Update bluemonday to v1.0.15 (#16379)

* update github.com/microcosm-cc/bluemonday

* add exec flag to contrib/update_dependencies.sh

* Fix TESTS

go.mod

@@ -80,7 +80,7 @@ require (
 	github.com/mattn/go-runewidth v0.0.13 // indirect
 	github.com/mattn/go-sqlite3 v1.14.7
 	github.com/mholt/archiver/v3 v3.5.0
-	github.com/microcosm-cc/bluemonday v1.0.14
+	github.com/microcosm-cc/bluemonday v1.0.15
 	github.com/miekg/dns v1.1.43 // indirect
 	github.com/minio/md5-simd v1.1.2 // indirect
 	github.com/minio/minio-go/v7 v7.0.12

go.sum

@@ -795,8 +795,8 @@ github.com/mholt/acmez v0.1.3 h1:J7MmNIk4Qf9b8mAGqAh4XkNeowv3f1zW816yf4zt7Qk=
 github.com/mholt/acmez v0.1.3/go.mod h1:8qnn8QA/Ewx8E3ZSsmscqsIjhhpxuy9vqdgbX2ceceM=
 github.com/mholt/archiver/v3 v3.5.0 h1:nE8gZIrw66cu4osS/U7UW7YDuGMHssxKutU8IfWxwWE=
 github.com/mholt/archiver/v3 v3.5.0/go.mod h1:qqTTPUK/HZPFgFQ/TJ3BzvTpF/dPtFVJXdQbCmeMxwc=
-github.com/microcosm-cc/bluemonday v1.0.14 h1:Djd+GeTanVeA23todvVC0AO5hsI+vAwQMLTy794Zr5I=
+github.com/microcosm-cc/bluemonday v1.0.15 h1:J4uN+qPng9rvkBZBoBb8YGR+ijuklIMpSOZZLjYpbeY=
-github.com/microcosm-cc/bluemonday v1.0.14/go.mod h1:beubO5lmWoy1tU8niaMyXNriNgROO37H3U/tsrcZsy0=
+github.com/microcosm-cc/bluemonday v1.0.15/go.mod h1:ZLvAzeakRwrGnzQEvstVzVt3ZpqOF2+sdFr0Om+ce30=
 github.com/miekg/dns v1.0.14/go.mod h1:W1PPwlIAgtquWBMBEV9nkV9Cazfe8ScdGz/Lj7v3Nrg=
 github.com/miekg/dns v1.1.42/go.mod h1:+evo5L0630/F6ca/Z9+GAqzhjGyn8/c+TBaOyfEl0V4=
 github.com/miekg/dns v1.1.43 h1:JKfpVSCB84vrAmHzyrsxB5NAr5kLoMXZArPSw7Qlgyg=
@@ -1235,7 +1235,6 @@ golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v
 golang.org/x/net v0.0.0-20210326060303-6b1517762897/go.mod h1:uSPa2vr4CLtc/ILN5odXGNXS6mhrKVzTaCXzk9m6W3k=
 golang.org/x/net v0.0.0-20210331060903-cb1fcc7394e5/go.mod h1:p54w0d4576C0XHj96bSt6lcn1PtDYWL6XObtHCRCNQM=
 golang.org/x/net v0.0.0-20210525063256-abc453219eb5/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
-golang.org/x/net v0.0.0-20210610132358-84b48f89b13b/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.0.0-20210614182718-04defd469f4e h1:XpT3nA5TvE525Ne3hInMh6+GETgn27Zfm9dxsThnX2Q=
 golang.org/x/net v0.0.0-20210614182718-04defd469f4e/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=

