mirror of
https://github.com/go-gitea/gitea.git
synced 2024-08-24 21:34:36 -04:00
Exclude HTML tags from Markdown post-processing
HTML tags are no longer processed for special links, etc Contents of <a>, <code> and <pre> are not processed for special links Processing for special links is done after Markdown conversion
This commit is contained in:
parent
5e763baa12
commit
127eb439d7
@ -7,6 +7,7 @@ package base
|
|||||||
import (
|
import (
|
||||||
"bytes"
|
"bytes"
|
||||||
"fmt"
|
"fmt"
|
||||||
|
"io"
|
||||||
"net/http"
|
"net/http"
|
||||||
"path"
|
"path"
|
||||||
"path/filepath"
|
"path/filepath"
|
||||||
@ -16,6 +17,8 @@ import (
|
|||||||
"github.com/russross/blackfriday"
|
"github.com/russross/blackfriday"
|
||||||
|
|
||||||
"github.com/gogits/gogs/modules/setting"
|
"github.com/gogits/gogs/modules/setting"
|
||||||
|
|
||||||
|
"golang.org/x/net/html"
|
||||||
)
|
)
|
||||||
|
|
||||||
func isletter(c byte) bool {
|
func isletter(c byte) bool {
|
||||||
@ -217,12 +220,57 @@ func RenderRawMarkdown(body []byte, urlPrefix string) []byte {
|
|||||||
}
|
}
|
||||||
|
|
||||||
func RenderMarkdown(rawBytes []byte, urlPrefix string) []byte {
|
func RenderMarkdown(rawBytes []byte, urlPrefix string) []byte {
|
||||||
body := RenderSpecialLink(rawBytes, urlPrefix)
|
result := RenderRawMarkdown(rawBytes, urlPrefix)
|
||||||
body = RenderRawMarkdown(body, urlPrefix)
|
result = PostProcessMarkdown(result, urlPrefix)
|
||||||
body = Sanitizer.SanitizeBytes(body)
|
result = Sanitizer.SanitizeBytes(result)
|
||||||
return body
|
return result
|
||||||
}
|
}
|
||||||
|
|
||||||
func RenderMarkdownString(raw, urlPrefix string) string {
|
func RenderMarkdownString(raw, urlPrefix string) string {
|
||||||
return string(RenderMarkdown([]byte(raw), urlPrefix))
|
return string(RenderMarkdown([]byte(raw), urlPrefix))
|
||||||
}
|
}
|
||||||
|
|
||||||
|
func PostProcessMarkdown(rawHtml []byte, urlPrefix string) []byte {
|
||||||
|
var buf bytes.Buffer
|
||||||
|
tokenizer := html.NewTokenizer(bytes.NewReader(rawHtml))
|
||||||
|
for html.ErrorToken != tokenizer.Next() {
|
||||||
|
|
||||||
|
// A parse error has occurred, so return the original input unmodified
|
||||||
|
return rawHtml
|
||||||
|
|
||||||
|
token := tokenizer.Token()
|
||||||
|
switch token.Type {
|
||||||
|
case html.TextToken:
|
||||||
|
text := []byte(token.String())
|
||||||
|
text = RenderSpecialLink(text, urlPrefix)
|
||||||
|
|
||||||
|
buf.Write(text)
|
||||||
|
|
||||||
|
case html.StartTagToken:
|
||||||
|
buf.WriteString(token.String())
|
||||||
|
|
||||||
|
tagName := token.Data
|
||||||
|
// If this is an excluded tag, we skip processing all output until a close tag is encountered
|
||||||
|
if strings.EqualFold("a", tagName) || strings.EqualFold("code", tagName) || strings.EqualFold("pre", tagName) {
|
||||||
|
for html.ErrorToken != tokenizer.Next() {
|
||||||
|
token = tokenizer.Token()
|
||||||
|
// Copy the token to the output verbatim
|
||||||
|
buf.WriteString(token.String())
|
||||||
|
// If this is the close tag, we are done
|
||||||
|
if html.EndTagToken == token.Type && strings.EqualFold(tagName, token.Data) { break }
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
default:
|
||||||
|
buf.WriteString(token.String())
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
if io.EOF == tokenizer.Err() {
|
||||||
|
return buf.Bytes()
|
||||||
|
}
|
||||||
|
|
||||||
|
// If we are not at the end of the input, then some other parsing error has occurred, so return
|
||||||
|
// the input verbatim.
|
||||||
|
return rawHtml
|
||||||
|
}
|
||||||
|
Loading…
Reference in New Issue
Block a user