrPredefinedEscaper: "predefined escaper ... disallowed in template"
// Example:
// <div class={{. | html}}>Hello<div>
// Discussion:
// Package html/template already contextually escapes all pipelines to
// produce HTML output safe against code injection. Manually escaping
// pipeline output using the predefined escapers "html" or "urlquery" is
// unnecessary, and may affect the correctness or safety of the escaped
// pipeline output in Go 1.8 and earlier.
//
// In most cases, such as the given example, this error can be resolved by
// simply removing the predefined escaper from the pipeline and letting the
// contextual autoescaper handle the escaping of the pipeline. In other
// instances, where the predefined escaper occurs in the middle of a
// pipeline where subsequent commands expect escaped input, e.g.
// {{.X | html | makeALink}}
// where makeALink does
// return `<a href="`+input+`">link</a>`
// consider refactoring the surrounding template to make use of the
// contextual autoescaper, i.e.
// <a href="{{.X}}">link</a>
//
// To ease migration to Go 1.9 and beyond, "html" and "urlquery" will
// continue to be allowed as the last command in a pipeline. However, if the
// pipeline occurs in an unquoted attribute value context, "html" is
// disallowed. Avoid using "html" and "urlquery" entirely in new templates.
ErrPredefinedEscaper
// ErrJSTemplate: "... appears in a JS template literal"
// Example:
// <script>var tmpl = `{{.Interp}}`</script>
// Discussion:
// Package html/template does not support actions inside of JS template
// literals.
//
// Deprecated: ErrJSTemplate is no longer returned when an action is present
// in a JS template literal. Actions inside of JS template literals are now
// escaped as expected.
ErrJSTemplate
)
func (e *Error) Error() string {
switch {
case e.Node != nil:
loc, _ := (*parse.Tree)(nil).ErrorContext(e.Node)
return fmt.Sprintf("html/template:%s: %s", loc, e.Description)
case e.Line != 0:
return fmt.Sprintf("html/template:%s:%d: %s", e.Name, e.Line, e.Description)
case e.Name != "":
return fmt.Sprintf("html/template:%s: %s", e.Name, e.Description)
}
return "html/template: " + e.Description
}
// errorf creates an error given a format string f and args.
// The template Name still needs to be supplied.
func errorf(k ErrorCode, node parse.Node, line int, f string, args ...any) *Error {
return &Error{k, node, "", line, fmt.Sprintf(f, args...)}
}

1004
src/html/template/escape.go Normal file

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,181 @@
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template_test
import (
"fmt"
"html/template"
"log"
"os"
"strings"
)
func Example() {
const tpl = `
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>{{.Title}}</title>
</head>
<body>
{{range .Items}}<div>{{ . }}</div>{{else}}<div><strong>no rows</strong></div>{{end}}
</body>
</html>`
check := func(err error) {
if err != nil {
log.Fatal(err)
}
}
t, err := template.New("webpage").Parse(tpl)
check(err)
data := struct {
Title string
Items []string
}{
Title: "My page",
Items: []string{
"My photos",
"My blog",
},
}
err = t.Execute(os.Stdout, data)
check(err)
noItems := struct {
Title string
Items []string
}{
Title: "My another page",
Items: []string{},
}
err = t.Execute(os.Stdout, noItems)
check(err)
// Output:
// <!DOCTYPE html>
// <html>
// <head>
// <meta charset="UTF-8">
// <title>My page</title>
// </head>
// <body>
// <div>My photos</div><div>My blog</div>
// </body>
// </html>
// <!DOCTYPE html>
// <html>
// <head>
// <meta charset="UTF-8">
// <title>My another page</title>
// </head>
// <body>
// <div><strong>no rows</strong></div>
// </body>
// </html>
}
func Example_autoescaping() {
check := func(err error) {
if err != nil {
log.Fatal(err)
}
}
t, err := template.New("foo").Parse(`{{define "T"}}Hello, {{.}}!{{end}}`)
check(err)
err = t.ExecuteTemplate(os.Stdout, "T", "<script>alert('you have been pwned')</script>")
check(err)
// Output:
// Hello, &lt;script&gt;alert(&#39;you have been pwned&#39;)&lt;/script&gt;!
}
func Example_escape() {
const s = `"Fran & Freddie's Diner" <tasty@example.com>`
v := []any{`"Fran & Freddie's Diner"`, ' ', `<tasty@example.com>`}
fmt.Println(template.HTMLEscapeString(s))
template.HTMLEscape(os.Stdout, []byte(s))
fmt.Fprintln(os.Stdout, "")
fmt.Println(template.HTMLEscaper(v...))
fmt.Println(template.JSEscapeString(s))
template.JSEscape(os.Stdout, []byte(s))
fmt.Fprintln(os.Stdout, "")
fmt.Println(template.JSEscaper(v...))
fmt.Println(template.URLQueryEscaper(v...))
// Output:
// &#34;Fran &amp; Freddie&#39;s Diner&#34; &lt;tasty@example.com&gt;
// &#34;Fran &amp; Freddie&#39;s Diner&#34; &lt;tasty@example.com&gt;
// &#34;Fran &amp; Freddie&#39;s Diner&#34;32&lt;tasty@example.com&gt;
// \"Fran \u0026 Freddie\'s Diner\" \u003Ctasty@example.com\u003E
// \"Fran \u0026 Freddie\'s Diner\" \u003Ctasty@example.com\u003E
// \"Fran \u0026 Freddie\'s Diner\"32\u003Ctasty@example.com\u003E
// %22Fran+%26+Freddie%27s+Diner%2232%3Ctasty%40example.com%3E
}
func ExampleTemplate_Delims() {
const text = "<<.Greeting>> {{.Name}}"
data := struct {
Greeting string
Name string
}{
Greeting: "Hello",
Name: "Joe",
}
t := template.Must(template.New("tpl").Delims("<<", ">>").Parse(text))
err := t.Execute(os.Stdout, data)
if err != nil {
log.Fatal(err)
}
// Output:
// Hello {{.Name}}
}
// The following example is duplicated in text/template; keep them in sync.
func ExampleTemplate_block() {
const (
master = `Names:{{block "list" .}}{{"\n"}}{{range .}}{{println "-" .}}{{end}}{{end}}`
overlay = `{{define "list"}} {{join . ", "}}{{end}} `
)
var (
funcs = template.FuncMap{"join": strings.Join}
guardians = []string{"Gamora", "Groot", "Nebula", "Rocket", "Star-Lord"}
)
masterTmpl, err := template.New("master").Funcs(funcs).Parse(master)
if err != nil {
log.Fatal(err)
}
overlayTmpl, err := template.Must(masterTmpl.Clone()).Parse(overlay)
if err != nil {
log.Fatal(err)
}
if err := masterTmpl.Execute(os.Stdout, guardians); err != nil {
log.Fatal(err)
}
if err := overlayTmpl.Execute(os.Stdout, guardians); err != nil {
log.Fatal(err)
}
// Output:
// Names:
// - Gamora
// - Groot
// - Nebula
// - Rocket
// - Star-Lord
// Names: Gamora, Groot, Nebula, Rocket, Star-Lord
}

View File

@@ -0,0 +1,225 @@
// Copyright 2016 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template_test
import (
"io"
"log"
"os"
"path/filepath"
"text/template"
)
// templateFile defines the contents of a template to be stored in a file, for testing.
type templateFile struct {
name string
contents string
}
func createTestDir(files []templateFile) string {
dir, err := os.MkdirTemp("", "template")
if err != nil {
log.Fatal(err)
}
for _, file := range files {
f, err := os.Create(filepath.Join(dir, file.name))
if err != nil {
log.Fatal(err)
}
defer f.Close()
_, err = io.WriteString(f, file.contents)
if err != nil {
log.Fatal(err)
}
}
return dir
}
// The following example is duplicated in text/template; keep them in sync.
// Here we demonstrate loading a set of templates from a directory.
func ExampleTemplate_glob() {
// Here we create a temporary directory and populate it with our sample
// template definition files; usually the template files would already
// exist in some location known to the program.
dir := createTestDir([]templateFile{
// T0.tmpl is a plain template file that just invokes T1.
{"T0.tmpl", `T0 invokes T1: ({{template "T1"}})`},
// T1.tmpl defines a template, T1 that invokes T2.
{"T1.tmpl", `{{define "T1"}}T1 invokes T2: ({{template "T2"}}){{end}}`},
// T2.tmpl defines a template T2.
{"T2.tmpl", `{{define "T2"}}This is T2{{end}}`},
})
// Clean up after the test; another quirk of running as an example.
defer os.RemoveAll(dir)
// pattern is the glob pattern used to find all the template files.
pattern := filepath.Join(dir, "*.tmpl")
// Here starts the example proper.
// T0.tmpl is the first name matched, so it becomes the starting template,
// the value returned by ParseGlob.
tmpl := template.Must(template.ParseGlob(pattern))
err := tmpl.Execute(os.Stdout, nil)
if err != nil {
log.Fatalf("template execution: %s", err)
}
// Output:
// T0 invokes T1: (T1 invokes T2: (This is T2))
}
// Here we demonstrate loading a set of templates from files in different directories
func ExampleTemplate_parsefiles() {
// Here we create different temporary directories and populate them with our sample
// template definition files; usually the template files would already
// exist in some location known to the program.
dir1 := createTestDir([]templateFile{
// T1.tmpl is a plain template file that just invokes T2.
{"T1.tmpl", `T1 invokes T2: ({{template "T2"}})`},
})
dir2 := createTestDir([]templateFile{
// T2.tmpl defines a template T2.
{"T2.tmpl", `{{define "T2"}}This is T2{{end}}`},
})
// Clean up after the test; another quirk of running as an example.
defer func(dirs ...string) {
for _, dir := range dirs {
os.RemoveAll(dir)
}
}(dir1, dir2)
// Here starts the example proper.
// Let's just parse only dir1/T0 and dir2/T2
paths := []string{
filepath.Join(dir1, "T1.tmpl"),
filepath.Join(dir2, "T2.tmpl"),
}
tmpl := template.Must(template.ParseFiles(paths...))
err := tmpl.Execute(os.Stdout, nil)
if err != nil {
log.Fatalf("template execution: %s", err)
}
// Output:
// T1 invokes T2: (This is T2)
}
// The following example is duplicated in text/template; keep them in sync.
// This example demonstrates one way to share some templates
// and use them in different contexts. In this variant we add multiple driver
// templates by hand to an existing bundle of templates.
func ExampleTemplate_helpers() {
// Here we create a temporary directory and populate it with our sample
// template definition files; usually the template files would already
// exist in some location known to the program.
dir := createTestDir([]templateFile{
// T1.tmpl defines a template, T1 that invokes T2.
{"T1.tmpl", `{{define "T1"}}T1 invokes T2: ({{template "T2"}}){{end}}`},
// T2.tmpl defines a template T2.
{"T2.tmpl", `{{define "T2"}}This is T2{{end}}`},
})
// Clean up after the test; another quirk of running as an example.
defer os.RemoveAll(dir)
// pattern is the glob pattern used to find all the template files.
pattern := filepath.Join(dir, "*.tmpl")
// Here starts the example proper.
// Load the helpers.
templates := template.Must(template.ParseGlob(pattern))
// Add one driver template to the bunch; we do this with an explicit template definition.
_, err := templates.Parse("{{define `driver1`}}Driver 1 calls T1: ({{template `T1`}})\n{{end}}")
if err != nil {
log.Fatal("parsing driver1: ", err)
}
// Add another driver template.
_, err = templates.Parse("{{define `driver2`}}Driver 2 calls T2: ({{template `T2`}})\n{{end}}")
if err != nil {
log.Fatal("parsing driver2: ", err)
}
// We load all the templates before execution. This package does not require
// that behavior but html/template's escaping does, so it's a good habit.
err = templates.ExecuteTemplate(os.Stdout, "driver1", nil)
if err != nil {
log.Fatalf("driver1 execution: %s", err)
}
err = templates.ExecuteTemplate(os.Stdout, "driver2", nil)
if err != nil {
log.Fatalf("driver2 execution: %s", err)
}
// Output:
// Driver 1 calls T1: (T1 invokes T2: (This is T2))
// Driver 2 calls T2: (This is T2)
}
// The following example is duplicated in text/template; keep them in sync.
// This example demonstrates how to use one group of driver
// templates with distinct sets of helper templates.
func ExampleTemplate_share() {
// Here we create a temporary directory and populate it with our sample
// template definition files; usually the template files would already
// exist in some location known to the program.
dir := createTestDir([]templateFile{
// T0.tmpl is a plain template file that just invokes T1.
{"T0.tmpl", "T0 ({{.}} version) invokes T1: ({{template `T1`}})\n"},
// T1.tmpl defines a template, T1 that invokes T2. Note T2 is not defined
{"T1.tmpl", `{{define "T1"}}T1 invokes T2: ({{template "T2"}}){{end}}`},
})
// Clean up after the test; another quirk of running as an example.
defer os.RemoveAll(dir)
// pattern is the glob pattern used to find all the template files.
pattern := filepath.Join(dir, "*.tmpl")
// Here starts the example proper.
// Load the drivers.
drivers := template.Must(template.ParseGlob(pattern))
// We must define an implementation of the T2 template. First we clone
// the drivers, then add a definition of T2 to the template name space.
// 1. Clone the helper set to create a new name space from which to run them.
first, err := drivers.Clone()
if err != nil {
log.Fatal("cloning helpers: ", err)
}
// 2. Define T2, version A, and parse it.
_, err = first.Parse("{{define `T2`}}T2, version A{{end}}")
if err != nil {
log.Fatal("parsing T2: ", err)
}
// Now repeat the whole thing, using a different version of T2.
// 1. Clone the drivers.
second, err := drivers.Clone()
if err != nil {
log.Fatal("cloning drivers: ", err)
}
// 2. Define T2, version B, and parse it.
_, err = second.Parse("{{define `T2`}}T2, version B{{end}}")
if err != nil {
log.Fatal("parsing T2: ", err)
}
// Execute the templates in the reverse order to verify the
// first is unaffected by the second.
err = second.ExecuteTemplate(os.Stdout, "T0.tmpl", "second")
if err != nil {
log.Fatalf("second execution: %s", err)
}
err = first.ExecuteTemplate(os.Stdout, "T0.tmpl", "first")
if err != nil {
log.Fatalf("first: execution: %s", err)
}
// Output:
// T0 (second version) invokes T1: (T1 invokes T2: (T2, version B))
// T0 (first version) invokes T1: (T1 invokes T2: (T2, version A))
}

