Skip to content

bpo-23689: re module, fix memory leak when a match is terminated by a signal or memory allocation failure #32283

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

Merged
serhiy-storchaka merged 11 commits into from Apr 3, 2022
Merged

bpo-23689: re module, fix memory leak when a match is terminated by a signal or memory allocation failure #32283

Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
e5317fe
1. add variables
wjssz Apr 3, 2022
28e4d2d
2. add _CompileData class to _compile.py
wjssz Apr 3, 2022
7493fad
3. add repeat_count parameter to _sre.compile() function
wjssz Apr 3, 2022
b928335
4. emit repeat_count in sre_compile.py
wjssz Apr 3, 2022
dfec05d
5. change _validate_outer() parameter
wjssz Apr 3, 2022
db7e88b
6. validate in _validate_inner() / _validate_outer()
wjssz Apr 3, 2022
a43cb6e
7. allocate repeats_array for SRE_STATE
wjssz Apr 3, 2022
1560a0c
8. support code in sre_lib.h
wjssz Apr 3, 2022
96be025
9. add unit-tests
wjssz Apr 3, 2022
7ee3e66
a. skip -= field_number
wjssz Apr 3, 2022
e95b19f
b. improve unit-test
wjssz Apr 3, 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 3, 2022
commit dfec05de5ca6e68303f2c6b6cbca5c34fc772219
34 changes: 17 additions & 17 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 @@ -1667,8 +1667,8 @@ _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) {
VTRACE(("arg=%d, groups=%d\n", (int)arg, (int)groups));
if (arg > 2 * (size_t)self->groups + 1) {
VTRACE(("arg=%d, groups=%d\n", (int)arg, (int)self->groups));
FAIL;
}
break;
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