mirror of
https://github.com/go-gitea/gitea
synced 2024-11-19 00:11:24 +01:00
99efa02edf
This PR rewrites the invisible unicode detection algorithm to more closely match that of the Monaco editor on the system. It provides a technique for detecting ambiguous characters and relaxes the detection of combining marks. Control characters are in addition detected as invisible in this implementation whereas they are not on monaco but this is related to font issues. Close #19913 Signed-off-by: Andrew Thornton <art27@cantab.net>
58 lines
2.2 KiB
Go
58 lines
2.2 KiB
Go
// Copyright 2022 The Gitea Authors. All rights reserved.
|
|
// Use of this source code is governed by a MIT-style
|
|
// license that can be found in the LICENSE file.
|
|
|
|
//go:generate go run invisible/generate.go -v -o ./invisible_gen.go
|
|
|
|
//go:generate go run ambiguous/generate.go -v -o ./ambiguous_gen.go ambiguous/ambiguous.json
|
|
|
|
package charset
|
|
|
|
import (
|
|
"io"
|
|
"strings"
|
|
|
|
"code.gitea.io/gitea/modules/log"
|
|
"code.gitea.io/gitea/modules/translation"
|
|
)
|
|
|
|
// RuneNBSP is the codepoint for NBSP
|
|
const RuneNBSP = 0xa0
|
|
|
|
// EscapeControlHTML escapes the unicode control sequences in a provided html document
|
|
func EscapeControlHTML(text string, locale translation.Locale, allowed ...rune) (escaped *EscapeStatus, output string) {
|
|
sb := &strings.Builder{}
|
|
outputStream := &HTMLStreamerWriter{Writer: sb}
|
|
streamer := NewEscapeStreamer(locale, outputStream, allowed...).(*escapeStreamer)
|
|
|
|
if err := StreamHTML(strings.NewReader(text), streamer); err != nil {
|
|
streamer.escaped.HasError = true
|
|
log.Error("Error whilst escaping: %v", err)
|
|
}
|
|
return streamer.escaped, sb.String()
|
|
}
|
|
|
|
// EscapeControlReaders escapes the unicode control sequences in a provider reader and writer in a locale and returns the findings as an EscapeStatus and the escaped []byte
|
|
func EscapeControlReader(reader io.Reader, writer io.Writer, locale translation.Locale, allowed ...rune) (escaped *EscapeStatus, err error) {
|
|
outputStream := &HTMLStreamerWriter{Writer: writer}
|
|
streamer := NewEscapeStreamer(locale, outputStream, allowed...).(*escapeStreamer)
|
|
|
|
if err = StreamHTML(reader, streamer); err != nil {
|
|
streamer.escaped.HasError = true
|
|
log.Error("Error whilst escaping: %v", err)
|
|
}
|
|
return streamer.escaped, err
|
|
}
|
|
|
|
// EscapeControlString escapes the unicode control sequences in a provided string and returns the findings as an EscapeStatus and the escaped string
|
|
func EscapeControlString(text string, locale translation.Locale, allowed ...rune) (escaped *EscapeStatus, output string) {
|
|
sb := &strings.Builder{}
|
|
outputStream := &HTMLStreamerWriter{Writer: sb}
|
|
streamer := NewEscapeStreamer(locale, outputStream, allowed...).(*escapeStreamer)
|
|
|
|
if err := streamer.Text(text); err != nil {
|
|
streamer.escaped.HasError = true
|
|
log.Error("Error whilst escaping: %v", err)
|
|
}
|
|
return streamer.escaped, sb.String()
|
|
}
|