Run processors on whole of text (#16155) (#16185)

Backport #16155

There is an inefficiency in the design of our processors which means that Emoji
and other processors run in order n^2 time.

This PR forces the processors to process the entirety of text node before passing
back up. The fundamental inefficiency remains but it should be significantly
ameliorated.

Signed-off-by: Andrew Thornton <art27@cantab.net>
mj-v1.14.3
zeripath 3 years ago committed by GitHub
parent 849d316d8d
commit 5ff807acde
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -6,6 +6,7 @@
package emoji
import (
"io"
"sort"
"strings"
"sync"
@ -145,6 +146,8 @@ func (n *rememberSecondWriteWriter) Write(p []byte) (int, error) {
if n.writecount == 2 {
n.idx = n.pos
n.end = n.pos + len(p)
n.pos += len(p)
return len(p), io.EOF
}
n.pos += len(p)
return len(p), nil
@ -155,6 +158,8 @@ func (n *rememberSecondWriteWriter) WriteString(s string) (int, error) {
if n.writecount == 2 {
n.idx = n.pos
n.end = n.pos + len(s)
n.pos += len(s)
return len(s), io.EOF
}
n.pos += len(s)
return len(s), nil

@ -87,6 +87,7 @@ func isLinkStr(link string) bool {
return validLinksPattern.MatchString(link)
}
// FIXME: This function is not concurrent safe
func getIssueFullPattern() *regexp.Regexp {
if issueFullPattern == nil {
issueFullPattern = regexp.MustCompile(regexp.QuoteMeta(setting.AppURL) +
@ -605,11 +606,16 @@ func replaceContentList(node *html.Node, i, j int, newNodes []*html.Node) {
}
func mentionProcessor(ctx *postProcessCtx, node *html.Node) {
start := 0
next := node.NextSibling
for node != nil && node != next && start < len(node.Data) {
// We replace only the first mention; other mentions will be addressed later
found, loc := references.FindFirstMentionBytes([]byte(node.Data))
found, loc := references.FindFirstMentionBytes([]byte(node.Data[start:]))
if !found {
return
}
loc.Start += start
loc.End += start
mention := node.Data[loc.Start:loc.End]
var teams string
teams, ok := ctx.metas["teams"]
@ -621,10 +627,17 @@ func mentionProcessor(ctx *postProcessCtx, node *html.Node) {
mentionOrgAndTeam := strings.Split(mention, "/")
if mentionOrgAndTeam[0][1:] == ctx.metas["org"] && strings.Contains(teams, ","+strings.ToLower(mentionOrgAndTeam[1])+",") {
replaceContent(node, loc.Start, loc.End, createLink(util.URLJoin(setting.AppURL, "org", ctx.metas["org"], "teams", mentionOrgAndTeam[1]), mention, "mention"))
node = node.NextSibling.NextSibling
start = 0
continue
}
return
start = loc.End
continue
}
replaceContent(node, loc.Start, loc.End, createLink(util.URLJoin(setting.AppURL, mention[1:]), mention, "mention"))
node = node.NextSibling.NextSibling
start = 0
}
}
func shortLinkProcessor(ctx *postProcessCtx, node *html.Node) {
@ -632,6 +645,8 @@ func shortLinkProcessor(ctx *postProcessCtx, node *html.Node) {
}
func shortLinkProcessorFull(ctx *postProcessCtx, node *html.Node, noLink bool) {
next := node.NextSibling
for node != nil && node != next {
m := shortLinkPattern.FindStringSubmatchIndex(node.Data)
if m == nil {
return
@ -711,7 +726,7 @@ func shortLinkProcessorFull(ctx *postProcessCtx, node *html.Node, noLink bool) {
switch ext := filepath.Ext(link); ext {
// fast path: empty string, ignore
case "":
break
// leave image as false
case ".jpg", ".jpeg", ".png", ".tif", ".tiff", ".webp", ".gif", ".bmp", ".ico", ".svg":
image = true
}
@ -787,12 +802,16 @@ func shortLinkProcessorFull(ctx *postProcessCtx, node *html.Node, noLink bool) {
