Repair widespread misuse of _PyString_Resize. Since it's clear people

don't understand how this function works, also beefed up the docs.  The
most common usage error is of this form (often spread out across gotos):

	if (_PyString_Resize(&s, n) < 0) {
		Py_DECREF(s);
		s = NULL;
		goto outtahere;
	}

The error is that if _PyString_Resize runs out of memory, it automatically
decrefs the input string object s (which also deallocates it, since its
refcount must be 1 upon entry), and sets s to NULL.  So if the "if"
branch ever triggers, it's an error to call Py_DECREF(s):  s is already
NULL!  A correct way to write the above is the simpler (and intended)

	if (_PyString_Resize(&s, n) < 0)
		goto outtahere;

Bugfix candidate.

Doc/api/concrete.tex

@@ -586,7 +586,16 @@ parameter and are called with a non-string parameter.
 \begin{cfuncdesc}{int}{_PyString_Resize}{PyObject **string, int newsize}
   A way to resize a string object even though it is ``immutable''.
   Only use this to build up a brand new string object; don't use this
-  if the string may already be known in other parts of the code.
+  if the string may already be known in other parts of the code.  It
+  is an error to call this function if the refcount on the input string
+  object is not one.
+  Pass the address of an existing string object as an lvalue (it may
+  be written into), and the new size desired.  On success, \var{*string}
+  holds the resized string object and 0 is returned; the address in
+  \var{*string} may differ from its input value.  If the
+  reallocation fails, the original string object at \var{*string} is
+  deallocated, \var{*string} is set to \NULL{}, a memory exception is set,
+  and -1 is returned.
 \end{cfuncdesc}
 
 \begin{cfuncdesc}{PyObject*}{PyString_Format}{PyObject *format,

Modules/_ssl.c

@@ -331,8 +331,8 @@ static PyObject *PySSL_SSLread(PySSLObject *self, PyObject *args)
 		Py_DECREF(buf);
 		return PySSL_SetError(self, count);
 	}
-	if (count != len && _PyString_Resize(&buf, count) < 0)
+	if (count != len)
-		return NULL;
+		_PyString_Resize(&buf, count);
 	return buf;
 }
 

Modules/cPickle.c

@@ -1252,14 +1252,8 @@ modified_EncodeRawUnicodeEscape(const Py_UNICODE *s, int size)
 			*p++ = (char) ch;
 	}
 	*p = '\0';
-	if (_PyString_Resize(&repr, p - q))
+	_PyString_Resize(&repr, p - q);
-		goto onError;
-
 	return repr;
-
-  onError:
-	Py_DECREF(repr);
-	return NULL;
 }
 
 

Modules/cdmodule.c

@@ -251,8 +251,7 @@ CD_readda(cdplayerobject *self, PyObject *args)
 		return NULL;
 	}
 	if (n < numframes)
-		if (_PyString_Resize(&result, n * sizeof(CDFRAME)))
+		_PyString_Resize(&result, n * sizeof(CDFRAME));
-			return NULL;
 
 	return result;
 }

Modules/clmodule.c

@@ -135,8 +135,7 @@ cl_CompressImage(PyObject *self, PyObject *args)
 	}
 
 	if (compressedBufferSize < frameBufferSize)
-		if (_PyString_Resize(&compressedBuffer, compressedBufferSize))
+		_PyString_Resize(&compressedBuffer, compressedBufferSize);
-			return NULL;
 
 	return compressedBuffer;
 }

Modules/linuxaudiodev.c

@@ -158,8 +158,7 @@ lad_read(lad_t *self, PyObject *args)
         return NULL;
     }
     self->x_icount += count;
-    if (_PyString_Resize(&rv, count) == -1)
+    _PyString_Resize(&rv, count);
-	return NULL;
     return rv;
 }
 

