Skip to content

bpo-23689: re module, fix memory leak when a match is terminated by a signal or memory allocation failure #32223

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
ghost wants to merge 8 commits into from
Closed

bpo-23689: re module, fix memory leak when a match is terminated by a signal or memory allocation failure #32223

Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
d43cefe
1. add variables
wjssz Mar 31, 2022
f0d2621
2. add CompileData class to sre_compile.py
wjssz Mar 31, 2022
b8b9921
3. add repeat_count parameter to _sre.compile() function
wjssz Mar 31, 2022
17907f5
4. emit repeat_count in sre_compile.py
wjssz Mar 31, 2022
787ca6f
5. change _validate_outer() parameter
wjssz Mar 31, 2022
6ec932b
6. validate in _validate_inner()
wjssz Mar 31, 2022
01c4c29
7. allocate repeats_array for SRE_STATE
wjssz Mar 31, 2022
7631b7d
8. support code in sre_lib.h
wjssz Mar 31, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
5. change _validate_outer() parameter 
-_validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
+_validate_inner(SRE_CODE *code, SRE_CODE *end, PatternObject *self)
  • Loading branch information
@wjssz
wjssz committed Apr 1, 2022
commit 787ca6f158d9f4db3b4692d0920bfbd66ed075eb
32 changes: 16 additions & 16 deletions Modules/_sre.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1646,7 +1646,7 @@ _validate_charset(SRE_CODE *code, SRE_CODE *end)
}

static int
_validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
_validate_inner(SRE_CODE *code, SRE_CODE *end, PatternObject *self)
{
/* Some variables are manipulated by the macros above */
SRE_CODE op;
Expand All @@ -1667,7 +1667,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
sre_match() code is robust even if they don't, and the worst
you can get is nonsensical match results. */
GET_ARG;
if (arg > 2 * (size_t)groups + 1) {
if (arg > 2 * (size_t)self->groups + 1) {
VTRACE(("arg=%d, groups=%d\n", (int)arg, (int)groups));
FAIL;
}
Expand Down Expand Up @@ -1797,7 +1797,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
if (skip == 0)
break;
/* Stop 2 before the end; we check the JUMP below */
if (!_validate_inner(code, code+skip-3, groups))
if (!_validate_inner(code, code+skip-3, self))
FAIL;
code += skip-3;
/* Check that it ends with a JUMP, and that each JUMP
Expand Down Expand Up @@ -1826,7 +1826,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
FAIL;
if (max > SRE_MAXREPEAT)
FAIL;
if (!_validate_inner(code, code+skip-4, groups))
if (!_validate_inner(code, code+skip-4, self))
FAIL;
code += skip-4;
GET_OP;
Expand All @@ -1846,7 +1846,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
FAIL;
if (max > SRE_MAXREPEAT)
FAIL;
if (!_validate_inner(code, code+skip-3, groups))
if (!_validate_inner(code, code+skip-3, self))
FAIL;
code += skip-3;
GET_OP;
Expand All @@ -1864,7 +1864,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
case SRE_OP_ATOMIC_GROUP:
{
GET_SKIP;
if (!_validate_inner(code, code+skip-2, groups))
if (!_validate_inner(code, code+skip-2, self))
FAIL;
code += skip-2;
GET_OP;
Expand All @@ -1878,7 +1878,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
case SRE_OP_GROUPREF_UNI_IGNORE:
case SRE_OP_GROUPREF_LOC_IGNORE:
GET_ARG;
if (arg >= (size_t)groups)
if (arg >= (size_t)self->groups)
FAIL;
break;

Expand All @@ -1887,7 +1887,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
'group' is either an integer group number or a group name,
'then' and 'else' are sub-regexes, and 'else' is optional. */
GET_ARG;
if (arg >= (size_t)groups)
if (arg >= (size_t)self->groups)
FAIL;
GET_SKIP_ADJ(1);
code--; /* The skip is relative to the first arg! */
Expand Down Expand Up @@ -1920,17 +1920,17 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
code[skip-3] == SRE_OP_JUMP)
{
VTRACE(("both then and else parts present\n"));
if (!_validate_inner(code+1, code+skip-3, groups))
if (!_validate_inner(code+1, code+skip-3, self))
FAIL;
code += skip-2; /* Position after JUMP, at <skipno> */
GET_SKIP;
if (!_validate_inner(code, code+skip-1, groups))
if (!_validate_inner(code, code+skip-1, self))
FAIL;
code += skip-1;
}
else {
VTRACE(("only a then part present\n"));
if (!_validate_inner(code+1, code+skip-1, groups))
if (!_validate_inner(code+1, code+skip-1, self))
FAIL;
code += skip-1;
}
Expand All @@ -1944,7 +1944,7 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
if (arg & 0x80000000)
FAIL; /* Width too large */
/* Stop 1 before the end; we check the SUCCESS below */
if (!_validate_inner(code+1, code+skip-2, groups))
if (!_validate_inner(code+1, code+skip-2, self))
FAIL;
code += skip-2;
GET_OP;
Expand All @@ -1963,18 +1963,18 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
}

static int
_validate_outer(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
_validate_outer(SRE_CODE *code, SRE_CODE *end, PatternObject *self)
{
if (groups < 0 || (size_t)groups > SRE_MAXGROUPS ||
if (self->groups < 0 || (size_t)self->groups > SRE_MAXGROUPS ||
code >= end || end[-1] != SRE_OP_SUCCESS)
FAIL;
return _validate_inner(code, end-1, groups);
return _validate_inner(code, end-1, self);
}

static int
_validate(PatternObject *self)
{
if (!_validate_outer(self->code, self->code+self->codesize, self->groups))
if (!_validate_outer(self->code, self->code+self->codesize, self))
{
PyErr_SetString(PyExc_RuntimeError, "invalid SRE code");
return 0;
Expand Down