From: Martin v. Löwis <martin@v.loewis.de>
Date: Sat, 17 Apr 2004 19:36:13 +0000 (+0000)
Subject: Special case normalization of empty strings. Fixes #924361.
X-Git-Tag: v2.3.4c1~57
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=a2f79361f8472877c79fbb7bd70615eb2e5b918f;p=thirdparty%2FPython%2Fcpython.git

Special case normalization of empty strings. Fixes #924361.
---

diff --git a/Misc/NEWS b/Misc/NEWS
index 792b41066aff..8d73eab69d87 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -42,6 +42,8 @@ Core and builtins
 Library
 -------
 
+- Bug #924361: Properly support normalization of empty unicode strings.
+
 - Fixed a caching bug in platform.platform() where the argument of 'terse' was
   not taken into consideration when caching value.
 
diff --git a/Modules/unicodedata.c b/Modules/unicodedata.c
index 311db296bdbc..ba218a3e58e8 100644
--- a/Modules/unicodedata.c
+++ b/Modules/unicodedata.c
@@ -515,6 +515,13 @@ unicodedata_normalize(PyObject *self, PyObject *args)
                          &form, &PyUnicode_Type, &input))
         return NULL;
 
+    if (PyUnicode_GetSize(input) == 0) {
+        /* Special case empty input strings, since resizing
+           them  later would cause internal errors. */
+        Py_INCREF(input);
+        return input;
+    }
+
     if (strcmp(form, "NFC") == 0)
         return nfc_nfkc(input, 0);
     if (strcmp(form, "NFKC") == 0)