Modules/regexmodule.c

@@ -516,11 +516,8 @@ symcomp(PyObject *pattern, PyObject *gdict)
 		return NULL;
 	}
 	/* _PyString_Resize() decrements npattern on failure */
-	if (_PyString_Resize(&npattern, n - v) == 0)
+	_PyString_Resize(&npattern, n - v);
-		return npattern;
+	return npattern;
-	else {
-		return NULL;
-	}
 
 }
 

Modules/socketmodule.c

@@ -1402,8 +1402,8 @@ PySocketSock_recv(PySocketSockObject *s, PyObject *args)
 		Py_DECREF(buf);
 		return s->errorhandler();
 	}
-	if (n != len && _PyString_Resize(&buf, n) < 0)
+	if (n != len)
-		return NULL;
+		_PyString_Resize(&buf, n);
 	return buf;
 }
 

Modules/stropmodule.c

@@ -215,10 +215,8 @@ strop_joinfields(PyObject *self, PyObject *args)
 			}
 			slen = PyString_GET_SIZE(item);
 			while (reslen + slen + seplen >= sz) {
-				if (_PyString_Resize(&res, sz * 2)) {
+				if (_PyString_Resize(&res, sz * 2) < 0)
-					Py_DECREF(res);
 					return NULL;
-				}
 				sz *= 2;
 				p = PyString_AsString(res) + reslen;
 			}
@@ -231,10 +229,7 @@ strop_joinfields(PyObject *self, PyObject *args)
 			p += slen;
 			reslen += slen;
 		}
-		if (_PyString_Resize(&res, reslen)) {
+		_PyString_Resize(&res, reslen);
-			Py_DECREF(res);
-			res = NULL;
-		}
 		return res;
 	}
 
@@ -257,8 +252,7 @@ strop_joinfields(PyObject *self, PyObject *args)
 		}
 		slen = PyString_GET_SIZE(item);
 		while (reslen + slen + seplen >= sz) {
-			if (_PyString_Resize(&res, sz * 2)) {
+			if (_PyString_Resize(&res, sz * 2) < 0) {
-				Py_DECREF(res);
 				Py_DECREF(item);
 				return NULL;
 			}
@@ -275,10 +269,7 @@ strop_joinfields(PyObject *self, PyObject *args)
 		reslen += slen;
 		Py_DECREF(item);
 	}
-	if (_PyString_Resize(&res, reslen)) {
+	_PyString_Resize(&res, reslen);
-		Py_DECREF(res);
-		res = NULL;
-	}
 	return res;
 }
 
@@ -989,8 +980,8 @@ strop_translate(PyObject *self, PyObject *args)
 		return input_obj;
 	}
 	/* Fix the size of the resulting string */
-	if (inlen > 0 &&_PyString_Resize(&result, output-output_start))
+	if (inlen > 0)
-		return NULL;
+		_PyString_Resize(&result, output - output_start);
 	return result;
 }
 

Modules/zlibmodule.c

@@ -255,9 +255,8 @@ PyZlib_decompress(PyObject *self, PyObject *args)
 	    /* fall through */
 	case(Z_OK):
 	    /* need more memory */
-	    if (_PyString_Resize(&result_str, r_strlen << 1) == -1) {
+	    if (_PyString_Resize(&result_str, r_strlen << 1) < 0) {
 		inflateEnd(&zst);
-		result_str = NULL;
 		goto error;
 	    }
 	    zst.next_out = (unsigned char *)PyString_AS_STRING(result_str) \
@@ -414,10 +413,8 @@ PyZlib_objcompress(compobject *self, PyObject *args)
     /* while Z_OK and the output buffer is full, there might be more output,
        so extend the output buffer and try again */
     while (err == Z_OK && self->zst.avail_out == 0) {
-	if (_PyString_Resize(&RetVal, length << 1) == -1) {
+	if (_PyString_Resize(&RetVal, length << 1) < 0)
-	    RetVal = NULL;
 	    goto error;
-	}
 	self->zst.next_out = (unsigned char *)PyString_AS_STRING(RetVal) \
 	    + length;
 	self->zst.avail_out = length;
@@ -438,9 +435,7 @@ PyZlib_objcompress(compobject *self, PyObject *args)
 	RetVal = NULL;
 	goto error;
     }
-    if (_PyString_Resize(&RetVal,
+    _PyString_Resize(&RetVal, self->zst.total_out - start_total_out);
-			 self->zst.total_out - start_total_out) < 0)
-	RetVal = NULL;
 
