Mirrors/uBlock
1
0
Fork 0
mirror of https://github.com/gorhill/uBlock.git synced 2024-11-02 00:42:45 +01:00
Code Issues Releases Wiki Activity

Handle properly Unicode characters in static network filters

Browse Source 
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/772

Unicode characters inside hostname part of a filter will
be converted to punycode.

Unicode characters anywhere else in the pattern will be
percent-encoded.

Unicode characters which cannot be encoded will cause a
filter to be invalid.
This commit is contained in:
Raymond Hill 2020-07-04 14:47:33 -04:00
parent a85a908f09
commit 0da34f7edf
No known key found for this signature in database
GPG Key ID: 25E1490B761470C2
2 changed files with 35 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
src/js/static-filtering-parser.js
View File

@ -104,6 +104,8 @@ const Parser = class {
this.maxTokenLength = Number.MAX_SAFE_INTEGER;
this.reIsLocalhostRedirect = /(?:0\.0\.0\.0|(?:broadcast|local)host|local|ip6-\w+)\b/;
this.reHostname = /^[^\x00-\x24\x26-\x29\x2B\x2C\x2F\x3A-\x5E\x60\x7B-\x7F]+/;
this.reUnicodeChar = /[^\x00-\x7F]/;
this.reUnicodeChars = /[^\x00-\x7F]/g;
this.punycoder = new URL(self.location);
this.selectorCompiler = new this.SelectorCompiler(this);
// TODO: reuse for network filtering analysis
@ -631,7 +633,7 @@ const Parser = class {
} else if (
this.patternIsDubious() || (
this.patternHasUnicode() &&
this.toPunycode(true) === false
this.toASCII(true) === false
)
) {
this.markSpan(this.patternSpan, BITError);
@ -1047,12 +1049,17 @@ const Parser = class {
// https://github.com/uBlockOrigin/uBlock-issues/issues/1118#issuecomment-650730158
// Be ready to deal with non-punycode-able Unicode characters.
toPunycode(dryrun = false) {
// https://github.com/uBlockOrigin/uBlock-issues/issues/772
// Encode Unicode characters beyond the hostname part.
toASCII(dryrun = false) {
if ( this.patternHasUnicode() === false ) { return true; }
const { i, len } = this.patternSpan;
if ( len === 0 ) { return true; }
const patternIsRegex = this.patternIsRegex();
let pattern = this.getNetPattern();
const match = this.reHostname.exec(this.pattern);
// Punycode hostname part of the pattern.
if ( patternIsRegex === false ) {
const match = this.reHostname.exec(pattern);
if ( match === null ) { return true; }
try {
this.punycoder.hostname = match[0].replace(/\*/g, '__asterisk__');
@ -1061,9 +1068,24 @@ const Parser = class {
}
const hn = this.punycoder.hostname;
if ( hn === '' ) { return false; }
if ( dryrun ) { return true; }
const punycoded = hn.replace(/__asterisk__/g, '*');
pattern = punycoded + this.pattern.slice(match.index + match[0].length);
pattern = punycoded + pattern.slice(match.index + match[0].length);
}
// Percent-encode remaining Unicode characters.
if ( this.reUnicodeChar.test(pattern) ) {
try {
pattern = pattern.replace(
this.reUnicodeChars,
s => encodeURIComponent(s)
);
} catch (ex) {
return false;
}
}
if ( dryrun ) { return true; }
if ( patternIsRegex ) {
pattern = `/${pattern}/`;
}
const beg = this.slices[i+1];
const end = this.slices[i+len+1];
const raw = this.raw.slice(0, beg) + pattern + this.raw.slice(end);

2
src/js/storage.js
View File

@ -825,7 +825,7 @@ self.addEventListener('hiddenSettingsChanged', ( ) => {
// https://github.com/gorhill/uBlock/issues/2599
// convert hostname to punycode if needed
if ( parser.patternHasUnicode() && parser.toPunycode() === false ) {
if ( parser.patternHasUnicode() && parser.toASCII() === false ) {
continue;
}
staticNetFilteringEngine.compile(parser, writer);