modules/markup/html_test.go

@@ -138,13 +138,13 @@ func TestRender_links(t *testing.T) {
 		`<p><a href="http://www.example.com/wpstyle/?p=364" rel="nofollow">http://www.example.com/wpstyle/?p=364</a></p>`)
 	test(
 		"https://www.example.com/foo/?bar=baz&inga=42&quux",
-		`<p><a href="https://www.example.com/foo/?bar=baz&inga=42&quux" rel="nofollow">https://www.example.com/foo/?bar=baz&amp;inga=42&amp;quux</a></p>`)
+		`<p><a href="https://www.example.com/foo/?bar=baz&amp;inga=42&amp;quux" rel="nofollow">https://www.example.com/foo/?bar=baz&amp;inga=42&amp;quux</a></p>`)
 	test(
 		"http://142.42.1.1/",
 		`<p><a href="http://142.42.1.1/" rel="nofollow">http://142.42.1.1/</a></p>`)
 	test(
 		"https://github.com/go-gitea/gitea/?p=aaa/bbb.html#ccc-ddd",
-		`<p><a href="https://github.com/go-gitea/gitea/?p=aaa%2Fbbb.html#ccc-ddd" rel="nofollow">https://github.com/go-gitea/gitea/?p=aaa/bbb.html#ccc-ddd</a></p>`)
+		`<p><a href="https://github.com/go-gitea/gitea/?p=aaa/bbb.html#ccc-ddd" rel="nofollow">https://github.com/go-gitea/gitea/?p=aaa/bbb.html#ccc-ddd</a></p>`)
 	test(
 		"https://en.wikipedia.org/wiki/URL_(disambiguation)",
 		`<p><a href="https://en.wikipedia.org/wiki/URL_(disambiguation)" rel="nofollow">https://en.wikipedia.org/wiki/URL_(disambiguation)</a></p>`)
@@ -162,7 +162,7 @@ func TestRender_links(t *testing.T) {
 		`<p><a href="ftp://gitea.com/file.txt" rel="nofollow">ftp://gitea.com/file.txt</a></p>`)
 	test(
 		"magnet:?xt=urn:btih:5dee65101db281ac9c46344cd6b175cdcadabcde&dn=download",
-		`<p><a href="magnet:?xt=urn%3Abtih%3A5dee65101db281ac9c46344cd6b175cdcadabcde&dn=download" rel="nofollow">magnet:?xt=urn:btih:5dee65101db281ac9c46344cd6b175cdcadabcde&amp;dn=download</a></p>`)
+		`<p><a href="magnet:?xt=urn:btih:5dee65101db281ac9c46344cd6b175cdcadabcde&amp;dn=download" rel="nofollow">magnet:?xt=urn:btih:5dee65101db281ac9c46344cd6b175cdcadabcde&amp;dn=download</a></p>`)
 
 	// Test that should *not* be turned into URL
 	test(

vendor/github.com/microcosm-cc/bluemonday/CONTRIBUTING.md

@@ -9,6 +9,7 @@ Third-party patches are essential for keeping bluemonday secure and offering the
 ## Guidelines
 
 1. Do not vendor dependencies. As a security package, were we to vendor dependencies the projects that then vendor bluemonday may not receive the latest security updates to the dependencies. By not vendoring dependencies the project that implements bluemonday will vendor the latest version of any dependent packages. Vendoring is a project problem, not a package problem. bluemonday will be tested against the latest version of dependencies periodically and during any PR/merge.
+2. I do not care about spelling mistakes or whitespace and I do not believe that you should either. PRs therefore must be functional in their nature or be substantial and impactful if documentation or examples.
 
 ## Submitting an Issue
 

vendor/github.com/microcosm-cc/bluemonday/go.mod

@@ -3,7 +3,8 @@ module github.com/microcosm-cc/bluemonday
 go 1.16
 
 require (
+	github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d
 	github.com/aymerick/douceur v0.2.0
 	github.com/gorilla/css v1.0.0 // indirect
-	golang.org/x/net v0.0.0-20210610132358-84b48f89b13b
+	golang.org/x/net v0.0.0-20210614182718-04defd469f4e
 )