File diff suppressed because it is too large Load Diff

270
src/html/template/html.go Normal file
View File

@@ -0,0 +1,270 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"fmt"
"strings"
"unicode/utf8"
)
// htmlNospaceEscaper escapes for inclusion in unquoted attribute values.
func htmlNospaceEscaper(args ...any) string {
s, t := stringify(args...)
if s == "" {
return filterFailsafe
}
if t == contentTypeHTML {
return htmlReplacer(stripTags(s), htmlNospaceNormReplacementTable, false)
}
return htmlReplacer(s, htmlNospaceReplacementTable, false)
}
// attrEscaper escapes for inclusion in quoted attribute values.
func attrEscaper(args ...any) string {
s, t := stringify(args...)
if t == contentTypeHTML {
return htmlReplacer(stripTags(s), htmlNormReplacementTable, true)
}
return htmlReplacer(s, htmlReplacementTable, true)
}
// rcdataEscaper escapes for inclusion in an RCDATA element body.
func rcdataEscaper(args ...any) string {
s, t := stringify(args...)
if t == contentTypeHTML {
return htmlReplacer(s, htmlNormReplacementTable, true)
}
return htmlReplacer(s, htmlReplacementTable, true)
}
// htmlEscaper escapes for inclusion in HTML text.
func htmlEscaper(args ...any) string {
s, t := stringify(args...)
if t == contentTypeHTML {
return s
}
return htmlReplacer(s, htmlReplacementTable, true)
}
// htmlReplacementTable contains the runes that need to be escaped
// inside a quoted attribute value or in a text node.
var htmlReplacementTable = []string{
// https://www.w3.org/TR/html5/syntax.html#attribute-value-(unquoted)-state
// U+0000 NULL Parse error. Append a U+FFFD REPLACEMENT
// CHARACTER character to the current attribute's value.
// "
// and similarly
// https://www.w3.org/TR/html5/syntax.html#before-attribute-value-state
0: "\uFFFD",
'"': "&#34;",
'&': "&amp;",
'\'': "&#39;",
'+': "&#43;",
'<': "&lt;",
'>': "&gt;",
}
// htmlNormReplacementTable is like htmlReplacementTable but without '&' to
// avoid over-encoding existing entities.
var htmlNormReplacementTable = []string{
0: "\uFFFD",
'"': "&#34;",
'\'': "&#39;",
'+': "&#43;",
'<': "&lt;",
'>': "&gt;",
}
// htmlNospaceReplacementTable contains the runes that need to be escaped
// inside an unquoted attribute value.
// The set of runes escaped is the union of the HTML specials and
// those determined by running the JS below in browsers:
// <div id=d></div>
// <script>(function () {
// var a = [], d = document.getElementById("d"), i, c, s;
// for (i = 0; i < 0x10000; ++i) {
//
// c = String.fromCharCode(i);
// d.innerHTML = "<span title=" + c + "lt" + c + "></span>"
// s = d.getElementsByTagName("SPAN")[0];
// if (!s || s.title !== c + "lt" + c) { a.push(i.toString(16)); }
//
// }
// document.write(a.join(", "));
// })()</script>
var htmlNospaceReplacementTable = []string{
0: "&#xfffd;",
'\t': "&#9;",
'\n': "&#10;",
'\v': "&#11;",
'\f': "&#12;",
'\r': "&#13;",
' ': "&#32;",
'"': "&#34;",
'&': "&amp;",
'\'': "&#39;",
'+': "&#43;",
'<': "&lt;",
'=': "&#61;",
'>': "&gt;",
// A parse error in the attribute value (unquoted) and
// before attribute value states.
// Treated as a quoting character by IE.
'`': "&#96;",
}
// htmlNospaceNormReplacementTable is like htmlNospaceReplacementTable but
// without '&' to avoid over-encoding existing entities.
var htmlNospaceNormReplacementTable = []string{
0: "&#xfffd;",
'\t': "&#9;",
'\n': "&#10;",
'\v': "&#11;",
'\f': "&#12;",
'\r': "&#13;",
' ': "&#32;",
'"': "&#34;",
'\'': "&#39;",
'+': "&#43;",
'<': "&lt;",
'=': "&#61;",
'>': "&gt;",
// A parse error in the attribute value (unquoted) and
// before attribute value states.
// Treated as a quoting character by IE.
'`': "&#96;",
}
// htmlReplacer returns s with runes replaced according to replacementTable
// and when badRunes is true, certain bad runes are allowed through unescaped.
func htmlReplacer(s string, replacementTable []string, badRunes bool) string {
written, b := 0, new(strings.Builder)
r, w := rune(0), 0
for i := 0; i < len(s); i += w {
// Cannot use 'for range s' because we need to preserve the width
// of the runes in the input. If we see a decoding error, the input
// width will not be utf8.Runelen(r) and we will overrun the buffer.
r, w = utf8.DecodeRuneInString(s[i:])
if int(r) < len(replacementTable) {
if repl := replacementTable[r]; len(repl) != 0 {
if written == 0 {
b.Grow(len(s))
}
b.WriteString(s[written:i])
b.WriteString(repl)
written = i + w
}
} else if badRunes {
// No-op.
// IE does not allow these ranges in unquoted attrs.
} else if 0xfdd0 <= r && r <= 0xfdef || 0xfff0 <= r && r <= 0xffff {
if written == 0 {
b.Grow(len(s))
}
fmt.Fprintf(b, "%s&#x%x;", s[written:i], r)
written = i + w
}
}
if written == 0 {
return s
}
b.WriteString(s[written:])
return b.String()
}
// stripTags takes a snippet of HTML and returns only the text content.
// For example, `<b>&iexcl;Hi!</b> <script>...</script>` -> `&iexcl;Hi! `.
func stripTags(html string) string {
var b strings.Builder
s, c, i, allText := []byte(html), context{}, 0, true
// Using the transition funcs helps us avoid mangling
// `<div title="1>2">` or `I <3 Ponies!`.
for i != len(s) {
if c.delim == delimNone {
st := c.state
// Use RCDATA instead of parsing into JS or CSS styles.
if c.element != elementNone && !isInTag(st) {
st = stateRCDATA
}
d, nread := transitionFunc[st](c, s[i:])
i1 := i + nread
if c.state == stateText || c.state == stateRCDATA {
// Emit text up to the start of the tag or comment.
j := i1
if d.state != c.state {
for j1 := j - 1; j1 >= i; j1-- {
if s[j1] == '<' {
j = j1
break
}
}
}
b.Write(s[i:j])
} else {
allText = false
}
c, i = d, i1
continue
}
i1 := i + bytes.IndexAny(s[i:], delimEnds[c.delim])
if i1 < i {
break
}
if c.delim != delimSpaceOrTagEnd {
// Consume any quote.
i1++
}
c, i = context{state: stateTag, element: c.element}, i1
}
if allText {
return html
} else if c.state == stateText || c.state == stateRCDATA {
b.Write(s[i:])
}
return b.String()
}
// htmlNameFilter accepts valid parts of an HTML attribute or tag name or
// a known-safe HTML attribute.
func htmlNameFilter(args ...any) string {
s, t := stringify(args...)
if t == contentTypeHTMLAttr {
return s
}
if len(s) == 0 {
// Avoid violation of structure preservation.
// <input checked {{.K}}={{.V}}>.
// Without this, if .K is empty then .V is the value of
// checked, but otherwise .V is the value of the attribute
// named .K.
return filterFailsafe
}
s = strings.ToLower(s)
if t := attrType(s); t != contentTypePlain {
// TODO: Split attr and element name part filters so we can recognize known attributes.
return filterFailsafe
}
for _, r := range s {
switch {
case '0' <= r && r <= '9':
case 'a' <= r && r <= 'z':
default:
return filterFailsafe
}
}
return s
}
// commentEscaper returns the empty string regardless of input.
// Comment content does not correspond to any parsed structure or
// human-readable content, so the simplest and most secure policy is to drop
// content interpolated into comments.
// This approach is equally valid whether or not static comment content is
// removed from the template.
func commentEscaper(args ...any) string {
return ""
}

View File

@@ -0,0 +1,97 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"html"
"strings"
"testing"
)
func TestHTMLNospaceEscaper(t *testing.T) {
input := ("\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f" +
"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !"#$%&'()*+,-./` +
`0123456789:;<=>?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\]^_` +
"`abcdefghijklmno" +
"pqrstuvwxyz{|}~\x7f" +
"\u00A0\u0100\u2028\u2029\ufeff\ufdec\U0001D11E" +
"erroneous\x960") // keep at the end
want := ("&#xfffd;\x01\x02\x03\x04\x05\x06\x07" +
"\x08&#9;&#10;&#11;&#12;&#13;\x0E\x0F" +
"\x10\x11\x12\x13\x14\x15\x16\x17" +
"\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
`&#32;!&#34;#$%&amp;&#39;()*&#43;,-./` +
`0123456789:;&lt;&#61;&gt;?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\]^_` +
`&#96;abcdefghijklmno` +
`pqrstuvwxyz{|}~` + "\u007f" +
"\u00A0\u0100\u2028\u2029\ufeff&#xfdec;\U0001D11E" +
"erroneous&#xfffd;0") // keep at the end
got := htmlNospaceEscaper(input)
if got != want {
t.Errorf("encode: want\n\t%q\nbut got\n\t%q", want, got)
}
r := strings.NewReplacer("\x00", "\ufffd", "\x96", "\ufffd")
got, want = html.UnescapeString(got), r.Replace(input)
if want != got {
t.Errorf("decode: want\n\t%q\nbut got\n\t%q", want, got)
}
}
func TestStripTags(t *testing.T) {
tests := []struct {
input, want string
}{
{"", ""},
{"Hello, World!", "Hello, World!"},
{"foo&amp;bar", "foo&amp;bar"},
{`Hello <a href="www.example.com/">World</a>!`, "Hello World!"},
{"Foo <textarea>Bar</textarea> Baz", "Foo Bar Baz"},
{"Foo <!-- Bar --> Baz", "Foo Baz"},
{"<", "<"},
{"foo < bar", "foo < bar"},
{`Foo<script type="text/javascript">alert(1337)</script>Bar`, "FooBar"},
{`Foo<div title="1>2">Bar`, "FooBar"},
{`I <3 Ponies!`, `I <3 Ponies!`},
{`<script>foo()</script>`, ``},
}
for _, test := range tests {
if got := stripTags(test.input); got != test.want {
t.Errorf("%q: want %q, got %q", test.input, test.want, got)
}
}
}
func BenchmarkHTMLNospaceEscaper(b *testing.B) {
for i := 0; i < b.N; i++ {
htmlNospaceEscaper("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}
func BenchmarkHTMLNospaceEscaperNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
htmlNospaceEscaper("The_quick,_brown_fox_jumps_over_the_lazy_dog.")
}
}
func BenchmarkStripTags(b *testing.B) {
for i := 0; i < b.N; i++ {
stripTags("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}
func BenchmarkStripTagsNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
stripTags("The quick, brown fox jumps over the lazy dog.")
}
}