linkNode.Attr = []html.Attribute{{Key: "href", Val: link}}
}
replaceContent(node, m[0], m[1], linkNode)
node = node.NextSibling.NextSibling
}
}
func fullIssuePatternProcessor(ctx *postProcessCtx, node *html.Node) {
if ctx.metas == nil {
return
}
next := node.NextSibling
for node != nil && node != next {
m := getIssueFullPattern().FindStringSubmatchIndex(node.Data)
if m == nil {
return
@ -810,23 +829,25 @@ func fullIssuePatternProcessor(ctx *postProcessCtx, node *html.Node) {
// TODO if m[4]:m[5] is not nil, then link is to a comment,
// and we should indicate that in the text somehow
replaceContent(node, m[0], m[1], createLink(link, id, "ref-issue"))
} else {
orgRepoID := matchOrg + "/" + matchRepo + id
replaceContent(node, m[0], m[1], createLink(link, orgRepoID, "ref-issue"))
}
node = node.NextSibling.NextSibling
}
}
func issueIndexPatternProcessor(ctx *postProcessCtx, node *html.Node) {
if ctx.metas == nil {
return
}
var (
found bool
ref *references.RenderizableReference
)
next := node.NextSibling
for node != nil && node != next {
_, exttrack := ctx.metas["format"]
alphanum := ctx.metas["style"] == IssueNameStyleAlphanumeric
@ -867,7 +888,8 @@ func issueIndexPatternProcessor(ctx *postProcessCtx, node *html.Node) {
if ref.Action == references.XRefActionNone {
replaceContent(node, ref.RefLocation.Start, ref.RefLocation.End, link)
return
node = node.NextSibling.NextSibling
continue
}
// Decorate action keywords if actionable
@ -885,6 +907,8 @@ func issueIndexPatternProcessor(ctx *postProcessCtx, node *html.Node) {
Data: node.Data[ref.ActionLocation.End:ref.RefLocation.Start],
}
replaceContentList(node, ref.ActionLocation.Start, ref.RefLocation.End, []*html.Node{keyword, spaces, link})
node = node.NextSibling.NextSibling.NextSibling.NextSibling
}
}
// fullSha1PatternProcessor renders SHA containing URLs
@ -892,6 +916,9 @@ func fullSha1PatternProcessor(ctx *postProcessCtx, node *html.Node) {
if ctx.metas == nil {
return
}
next := node.NextSibling
for node != nil && node != next {
m := anySHA1Pattern.FindStringSubmatchIndex(node.Data)
if m == nil {
return
@ -936,15 +963,23 @@ func fullSha1PatternProcessor(ctx *postProcessCtx, node *html.Node) {
}
replaceContent(node, start, end, createCodeLink(urlFull, text, "commit"))
node = node.NextSibling.NextSibling
}
}
// emojiShortCodeProcessor for rendering text like :smile: into emoji
func emojiShortCodeProcessor(ctx *postProcessCtx, node *html.Node) {
m := EmojiShortCodeRegex.FindStringSubmatchIndex(node.Data)
start := 0
next := node.NextSibling
for node != nil && node != next && start < len(node.Data) {
m := EmojiShortCodeRegex.FindStringSubmatchIndex(node.Data[start:])
if m == nil {
return
}
m[0] += start
m[1] += start
start = m[1]
alias := node.Data[m[0]:m[1]]
alias = strings.ReplaceAll(alias, ":", "")
@ -954,25 +989,39 @@ func emojiShortCodeProcessor(ctx *postProcessCtx, node *html.Node) {
s := strings.Join(setting.UI.Reactions, " ") + "gitea"
if strings.Contains(s, alias) {
replaceContent(node, m[0], m[1], createCustomEmoji(alias, "emoji"))
return
node = node.NextSibling.NextSibling
start = 0
continue
}
return
continue
}
replaceContent(node, m[0], m[1], createEmoji(converted.Emoji, "emoji", converted.Description))
node = node.NextSibling.NextSibling
start = 0
}
}
// emoji processor to match emoji and add emoji class
func emojiProcessor(ctx *postProcessCtx, node *html.Node) {