vendor/github.com/microcosm-cc/bluemonday/go.sum

@@ -1,3 +1,5 @@
+github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d h1:Byv0BzEl3/e6D5CLfI0j/7hiIEtvGVFPCZ7Ei2oq8iQ=
+github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d/go.mod h1:WaHUgvxTVq04UNunO+XhnAqY/wQc+bxr74GqbsZ/Jqw=
 github.com/aymerick/douceur v0.2.0 h1:Mv+mAeH1Q+n9Fr+oyamOlAkUNPWPlA8PPGR0QAaYuPk=
 github.com/aymerick/douceur v0.2.0/go.mod h1:wlT5vV2O3h55X9m7iVYN0TBM0NH/MmbLnd30/FjWUq4=
 github.com/gorilla/css v1.0.0 h1:BQqNyPTi50JCFMTw/b67hByjMVXZRwGha6wxVGkeihY=
@@ -6,6 +8,8 @@ golang.org/x/net v0.0.0-20210421230115-4e50805a0758 h1:aEpZnXcAmXkd6AvLb2OPt+EN1
 golang.org/x/net v0.0.0-20210421230115-4e50805a0758/go.mod h1:72T/g9IO56b78aLF+1Kcs5dz7/ng1VjMUvfKvpfy+jM=
 golang.org/x/net v0.0.0-20210610132358-84b48f89b13b h1:k+E048sYJHyVnsr1GDrRZWQ32D2C7lWs9JRc0bel53A=
 golang.org/x/net v0.0.0-20210610132358-84b48f89b13b/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
+golang.org/x/net v0.0.0-20210614182718-04defd469f4e h1:XpT3nA5TvE525Ne3hInMh6+GETgn27Zfm9dxsThnX2Q=
+golang.org/x/net v0.0.0-20210614182718-04defd469f4e/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/sys v0.0.0-20201119102817-f84b799fce68/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20210420072515-93ed5bcd2bfe/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20210423082822-04245dca01da/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=

vendor/github.com/microcosm-cc/bluemonday/sanitize.go

@@ -130,7 +130,7 @@ func escapeUrlComponent(w stringWriterWriter, val string) error {
 	return err
 }
 
-// Query represents a query
+// Query represents a single part of the query string, a query param 
 type Query struct {
 	Key      string
 	Value    string
@@ -138,6 +138,10 @@ type Query struct {
 }
 
 func parseQuery(query string) (values []Query, err error) {
+	// This is essentially a copy of parseQuery from
+	// https://golang.org/src/net/url/url.go but adjusted to build our values
+	// based on our type, which we need to preserve the ordering of the query
+	// string
 	for query != "" {
 		key := query
 		if i := strings.IndexAny(key, "&;"); i >= 0 {
@@ -213,43 +217,6 @@ func sanitizedURL(val string) (string, error) {
 	return u.String(), nil
 }
 
-func (p *Policy) writeLinkableBuf(buff stringWriterWriter, token *html.Token) (int, error) {
-	// do not escape multiple query parameters
-	tokenBuff := bytes.NewBuffer(make([]byte, 0, 1024)) // This should stay on the stack unless it gets too big
-
-	tokenBuff.WriteByte('<')
-	tokenBuff.WriteString(token.Data)
-	for _, attr := range token.Attr {
-		tokenBuff.WriteByte(' ')
-		tokenBuff.WriteString(attr.Key)
-		tokenBuff.Write([]byte{'=', '"'})
-		switch attr.Key {
-		case "href", "src":
-			u, ok := p.validURL(attr.Val)
-			if !ok {
-				tokenBuff.WriteString(html.EscapeString(attr.Val))
-				continue
-			}
-			u, err := sanitizedURL(u)
-			if err == nil {
-				tokenBuff.WriteString(u)
-			} else {
-				// fallthrough
-				tokenBuff.WriteString(html.EscapeString(attr.Val))
-			}
-		default:
-			// re-apply
-			tokenBuff.WriteString(html.EscapeString(attr.Val))
-		}
-		tokenBuff.WriteByte('"')
-	}
-	if token.Type == html.SelfClosingTagToken {
-		tokenBuff.WriteString("/")
-	}
-	tokenBuff.WriteString(">")
-	return buff.Write(tokenBuff.Bytes())
-}
-
 // Performs the actual sanitization process.
 func (p *Policy) sanitizeWithBuff(r io.Reader) *bytes.Buffer {
 	var buff bytes.Buffer
@@ -344,7 +311,9 @@ func (p *Policy) sanitize(r io.Reader, w io.Writer) error {
 				aps = aa
 			}
 			if len(token.Attr) != 0 {
-				token.Attr = p.sanitizeAttrs(token.Data, token.Attr, aps)
+				token.Attr = escapeAttributes(
+					p.sanitizeAttrs(token.Data, token.Attr, aps),
+				)
 			}
 
