Fix markdown URL parsing (#17924)

Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: zeripath <art27@cantab.net>
This commit is contained in:
wxiaoguang 2021-12-12 01:21:36 +08:00 committed by GitHub
parent 379a5241c6
commit 6d4172987e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 127 additions and 75 deletions

View file

@ -41,16 +41,21 @@ var (
// While fast, this is also incorrect and lead to false positives.
// TODO: fix invalid linking issue
// valid chars in encoded path and parameter: [-+~_%.a-zA-Z0-9/]
// sha1CurrentPattern matches string that represents a commit SHA, e.g. d8a994ef243349f321568f9e36d5c3f444b99cae
// Although SHA1 hashes are 40 chars long, the regex matches the hash from 7 to 40 chars in length
// so that abbreviated hash links can be used as well. This matches git and github useability.
// so that abbreviated hash links can be used as well. This matches git and GitHub usability.
sha1CurrentPattern = regexp.MustCompile(`(?:\s|^|\(|\[)([0-9a-f]{7,40})(?:\s|$|\)|\]|[.,](\s|$))`)
// shortLinkPattern matches short but difficult to parse [[name|link|arg=test]] syntax
shortLinkPattern = regexp.MustCompile(`\[\[(.*?)\]\](\w*)`)
// anySHA1Pattern allows to split url containing SHA into parts
anySHA1Pattern = regexp.MustCompile(`https?://(?:\S+/){4}([0-9a-f]{40})(/[^#\s]+)?(#\S+)?`)
// anySHA1Pattern splits url containing SHA into parts
anySHA1Pattern = regexp.MustCompile(`https?://(?:\S+/){4,5}([0-9a-f]{40})(/[-+~_%.a-zA-Z0-9/]+)?(#[-+~_%.a-zA-Z0-9]+)?`)
// comparePattern matches "http://domain/org/repo/compare/COMMIT1...COMMIT2#hash"
comparePattern = regexp.MustCompile(`https?://(?:\S+/){4,5}([0-9a-f]{40})(\.\.\.?)([0-9a-f]{40})?(#[-+~_%.a-zA-Z0-9]+)?`)
validLinksPattern = regexp.MustCompile(`^[a-z][\w-]+://`)
@ -65,7 +70,7 @@ var (
blackfridayExtRegex = regexp.MustCompile(`[^:]*:user-content-`)
// EmojiShortCodeRegex find emoji by alias like :smile:
EmojiShortCodeRegex = regexp.MustCompile(`:[\w\+\-]+:`)
EmojiShortCodeRegex = regexp.MustCompile(`:[-+\w]+:`)
)
// CSS class for action keywords (e.g. "closes: #1")
@ -152,6 +157,7 @@ type processor func(ctx *RenderContext, node *html.Node)
var defaultProcessors = []processor{
fullIssuePatternProcessor,
comparePatternProcessor,
fullSha1PatternProcessor,
shortLinkProcessor,
linkProcessor,
@ -178,6 +184,7 @@ func PostProcess(
var commitMessageProcessors = []processor{
fullIssuePatternProcessor,
comparePatternProcessor,
fullSha1PatternProcessor,
linkProcessor,
mentionProcessor,
@ -208,6 +215,7 @@ func RenderCommitMessage(
var commitMessageSubjectProcessors = []processor{
fullIssuePatternProcessor,
comparePatternProcessor,
fullSha1PatternProcessor,
linkProcessor,
mentionProcessor,
@ -920,12 +928,57 @@ func fullSha1PatternProcessor(ctx *RenderContext, node *html.Node) {
if hash != "" {
text += " (" + hash + ")"
}
replaceContent(node, start, end, createCodeLink(urlFull, text, "commit"))
node = node.NextSibling.NextSibling
}
}
func comparePatternProcessor(ctx *RenderContext, node *html.Node) {
if ctx.Metas == nil {
return
}
next := node.NextSibling
for node != nil && node != next {
m := comparePattern.FindStringSubmatchIndex(node.Data)
if m == nil {
return
}
urlFull := node.Data[m[0]:m[1]]
text1 := base.ShortSha(node.Data[m[2]:m[3]])
textDots := base.ShortSha(node.Data[m[4]:m[5]])
text2 := base.ShortSha(node.Data[m[6]:m[7]])
hash := ""
if m[9] > 0 {
hash = node.Data[m[8]:m[9]][1:]
}
start := m[0]
end := m[1]
// If url ends in '.', it's very likely that it is not part of the
// actual url but used to finish a sentence.
if strings.HasSuffix(urlFull, ".") {
end--
urlFull = urlFull[:len(urlFull)-1]
if hash != "" {
hash = hash[:len(hash)-1]
} else if text2 != "" {
text2 = text2[:len(text2)-1]
}
}
text := text1 + textDots + text2
if hash != "" {
text += " (" + hash + ")"
}
replaceContent(node, start, end, createCodeLink(urlFull, text, "compare"))
node = node.NextSibling.NextSibling
}
}
// emojiShortCodeProcessor for rendering text like :smile: into emoji
func emojiShortCodeProcessor(ctx *RenderContext, node *html.Node) {
start := 0
@ -1038,8 +1091,8 @@ func sha1CurrentPatternProcessor(ctx *RenderContext, node *html.Node) {
continue
}
replaceContent(node, m[2], m[3],
createCodeLink(util.URLJoin(setting.AppURL, ctx.Metas["user"], ctx.Metas["repo"], "commit", hash), base.ShortSha(hash), "commit"))
link := util.URLJoin(setting.AppURL, ctx.Metas["user"], ctx.Metas["repo"], "commit", hash)
replaceContent(node, m[2], m[3], createCodeLink(link, base.ShortSha(hash), "commit"))
start = 0
node = node.NextSibling.NextSibling
}