mirrors/gotosocial
1
0
Fork 0
mirror of https://github.com/superseriousbusiness/gotosocial.git synced 2025-11-01 21:42:25 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[bugfix] Fix unicode-unaware word boundary check in hashtags (#1049)

Browse source 
* [bugfix] Fix unicode-unaware word boundary check in hashtag regex

Go `\b` does not care for Unicode, and without lookahead, the workarounds got
very ugly. So I replaced the regex with a parser.

The parser runs in O(n) time and performance should not be affected.

* [bugfix] Add back hashtag max length and add tests for it
This commit is contained in:
ugla 2022-11-15 16:05:34 +01:00 • committed by GitHub
commit 52109776f6
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 146 additions and 45 deletions
Download patch file Download diff file Expand all files Collapse all files

48
internal/text/common.go
View file

@ -27,36 +27,46 @@ import (
"github.com/superseriousbusiness/gotosocial/internal/gtsmodel"
"github.com/superseriousbusiness/gotosocial/internal/log"
"github.com/superseriousbusiness/gotosocial/internal/regexes"
"github.com/superseriousbusiness/gotosocial/internal/util"
)
func (f *formatter) ReplaceTags(ctx context.Context, in string, tags []*gtsmodel.Tag) string {
return regexes.ReplaceAllStringFunc(regexes.HashtagFinder, in, func(match string, buf *bytes.Buffer) string {
// we have a match
matchTrimmed := strings.TrimSpace(match)
tagAsEntered := matchTrimmed[1:]
spans := util.FindHashtagSpansInText(in)
if len(spans) == 0 {
return in
}
var b strings.Builder
i := 0
spans:
for _, t := range spans {
b.WriteString(in[i:t.First])
i = t.Second
tagAsEntered := in[t.First+1 : t.Second]
// check through the tags to find what we're matching
for _, tag := range tags {
if strings.EqualFold(tagAsEntered, tag.Name) {
// Add any dropped space from match
if unicode.IsSpace(rune(match[0])) {
buf.WriteByte(match[0])
}
// replace the #tag with the formatted tag content
// `<a href="tag.URL" class="mention hashtag" rel="tag">#<span>tagAsEntered</span></a>
buf.WriteString(`<a href="`)
buf.WriteString(tag.URL)
buf.WriteString(`" class="mention hashtag" rel="tag">#<span>`)
buf.WriteString(tagAsEntered)
buf.WriteString(`</span></a>`)
return buf.String()
b.WriteString(`<a href="`)
b.WriteString(tag.URL)
b.WriteString(`" class="mention hashtag" rel="tag">#<span>`)
b.WriteString(tagAsEntered)
b.WriteString(`</span></a>`)
continue spans
}
}
// the match wasn't in the list of tags for whatever reason, so just return the match as we found it so nothing changes
return match
})
b.WriteString(in[t.First:t.Second])
}
// Get the last bits.
i = spans[len(spans)-1].Second
b.WriteString(in[i:])
return b.String()
}
func (f *formatter) ReplaceMentions(ctx context.Context, in string, mentions []*gtsmodel.Mention) string {