  error:
     LEAVE_ZLIB
@@ -510,10 +505,8 @@ PyZlib_objdecompress(compobject *self, PyObject *args)
 	if (max_length && length > max_length)
 	    length = max_length;
 
-	if (_PyString_Resize(&RetVal, length) == -1) {
+	if (_PyString_Resize(&RetVal, length) < 0)
-	    RetVal = NULL;
 	    goto error;
-	}
 	self->zst.next_out = (unsigned char *)PyString_AS_STRING(RetVal) \
 	    + old_length;
 	self->zst.avail_out = length - old_length;
@@ -561,8 +554,7 @@ PyZlib_objdecompress(compobject *self, PyObject *args)
 	goto error;
     }
 
-    if (_PyString_Resize(&RetVal, self->zst.total_out - start_total_out) < 0)
+    _PyString_Resize(&RetVal, self->zst.total_out - start_total_out);
-	RetVal = NULL;
 
  error:
     LEAVE_ZLIB
@@ -612,10 +604,8 @@ PyZlib_flush(compobject *self, PyObject *args)
     /* while Z_OK and the output buffer is full, there might be more output,
        so extend the output buffer and try again */
     while (err == Z_OK && self->zst.avail_out == 0) {
-	if (_PyString_Resize(&RetVal, length << 1) == -1)  {
+	if (_PyString_Resize(&RetVal, length << 1) < 0)
-	    RetVal = NULL;
 	    goto error;
-	}
 	self->zst.next_out = (unsigned char *)PyString_AS_STRING(RetVal) \
 	    + length;
 	self->zst.avail_out = length;
@@ -651,8 +641,7 @@ PyZlib_flush(compobject *self, PyObject *args)
 	goto error;
     }
 
-    if (_PyString_Resize(&RetVal, self->zst.total_out - start_total_out) < 0)
+    _PyString_Resize(&RetVal, self->zst.total_out - start_total_out);
-	RetVal = NULL;
 
  error:
     LEAVE_ZLIB

Objects/fileobject.c

@@ -1317,9 +1317,7 @@ file_readlines(PyFileObject *f, PyObject *args)
 			goto error;
 	}
   cleanup:
-	if (big_buffer) {
+	Py_XDECREF(big_buffer);
-		Py_DECREF(big_buffer);
-	}
 	return list;
 }
 

Objects/stringobject.c

@@ -1869,8 +1869,8 @@ string_translate(PyStringObject *self, PyObject *args)
 		return input_obj;
 	}
 	/* Fix the size of the resulting string */
-	if (inlen > 0 &&_PyString_Resize(&result, output-output_start))
+	if (inlen > 0)
-		return NULL;
+		_PyString_Resize(&result, output - output_start);
 	return result;
 }
 
@@ -2927,7 +2927,14 @@ PyString_ConcatAndDel(register PyObject **pv, register PyObject *w)
    is only one module referencing the object.  You can also think of it
    as creating a new string object and destroying the old one, only
    more efficiently.  In any case, don't use this if the string may
-   already be known to some other part of the code... */
+   already be known to some other part of the code...
+   Note that if there's not enough memory to resize the string, the original
+   string object at *pv is deallocated, *pv is set to NULL, an "out of
+   memory" exception is set, and -1 is returned.  Else (on success) 0 is
+   returned, and the value in *pv may or may not be the same as on input.
+   As always, an extra byte is allocated for a trailing \0 byte (newsize
+   does *not* include that), and a trailing \0 byte is stored.
+*/
 
 int
 _PyString_Resize(PyObject **pv, int newsize)