m := emoji.FindEmojiSubmatchIndex(node.Data)
start := 0
next := node.NextSibling
for node != nil && node != next && start < len(node.Data) {
m := emoji.FindEmojiSubmatchIndex(node.Data[start:])
if m == nil {
return
}
m[0] += start
m[1] += start
codepoint := node.Data[m[0]:m[1]]
start = m[1]
val := emoji.FromCode(codepoint)
if val != nil {
replaceContent(node, m[0], m[1], createEmoji(codepoint, "emoji", val.Description))
node = node.NextSibling.NextSibling
start = 0
}
}
}
@ -982,10 +1031,17 @@ func sha1CurrentPatternProcessor(ctx *postProcessCtx, node *html.Node) {
if ctx.metas == nil || ctx.metas["user"] == "" || ctx.metas["repo"] == "" || ctx.metas["repoPath"] == "" {
return
}
m := sha1CurrentPattern.FindStringSubmatchIndex(node.Data)
start := 0
next := node.NextSibling
for node != nil && node != next && start < len(node.Data) {
m := sha1CurrentPattern.FindStringSubmatchIndex(node.Data[start:])
if m == nil {
return
}
m[2] += start
m[3] += start
hash := node.Data[m[2]:m[3]]
// The regex does not lie, it matches the hash pattern.
// However, a regex cannot know if a hash actually exists or not.
@ -999,32 +1055,45 @@ func sha1CurrentPatternProcessor(ctx *postProcessCtx, node *html.Node) {
if !strings.Contains(err.Error(), "fatal: Needed a single revision") {
log.Debug("sha1CurrentPatternProcessor git rev-parse: %v", err)
}
return
start = m[3]
continue
}
replaceContent(node, m[2], m[3],
createCodeLink(util.URLJoin(setting.AppURL, ctx.metas["user"], ctx.metas["repo"], "commit", hash), base.ShortSha(hash), "commit"))
start = 0
node = node.NextSibling.NextSibling
}
}
// emailAddressProcessor replaces raw email addresses with a mailto: link.
func emailAddressProcessor(ctx *postProcessCtx, node *html.Node) {
next := node.NextSibling
for node != nil && node != next {
m := emailRegex.FindStringSubmatchIndex(node.Data)
if m == nil {
return
}
mail := node.Data[m[2]:m[3]]
replaceContent(node, m[2], m[3], createLink("mailto:"+mail, mail, "mailto"))
node = node.NextSibling.NextSibling
}
}
// linkProcessor creates links for any HTTP or HTTPS URL not captured by
// markdown.
func linkProcessor(ctx *postProcessCtx, node *html.Node) {
next := node.NextSibling
for node != nil && node != next {
m := common.LinkRegex.FindStringIndex(node.Data)
if m == nil {
return
}
uri := node.Data[m[0]:m[1]]
replaceContent(node, m[0], m[1], createLink(uri, uri, "link"))
node = node.NextSibling.NextSibling
}
}
func genDefaultLinkProcessor(defaultLink string) processor {
@ -1048,12 +1117,17 @@ func genDefaultLinkProcessor(defaultLink string) processor {
// descriptionLinkProcessor creates links for DescriptionHTML
func descriptionLinkProcessor(ctx *postProcessCtx, node *html.Node) {
next := node.NextSibling
for node != nil && node != next {
m := common.LinkRegex.FindStringIndex(node.Data)
if m == nil {
return
}
uri := node.Data[m[0]:m[1]]
replaceContent(node, m[0], m[1], createDescriptionLink(uri, uri))
node = node.NextSibling.NextSibling
}
}
func createDescriptionLink(href, content string) *html.Node {

@ -425,3 +425,19 @@ func TestIssue16020(t *testing.T) {
assert.NoError(t, err)
assert.Equal(t, data, string(res))
}
func BenchmarkEmojiPostprocess(b *testing.B) {
data := "🥰 "
for len(data) < 1<<16 {
data += data
}
b.ResetTimer()
for i := 0; i < b.N; i++ {
_, err := PostProcess(
[]byte(data),
"https://example.com",
localMetas,
false)
assert.NoError(b, err)
}
}

Loading…
Cancel
Save