lib/libxml2
1
0
Fork 0
mirror of https://gitlab.gnome.org/GNOME/libxml2.git synced 2026-10-24 01:09:01 +03:00
Code Activity

xmlregexp: Add support for non-standard escapes in character groups

Browse Source 
Continue from https://gitlab.gnome.org/GNOME/libxml2/-/merge_requests/116
This commit is contained in:
Hieu Le Minh
2026-04-04 12:21:01 +07:00
parent 7736789aba
commit 07b13346cb
2 changed files with 39 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
python/tests/regexp.py
View File

@@ -29,7 +29,11 @@ check_regex(r"[\t-\r]", "\t\n\r", ["a", "(", ""])
check_regex("[\\t-\\r]", "\t\n\r", ["a", "(", ""])
check_regex(r"[\[-\]]", "[]\\", "Z^")
check_regex(r"[\*-/]", "*+,-./", "()0")
# Non-standard escape sequences
check_regex(r"[\u0041-\u005A]", "ABCMYZ", ["a", "0", "[", ""])
check_regex(r"[\u0009\u0061-\u007A]", "\tabcmyz", ["A", "0", "{", "", "\n"])
check_regex(r"[\!-\%]", '!"#$%', ["&", " ", "a", ""])
check_regex(r"[\:-\>]", ':;<=>', ["?", "9", "@", ""])
# Memory debug specific
libxml2.cleanupParser()

37
xmlregexp.c
View File

@@ -4896,6 +4896,12 @@ xmlFAParseCharClassEsc(xmlRegParserCtxtPtr ctxt) {
case 't':
cur = '\t';
break;
case 'u':
cur = parse_escaped_codepoint(ctxt);
if (cur < 0) {
return;
}
break;
}
xmlRegAtomAddRange(ctxt, ctxt->atom, ctxt->neg,
XML_REGEXP_CHARVAL, cur, cur, NULL);
@@ -4980,9 +4986,17 @@ xmlFAParseCharRange(xmlRegParserCtxtPtr ctxt) {
case 'n': start = 0xA; break;
case 'r': start = 0xD; break;
case 't': start = 0x9; break;
case 'u':
start = parse_escaped_codepoint(ctxt);
if (start < 0) {
return;
}
break;
case '\\': case '|': case '.': case '-': case '^': case '?':
case '*': case '+': case '{': case '}': case '(': case ')':
case '[': case ']':
case '[': case ']': case '!': case '"': case '#': case '$':
case '%': case ',': case '/': case ':': case ';': case '=':
case '>': case '@': case '`': case '~':
start = cur; break;
default:
ERROR("Invalid escape value");
@@ -5025,9 +5039,17 @@ xmlFAParseCharRange(xmlRegParserCtxtPtr ctxt) {
case 'n': end = 0xA; break;
case 'r': end = 0xD; break;
case 't': end = 0x9; break;
case 'u':
end = parse_escaped_codepoint(ctxt);
if (end < 0) {
return;
}
break;
case '\\': case '|': case '.': case '-': case '^': case '?':
case '*': case '+': case '{': case '}': case '(': case ')':
case '[': case ']':
case '[': case ']': case '!': case '"': case '#': case '$':
case '%': case ',': case '/': case ':': case ';': case '=':
case '>': case '@': case '`': case '~':
end = cur; break;
default:
ERROR("Invalid escape value");
@@ -5069,13 +5091,22 @@ xmlFAParsePosCharGroup(xmlRegParserCtxtPtr ctxt) {
case 'n': case 'r': case 't':
case '\\': case '|': case '.': case '-': case '^': case '?':
case '*': case '+': case '{': case '}': case '(': case ')':
case '[': case ']':
case '[': case ']': case '!': case '"': case '#': case '$':
case '%': case ',': case '/': case ':': case ';': case '=':
case '>': case '@': case '`': case '~':
if (NXT(2) == '-') {
xmlFAParseCharRange(ctxt);
} else {
xmlFAParseCharClassEsc(ctxt);
}
break;
case 'u':
if (NXT(6) == '-') {
xmlFAParseCharRange(ctxt);
} else {
xmlFAParseCharClassEsc(ctxt);
}
break;
default:
xmlFAParseCharClassEsc(ctxt);
}