485
src/html/template/js.go Normal file
View File

@@ -0,0 +1,485 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"encoding/json"
"fmt"
"reflect"
"strings"
"unicode/utf8"
)
// jsWhitespace contains all of the JS whitespace characters, as defined
// by the \s character class.
// See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_expressions/Character_classes.
const jsWhitespace = "\f\n\r\t\v\u0020\u00a0\u1680\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u2028\u2029\u202f\u205f\u3000\ufeff"
// nextJSCtx returns the context that determines whether a slash after the
// given run of tokens starts a regular expression instead of a division
// operator: / or /=.
//
// This assumes that the token run does not include any string tokens, comment
// tokens, regular expression literal tokens, or division operators.
//
// This fails on some valid but nonsensical JavaScript programs like
// "x = ++/foo/i" which is quite different than "x++/foo/i", but is not known to
// fail on any known useful programs. It is based on the draft
// JavaScript 2.0 lexical grammar and requires one token of lookbehind:
// https://www.mozilla.org/js/language/js20-2000-07/rationale/syntax.html
func nextJSCtx(s []byte, preceding jsCtx) jsCtx {
// Trim all JS whitespace characters
s = bytes.TrimRight(s, jsWhitespace)
if len(s) == 0 {
return preceding
}
// All cases below are in the single-byte UTF-8 group.
switch c, n := s[len(s)-1], len(s); c {
case '+', '-':
// ++ and -- are not regexp preceders, but + and - are whether
// they are used as infix or prefix operators.
start := n - 1
// Count the number of adjacent dashes or pluses.
for start > 0 && s[start-1] == c {
start--
}
if (n-start)&1 == 1 {
// Reached for trailing minus signs since "---" is the
// same as "-- -".
return jsCtxRegexp
}
return jsCtxDivOp
case '.':
// Handle "42."
if n != 1 && '0' <= s[n-2] && s[n-2] <= '9' {
return jsCtxDivOp
}
return jsCtxRegexp
// Suffixes for all punctuators from section 7.7 of the language spec
// that only end binary operators not handled above.
case ',', '<', '>', '=', '*', '%', '&', '|', '^', '?':
return jsCtxRegexp
// Suffixes for all punctuators from section 7.7 of the language spec
// that are prefix operators not handled above.
case '!', '~':
return jsCtxRegexp
// Matches all the punctuators from section 7.7 of the language spec
// that are open brackets not handled above.
case '(', '[':
return jsCtxRegexp
// Matches all the punctuators from section 7.7 of the language spec
// that precede expression starts.
case ':', ';', '{':
return jsCtxRegexp
// CAVEAT: the close punctuators ('}', ']', ')') precede div ops and
// are handled in the default except for '}' which can precede a
// division op as in
// ({ valueOf: function () { return 42 } } / 2
// which is valid, but, in practice, developers don't divide object
// literals, so our heuristic works well for code like
// function () { ... } /foo/.test(x) && sideEffect();
// The ')' punctuator can precede a regular expression as in
// if (b) /foo/.test(x) && ...
// but this is much less likely than
// (a + b) / c
case '}':
return jsCtxRegexp
default:
// Look for an IdentifierName and see if it is a keyword that
// can precede a regular expression.
j := n
for j > 0 && isJSIdentPart(rune(s[j-1])) {
j--
}
if regexpPrecederKeywords[string(s[j:])] {
return jsCtxRegexp
}
}
// Otherwise is a punctuator not listed above, or
// a string which precedes a div op, or an identifier
// which precedes a div op.
return jsCtxDivOp
}
// regexpPrecederKeywords is a set of reserved JS keywords that can precede a
// regular expression in JS source.
var regexpPrecederKeywords = map[string]bool{
"break": true,
"case": true,
"continue": true,
"delete": true,
"do": true,
"else": true,
"finally": true,
"in": true,
"instanceof": true,
"return": true,
"throw": true,
"try": true,
"typeof": true,
"void": true,
}
var jsonMarshalType = reflect.TypeFor[json.Marshaler]()
// indirectToJSONMarshaler returns the value, after dereferencing as many times
// as necessary to reach the base type (or nil) or an implementation of json.Marshal.
func indirectToJSONMarshaler(a any) any {
// text/template now supports passing untyped nil as a func call
// argument, so we must support it. Otherwise we'd panic below, as one
// cannot call the Type or Interface methods on an invalid
// reflect.Value. See golang.org/issue/18716.
if a == nil {
return nil
}
v := reflect.ValueOf(a)
for !v.Type().Implements(jsonMarshalType) && v.Kind() == reflect.Pointer && !v.IsNil() {
v = v.Elem()
}
return v.Interface()
}
// jsValEscaper escapes its inputs to a JS Expression (section 11.14) that has
// neither side-effects nor free variables outside (NaN, Infinity).
func jsValEscaper(args ...any) string {
var a any
if len(args) == 1 {
a = indirectToJSONMarshaler(args[0])
switch t := a.(type) {
case JS:
return string(t)
case JSStr:
// TODO: normalize quotes.
return `"` + string(t) + `"`
case json.Marshaler:
// Do not treat as a Stringer.
case fmt.Stringer:
a = t.String()
}
} else {
for i, arg := range args {
args[i] = indirectToJSONMarshaler(arg)
}
a = fmt.Sprint(args...)
}
// TODO: detect cycles before calling Marshal which loops infinitely on
// cyclic data. This may be an unacceptable DoS risk.
b, err := json.Marshal(a)
if err != nil {
// While the standard JSON marshaler does not include user controlled
// information in the error message, if a type has a MarshalJSON method,
// the content of the error message is not guaranteed. Since we insert
// the error into the template, as part of a comment, we attempt to
// prevent the error from either terminating the comment, or the script
// block itself.
//
// In particular we:
// * replace "*/" comment end tokens with "* /", which does not
// terminate the comment
// * replace "</script" with "\x3C/script", and "<!--" with
// "\x3C!--", which prevents confusing script block termination
// semantics
//
// We also put a space before the comment so that if it is flush against
// a division operator it is not turned into a line comment:
// x/{{y}}
// turning into
// x//* error marshaling y:
// second line of error message */null
errStr := err.Error()
errStr = strings.ReplaceAll(errStr, "*/", "* /")
errStr = strings.ReplaceAll(errStr, "</script", `\x3C/script`)
errStr = strings.ReplaceAll(errStr, "<!--", `\x3C!--`)
return fmt.Sprintf(" /* %s */null ", errStr)
}
// TODO: maybe post-process output to prevent it from containing
// "<!--", "-->", "<![CDATA[", "]]>", or "</script"
// in case custom marshalers produce output containing those.
// Note: Do not use \x escaping to save bytes because it is not JSON compatible and this escaper
// supports ld+json content-type.
if len(b) == 0 {
// In, `x=y/{{.}}*z` a json.Marshaler that produces "" should
// not cause the output `x=y/*z`.
return " null "
}
first, _ := utf8.DecodeRune(b)
last, _ := utf8.DecodeLastRune(b)
var buf strings.Builder
// Prevent IdentifierNames and NumericLiterals from running into
// keywords: in, instanceof, typeof, void
pad := isJSIdentPart(first) || isJSIdentPart(last)
if pad {
buf.WriteByte(' ')
}
written := 0
// Make sure that json.Marshal escapes codepoints U+2028 & U+2029
// so it falls within the subset of JSON which is valid JS.
for i := 0; i < len(b); {
rune, n := utf8.DecodeRune(b[i:])
repl := ""
if rune == 0x2028 {
repl = `\u2028`
} else if rune == 0x2029 {
repl = `\u2029`
}
if repl != "" {
buf.Write(b[written:i])
buf.WriteString(repl)
written = i + n
}
i += n
}
if buf.Len() != 0 {
buf.Write(b[written:])
if pad {
buf.WriteByte(' ')
}
return buf.String()
}
return string(b)
}
// jsStrEscaper produces a string that can be included between quotes in
// JavaScript source, in JavaScript embedded in an HTML5 <script> element,
// or in an HTML5 event handler attribute such as onclick.
func jsStrEscaper(args ...any) string {
s, t := stringify(args...)
if t == contentTypeJSStr {
return replace(s, jsStrNormReplacementTable)
}
return replace(s, jsStrReplacementTable)
}
func jsTmplLitEscaper(args ...any) string {
s, _ := stringify(args...)
return replace(s, jsBqStrReplacementTable)
}
// jsRegexpEscaper behaves like jsStrEscaper but escapes regular expression
// specials so the result is treated literally when included in a regular
// expression literal. /foo{{.X}}bar/ matches the string "foo" followed by
// the literal text of {{.X}} followed by the string "bar".
func jsRegexpEscaper(args ...any) string {
s, _ := stringify(args...)
s = replace(s, jsRegexpReplacementTable)
if s == "" {
// /{{.X}}/ should not produce a line comment when .X == "".
return "(?:)"
}
return s
}
// replace replaces each rune r of s with replacementTable[r], provided that
// r < len(replacementTable). If replacementTable[r] is the empty string then
// no replacement is made.
// It also replaces runes U+2028 and U+2029 with the raw strings `\u2028` and
// `\u2029`.
func replace(s string, replacementTable []string) string {
var b strings.Builder
r, w, written := rune(0), 0, 0
for i := 0; i < len(s); i += w {
// See comment in htmlEscaper.
r, w = utf8.DecodeRuneInString(s[i:])
var repl string
switch {
case int(r) < len(lowUnicodeReplacementTable):
repl = lowUnicodeReplacementTable[r]
case int(r) < len(replacementTable) && replacementTable[r] != "":
repl = replacementTable[r]
case r == '\u2028':
repl = `\u2028`
case r == '\u2029':
repl = `\u2029`
default:
continue
}
if written == 0 {
b.Grow(len(s))
}
b.WriteString(s[written:i])
b.WriteString(repl)
written = i + w
}
if written == 0 {
return s
}
b.WriteString(s[written:])
return b.String()
}
var lowUnicodeReplacementTable = []string{
0: `\u0000`, 1: `\u0001`, 2: `\u0002`, 3: `\u0003`, 4: `\u0004`, 5: `\u0005`, 6: `\u0006`,
'\a': `\u0007`,
'\b': `\u0008`,
'\t': `\t`,
'\n': `\n`,
'\v': `\u000b`, // "\v" == "v" on IE 6.
'\f': `\f`,
'\r': `\r`,
0xe: `\u000e`, 0xf: `\u000f`, 0x10: `\u0010`, 0x11: `\u0011`, 0x12: `\u0012`, 0x13: `\u0013`,
0x14: `\u0014`, 0x15: `\u0015`, 0x16: `\u0016`, 0x17: `\u0017`, 0x18: `\u0018`, 0x19: `\u0019`,
0x1a: `\u001a`, 0x1b: `\u001b`, 0x1c: `\u001c`, 0x1d: `\u001d`, 0x1e: `\u001e`, 0x1f: `\u001f`,
}
var jsStrReplacementTable = []string{
0: `\u0000`,
'\t': `\t`,
'\n': `\n`,
'\v': `\u000b`, // "\v" == "v" on IE 6.
'\f': `\f`,
'\r': `\r`,
// Encode HTML specials as hex so the output can be embedded
// in HTML attributes without further encoding.
'"': `\u0022`,
'`': `\u0060`,
'&': `\u0026`,
'\'': `\u0027`,
'+': `\u002b`,
'/': `\/`,
'<': `\u003c`,
'>': `\u003e`,
'\\': `\\`,
}
// jsBqStrReplacementTable is like jsStrReplacementTable except it also contains
// the special characters for JS template literals: $, {, and }.
var jsBqStrReplacementTable = []string{
0: `\u0000`,
'\t': `\t`,
'\n': `\n`,
'\v': `\u000b`, // "\v" == "v" on IE 6.
'\f': `\f`,
'\r': `\r`,
// Encode HTML specials as hex so the output can be embedded
// in HTML attributes without further encoding.
'"': `\u0022`,
'`': `\u0060`,
'&': `\u0026`,
'\'': `\u0027`,
'+': `\u002b`,
'/': `\/`,
'<': `\u003c`,
'>': `\u003e`,
'\\': `\\`,
'$': `\u0024`,
'{': `\u007b`,
'}': `\u007d`,
}
// jsStrNormReplacementTable is like jsStrReplacementTable but does not
// overencode existing escapes since this table has no entry for `\`.
var jsStrNormReplacementTable = []string{
0: `\u0000`,
'\t': `\t`,
'\n': `\n`,
'\v': `\u000b`, // "\v" == "v" on IE 6.
'\f': `\f`,
'\r': `\r`,
// Encode HTML specials as hex so the output can be embedded
// in HTML attributes without further encoding.
'"': `\u0022`,
'&': `\u0026`,
'\'': `\u0027`,
'`': `\u0060`,
'+': `\u002b`,
'/': `\/`,
'<': `\u003c`,
'>': `\u003e`,
}
var jsRegexpReplacementTable = []string{
0: `\u0000`,
'\t': `\t`,
'\n': `\n`,
'\v': `\u000b`, // "\v" == "v" on IE 6.
'\f': `\f`,
'\r': `\r`,
// Encode HTML specials as hex so the output can be embedded
// in HTML attributes without further encoding.
'"': `\u0022`,
'$': `\$`,
'&': `\u0026`,
'\'': `\u0027`,
'(': `\(`,
')': `\)`,
'*': `\*`,
'+': `\u002b`,
'-': `\-`,
'.': `\.`,
'/': `\/`,
'<': `\u003c`,
'>': `\u003e`,
'?': `\?`,
'[': `\[`,
'\\': `\\`,
']': `\]`,
'^': `\^`,
'{': `\{`,
'|': `\|`,
'}': `\}`,
}
// isJSIdentPart reports whether the given rune is a JS identifier part.
// It does not handle all the non-Latin letters, joiners, and combining marks,
// but it does handle every codepoint that can occur in a numeric literal or
// a keyword.
func isJSIdentPart(r rune) bool {
switch {
case r == '$':
return true
case '0' <= r && r <= '9':
return true
case 'A' <= r && r <= 'Z':
return true
case r == '_':
return true
case 'a' <= r && r <= 'z':
return true
}
return false
}
// isJSType reports whether the given MIME type should be considered JavaScript.
//
// It is used to determine whether a script tag with a type attribute is a javascript container.
func isJSType(mimeType string) bool {
// per
// https://www.w3.org/TR/html5/scripting-1.html#attr-script-type
// https://tools.ietf.org/html/rfc7231#section-3.1.1
// https://tools.ietf.org/html/rfc4329#section-3
// https://www.ietf.org/rfc/rfc4627.txt
// discard parameters
mimeType, _, _ = strings.Cut(mimeType, ";")
mimeType = strings.ToLower(mimeType)
mimeType = strings.TrimSpace(mimeType)
switch mimeType {
case
"application/ecmascript",
"application/javascript",
"application/json",
"application/ld+json",
"application/x-ecmascript",
"application/x-javascript",
"module",
"text/ecmascript",
"text/javascript",
"text/javascript1.0",
"text/javascript1.1",
"text/javascript1.2",
"text/javascript1.3",
"text/javascript1.4",
"text/javascript1.5",
"text/jscript",
"text/livescript",
"text/x-ecmascript",
"text/x-javascript":
return true
default:
return false
}
}