Objects/unicodeobject.c

@@ -927,10 +927,7 @@ PyObject *PyUnicode_EncodeUTF7(const Py_UNICODE *s,
         *out++ = '-';
     }
 
-    if (_PyString_Resize(&v, out - start)) {
+    _PyString_Resize(&v, out - start);
-        Py_DECREF(v);
-        return NULL;
-    }
     return v;
 }
 
@@ -1764,7 +1761,7 @@ PyObject *unicodeescape_string(const Py_UNICODE *s,
 	    /* Resize the string if necessary */
 	    if (offset + 12 > PyString_GET_SIZE(repr)) {
 		if (_PyString_Resize(&repr, PyString_GET_SIZE(repr) + 100))
-		    goto onError;
+		    return NULL;
 		p = PyString_AS_STRING(repr) + offset;
 	    }
 
@@ -1847,14 +1844,8 @@ PyObject *unicodeescape_string(const Py_UNICODE *s,
         *p++ = PyString_AS_STRING(repr)[1];
 
     *p = '\0';
-    if (_PyString_Resize(&repr, p - PyString_AS_STRING(repr)))
+    _PyString_Resize(&repr, p - PyString_AS_STRING(repr));
-	goto onError;
-
     return repr;
-
- onError:
-    Py_DECREF(repr);
-    return NULL;
 }
 
 PyObject *PyUnicode_EncodeUnicodeEscape(const Py_UNICODE *s,
@@ -1985,14 +1976,8 @@ PyObject *PyUnicode_EncodeRawUnicodeEscape(const Py_UNICODE *s,
             *p++ = (char) ch;
     }
     *p = '\0';
-    if (_PyString_Resize(&repr, p - q))
+    _PyString_Resize(&repr, p - q);
-	goto onError;
-
     return repr;
-
- onError:
-    Py_DECREF(repr);
-    return NULL;
 }
 
 PyObject *PyUnicode_AsRawUnicodeEscapeString(PyObject *unicode)
@@ -2092,8 +2077,7 @@ PyObject *PyUnicode_EncodeLatin1(const Py_UNICODE *p,
     }
     /* Resize if error handling skipped some characters */
     if (s - start < PyString_GET_SIZE(repr))
-	if (_PyString_Resize(&repr, s - start))
+	_PyString_Resize(&repr, s - start);
-	    goto onError;
     return repr;
 
  onError:
@@ -2240,8 +2224,7 @@ PyObject *PyUnicode_EncodeASCII(const Py_UNICODE *p,
     }
     /* Resize if error handling skipped some characters */
     if (s - start < PyString_GET_SIZE(repr))
-	if (_PyString_Resize(&repr, s - start))
+	_PyString_Resize(&repr, s - start);
-	    goto onError;
     return repr;
 
  onError:
@@ -2588,12 +2571,11 @@ PyObject *PyUnicode_EncodeCharmap(const Py_UNICODE *p,
 	Py_DECREF(x);
     }
     if (s - PyString_AS_STRING(v) < PyString_GET_SIZE(v))
-	if (_PyString_Resize(&v, (int)(s - PyString_AS_STRING(v))))
+	_PyString_Resize(&v, (int)(s - PyString_AS_STRING(v)));
-	    goto onError;
     return v;
 
  onError:
-    Py_DECREF(v);
+    Py_XDECREF(v);
     return NULL;
 }
 

Python/bltinmodule.c

@@ -1993,8 +1993,8 @@ filterstring(PyObject *func, PyObject *strobj)
 		Py_DECREF(item);
 	}
 
-	if (j < len && _PyString_Resize(&result, j) < 0)
+	if (j < len)
-		return NULL;
+		_PyString_Resize(&result, j);
 
 	return result;