 			if len(token.Attr) == 0 {
@@ -361,17 +330,10 @@ func (p *Policy) sanitize(r io.Reader, w io.Writer) error {
 			}
 
 			if !skipElementContent {
-				// do not escape multiple query parameters
-				if linkable(token.Data) {
-					if _, err := p.writeLinkableBuf(buff, &token); err != nil {
-						return err
-					}
-				} else {
 				if _, err := buff.WriteString(token.String()); err != nil {
 					return err
 				}
 			}
-			}
 
 		case html.EndTagToken:
 
@@ -439,7 +401,7 @@ func (p *Policy) sanitize(r io.Reader, w io.Writer) error {
 			}
 
 			if len(token.Attr) != 0 {
-				token.Attr = p.sanitizeAttrs(token.Data, token.Attr, aps)
+				token.Attr = escapeAttributes(p.sanitizeAttrs(token.Data, token.Attr, aps))
 			}
 
 			if len(token.Attr) == 0 && !p.allowNoAttrs(token.Data) {
@@ -451,17 +413,10 @@ func (p *Policy) sanitize(r io.Reader, w io.Writer) error {
 				}
 			}
 			if !skipElementContent {
-				// do not escape multiple query parameters
-				if linkable(token.Data) {
-					if _, err := p.writeLinkableBuf(buff, &token); err != nil {
-						return err
-					}
-				} else {
 				if _, err := buff.WriteString(token.String()); err != nil {
 					return err
 				}
 			}
-			}
 
 		case html.TextToken:
 
@@ -569,9 +524,11 @@ attrsLoop:
 			for _, ap := range apl {
 				if ap.regexp != nil {
 					if ap.regexp.MatchString(htmlAttr.Val) {
+				htmlAttr.Val = escapeAttribute(htmlAttr.Val)
 						cleanAttrs = append(cleanAttrs, htmlAttr)
 					}
 				} else {
+				htmlAttr.Val = escapeAttribute(htmlAttr.Val)
 					cleanAttrs = append(cleanAttrs, htmlAttr)
 				}
 			}
@@ -1087,3 +1044,18 @@ func normaliseElementName(str string) string {
 		`"`,
 	)
 }
+
+func escapeAttributes(attrs []html.Attribute) []html.Attribute {
+	escapedAttrs := []html.Attribute{}
+	for _, attr := range attrs {
+		attr.Val = escapeAttribute(attr.Val)
+		escapedAttrs = append(escapedAttrs, attr)
+	}
+	return escapedAttrs
+}
+
+func escapeAttribute(val string) string {
+	val = strings.Replace(val, string([]rune{'\u00A0'}), `&nbsp;`, -1)
+	val = strings.Replace(val, `"`, `&quot;`, -1)
+	return val
+}

vendor/modules.txt

@@ -604,7 +604,7 @@ github.com/mholt/acmez/acme
 # github.com/mholt/archiver/v3 v3.5.0
 ## explicit
 github.com/mholt/archiver/v3
-# github.com/microcosm-cc/bluemonday v1.0.14
+# github.com/microcosm-cc/bluemonday v1.0.15
 ## explicit
 github.com/microcosm-cc/bluemonday
 github.com/microcosm-cc/bluemonday/css