View File

@@ -0,0 +1,437 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"errors"
"math"
"strings"
"testing"
)
func TestNextJsCtx(t *testing.T) {
tests := []struct {
jsCtx jsCtx
s string
}{
// Statement terminators precede regexps.
{jsCtxRegexp, ";"},
// This is not airtight.
// ({ valueOf: function () { return 1 } } / 2)
// is valid JavaScript but in practice, devs do not do this.
// A block followed by a statement starting with a RegExp is
// much more common:
// while (x) {...} /foo/.test(x) || panic()
{jsCtxRegexp, "}"},
// But member, call, grouping, and array expression terminators
// precede div ops.
{jsCtxDivOp, ")"},
{jsCtxDivOp, "]"},
// At the start of a primary expression, array, or expression
// statement, expect a regexp.
{jsCtxRegexp, "("},
{jsCtxRegexp, "["},
{jsCtxRegexp, "{"},
// Assignment operators precede regexps as do all exclusively
// prefix and binary operators.
{jsCtxRegexp, "="},
{jsCtxRegexp, "+="},
{jsCtxRegexp, "*="},
{jsCtxRegexp, "*"},
{jsCtxRegexp, "!"},
// Whether the + or - is infix or prefix, it cannot precede a
// div op.
{jsCtxRegexp, "+"},
{jsCtxRegexp, "-"},
// An incr/decr op precedes a div operator.
// This is not airtight. In (g = ++/h/i) a regexp follows a
// pre-increment operator, but in practice devs do not try to
// increment or decrement regular expressions.
// (g++/h/i) where ++ is a postfix operator on g is much more
// common.
{jsCtxDivOp, "--"},
{jsCtxDivOp, "++"},
{jsCtxDivOp, "x--"},
// When we have many dashes or pluses, then they are grouped
// left to right.
{jsCtxRegexp, "x---"}, // A postfix -- then a -.
// return followed by a slash returns the regexp literal or the
// slash starts a regexp literal in an expression statement that
// is dead code.
{jsCtxRegexp, "return"},
{jsCtxRegexp, "return "},
{jsCtxRegexp, "return\t"},
{jsCtxRegexp, "return\n"},
{jsCtxRegexp, "return\u2028"},
// Identifiers can be divided and cannot validly be preceded by
// a regular expressions. Semicolon insertion cannot happen
// between an identifier and a regular expression on a new line
// because the one token lookahead for semicolon insertion has
// to conclude that it could be a div binary op and treat it as
// such.
{jsCtxDivOp, "x"},
{jsCtxDivOp, "x "},
{jsCtxDivOp, "x\t"},
{jsCtxDivOp, "x\n"},
{jsCtxDivOp, "x\u2028"},
{jsCtxDivOp, "preturn"},
// Numbers precede div ops.
{jsCtxDivOp, "0"},
// Dots that are part of a number are div preceders.
{jsCtxDivOp, "0."},
// Some JS interpreters treat NBSP as a normal space, so
// we must too in order to properly escape things.
{jsCtxRegexp, "=\u00A0"},
}
for _, test := range tests {
if ctx := nextJSCtx([]byte(test.s), jsCtxRegexp); ctx != test.jsCtx {
t.Errorf("%q: want %s got %s", test.s, test.jsCtx, ctx)
}
if ctx := nextJSCtx([]byte(test.s), jsCtxDivOp); ctx != test.jsCtx {
t.Errorf("%q: want %s got %s", test.s, test.jsCtx, ctx)
}
}
if nextJSCtx([]byte(" "), jsCtxRegexp) != jsCtxRegexp {
t.Error("Blank tokens")
}
if nextJSCtx([]byte(" "), jsCtxDivOp) != jsCtxDivOp {
t.Error("Blank tokens")
}
}
type jsonErrType struct{}
func (e *jsonErrType) MarshalJSON() ([]byte, error) {
return nil, errors.New("beep */ boop </script blip <!--")
}
func TestJSValEscaper(t *testing.T) {
tests := []struct {
x any
js string
skipNest bool
}{
{int(42), " 42 ", false},
{uint(42), " 42 ", false},
{int16(42), " 42 ", false},
{uint16(42), " 42 ", false},
{int32(-42), " -42 ", false},
{uint32(42), " 42 ", false},
{int16(-42), " -42 ", false},
{uint16(42), " 42 ", false},
{int64(-42), " -42 ", false},
{uint64(42), " 42 ", false},
{uint64(1) << 53, " 9007199254740992 ", false},
// ulp(1 << 53) > 1 so this loses precision in JS
// but it is still a representable integer literal.
{uint64(1)<<53 + 1, " 9007199254740993 ", false},
{float32(1.0), " 1 ", false},
{float32(-1.0), " -1 ", false},
{float32(0.5), " 0.5 ", false},
{float32(-0.5), " -0.5 ", false},
{float32(1.0) / float32(256), " 0.00390625 ", false},
{float32(0), " 0 ", false},
{math.Copysign(0, -1), " -0 ", false},
{float64(1.0), " 1 ", false},
{float64(-1.0), " -1 ", false},
{float64(0.5), " 0.5 ", false},
{float64(-0.5), " -0.5 ", false},
{float64(0), " 0 ", false},
{math.Copysign(0, -1), " -0 ", false},
{"", `""`, false},
{"foo", `"foo"`, false},
// Newlines.
{"\r\n\u2028\u2029", `"\r\n\u2028\u2029"`, false},
// "\v" == "v" on IE 6 so use "\u000b" instead.
{"\t\x0b", `"\t\u000b"`, false},
{struct{ X, Y int }{1, 2}, `{"X":1,"Y":2}`, false},
{[]any{}, "[]", false},
{[]any{42, "foo", nil}, `[42,"foo",null]`, false},
{[]string{"<!--", "</script>", "-->"}, `["\u003c!--","\u003c/script\u003e","--\u003e"]`, false},
{"<!--", `"\u003c!--"`, false},
{"-->", `"--\u003e"`, false},
{"<![CDATA[", `"\u003c![CDATA["`, false},
{"]]>", `"]]\u003e"`, false},
{"</script", `"\u003c/script"`, false},
{"\U0001D11E", "\"\U0001D11E\"", false}, // or "\uD834\uDD1E"
{nil, " null ", false},
{&jsonErrType{}, " /* json: error calling MarshalJSON for type *template.jsonErrType: beep * / boop \\x3C/script blip \\x3C!-- */null ", true},
}
for _, test := range tests {
if js := jsValEscaper(test.x); js != test.js {
t.Errorf("%+v: want\n\t%q\ngot\n\t%q", test.x, test.js, js)
}
if test.skipNest {
continue
}
// Make sure that escaping corner cases are not broken
// by nesting.
a := []any{test.x}
want := "[" + strings.TrimSpace(test.js) + "]"
if js := jsValEscaper(a); js != want {
t.Errorf("%+v: want\n\t%q\ngot\n\t%q", a, want, js)
}
}
}
func TestJSStrEscaper(t *testing.T) {
tests := []struct {
x any
esc string
}{
{"", ``},
{"foo", `foo`},
{"\u0000", `\u0000`},
{"\t", `\t`},
{"\n", `\n`},
{"\r", `\r`},
{"\u2028", `\u2028`},
{"\u2029", `\u2029`},
{"\\", `\\`},
{"\\n", `\\n`},
{"foo\r\nbar", `foo\r\nbar`},
// Preserve attribute boundaries.
{`"`, `\u0022`},
{`'`, `\u0027`},
// Allow embedding in HTML without further escaping.
{`&amp;`, `\u0026amp;`},
// Prevent breaking out of text node and element boundaries.
{"</script>", `\u003c\/script\u003e`},
{"<![CDATA[", `\u003c![CDATA[`},
{"]]>", `]]\u003e`},
// https://dev.w3.org/html5/markup/aria/syntax.html#escaping-text-span
// "The text in style, script, title, and textarea elements
// must not have an escaping text span start that is not
// followed by an escaping text span end."
// Furthermore, spoofing an escaping text span end could lead
// to different interpretation of a </script> sequence otherwise
// masked by the escaping text span, and spoofing a start could
// allow regular text content to be interpreted as script
// allowing script execution via a combination of a JS string
// injection followed by an HTML text injection.
{"<!--", `\u003c!--`},
{"-->", `--\u003e`},
// From https://code.google.com/p/doctype/wiki/ArticleUtf7
{"+ADw-script+AD4-alert(1)+ADw-/script+AD4-",
`\u002bADw-script\u002bAD4-alert(1)\u002bADw-\/script\u002bAD4-`,
},
// Invalid UTF-8 sequence
{"foo\xA0bar", "foo\xA0bar"},
// Invalid unicode scalar value.
{"foo\xed\xa0\x80bar", "foo\xed\xa0\x80bar"},
}
for _, test := range tests {
esc := jsStrEscaper(test.x)
if esc != test.esc {
t.Errorf("%q: want %q got %q", test.x, test.esc, esc)
}
}
}
func TestJSRegexpEscaper(t *testing.T) {
tests := []struct {
x any
esc string
}{
{"", `(?:)`},
{"foo", `foo`},
{"\u0000", `\u0000`},
{"\t", `\t`},
{"\n", `\n`},
{"\r", `\r`},
{"\u2028", `\u2028`},
{"\u2029", `\u2029`},
{"\\", `\\`},
{"\\n", `\\n`},
{"foo\r\nbar", `foo\r\nbar`},
// Preserve attribute boundaries.
{`"`, `\u0022`},
{`'`, `\u0027`},
// Allow embedding in HTML without further escaping.
{`&amp;`, `\u0026amp;`},
// Prevent breaking out of text node and element boundaries.
{"</script>", `\u003c\/script\u003e`},
{"<![CDATA[", `\u003c!\[CDATA\[`},
{"]]>", `\]\]\u003e`},
// Escaping text spans.
{"<!--", `\u003c!\-\-`},
{"-->", `\-\-\u003e`},
{"*", `\*`},
{"+", `\u002b`},
{"?", `\?`},
{"[](){}", `\[\]\(\)\{\}`},
{"$foo|x.y", `\$foo\|x\.y`},
{"x^y", `x\^y`},
}
for _, test := range tests {
esc := jsRegexpEscaper(test.x)
if esc != test.esc {
t.Errorf("%q: want %q got %q", test.x, test.esc, esc)
}
}
}
func TestEscapersOnLower7AndSelectHighCodepoints(t *testing.T) {
input := ("\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f" +
"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !"#$%&'()*+,-./` +
`0123456789:;<=>?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\]^_` +
"`abcdefghijklmno" +
"pqrstuvwxyz{|}~\x7f" +
"\u00A0\u0100\u2028\u2029\ufeff\U0001D11E")
tests := []struct {
name string
escaper func(...any) string
escaped string
}{
{
"jsStrEscaper",
jsStrEscaper,
`\u0000\u0001\u0002\u0003\u0004\u0005\u0006\u0007` +
`\u0008\t\n\u000b\f\r\u000e\u000f` +
`\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017` +
`\u0018\u0019\u001a\u001b\u001c\u001d\u001e\u001f` +
` !\u0022#$%\u0026\u0027()*\u002b,-.\/` +
`0123456789:;\u003c=\u003e?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\\]^_` +
"\\u0060abcdefghijklmno" +
"pqrstuvwxyz{|}~\u007f" +
"\u00A0\u0100\\u2028\\u2029\ufeff\U0001D11E",
},
{
"jsRegexpEscaper",
jsRegexpEscaper,
`\u0000\u0001\u0002\u0003\u0004\u0005\u0006\u0007` +
`\u0008\t\n\u000b\f\r\u000e\u000f` +
`\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017` +
`\u0018\u0019\u001a\u001b\u001c\u001d\u001e\u001f` +
` !\u0022#\$%\u0026\u0027\(\)\*\u002b,\-\.\/` +
`0123456789:;\u003c=\u003e\?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ\[\\\]\^_` +
"`abcdefghijklmno" +
`pqrstuvwxyz\{\|\}~` + "\u007f" +
"\u00A0\u0100\\u2028\\u2029\ufeff\U0001D11E",
},
}
for _, test := range tests {
if s := test.escaper(input); s != test.escaped {
t.Errorf("%s once: want\n\t%q\ngot\n\t%q", test.name, test.escaped, s)
continue
}
// Escape it rune by rune to make sure that any
// fast-path checking does not break escaping.
var buf strings.Builder
for _, c := range input {
buf.WriteString(test.escaper(string(c)))
}
if s := buf.String(); s != test.escaped {
t.Errorf("%s rune-wise: want\n\t%q\ngot\n\t%q", test.name, test.escaped, s)
continue
}
}
}
func TestIsJsMimeType(t *testing.T) {
tests := []struct {
in string
out bool
}{
{"application/javascript;version=1.8", true},
{"application/javascript;version=1.8;foo=bar", true},
{"application/javascript/version=1.8", false},
{"text/javascript", true},
{"application/json", true},
{"application/ld+json", true},
{"module", true},
}
for _, test := range tests {
if isJSType(test.in) != test.out {
t.Errorf("isJSType(%q) = %v, want %v", test.in, !test.out, test.out)
}
}
}
func BenchmarkJSValEscaperWithNum(b *testing.B) {
for i := 0; i < b.N; i++ {
jsValEscaper(3.141592654)
}
}
func BenchmarkJSValEscaperWithStr(b *testing.B) {
for i := 0; i < b.N; i++ {
jsValEscaper("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}
func BenchmarkJSValEscaperWithStrNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
jsValEscaper("The quick, brown fox jumps over the lazy dog")
}
}
func BenchmarkJSValEscaperWithObj(b *testing.B) {
o := struct {
S string
N int
}{
"The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>\u2028",
42,
}
for i := 0; i < b.N; i++ {
jsValEscaper(o)
}
}
func BenchmarkJSValEscaperWithObjNoSpecials(b *testing.B) {
o := struct {
S string
N int
}{
"The quick, brown fox jumps over the lazy dog",
42,
}
for i := 0; i < b.N; i++ {
jsValEscaper(o)
}
}
func BenchmarkJSStrEscaperNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
jsStrEscaper("The quick, brown fox jumps over the lazy dog.")
}
}
func BenchmarkJSStrEscaper(b *testing.B) {
for i := 0; i < b.N; i++ {
jsStrEscaper("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}
func BenchmarkJSRegexpEscaperNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
jsRegexpEscaper("The quick, brown fox jumps over the lazy dog")
}
}
func BenchmarkJSRegexpEscaper(b *testing.B) {
for i := 0; i < b.N; i++ {
jsRegexpEscaper("The <i>quick</i>,\r\n<span style='color:brown'>brown</span> fox jumps\u2028over the <canine class=\"lazy\">dog</canine>")
}
}

View File

@@ -0,0 +1,25 @@
// Code generated by "stringer -type jsCtx"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[jsCtxRegexp-0]
_ = x[jsCtxDivOp-1]
_ = x[jsCtxUnknown-2]
}
const _jsCtx_name = "jsCtxRegexpjsCtxDivOpjsCtxUnknown"
var _jsCtx_index = [...]uint8{0, 11, 21, 33}
func (i jsCtx) String() string {
if i >= jsCtx(len(_jsCtx_index)-1) {
return "jsCtx(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _jsCtx_name[_jsCtx_index[i]:_jsCtx_index[i+1]]
}

View File

@@ -0,0 +1,289 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Tests for multiple-template execution, copied from text/template.
package template
import (
"archive/zip"
"os"
"strings"
"testing"
"text/template/parse"
)
var multiExecTests = []execTest{
{"empty", "", "", nil, true},
{"text", "some text", "some text", nil, true},
{"invoke x", `{{template "x" .SI}}`, "TEXT", tVal, true},
{"invoke x no args", `{{template "x"}}`, "TEXT", tVal, true},
{"invoke dot int", `{{template "dot" .I}}`, "17", tVal, true},
{"invoke dot []int", `{{template "dot" .SI}}`, "[3 4 5]", tVal, true},
{"invoke dotV", `{{template "dotV" .U}}`, "v", tVal, true},
{"invoke nested int", `{{template "nested" .I}}`, "17", tVal, true},
{"variable declared by template", `{{template "nested" $x:=.SI}},{{index $x 1}}`, "[3 4 5],4", tVal, true},
// User-defined function: test argument evaluator.
{"testFunc literal", `{{oneArg "joe"}}`, "oneArg=joe", tVal, true},
{"testFunc .", `{{oneArg .}}`, "oneArg=joe", "joe", true},
}
// These strings are also in testdata/*.
const multiText1 = `
{{define "x"}}TEXT{{end}}
{{define "dotV"}}{{.V}}{{end}}
`
const multiText2 = `
{{define "dot"}}{{.}}{{end}}
{{define "nested"}}{{template "dot" .}}{{end}}
`
func TestMultiExecute(t *testing.T) {
// Declare a couple of templates first.
template, err := New("root").Parse(multiText1)
if err != nil {
t.Fatalf("parse error for 1: %s", err)
}
_, err = template.Parse(multiText2)
if err != nil {
t.Fatalf("parse error for 2: %s", err)
}
testExecute(multiExecTests, template, t)
}
func TestParseFiles(t *testing.T) {
_, err := ParseFiles("DOES NOT EXIST")
if err == nil {
t.Error("expected error for non-existent file; got none")
}
template := New("root")
_, err = template.ParseFiles("testdata/file1.tmpl", "testdata/file2.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(multiExecTests, template, t)
}
func TestParseGlob(t *testing.T) {
_, err := ParseGlob("DOES NOT EXIST")
if err == nil {
t.Error("expected error for non-existent file; got none")
}
_, err = New("error").ParseGlob("[x")
if err == nil {
t.Error("expected error for bad pattern; got none")
}
template := New("root")
_, err = template.ParseGlob("testdata/file*.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(multiExecTests, template, t)
}
func TestParseFS(t *testing.T) {
fs := os.DirFS("testdata")
{
_, err := ParseFS(fs, "DOES NOT EXIST")
if err == nil {
t.Error("expected error for non-existent file; got none")
}
}
{
template := New("root")
_, err := template.ParseFS(fs, "file1.tmpl", "file2.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(multiExecTests, template, t)
}
{
template := New("root")
_, err := template.ParseFS(fs, "file*.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(multiExecTests, template, t)
}
}
// In these tests, actual content (not just template definitions) comes from the parsed files.
var templateFileExecTests = []execTest{
{"test", `{{template "tmpl1.tmpl"}}{{template "tmpl2.tmpl"}}`, "template1\n\ny\ntemplate2\n\nx\n", 0, true},
}
func TestParseFilesWithData(t *testing.T) {
template, err := New("root").ParseFiles("testdata/tmpl1.tmpl", "testdata/tmpl2.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(templateFileExecTests, template, t)
}
func TestParseGlobWithData(t *testing.T) {
template, err := New("root").ParseGlob("testdata/tmpl*.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(templateFileExecTests, template, t)
}
func TestParseZipFS(t *testing.T) {
z, err := zip.OpenReader("testdata/fs.zip")
if err != nil {
t.Fatalf("error parsing zip: %v", err)
}
template, err := New("root").ParseFS(z, "tmpl*.tmpl")
if err != nil {
t.Fatalf("error parsing files: %v", err)
}
testExecute(templateFileExecTests, template, t)
}
const (
cloneText1 = `{{define "a"}}{{template "b"}}{{template "c"}}{{end}}`
cloneText2 = `{{define "b"}}b{{end}}`
cloneText3 = `{{define "c"}}root{{end}}`
cloneText4 = `{{define "c"}}clone{{end}}`
)
// Issue 7032
func TestAddParseTreeToUnparsedTemplate(t *testing.T) {
master := "{{define \"master\"}}{{end}}"
tmpl := New("master")
tree, err := parse.Parse("master", master, "", "", nil)
if err != nil {
t.Fatalf("unexpected parse err: %v", err)
}
masterTree := tree["master"]
tmpl.AddParseTree("master", masterTree) // used to panic
}
func TestRedefinition(t *testing.T) {
var tmpl *Template
var err error
if tmpl, err = New("tmpl1").Parse(`{{define "test"}}foo{{end}}`); err != nil {
t.Fatalf("parse 1: %v", err)
}
if _, err = tmpl.Parse(`{{define "test"}}bar{{end}}`); err != nil {
t.Fatalf("got error %v, expected nil", err)
}
if _, err = tmpl.New("tmpl2").Parse(`{{define "test"}}bar{{end}}`); err != nil {
t.Fatalf("got error %v, expected nil", err)
}
}
// Issue 10879
func TestEmptyTemplateCloneCrash(t *testing.T) {
t1 := New("base")
t1.Clone() // used to panic
}
// Issue 10910, 10926
func TestTemplateLookUp(t *testing.T) {
t.Skip("broken on html/template") // TODO
t1 := New("foo")
if t1.Lookup("foo") != nil {
t.Error("Lookup returned non-nil value for undefined template foo")
}
t1.New("bar")
if t1.Lookup("bar") != nil {
t.Error("Lookup returned non-nil value for undefined template bar")
}
t1.Parse(`{{define "foo"}}test{{end}}`)
if t1.Lookup("foo") == nil {
t.Error("Lookup returned nil value for defined template")
}
}
func TestParse(t *testing.T) {
// In multiple calls to Parse with the same receiver template, only one call
// can contain text other than space, comments, and template definitions
t1 := New("test")
if _, err := t1.Parse(`{{define "test"}}{{end}}`); err != nil {
t.Fatalf("parsing test: %s", err)
}
if _, err := t1.Parse(`{{define "test"}}{{/* this is a comment */}}{{end}}`); err != nil {
t.Fatalf("parsing test: %s", err)
}
if _, err := t1.Parse(`{{define "test"}}foo{{end}}`); err != nil {
t.Fatalf("parsing test: %s", err)
}
}
func TestEmptyTemplate(t *testing.T) {
cases := []struct {
defn []string
in string
want string
}{
{[]string{"x", "y"}, "", "y"},
{[]string{""}, "once", ""},
{[]string{"", ""}, "twice", ""},
{[]string{"{{.}}", "{{.}}"}, "twice", "twice"},
{[]string{"{{/* a comment */}}", "{{/* a comment */}}"}, "comment", ""},
{[]string{"{{.}}", ""}, "twice", "twice"}, // TODO: should want "" not "twice"
}
for i, c := range cases {
root := New("root")
var (
m *Template
err error
)
for _, d := range c.defn {
m, err = root.New(c.in).Parse(d)
if err != nil {
t.Fatal(err)
}
}
buf := &strings.Builder{}
if err := m.Execute(buf, c.in); err != nil {
t.Error(i, err)
continue
}
if buf.String() != c.want {
t.Errorf("expected string %q: got %q", c.want, buf.String())
}
}
}
// Issue 19249 was a regression in 1.8 caused by the handling of empty
// templates added in that release, which got different answers depending
// on the order templates appeared in the internal map.
func TestIssue19294(t *testing.T) {
// The empty block in "xhtml" should be replaced during execution
// by the contents of "stylesheet", but if the internal map associating
// names with templates is built in the wrong order, the empty block
// looks non-empty and this doesn't happen.
var inlined = map[string]string{
"stylesheet": `{{define "stylesheet"}}stylesheet{{end}}`,
"xhtml": `{{block "stylesheet" .}}{{end}}`,
}
all := []string{"stylesheet", "xhtml"}
for i := 0; i < 100; i++ {
res, err := New("title.xhtml").Parse(`{{template "xhtml" .}}`)
if err != nil {
t.Fatal(err)
}
for _, name := range all {
_, err := res.New(name).Parse(inlined[name])
if err != nil {
t.Fatal(err)
}
}
var buf strings.Builder
res.Execute(&buf, 0)
if buf.String() != "stylesheet" {
t.Fatalf("iteration %d: got %q; expected %q", i, buf.String(), "stylesheet")
}
}
}

