#3773: Check for errors around the use of PyTokenizer_FindEncoding().
reviewed by Brett Cannon.
diff --git a/Misc/NEWS b/Misc/NEWS
index 0e51c80..889cead 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,9 @@
Core and Builtins
-----------------
+- Issue 3774: Added a few more checks in PyTokenizer_FindEncoding to handle
+ error conditions.
+
- Issue 3594: Fix Parser/tokenizer.c:fp_setreadl() to open the file being
tokenized by either a file path or file pointer for the benefit of
PyTokenizer_FindEncoding().
diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c
index e4cf8e4..a040696 100644
--- a/Parser/tokenizer.c
+++ b/Parser/tokenizer.c
@@ -1610,7 +1610,10 @@
fclose(fp);
if (tok->encoding) {
encoding = (char *)PyMem_MALLOC(strlen(tok->encoding) + 1);
- strcpy(encoding, tok->encoding);
+ if (encoding)
+ strcpy(encoding, tok->encoding);
+ else
+ PyErr_NoMemory();
}
PyTokenizer_Free(tok);
return encoding;
diff --git a/Python/import.c b/Python/import.c
index d87d751..9c077fe 100644
--- a/Python/import.c
+++ b/Python/import.c
@@ -2830,6 +2830,8 @@
memory. */
found_encoding = PyTokenizer_FindEncoding(fd);
lseek(fd, 0, 0); /* Reset position */
+ if (found_encoding == NULL && PyErr_Occurred())
+ return NULL;
encoding = (found_encoding != NULL) ? found_encoding :
(char*)PyUnicode_GetDefaultEncoding();
}