portugalfuturista/rizin
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Don't match numbers with a '.', '*' and '\w' prefix.

Browse source 
The exclusion pattern [^\w.*] excludes all numbers with a '.' and '*' prefix.
This is only done because the previous parsing did it like this.
And not doing it, breaks substitutions from before. E.g. when addresses,
were replaced with flag names which include numbers (case labels like: case.<switch-address>.<case-number>).
See: https://github.com/rizinorg/rizin/issues/4238 for more details of this problem.
This commit is contained in:
Rot127 2024-02-16 00:30:50 -05:00 committed by Anton Kochkov
parent 6fcc9f6c6a
commit 2fde7808a0
1 changed files with 10 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
librz/parse/filter.c
View file

@ -82,8 +82,9 @@ static void replaceWords(char *s, const char *k, const char *v) {
}
/**
* \brief Returns a pointer to the next number in the string.
* \brief Returns a pointer to the next number (hex and decimal) in the string.
* This function ignores numbers from ansi excape codes.
* It also ignroes numbers with a prefix of: '\w', '*', '.'.
*
* \param str The string to search the number in.
*
@ -95,7 +96,14 @@ static char *find_next_number(char *str) {
if (rz_regex_contains("\x1b\\[[0-9;]*m", str, RZ_REGEX_ZERO_TERMINATED, RZ_REGEX_EXTENDED, 0)) {
ansi = true;
}
const char *search = ansi ? "(\x1b\\[[0-9;]*m)(?<number>(0x[a-fA-F0-9]+)|\\d+)" : "(\\W(?<number>(0x[a-fA-F0-9]+)|\\d+))";
const char *search = (ansi ? "(\x1b\\[[0-9;]*m)(?<number>(0x[a-fA-F0-9]+)|\\d+)" :
// The exclusion pattern [^\w.*] excludes all numbers with a '\w', '.' and '*' prefix.
// This is only done because the previous parsing did it like this.
// And not doing it, breaks substitutions from before.
// Because previously inserted flag names in the opstr, can contain numbers
// (e.g. case labels like: case.<switch-address>.<case-number>).
// See: https://github.com/rizinorg/rizin/issues/4238 for more details of this problem.
"(([^\\w.*]|^)(?<number>(0x[a-fA-F0-9]+)|\\d+))");
RzRegex *re = rz_regex_new(search, RZ_REGEX_EXTENDED, RZ_REGEX_DEFAULT);
RzPVector *match = rz_regex_match_first(re, str, RZ_REGEX_ZERO_TERMINATED, 0, RZ_REGEX_DEFAULT);
if (rz_pvector_empty(match)) {