View File

@@ -0,0 +1,51 @@
// Code generated by "stringer -type state"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[stateText-0]
_ = x[stateTag-1]
_ = x[stateAttrName-2]
_ = x[stateAfterName-3]
_ = x[stateBeforeValue-4]
_ = x[stateHTMLCmt-5]
_ = x[stateRCDATA-6]
_ = x[stateAttr-7]
_ = x[stateURL-8]
_ = x[stateSrcset-9]
_ = x[stateJS-10]
_ = x[stateJSDqStr-11]
_ = x[stateJSSqStr-12]
_ = x[stateJSTmplLit-13]
_ = x[stateJSRegexp-14]
_ = x[stateJSBlockCmt-15]
_ = x[stateJSLineCmt-16]
_ = x[stateJSHTMLOpenCmt-17]
_ = x[stateJSHTMLCloseCmt-18]
_ = x[stateCSS-19]
_ = x[stateCSSDqStr-20]
_ = x[stateCSSSqStr-21]
_ = x[stateCSSDqURL-22]
_ = x[stateCSSSqURL-23]
_ = x[stateCSSURL-24]
_ = x[stateCSSBlockCmt-25]
_ = x[stateCSSLineCmt-26]
_ = x[stateError-27]
_ = x[stateDead-28]
}
const _state_name = "stateTextstateTagstateAttrNamestateAfterNamestateBeforeValuestateHTMLCmtstateRCDATAstateAttrstateURLstateSrcsetstateJSstateJSDqStrstateJSSqStrstateJSTmplLitstateJSRegexpstateJSBlockCmtstateJSLineCmtstateJSHTMLOpenCmtstateJSHTMLCloseCmtstateCSSstateCSSDqStrstateCSSSqStrstateCSSDqURLstateCSSSqURLstateCSSURLstateCSSBlockCmtstateCSSLineCmtstateErrorstateDead"
var _state_index = [...]uint16{0, 9, 17, 30, 44, 60, 72, 83, 92, 100, 111, 118, 130, 142, 156, 169, 184, 198, 216, 235, 243, 256, 269, 282, 295, 306, 322, 337, 347, 356}
func (i state) String() string {
if i >= state(len(_state_index)-1) {
return "state(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _state_name[_state_index[i]:_state_index[i+1]]
}

View File

@@ -0,0 +1,530 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"fmt"
"io"
"io/fs"
"os"
"path"
"path/filepath"
"sync"
"text/template"
"text/template/parse"
)
// Template is a specialized Template from "text/template" that produces a safe
// HTML document fragment.
type Template struct {
// Sticky error if escaping fails, or escapeOK if succeeded.
escapeErr error
// We could embed the text/template field, but it's safer not to because
// we need to keep our version of the name space and the underlying
// template's in sync.
text *template.Template
// The underlying template's parse tree, updated to be HTML-safe.
Tree *parse.Tree
*nameSpace // common to all associated templates
}
// escapeOK is a sentinel value used to indicate valid escaping.
var escapeOK = fmt.Errorf("template escaped correctly")
// nameSpace is the data structure shared by all templates in an association.
type nameSpace struct {
mu sync.Mutex
set map[string]*Template
escaped bool
esc escaper
}
// Templates returns a slice of the templates associated with t, including t
// itself.
func (t *Template) Templates() []*Template {
ns := t.nameSpace
ns.mu.Lock()
defer ns.mu.Unlock()
// Return a slice so we don't expose the map.
m := make([]*Template, 0, len(ns.set))
for _, v := range ns.set {
m = append(m, v)
}
return m
}
// Option sets options for the template. Options are described by
// strings, either a simple string or "key=value". There can be at
// most one equals sign in an option string. If the option string
// is unrecognized or otherwise invalid, Option panics.
//
// Known options:
//
// missingkey: Control the behavior during execution if a map is
// indexed with a key that is not present in the map.
//
// "missingkey=default" or "missingkey=invalid"
// The default behavior: Do nothing and continue execution.
// If printed, the result of the index operation is the string
// "<no value>".
// "missingkey=zero"
// The operation returns the zero value for the map type's element.
// "missingkey=error"
// Execution stops immediately with an error.
func (t *Template) Option(opt ...string) *Template {
t.text.Option(opt...)
return t
}
// checkCanParse checks whether it is OK to parse templates.
// If not, it returns an error.
func (t *Template) checkCanParse() error {
if t == nil {
return nil
}
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
if t.nameSpace.escaped {
return fmt.Errorf("html/template: cannot Parse after Execute")
}
return nil
}
// escape escapes all associated templates.
func (t *Template) escape() error {
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
t.nameSpace.escaped = true
if t.escapeErr == nil {
if t.Tree == nil {
return fmt.Errorf("template: %q is an incomplete or empty template", t.Name())
}
if err := escapeTemplate(t, t.text.Root, t.Name()); err != nil {
return err
}
} else if t.escapeErr != escapeOK {
return t.escapeErr
}
return nil
}
// Execute applies a parsed template to the specified data object,
// writing the output to wr.
// If an error occurs executing the template or writing its output,
// execution stops, but partial results may already have been written to
// the output writer.
// A template may be executed safely in parallel, although if parallel
// executions share a Writer the output may be interleaved.
func (t *Template) Execute(wr io.Writer, data any) error {
if err := t.escape(); err != nil {
return err
}
return t.text.Execute(wr, data)
}
// ExecuteTemplate applies the template associated with t that has the given
// name to the specified data object and writes the output to wr.
// If an error occurs executing the template or writing its output,
// execution stops, but partial results may already have been written to
// the output writer.
// A template may be executed safely in parallel, although if parallel
// executions share a Writer the output may be interleaved.
func (t *Template) ExecuteTemplate(wr io.Writer, name string, data any) error {
tmpl, err := t.lookupAndEscapeTemplate(name)
if err != nil {
return err
}
return tmpl.text.Execute(wr, data)
}
// lookupAndEscapeTemplate guarantees that the template with the given name
// is escaped, or returns an error if it cannot be. It returns the named
// template.
func (t *Template) lookupAndEscapeTemplate(name string) (tmpl *Template, err error) {
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
t.nameSpace.escaped = true
tmpl = t.set[name]
if tmpl == nil {
return nil, fmt.Errorf("html/template: %q is undefined", name)
}
if tmpl.escapeErr != nil && tmpl.escapeErr != escapeOK {
return nil, tmpl.escapeErr
}
if tmpl.text.Tree == nil || tmpl.text.Root == nil {
return nil, fmt.Errorf("html/template: %q is an incomplete template", name)
}
if t.text.Lookup(name) == nil {
panic("html/template internal error: template escaping out of sync")
}
if tmpl.escapeErr == nil {
err = escapeTemplate(tmpl, tmpl.text.Root, name)
}
return tmpl, err
}
// DefinedTemplates returns a string listing the defined templates,
// prefixed by the string "; defined templates are: ". If there are none,
// it returns the empty string. Used to generate an error message.
func (t *Template) DefinedTemplates() string {
return t.text.DefinedTemplates()
}
// Parse parses text as a template body for t.
// Named template definitions ({{define ...}} or {{block ...}} statements) in text
// define additional templates associated with t and are removed from the
// definition of t itself.
//
// Templates can be redefined in successive calls to Parse,
// before the first use of [Template.Execute] on t or any associated template.
// A template definition with a body containing only white space and comments
// is considered empty and will not replace an existing template's body.
// This allows using Parse to add new named template definitions without
// overwriting the main template body.
func (t *Template) Parse(text string) (*Template, error) {
if err := t.checkCanParse(); err != nil {
return nil, err
}
ret, err := t.text.Parse(text)
if err != nil {
return nil, err
}
// In general, all the named templates might have changed underfoot.
// Regardless, some new ones may have been defined.
// The template.Template set has been updated; update ours.
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
for _, v := range ret.Templates() {
name := v.Name()
tmpl := t.set[name]
if tmpl == nil {
tmpl = t.new(name)
}
tmpl.text = v
tmpl.Tree = v.Tree
}
return t, nil
}
// AddParseTree creates a new template with the name and parse tree
// and associates it with t.
//
// It returns an error if t or any associated template has already been executed.
func (t *Template) AddParseTree(name string, tree *parse.Tree) (*Template, error) {
if err := t.checkCanParse(); err != nil {
return nil, err
}
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
text, err := t.text.AddParseTree(name, tree)
if err != nil {
return nil, err
}
ret := &Template{
nil,
text,
text.Tree,
t.nameSpace,
}
t.set[name] = ret
return ret, nil
}
// Clone returns a duplicate of the template, including all associated
// templates. The actual representation is not copied, but the name space of
// associated templates is, so further calls to [Template.Parse] in the copy will add
// templates to the copy but not to the original. [Template.Clone] can be used to prepare
// common templates and use them with variant definitions for other templates
// by adding the variants after the clone is made.
//
// It returns an error if t has already been executed.
func (t *Template) Clone() (*Template, error) {
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
if t.escapeErr != nil {
return nil, fmt.Errorf("html/template: cannot Clone %q after it has executed", t.Name())
}
textClone, err := t.text.Clone()
if err != nil {
return nil, err
}
ns := &nameSpace{set: make(map[string]*Template)}
ns.esc = makeEscaper(ns)
ret := &Template{
nil,
textClone,
textClone.Tree,
ns,
}
ret.set[ret.Name()] = ret
for _, x := range textClone.Templates() {
name := x.Name()
src := t.set[name]
if src == nil || src.escapeErr != nil {
return nil, fmt.Errorf("html/template: cannot Clone %q after it has executed", t.Name())
}
x.Tree = x.Tree.Copy()
ret.set[name] = &Template{
nil,
x,
x.Tree,
ret.nameSpace,
}
}
// Return the template associated with the name of this template.
return ret.set[ret.Name()], nil
}
// New allocates a new HTML template with the given name.
func New(name string) *Template {
ns := &nameSpace{set: make(map[string]*Template)}
ns.esc = makeEscaper(ns)
tmpl := &Template{
nil,
template.New(name),
nil,
ns,
}
tmpl.set[name] = tmpl
return tmpl
}
// New allocates a new HTML template associated with the given one
// and with the same delimiters. The association, which is transitive,
// allows one template to invoke another with a {{template}} action.
//
// If a template with the given name already exists, the new HTML template
// will replace it. The existing template will be reset and disassociated with
// t.
func (t *Template) New(name string) *Template {
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
return t.new(name)
}
// new is the implementation of New, without the lock.
func (t *Template) new(name string) *Template {
tmpl := &Template{
nil,
t.text.New(name),
nil,
t.nameSpace,
}
if existing, ok := tmpl.set[name]; ok {
emptyTmpl := New(existing.Name())
*existing = *emptyTmpl
}
tmpl.set[name] = tmpl
return tmpl
}
// Name returns the name of the template.
func (t *Template) Name() string {
return t.text.Name()
}
type FuncMap = template.FuncMap
// Funcs adds the elements of the argument map to the template's function map.
// It must be called before the template is parsed.
// It panics if a value in the map is not a function with appropriate return
// type. However, it is legal to overwrite elements of the map. The return
// value is the template, so calls can be chained.
func (t *Template) Funcs(funcMap FuncMap) *Template {
t.text.Funcs(template.FuncMap(funcMap))
return t
}
// Delims sets the action delimiters to the specified strings, to be used in
// subsequent calls to [Template.Parse], [ParseFiles], or [ParseGlob]. Nested template
// definitions will inherit the settings. An empty delimiter stands for the
// corresponding default: {{ or }}.
// The return value is the template, so calls can be chained.
func (t *Template) Delims(left, right string) *Template {
t.text.Delims(left, right)
return t
}
// Lookup returns the template with the given name that is associated with t,
// or nil if there is no such template.
func (t *Template) Lookup(name string) *Template {
t.nameSpace.mu.Lock()
defer t.nameSpace.mu.Unlock()
return t.set[name]
}
// Must is a helper that wraps a call to a function returning ([*Template], error)
// and panics if the error is non-nil. It is intended for use in variable initializations
// such as
//
// var t = template.Must(template.New("name").Parse("html"))
func Must(t *Template, err error) *Template {
if err != nil {
panic(err)
}
return t
}
// ParseFiles creates a new [Template] and parses the template definitions from
// the named files. The returned template's name will have the (base) name and
// (parsed) contents of the first file. There must be at least one file.
// If an error occurs, parsing stops and the returned [*Template] is nil.
//
// When parsing multiple files with the same name in different directories,
// the last one mentioned will be the one that results.
// For instance, ParseFiles("a/foo", "b/foo") stores "b/foo" as the template
// named "foo", while "a/foo" is unavailable.
func ParseFiles(filenames ...string) (*Template, error) {
return parseFiles(nil, readFileOS, filenames...)
}
// ParseFiles parses the named files and associates the resulting templates with
// t. If an error occurs, parsing stops and the returned template is nil;
// otherwise it is t. There must be at least one file.
//
// When parsing multiple files with the same name in different directories,
// the last one mentioned will be the one that results.
//
// ParseFiles returns an error if t or any associated template has already been executed.
func (t *Template) ParseFiles(filenames ...string) (*Template, error) {
return parseFiles(t, readFileOS, filenames...)
}
// parseFiles is the helper for the method and function. If the argument
// template is nil, it is created from the first file.
func parseFiles(t *Template, readFile func(string) (string, []byte, error), filenames ...string) (*Template, error) {
if err := t.checkCanParse(); err != nil {
return nil, err
}
if len(filenames) == 0 {
// Not really a problem, but be consistent.
return nil, fmt.Errorf("html/template: no files named in call to ParseFiles")
}
for _, filename := range filenames {
name, b, err := readFile(filename)
if err != nil {
return nil, err
}
s := string(b)
// First template becomes return value if not already defined,
// and we use that one for subsequent New calls to associate
// all the templates together. Also, if this file has the same name
// as t, this file becomes the contents of t, so
// t, err := New(name).Funcs(xxx).ParseFiles(name)
// works. Otherwise we create a new template associated with t.
var tmpl *Template
if t == nil {
t = New(name)
}
if name == t.Name() {
tmpl = t
} else {
tmpl = t.New(name)
}
_, err = tmpl.Parse(s)
if err != nil {
return nil, err
}
}
return t, nil
}
// ParseGlob creates a new [Template] and parses the template definitions from
// the files identified by the pattern. The files are matched according to the
// semantics of filepath.Match, and the pattern must match at least one file.
// The returned template will have the (base) name and (parsed) contents of the
// first file matched by the pattern. ParseGlob is equivalent to calling
// [ParseFiles] with the list of files matched by the pattern.
//
// When parsing multiple files with the same name in different directories,
// the last one mentioned will be the one that results.
func ParseGlob(pattern string) (*Template, error) {
return parseGlob(nil, pattern)
}
// ParseGlob parses the template definitions in the files identified by the
// pattern and associates the resulting templates with t. The files are matched
// according to the semantics of filepath.Match, and the pattern must match at
// least one file. ParseGlob is equivalent to calling t.ParseFiles with the
// list of files matched by the pattern.
//
// When parsing multiple files with the same name in different directories,
// the last one mentioned will be the one that results.
//
// ParseGlob returns an error if t or any associated template has already been executed.
func (t *Template) ParseGlob(pattern string) (*Template, error) {
return parseGlob(t, pattern)
}
// parseGlob is the implementation of the function and method ParseGlob.
func parseGlob(t *Template, pattern string) (*Template, error) {
if err := t.checkCanParse(); err != nil {
return nil, err
}
filenames, err := filepath.Glob(pattern)
if err != nil {
return nil, err
}
if len(filenames) == 0 {
return nil, fmt.Errorf("html/template: pattern matches no files: %#q", pattern)
}
return parseFiles(t, readFileOS, filenames...)
}
// IsTrue reports whether the value is 'true', in the sense of not the zero of its type,
// and whether the value has a meaningful truth value. This is the definition of
// truth used by if and other such actions.
func IsTrue(val any) (truth, ok bool) {
return template.IsTrue(val)
}
// ParseFS is like [ParseFiles] or [ParseGlob] but reads from the file system fs
// instead of the host operating system's file system.
// It accepts a list of glob patterns.
// (Note that most file names serve as glob patterns matching only themselves.)
func ParseFS(fs fs.FS, patterns ...string) (*Template, error) {
return parseFS(nil, fs, patterns)
}
// ParseFS is like [Template.ParseFiles] or [Template.ParseGlob] but reads from the file system fs
// instead of the host operating system's file system.
// It accepts a list of glob patterns.
// (Note that most file names serve as glob patterns matching only themselves.)
func (t *Template) ParseFS(fs fs.FS, patterns ...string) (*Template, error) {
return parseFS(t, fs, patterns)
}
func parseFS(t *Template, fsys fs.FS, patterns []string) (*Template, error) {
var filenames []string
for _, pattern := range patterns {
list, err := fs.Glob(fsys, pattern)
if err != nil {
return nil, err
}
if len(list) == 0 {
return nil, fmt.Errorf("template: pattern matches no files: %#q", pattern)
}
filenames = append(filenames, list...)
}
return parseFiles(t, readFileFS(fsys), filenames...)
}
func readFileOS(file string) (name string, b []byte, err error) {
name = filepath.Base(file)
b, err = os.ReadFile(file)
return
}
func readFileFS(fsys fs.FS) func(string) (string, []byte, error) {
return func(file string) (name string, b []byte, err error) {
name = path.Base(file)
b, err = fs.ReadFile(fsys, file)
return
}
}

