From 7c924ec9258fd8e324dedcdcc35bb8706aea7922 Mon Sep 17 00:00:00 2001 From: Victor Stinner Date: Thu, 4 Mar 2010 12:09:33 +0000 Subject: [PATCH] Issue #1054943: Fix unicodedata.normalize('NFC', text) for the Public Review Issue #29. PR #29 was released in february 2004! --- Lib/test/test_unicodedata.py | 5 +++++ Misc/NEWS | 3 +++ Modules/unicodedata.c | 2 +- 3 files changed, 9 insertions(+), 1 deletion(-) diff --git a/Lib/test/test_unicodedata.py b/Lib/test/test_unicodedata.py index c82e4afb1077..6fbb6dd4509f 100644 --- a/Lib/test/test_unicodedata.py +++ b/Lib/test/test_unicodedata.py @@ -186,6 +186,11 @@ def test_normalize(self): # The rest can be found in test_normalization.py # which requires an external file. + def test_pr29(self): + # http://www.unicode.org/review/pr-29.html + for text in (u"\u0b47\u0300\u0b3e", u"\u1100\u0300\u1161"): + self.assertEqual(self.db.normalize('NFC', text), text) + def test_east_asian_width(self): eaw = self.db.east_asian_width self.assertRaises(TypeError, eaw, 'a') diff --git a/Misc/NEWS b/Misc/NEWS index fb45e5a59c4e..107c9912762c 100644 --- a/Misc/NEWS +++ b/Misc/NEWS @@ -41,6 +41,9 @@ Core and Builtins Library ------- +- Issue #1054943: Fix unicodedata.normalize('NFC', text) for the Public Review + Issue #29 + - Issue #7494: fix a crash in _lsprof (cProfile) after clearing the profiler, reset also the pointer to the current pointer context. diff --git a/Modules/unicodedata.c b/Modules/unicodedata.c index 2eb1911299c1..0269175c1ec4 100644 --- a/Modules/unicodedata.c +++ b/Modules/unicodedata.c @@ -682,7 +682,7 @@ nfc_nfkc(PyObject *self, PyObject *input, int k) comb = 0; while (i1 < end) { int comb1 = _getrecord_ex(*i1)->combining; - if (comb1 && comb == comb1) { + if (comb && (comb1 == 0 || comb == comb1)) { /* Character is blocked. */ i1++; continue;