hohn/codeql
1
0
Fork 0
mirror of https://github.com/github/codeql.git synced 2026-04-29 18:55:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #13779 from geoffw0/pythonparsemode

Browse Source 
Python: Understand multiple parse mode flags specified in a regular expression string
This commit is contained in:
yoff
2023-08-24 21:20:45 +02:00
committed by GitHub
parent 763216b932 f07f97a94e
commit a834703195
10 changed files with 54 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/ql/lib/change-notes/2023-07-20-regex-parse-modes.md Normal file
View File

@@ -0,0 +1,4 @@
---
category: minorAnalysis
---
* Regular expressions containing multiple parse mode flags are now interpretted correctly. For example `"(?is)abc.*"` with both the `i` and `s` flags.

29
python/ql/lib/semmle/python/regexp/internal/ParseRegExp.qll
View File

@@ -617,7 +617,7 @@ class RegExp extends Expr instanceof StrConst {
private predicate group_start(int start, int end) {
this.non_capturing_group_start(start, end)
or
this.flag_group_start(start, end, _)
this.flag_group_start(start, end)
or
this.named_group_start(start, end)
or
@@ -679,12 +679,29 @@ class RegExp extends Expr instanceof StrConst {
end = min(int i | i > start + 4 and this.getChar(i) = "?")
}
private predicate flag_group_start(int start, int end, string c) {
/**
* Holds if a parse mode starts between `start` and `end`.
*/
private predicate flag_group_start(int start, int end) {
this.isGroupStart(start) and
this.getChar(start + 1) = "?" and
end = start + 3 and
c = this.getChar(start + 2) and
c in ["i", "L", "m", "s", "u", "x"]
this.getChar(start + 2) in ["i", "L", "m", "s", "u", "x"] and
end = start + 2
}
/**
* Holds if a parse mode group is between `start` and `end`, and includes the
* mode flag `c`. For example the following span, with mode flag `i`:
* ```
* (?i)
* ```
*/
private predicate flag_group(int start, int end, string c) {
exists(int inStart, int inEnd |
this.flag_group_start(start, inStart) and
this.groupContents(start, end, inStart, inEnd) and
this.getChar([inStart .. inEnd - 1]) = c
)
}
/**
@@ -692,7 +709,7 @@ class RegExp extends Expr instanceof StrConst {
* it is defined by a prefix.
*/
string getModeFromPrefix() {
exists(string c | this.flag_group_start(_, _, c) |
exists(string c | this.flag_group(_, _, c) |
c = "i" and result = "IGNORECASE"
or
c = "L" and result = "LOCALE"