View File

@@ -0,0 +1,218 @@
// Copyright 2016 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template_test
import (
"bytes"
"encoding/json"
. "html/template"
"strings"
"testing"
"text/template/parse"
)
func TestTemplateClone(t *testing.T) {
// https://golang.org/issue/12996
orig := New("name")
clone, err := orig.Clone()
if err != nil {
t.Fatal(err)
}
if len(clone.Templates()) != len(orig.Templates()) {
t.Fatalf("Invalid length of t.Clone().Templates()")
}
const want = "stuff"
parsed := Must(clone.Parse(want))
var buf strings.Builder
err = parsed.Execute(&buf, nil)
if err != nil {
t.Fatal(err)
}
if got := buf.String(); got != want {
t.Fatalf("got %q; want %q", got, want)
}
}
func TestRedefineNonEmptyAfterExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `foo`)
c.mustExecute(c.root, nil, "foo")
c.mustNotParse(c.root, `bar`)
}
func TestRedefineEmptyAfterExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, ``)
c.mustExecute(c.root, nil, "")
c.mustNotParse(c.root, `foo`)
c.mustExecute(c.root, nil, "")
}
func TestRedefineAfterNonExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `{{if .}}<{{template "X"}}>{{end}}{{define "X"}}foo{{end}}`)
c.mustExecute(c.root, 0, "")
c.mustNotParse(c.root, `{{define "X"}}bar{{end}}`)
c.mustExecute(c.root, 1, "&lt;foo>")
}
func TestRedefineAfterNamedExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `<{{template "X" .}}>{{define "X"}}foo{{end}}`)
c.mustExecute(c.root, nil, "&lt;foo>")
c.mustNotParse(c.root, `{{define "X"}}bar{{end}}`)
c.mustExecute(c.root, nil, "&lt;foo>")
}
func TestRedefineNestedByNameAfterExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `{{define "X"}}foo{{end}}`)
c.mustExecute(c.lookup("X"), nil, "foo")
c.mustNotParse(c.root, `{{define "X"}}bar{{end}}`)
c.mustExecute(c.lookup("X"), nil, "foo")
}
func TestRedefineNestedByTemplateAfterExecution(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `{{define "X"}}foo{{end}}`)
c.mustExecute(c.lookup("X"), nil, "foo")
c.mustNotParse(c.lookup("X"), `bar`)
c.mustExecute(c.lookup("X"), nil, "foo")
}
func TestRedefineSafety(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `<html><a href="{{template "X"}}">{{define "X"}}{{end}}`)
c.mustExecute(c.root, nil, `<html><a href="">`)
// Note: Every version of Go prior to Go 1.8 accepted the redefinition of "X"
// on the next line, but luckily kept it from being used in the outer template.
// Now we reject it, which makes clearer that we're not going to use it.
c.mustNotParse(c.root, `{{define "X"}}" bar="baz{{end}}`)
c.mustExecute(c.root, nil, `<html><a href="">`)
}
func TestRedefineTopUse(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `{{template "X"}}{{.}}{{define "X"}}{{end}}`)
c.mustExecute(c.root, 42, `42`)
c.mustNotParse(c.root, `{{define "X"}}<script>{{end}}`)
c.mustExecute(c.root, 42, `42`)
}
func TestRedefineOtherParsers(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, ``)
c.mustExecute(c.root, nil, ``)
if _, err := c.root.ParseFiles("no.template"); err == nil || !strings.Contains(err.Error(), "Execute") {
t.Errorf("ParseFiles: %v\nwanted error about already having Executed", err)
}
if _, err := c.root.ParseGlob("*.no.template"); err == nil || !strings.Contains(err.Error(), "Execute") {
t.Errorf("ParseGlob: %v\nwanted error about already having Executed", err)
}
if _, err := c.root.AddParseTree("t1", c.root.Tree); err == nil || !strings.Contains(err.Error(), "Execute") {
t.Errorf("AddParseTree: %v\nwanted error about already having Executed", err)
}
}
func TestNumbers(t *testing.T) {
c := newTestCase(t)
c.mustParse(c.root, `{{print 1_2.3_4}} {{print 0x0_1.e_0p+02}}`)
c.mustExecute(c.root, nil, "12.34 7.5")
}
func TestStringsInScriptsWithJsonContentTypeAreCorrectlyEscaped(t *testing.T) {
// See #33671 and #37634 for more context on this.
tests := []struct{ name, in string }{
{"empty", ""},
{"invalid", string(rune(-1))},
{"null", "\u0000"},
{"unit separator", "\u001F"},
{"tab", "\t"},
{"gt and lt", "<>"},
{"quotes", `'"`},
{"ASCII letters", "ASCII letters"},
{"Unicode", "ʕ⊙ϖ⊙ʔ"},
{"Pizza", "🍕"},
}
const (
prefix = `<script type="application/ld+json">`
suffix = `</script>`
templ = prefix + `"{{.}}"` + suffix
)
tpl := Must(New("JS string is JSON string").Parse(templ))
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
var buf bytes.Buffer
if err := tpl.Execute(&buf, tt.in); err != nil {
t.Fatalf("Cannot render template: %v", err)
}
trimmed := bytes.TrimSuffix(bytes.TrimPrefix(buf.Bytes(), []byte(prefix)), []byte(suffix))
var got string
if err := json.Unmarshal(trimmed, &got); err != nil {
t.Fatalf("Cannot parse JS string %q as JSON: %v", trimmed[1:len(trimmed)-1], err)
}
if got != tt.in {
t.Errorf("Serialization changed the string value: got %q want %q", got, tt.in)
}
})
}
}
func TestSkipEscapeComments(t *testing.T) {
c := newTestCase(t)
tr := parse.New("root")
tr.Mode = parse.ParseComments
newT, err := tr.Parse("{{/* A comment */}}{{ 1 }}{{/* Another comment */}}", "", "", make(map[string]*parse.Tree))
if err != nil {
t.Fatalf("Cannot parse template text: %v", err)
}
c.root, err = c.root.AddParseTree("root", newT)
if err != nil {
t.Fatalf("Cannot add parse tree to template: %v", err)
}
c.mustExecute(c.root, nil, "1")
}
type testCase struct {
t *testing.T
root *Template
}
func newTestCase(t *testing.T) *testCase {
return &testCase{
t: t,
root: New("root"),
}
}
func (c *testCase) lookup(name string) *Template {
return c.root.Lookup(name)
}
func (c *testCase) mustParse(t *Template, text string) {
_, err := t.Parse(text)
if err != nil {
c.t.Fatalf("parse: %v", err)
}
}
func (c *testCase) mustNotParse(t *Template, text string) {
_, err := t.Parse(text)
if err == nil {
c.t.Fatalf("parse: unexpected success")
}
}
func (c *testCase) mustExecute(t *Template, val any, want string) {
var buf strings.Builder
err := t.Execute(&buf, val)
if err != nil {
c.t.Fatalf("execute: %v", err)
}
if buf.String() != want {
c.t.Fatalf("template output:\n%s\nwant:\n%s", buf.String(), want)
}
}

2
src/html/template/testdata/file1.tmpl vendored Normal file
View File

@@ -0,0 +1,2 @@
{{define "x"}}TEXT{{end}}
{{define "dotV"}}{{.V}}{{end}}

2
src/html/template/testdata/file2.tmpl vendored Normal file
View File

@@ -0,0 +1,2 @@
{{define "dot"}}{{.}}{{end}}
{{define "nested"}}{{template "dot" .}}{{end}}

BIN
src/html/template/testdata/fs.zip vendored Normal file

Binary file not shown.

3
src/html/template/testdata/tmpl1.tmpl vendored Normal file
View File

@@ -0,0 +1,3 @@
template1
{{define "x"}}x{{end}}
{{template "y"}}

3
src/html/template/testdata/tmpl2.tmpl vendored Normal file
View File

@@ -0,0 +1,3 @@
template2
{{define "y"}}y{{end}}
{{template "x"}}

View File

@@ -0,0 +1,686 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"strings"
)
// transitionFunc is the array of context transition functions for text nodes.
// A transition function takes a context and template text input, and returns
// the updated context and the number of bytes consumed from the front of the
// input.
var transitionFunc = [...]func(context, []byte) (context, int){
stateText: tText,
stateTag: tTag,
stateAttrName: tAttrName,
stateAfterName: tAfterName,
stateBeforeValue: tBeforeValue,
stateHTMLCmt: tHTMLCmt,
stateRCDATA: tSpecialTagEnd,
stateAttr: tAttr,
stateURL: tURL,
stateSrcset: tURL,
stateJS: tJS,
stateJSDqStr: tJSDelimited,
stateJSSqStr: tJSDelimited,
stateJSRegexp: tJSDelimited,
stateJSTmplLit: tJSTmpl,
stateJSBlockCmt: tBlockCmt,
stateJSLineCmt: tLineCmt,
stateJSHTMLOpenCmt: tLineCmt,
stateJSHTMLCloseCmt: tLineCmt,
stateCSS: tCSS,
stateCSSDqStr: tCSSStr,
stateCSSSqStr: tCSSStr,
stateCSSDqURL: tCSSStr,
stateCSSSqURL: tCSSStr,
stateCSSURL: tCSSStr,
stateCSSBlockCmt: tBlockCmt,
stateCSSLineCmt: tLineCmt,
stateError: tError,
}
var commentStart = []byte("<!--")
var commentEnd = []byte("-->")
// tText is the context transition function for the text state.
func tText(c context, s []byte) (context, int) {
k := 0
for {
i := k + bytes.IndexByte(s[k:], '<')
if i < k || i+1 == len(s) {
return c, len(s)
} else if i+4 <= len(s) && bytes.Equal(commentStart, s[i:i+4]) {
return context{state: stateHTMLCmt}, i + 4
}
i++
end := false
if s[i] == '/' {
if i+1 == len(s) {
return c, len(s)
}
end, i = true, i+1
}
j, e := eatTagName(s, i)
if j != i {
if end {
e = elementNone
}
// We've found an HTML tag.
return context{state: stateTag, element: e}, j
}
k = j
}
}
var elementContentType = [...]state{
elementNone: stateText,
elementScript: stateJS,
elementStyle: stateCSS,
elementTextarea: stateRCDATA,
elementTitle: stateRCDATA,
}
// tTag is the context transition function for the tag state.
func tTag(c context, s []byte) (context, int) {
// Find the attribute name.
i := eatWhiteSpace(s, 0)
if i == len(s) {
return c, len(s)
}
if s[i] == '>' {
return context{
state: elementContentType[c.element],
element: c.element,
}, i + 1
}
j, err := eatAttrName(s, i)
if err != nil {
return context{state: stateError, err: err}, len(s)
}
state, attr := stateTag, attrNone
if i == j {
return context{
state: stateError,
err: errorf(ErrBadHTML, nil, 0, "expected space, attr name, or end of tag, but got %q", s[i:]),
}, len(s)
}
attrName := strings.ToLower(string(s[i:j]))
if c.element == elementScript && attrName == "type" {
attr = attrScriptType
} else {
switch attrType(attrName) {
case contentTypeURL:
attr = attrURL
case contentTypeCSS:
attr = attrStyle
case contentTypeJS:
attr = attrScript
case contentTypeSrcset:
attr = attrSrcset
}
}
if j == len(s) {
state = stateAttrName
} else {
state = stateAfterName
}
return context{state: state, element: c.element, attr: attr}, j
}
// tAttrName is the context transition function for stateAttrName.
func tAttrName(c context, s []byte) (context, int) {
i, err := eatAttrName(s, 0)
if err != nil {
return context{state: stateError, err: err}, len(s)
} else if i != len(s) {
c.state = stateAfterName
}
return c, i
}
// tAfterName is the context transition function for stateAfterName.
func tAfterName(c context, s []byte) (context, int) {
// Look for the start of the value.
i := eatWhiteSpace(s, 0)
if i == len(s) {
return c, len(s)
} else if s[i] != '=' {
// Occurs due to tag ending '>', and valueless attribute.
c.state = stateTag
return c, i
}
c.state = stateBeforeValue
// Consume the "=".
return c, i + 1
}
var attrStartStates = [...]state{
attrNone: stateAttr,
attrScript: stateJS,
attrScriptType: stateAttr,
attrStyle: stateCSS,
attrURL: stateURL,
attrSrcset: stateSrcset,
}
// tBeforeValue is the context transition function for stateBeforeValue.
func tBeforeValue(c context, s []byte) (context, int) {
i := eatWhiteSpace(s, 0)
if i == len(s) {
return c, len(s)
}
// Find the attribute delimiter.
delim := delimSpaceOrTagEnd
switch s[i] {
case '\'':
delim, i = delimSingleQuote, i+1
case '"':
delim, i = delimDoubleQuote, i+1
}
c.state, c.delim = attrStartStates[c.attr], delim
return c, i
}
// tHTMLCmt is the context transition function for stateHTMLCmt.
func tHTMLCmt(c context, s []byte) (context, int) {
if i := bytes.Index(s, commentEnd); i != -1 {
return context{}, i + 3
}
return c, len(s)
}
// specialTagEndMarkers maps element types to the character sequence that
// case-insensitively signals the end of the special tag body.
var specialTagEndMarkers = [...][]byte{
elementScript: []byte("script"),
elementStyle: []byte("style"),
elementTextarea: []byte("textarea"),
elementTitle: []byte("title"),
}
var (
specialTagEndPrefix = []byte("</")
tagEndSeparators = []byte("> \t\n\f/")
)
// tSpecialTagEnd is the context transition function for raw text and RCDATA
// element states.
func tSpecialTagEnd(c context, s []byte) (context, int) {
if c.element != elementNone {
// script end tags ("</script") within script literals are ignored, so that
// we can properly escape them.
if c.element == elementScript && (isInScriptLiteral(c.state) || isComment(c.state)) {
return c, len(s)
}
if i := indexTagEnd(s, specialTagEndMarkers[c.element]); i != -1 {
return context{}, i
}
}
return c, len(s)
}
// indexTagEnd finds the index of a special tag end in a case insensitive way, or returns -1
func indexTagEnd(s []byte, tag []byte) int {
res := 0
plen := len(specialTagEndPrefix)
for len(s) > 0 {
// Try to find the tag end prefix first
i := bytes.Index(s, specialTagEndPrefix)
if i == -1 {
return i
}
s = s[i+plen:]
// Try to match the actual tag if there is still space for it
if len(tag) <= len(s) && bytes.EqualFold(tag, s[:len(tag)]) {
s = s[len(tag):]
// Check the tag is followed by a proper separator
if len(s) > 0 && bytes.IndexByte(tagEndSeparators, s[0]) != -1 {
return res + i
}
res += len(tag)
}
res += i + plen
}
return -1
}
// tAttr is the context transition function for the attribute state.
func tAttr(c context, s []byte) (context, int) {
return c, len(s)
}
// tURL is the context transition function for the URL state.
func tURL(c context, s []byte) (context, int) {
if bytes.ContainsAny(s, "#?") {
c.urlPart = urlPartQueryOrFrag
} else if len(s) != eatWhiteSpace(s, 0) && c.urlPart == urlPartNone {
// HTML5 uses "Valid URL potentially surrounded by spaces" for
// attrs: https://www.w3.org/TR/html5/index.html#attributes-1
c.urlPart = urlPartPreQuery
}
return c, len(s)
}
// tJS is the context transition function for the JS state.
func tJS(c context, s []byte) (context, int) {
i := bytes.IndexAny(s, "\"`'/{}<-#")
if i == -1 {
// Entire input is non string, comment, regexp tokens.
c.jsCtx = nextJSCtx(s, c.jsCtx)
return c, len(s)
}
c.jsCtx = nextJSCtx(s[:i], c.jsCtx)
switch s[i] {
case '"':
c.state, c.jsCtx = stateJSDqStr, jsCtxRegexp
case '\'':
c.state, c.jsCtx = stateJSSqStr, jsCtxRegexp
case '`':
c.state, c.jsCtx = stateJSTmplLit, jsCtxRegexp
case '/':
switch {
case i+1 < len(s) && s[i+1] == '/':
c.state, i = stateJSLineCmt, i+1
case i+1 < len(s) && s[i+1] == '*':
c.state, i = stateJSBlockCmt, i+1
case c.jsCtx == jsCtxRegexp:
c.state = stateJSRegexp
case c.jsCtx == jsCtxDivOp:
c.jsCtx = jsCtxRegexp
default:
return context{
state: stateError,
err: errorf(ErrSlashAmbig, nil, 0, "'/' could start a division or regexp: %.32q", s[i:]),
}, len(s)
}
// ECMAScript supports HTML style comments for legacy reasons, see Appendix
// B.1.1 "HTML-like Comments". The handling of these comments is somewhat
// confusing. Multi-line comments are not supported, i.e. anything on lines
// between the opening and closing tokens is not considered a comment, but
// anything following the opening or closing token, on the same line, is
// ignored. As such we simply treat any line prefixed with "<!--" or "-->"
// as if it were actually prefixed with "//" and move on.
case '<':
if i+3 < len(s) && bytes.Equal(commentStart, s[i:i+4]) {
c.state, i = stateJSHTMLOpenCmt, i+3
}
case '-':
if i+2 < len(s) && bytes.Equal(commentEnd, s[i:i+3]) {
c.state, i = stateJSHTMLCloseCmt, i+2
}
// ECMAScript also supports "hashbang" comment lines, see Section 12.5.
case '#':
if i+1 < len(s) && s[i+1] == '!' {
c.state, i = stateJSLineCmt, i+1
}
case '{':
// We only care about tracking brace depth if we are inside of a
// template literal.
if len(c.jsBraceDepth) == 0 {
return c, i + 1
}
c.jsBraceDepth[len(c.jsBraceDepth)-1]++
case '}':
if len(c.jsBraceDepth) == 0 {
return c, i + 1
}
// There are no cases where a brace can be escaped in the JS context
// that are not syntax errors, it seems. Because of this we can just
// count "\}" as "}" and move on, the script is already broken as
// fully fledged parsers will just fail anyway.
c.jsBraceDepth[len(c.jsBraceDepth)-1]--
if c.jsBraceDepth[len(c.jsBraceDepth)-1] >= 0 {
return c, i + 1
}
c.jsBraceDepth = c.jsBraceDepth[:len(c.jsBraceDepth)-1]
c.state = stateJSTmplLit
default:
panic("unreachable")
}
return c, i + 1
}
func tJSTmpl(c context, s []byte) (context, int) {
var k int
for {
i := k + bytes.IndexAny(s[k:], "`\\$")
if i < k {
break
}
switch s[i] {
case '\\':
i++
if i == len(s) {
return context{
state: stateError,
err: errorf(ErrPartialEscape, nil, 0, "unfinished escape sequence in JS string: %q", s),
}, len(s)
}
case '$':
if len(s) >= i+2 && s[i+1] == '{' {
c.jsBraceDepth = append(c.jsBraceDepth, 0)
c.state = stateJS
return c, i + 2
}
case '`':
// end
c.state = stateJS
return c, i + 1
}
k = i + 1
}
return c, len(s)
}
// tJSDelimited is the context transition function for the JS string and regexp
// states.
func tJSDelimited(c context, s []byte) (context, int) {
specials := `\"`
switch c.state {
case stateJSSqStr:
specials = `\'`
case stateJSRegexp:
specials = `\/[]`
}
k, inCharset := 0, false
for {
i := k + bytes.IndexAny(s[k:], specials)
if i < k {
break
}
switch s[i] {
case '\\':
i++
if i == len(s) {
return context{
state: stateError,
err: errorf(ErrPartialEscape, nil, 0, "unfinished escape sequence in JS string: %q", s),
}, len(s)
}
case '[':
inCharset = true
case ']':
inCharset = false
case '/':
// If "</script" appears in a regex literal, the '/' should not
// close the regex literal, and it will later be escaped to
// "\x3C/script" in escapeText.
if i > 0 && i+7 <= len(s) && bytes.Equal(bytes.ToLower(s[i-1:i+7]), []byte("</script")) {
i++
} else if !inCharset {
c.state, c.jsCtx = stateJS, jsCtxDivOp
return c, i + 1
}
default:
// end delimiter
if !inCharset {
c.state, c.jsCtx = stateJS, jsCtxDivOp
return c, i + 1
}
}
k = i + 1
}
if inCharset {
// This can be fixed by making context richer if interpolation
// into charsets is desired.
return context{
state: stateError,
err: errorf(ErrPartialCharset, nil, 0, "unfinished JS regexp charset: %q", s),
}, len(s)
}
return c, len(s)
}
var blockCommentEnd = []byte("*/")
// tBlockCmt is the context transition function for /*comment*/ states.
func tBlockCmt(c context, s []byte) (context, int) {
i := bytes.Index(s, blockCommentEnd)
if i == -1 {
return c, len(s)
}
switch c.state {
case stateJSBlockCmt:
c.state = stateJS
case stateCSSBlockCmt:
c.state = stateCSS
default:
panic(c.state.String())
}
return c, i + 2
}
// tLineCmt is the context transition function for //comment states, and the JS HTML-like comment state.
func tLineCmt(c context, s []byte) (context, int) {
var lineTerminators string
var endState state
switch c.state {
case stateJSLineCmt, stateJSHTMLOpenCmt, stateJSHTMLCloseCmt:
lineTerminators, endState = "\n\r\u2028\u2029", stateJS
case stateCSSLineCmt:
lineTerminators, endState = "\n\f\r", stateCSS
// Line comments are not part of any published CSS standard but
// are supported by the 4 major browsers.
// This defines line comments as
// LINECOMMENT ::= "//" [^\n\f\d]*
// since https://www.w3.org/TR/css3-syntax/#SUBTOK-nl defines
// newlines:
// nl ::= #xA | #xD #xA | #xD | #xC
default:
panic(c.state.String())
}
i := bytes.IndexAny(s, lineTerminators)
if i == -1 {
return c, len(s)
}
c.state = endState
// Per section 7.4 of EcmaScript 5 : https://es5.github.io/#x7.4
// "However, the LineTerminator at the end of the line is not
// considered to be part of the single-line comment; it is
// recognized separately by the lexical grammar and becomes part
// of the stream of input elements for the syntactic grammar."
return c, i
}
// tCSS is the context transition function for the CSS state.
func tCSS(c context, s []byte) (context, int) {
// CSS quoted strings are almost never used except for:
// (1) URLs as in background: "/foo.png"
// (2) Multiword font-names as in font-family: "Times New Roman"
// (3) List separators in content values as in inline-lists:
// <style>
// ul.inlineList { list-style: none; padding:0 }
// ul.inlineList > li { display: inline }
// ul.inlineList > li:before { content: ", " }
// ul.inlineList > li:first-child:before { content: "" }
// </style>
// <ul class=inlineList><li>One<li>Two<li>Three</ul>
// (4) Attribute value selectors as in a[href="http://example.com/"]
//
// We conservatively treat all strings as URLs, but make some
// allowances to avoid confusion.
//
// In (1), our conservative assumption is justified.
// In (2), valid font names do not contain ':', '?', or '#', so our
// conservative assumption is fine since we will never transition past
// urlPartPreQuery.
// In (3), our protocol heuristic should not be tripped, and there
// should not be non-space content after a '?' or '#', so as long as
// we only %-encode RFC 3986 reserved characters we are ok.
// In (4), we should URL escape for URL attributes, and for others we
// have the attribute name available if our conservative assumption
// proves problematic for real code.
k := 0
for {
i := k + bytes.IndexAny(s[k:], `("'/`)
if i < k {
return c, len(s)
}
switch s[i] {
case '(':
// Look for url to the left.
p := bytes.TrimRight(s[:i], "\t\n\f\r ")
if endsWithCSSKeyword(p, "url") {
j := len(s) - len(bytes.TrimLeft(s[i+1:], "\t\n\f\r "))
switch {
case j != len(s) && s[j] == '"':
c.state, j = stateCSSDqURL, j+1
case j != len(s) && s[j] == '\'':
c.state, j = stateCSSSqURL, j+1
default:
c.state = stateCSSURL
}
return c, j
}
case '/':
if i+1 < len(s) {
switch s[i+1] {
case '/':
c.state = stateCSSLineCmt
return c, i + 2
case '*':
c.state = stateCSSBlockCmt
return c, i + 2
}
}
case '"':
c.state = stateCSSDqStr
return c, i + 1
case '\'':
c.state = stateCSSSqStr
return c, i + 1
}
k = i + 1
}
}
// tCSSStr is the context transition function for the CSS string and URL states.
func tCSSStr(c context, s []byte) (context, int) {
var endAndEsc string
switch c.state {
case stateCSSDqStr, stateCSSDqURL:
endAndEsc = `\"`
case stateCSSSqStr, stateCSSSqURL:
endAndEsc = `\'`
case stateCSSURL:
// Unquoted URLs end with a newline or close parenthesis.
// The below includes the wc (whitespace character) and nl.
endAndEsc = "\\\t\n\f\r )"
default:
panic(c.state.String())
}
k := 0
for {
i := k + bytes.IndexAny(s[k:], endAndEsc)
if i < k {
c, nread := tURL(c, decodeCSS(s[k:]))
return c, k + nread
}
if s[i] == '\\' {
i++
if i == len(s) {
return context{
state: stateError,
err: errorf(ErrPartialEscape, nil, 0, "unfinished escape sequence in CSS string: %q", s),
}, len(s)
}
} else {
c.state = stateCSS
return c, i + 1
}
c, _ = tURL(c, decodeCSS(s[:i+1]))
k = i + 1
}
}
// tError is the context transition function for the error state.
func tError(c context, s []byte) (context, int) {
return c, len(s)
}
// eatAttrName returns the largest j such that s[i:j] is an attribute name.
// It returns an error if s[i:] does not look like it begins with an
// attribute name, such as encountering a quote mark without a preceding
// equals sign.
func eatAttrName(s []byte, i int) (int, *Error) {
for j := i; j < len(s); j++ {
switch s[j] {
case ' ', '\t', '\n', '\f', '\r', '=', '>':
return j, nil
case '\'', '"', '<':
// These result in a parse warning in HTML5 and are
// indicative of serious problems if seen in an attr
// name in a template.
return -1, errorf(ErrBadHTML, nil, 0, "%q in attribute name: %.32q", s[j:j+1], s)
default:
// No-op.
}
}
return len(s), nil
}
var elementNameMap = map[string]element{
"script": elementScript,
"style": elementStyle,
"textarea": elementTextarea,
"title": elementTitle,
}
// asciiAlpha reports whether c is an ASCII letter.
func asciiAlpha(c byte) bool {
return 'A' <= c && c <= 'Z' || 'a' <= c && c <= 'z'
}
// asciiAlphaNum reports whether c is an ASCII letter or digit.
func asciiAlphaNum(c byte) bool {
return asciiAlpha(c) || '0' <= c && c <= '9'
}
// eatTagName returns the largest j such that s[i:j] is a tag name and the tag type.
func eatTagName(s []byte, i int) (int, element) {
if i == len(s) || !asciiAlpha(s[i]) {
return i, elementNone
}
j := i + 1
for j < len(s) {
x := s[j]
if asciiAlphaNum(x) {
j++
continue
}
// Allow "x-y" or "x:y" but not "x-", "-y", or "x--y".
if (x == ':' || x == '-') && j+1 < len(s) && asciiAlphaNum(s[j+1]) {
j += 2
continue
}
break
}
return j, elementNameMap[strings.ToLower(string(s[i:j]))]
}
// eatWhiteSpace returns the largest j such that s[i:j] is white space.
func eatWhiteSpace(s []byte, i int) int {
for j := i; j < len(s); j++ {
switch s[j] {
case ' ', '\t', '\n', '\f', '\r':
// No-op.
default:
return j
}
}
return len(s)
}

View File

@@ -0,0 +1,60 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"bytes"
"strings"
"testing"
)
func TestFindEndTag(t *testing.T) {
tests := []struct {
s, tag string
want int
}{
{"", "tag", -1},
{"hello </textarea> hello", "textarea", 6},
{"hello </TEXTarea> hello", "textarea", 6},
{"hello </textAREA>", "textarea", 6},
{"hello </textarea", "textareax", -1},
{"hello </textarea>", "tag", -1},
{"hello tag </textarea", "tag", -1},
{"hello </tag> </other> </textarea> <other>", "textarea", 22},
{"</textarea> <other>", "textarea", 0},
{"<div> </div> </TEXTAREA>", "textarea", 13},
{"<div> </div> </TEXTAREA\t>", "textarea", 13},
{"<div> </div> </TEXTAREA >", "textarea", 13},
{"<div> </div> </TEXTAREAfoo", "textarea", -1},
{"</TEXTAREAfoo </textarea>", "textarea", 14},
{"<</script >", "script", 1},
{"</script>", "textarea", -1},
}
for _, test := range tests {
if got := indexTagEnd([]byte(test.s), []byte(test.tag)); test.want != got {
t.Errorf("%q/%q: want\n\t%d\nbut got\n\t%d", test.s, test.tag, test.want, got)
}
}
}
func BenchmarkTemplateSpecialTags(b *testing.B) {
r := struct {
Name, Gift string
}{"Aunt Mildred", "bone china tea set"}
h1 := "<textarea> Hello Hello Hello </textarea> "
h2 := "<textarea> <p> Dear {{.Name}},\n{{with .Gift}}Thank you for the lovely {{.}}. {{end}}\nBest wishes. </p>\n</textarea>"
html := strings.Repeat(h1, 100) + h2 + strings.Repeat(h1, 100) + h2
var buf bytes.Buffer
for i := 0; i < b.N; i++ {
tmpl := Must(New("foo").Parse(html))
if err := tmpl.Execute(&buf, r); err != nil {
b.Fatal(err)
}
buf.Reset()
}
}

216
src/html/template/url.go Normal file
View File

@@ -0,0 +1,216 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"fmt"
"strings"
)
// urlFilter returns its input unless it contains an unsafe scheme in which
// case it defangs the entire URL.
//
// Schemes that cause unintended side effects that are irreversible without user
// interaction are considered unsafe. For example, clicking on a "javascript:"
// link can immediately trigger JavaScript code execution.
//
// This filter conservatively assumes that all schemes other than the following
// are unsafe:
// - http: Navigates to a new website, and may open a new window or tab.
// These side effects can be reversed by navigating back to the
// previous website, or closing the window or tab. No irreversible
// changes will take place without further user interaction with
// the new website.
// - https: Same as http.
// - mailto: Opens an email program and starts a new draft. This side effect
// is not irreversible until the user explicitly clicks send; it
// can be undone by closing the email program.
//
// To allow URLs containing other schemes to bypass this filter, developers must
// explicitly indicate that such a URL is expected and safe by encapsulating it
// in a template.URL value.
func urlFilter(args ...any) string {
s, t := stringify(args...)
if t == contentTypeURL {
return s
}
if !isSafeURL(s) {
return "#" + filterFailsafe
}
return s
}
// isSafeURL is true if s is a relative URL or if URL has a protocol in
// (http, https, mailto).
func isSafeURL(s string) bool {
if protocol, _, ok := strings.Cut(s, ":"); ok && !strings.Contains(protocol, "/") {
if !strings.EqualFold(protocol, "http") && !strings.EqualFold(protocol, "https") && !strings.EqualFold(protocol, "mailto") {
return false
}
}
return true
}
// urlEscaper produces an output that can be embedded in a URL query.
// The output can be embedded in an HTML attribute without further escaping.
func urlEscaper(args ...any) string {
return urlProcessor(false, args...)
}
// urlNormalizer normalizes URL content so it can be embedded in a quote-delimited
// string or parenthesis delimited url(...).
// The normalizer does not encode all HTML specials. Specifically, it does not
// encode '&' so correct embedding in an HTML attribute requires escaping of
// '&' to '&amp;'.
func urlNormalizer(args ...any) string {
return urlProcessor(true, args...)
}
// urlProcessor normalizes (when norm is true) or escapes its input to produce
// a valid hierarchical or opaque URL part.
func urlProcessor(norm bool, args ...any) string {
s, t := stringify(args...)
if t == contentTypeURL {
norm = true
}
var b strings.Builder
if processURLOnto(s, norm, &b) {
return b.String()
}
return s
}
// processURLOnto appends a normalized URL corresponding to its input to b
// and reports whether the appended content differs from s.
func processURLOnto(s string, norm bool, b *strings.Builder) bool {
b.Grow(len(s) + 16)
written := 0
// The byte loop below assumes that all URLs use UTF-8 as the
// content-encoding. This is similar to the URI to IRI encoding scheme
// defined in section 3.1 of RFC 3987, and behaves the same as the
// EcmaScript builtin encodeURIComponent.
// It should not cause any misencoding of URLs in pages with
// Content-type: text/html;charset=UTF-8.
for i, n := 0, len(s); i < n; i++ {
c := s[i]
switch c {
// Single quote and parens are sub-delims in RFC 3986, but we
// escape them so the output can be embedded in single
// quoted attributes and unquoted CSS url(...) constructs.
// Single quotes are reserved in URLs, but are only used in
// the obsolete "mark" rule in an appendix in RFC 3986
// so can be safely encoded.
case '!', '#', '$', '&', '*', '+', ',', '/', ':', ';', '=', '?', '@', '[', ']':
if norm {
continue
}
// Unreserved according to RFC 3986 sec 2.3
// "For consistency, percent-encoded octets in the ranges of
// ALPHA (%41-%5A and %61-%7A), DIGIT (%30-%39), hyphen (%2D),
// period (%2E), underscore (%5F), or tilde (%7E) should not be
// created by URI producers
case '-', '.', '_', '~':
continue
case '%':
// When normalizing do not re-encode valid escapes.
if norm && i+2 < len(s) && isHex(s[i+1]) && isHex(s[i+2]) {
continue
}
default:
// Unreserved according to RFC 3986 sec 2.3
if 'a' <= c && c <= 'z' {
continue
}
if 'A' <= c && c <= 'Z' {
continue
}
if '0' <= c && c <= '9' {
continue
}
}
b.WriteString(s[written:i])
fmt.Fprintf(b, "%%%02x", c)
written = i + 1
}
b.WriteString(s[written:])
return written != 0
}
// Filters and normalizes srcset values which are comma separated
// URLs followed by metadata.
func srcsetFilterAndEscaper(args ...any) string {
s, t := stringify(args...)
switch t {
case contentTypeSrcset:
return s
case contentTypeURL:
// Normalizing gets rid of all HTML whitespace
// which separate the image URL from its metadata.
var b strings.Builder
if processURLOnto(s, true, &b) {
s = b.String()
}
// Additionally, commas separate one source from another.
return strings.ReplaceAll(s, ",", "%2c")
}
var b strings.Builder
written := 0
for i := 0; i < len(s); i++ {
if s[i] == ',' {
filterSrcsetElement(s, written, i, &b)
b.WriteString(",")
written = i + 1
}
}
filterSrcsetElement(s, written, len(s), &b)
return b.String()
}
// Derived from https://play.golang.org/p/Dhmj7FORT5
const htmlSpaceAndASCIIAlnumBytes = "\x00\x36\x00\x00\x01\x00\xff\x03\xfe\xff\xff\x07\xfe\xff\xff\x07"
// isHTMLSpace is true iff c is a whitespace character per
// https://infra.spec.whatwg.org/#ascii-whitespace
func isHTMLSpace(c byte) bool {
return (c <= 0x20) && 0 != (htmlSpaceAndASCIIAlnumBytes[c>>3]&(1<<uint(c&0x7)))
}
func isHTMLSpaceOrASCIIAlnum(c byte) bool {
return (c < 0x80) && 0 != (htmlSpaceAndASCIIAlnumBytes[c>>3]&(1<<uint(c&0x7)))
}
func filterSrcsetElement(s string, left int, right int, b *strings.Builder) {
start := left
for start < right && isHTMLSpace(s[start]) {
start++
}
end := right
for i := start; i < right; i++ {
if isHTMLSpace(s[i]) {
end = i
break
}
}
if url := s[start:end]; isSafeURL(url) {
// If image metadata is only spaces or alnums then
// we don't need to URL normalize it.
metadataOk := true
for i := end; i < right; i++ {
if !isHTMLSpaceOrASCIIAlnum(s[i]) {
metadataOk = false
break
}
}
if metadataOk {
b.WriteString(s[left:start])
processURLOnto(url, true, b)
b.WriteString(s[end:right])
return
}
}
b.WriteString("#")
b.WriteString(filterFailsafe)
}

View File

@@ -0,0 +1,169 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package template
import (
"testing"
)
func TestURLNormalizer(t *testing.T) {
tests := []struct {
url, want string
}{
{"", ""},
{
"http://example.com:80/foo/bar?q=foo%20&bar=x+y#frag",
"http://example.com:80/foo/bar?q=foo%20&bar=x+y#frag",
},
{" ", "%20"},
{"%7c", "%7c"},
{"%7C", "%7C"},
{"%2", "%252"},
{"%", "%25"},
{"%z", "%25z"},
{"/foo|bar/%5c\u1234", "/foo%7cbar/%5c%e1%88%b4"},
}
for _, test := range tests {
if got := urlNormalizer(test.url); test.want != got {
t.Errorf("%q: want\n\t%q\nbut got\n\t%q", test.url, test.want, got)
}
if test.want != urlNormalizer(test.want) {
t.Errorf("not idempotent: %q", test.want)
}
}
}
func TestURLFilters(t *testing.T) {
input := ("\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f" +
"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !"#$%&'()*+,-./` +
`0123456789:;<=>?` +
`@ABCDEFGHIJKLMNO` +
`PQRSTUVWXYZ[\]^_` +
"`abcdefghijklmno" +
"pqrstuvwxyz{|}~\x7f" +
"\u00A0\u0100\u2028\u2029\ufeff\U0001D11E")
tests := []struct {
name string
escaper func(...any) string
escaped string
}{
{
"urlEscaper",
urlEscaper,
"%00%01%02%03%04%05%06%07%08%09%0a%0b%0c%0d%0e%0f" +
"%10%11%12%13%14%15%16%17%18%19%1a%1b%1c%1d%1e%1f" +
"%20%21%22%23%24%25%26%27%28%29%2a%2b%2c-.%2f" +
"0123456789%3a%3b%3c%3d%3e%3f" +
"%40ABCDEFGHIJKLMNO" +
"PQRSTUVWXYZ%5b%5c%5d%5e_" +
"%60abcdefghijklmno" +
"pqrstuvwxyz%7b%7c%7d~%7f" +
"%c2%a0%c4%80%e2%80%a8%e2%80%a9%ef%bb%bf%f0%9d%84%9e",
},
{
"urlNormalizer",
urlNormalizer,
"%00%01%02%03%04%05%06%07%08%09%0a%0b%0c%0d%0e%0f" +
"%10%11%12%13%14%15%16%17%18%19%1a%1b%1c%1d%1e%1f" +
"%20!%22#$%25&%27%28%29*+,-./" +
"0123456789:;%3c=%3e?" +
"@ABCDEFGHIJKLMNO" +
"PQRSTUVWXYZ[%5c]%5e_" +
"%60abcdefghijklmno" +
"pqrstuvwxyz%7b%7c%7d~%7f" +
"%c2%a0%c4%80%e2%80%a8%e2%80%a9%ef%bb%bf%f0%9d%84%9e",
},
}
for _, test := range tests {
if s := test.escaper(input); s != test.escaped {
t.Errorf("%s: want\n\t%q\ngot\n\t%q", test.name, test.escaped, s)
continue
}
}
}
func TestSrcsetFilter(t *testing.T) {
tests := []struct {
name string
input string
want string
}{
{
"one ok",
"http://example.com/img.png",
"http://example.com/img.png",
},
{
"one ok with metadata",
" /img.png 200w",
" /img.png 200w",
},
{
"one bad",
"javascript:alert(1) 200w",
"#ZgotmplZ",
},
{
"two ok",
"foo.png, bar.png",
"foo.png, bar.png",
},
{
"left bad",
"javascript:alert(1), /foo.png",
"#ZgotmplZ, /foo.png",
},
{
"right bad",
"/bogus#, javascript:alert(1)",
"/bogus#,#ZgotmplZ",
},
}
for _, test := range tests {
if got := srcsetFilterAndEscaper(test.input); got != test.want {
t.Errorf("%s: srcsetFilterAndEscaper(%q) want %q != %q", test.name, test.input, test.want, got)
}
}
}
func BenchmarkURLEscaper(b *testing.B) {
for i := 0; i < b.N; i++ {
urlEscaper("http://example.com:80/foo?q=bar%20&baz=x+y#frag")
}
}
func BenchmarkURLEscaperNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
urlEscaper("TheQuickBrownFoxJumpsOverTheLazyDog.")
}
}
func BenchmarkURLNormalizer(b *testing.B) {
for i := 0; i < b.N; i++ {
urlNormalizer("The quick brown fox jumps over the lazy dog.\n")
}
}
func BenchmarkURLNormalizerNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
urlNormalizer("http://example.com:80/foo?q=bar%20&baz=x+y#frag")
}
}
func BenchmarkSrcsetFilter(b *testing.B) {
for i := 0; i < b.N; i++ {
srcsetFilterAndEscaper(" /foo/bar.png 200w, /baz/boo(1).png")
}
}
func BenchmarkSrcsetFilterNoSpecials(b *testing.B) {
for i := 0; i < b.N; i++ {
srcsetFilterAndEscaper("http://example.com:80/foo?q=bar%20&baz=x+y#frag")
}
}

View File

@@ -0,0 +1,26 @@
// Code generated by "stringer -type urlPart"; DO NOT EDIT.
package template
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[urlPartNone-0]
_ = x[urlPartPreQuery-1]
_ = x[urlPartQueryOrFrag-2]
_ = x[urlPartUnknown-3]
}
const _urlPart_name = "urlPartNoneurlPartPreQueryurlPartQueryOrFragurlPartUnknown"
var _urlPart_index = [...]uint8{0, 11, 26, 44, 58}
func (i urlPart) String() string {
if i >= urlPart(len(_urlPart_index)-1) {
return "urlPart(" + strconv.FormatInt(int64(i), 10) + ")"
}
return _urlPart_name[_urlPart_index[i]:_urlPart_index[i+1]]
}