From 4b0b1accb57aabdd1b30dd445c2d530e9cdda65d Mon Sep 17 00:00:00 2001 From: Raymond Hettinger Date: Sat, 3 May 2014 16:41:19 -0700 Subject: [PATCH] Issue #21101: Eliminate double hashing in the C code for collections.Counter(). --- Misc/NEWS | 3 +++ Modules/_collectionsmodule.c | 17 ++++++++++++++--- 2 files changed, 17 insertions(+), 3 deletions(-) diff --git a/Misc/NEWS b/Misc/NEWS index a50ae07fe53..18c943e720d 100644 --- a/Misc/NEWS +++ b/Misc/NEWS @@ -73,6 +73,9 @@ Library Decimal.quantize() method in the Python version. It had never been present in the C version. +- Issue #21101: Eliminate double hashing in the C speed-up code for + collections.Counter(). + - Issue #21321: itertools.islice() now releases the reference to the source iterator when the slice is exhausted. Patch by Anton Afanasyev. diff --git a/Modules/_collectionsmodule.c b/Modules/_collectionsmodule.c index 0ab4156a247..c6c8666d813 100644 --- a/Modules/_collectionsmodule.c +++ b/Modules/_collectionsmodule.c @@ -1831,18 +1831,29 @@ _count_elements(PyObject *self, PyObject *args) if (mapping_get != NULL && mapping_get == dict_get && mapping_setitem != NULL && mapping_setitem == dict_setitem) { while (1) { + Py_hash_t hash; + key = PyIter_Next(it); if (key == NULL) break; - oldval = PyDict_GetItem(mapping, key); + + if (!PyUnicode_CheckExact(key) || + (hash = ((PyASCIIObject *) key)->hash) == -1) + { + hash = PyObject_Hash(key); + if (hash == -1) + goto done; + } + + oldval = _PyDict_GetItem_KnownHash(mapping, key, hash); if (oldval == NULL) { - if (PyDict_SetItem(mapping, key, one) == -1) + if (_PyDict_SetItem_KnownHash(mapping, key, one, hash) == -1) break; } else { newval = PyNumber_Add(oldval, one); if (newval == NULL) break; - if (PyDict_SetItem(mapping, key, newval) == -1) + if (_PyDict_SetItem_KnownHash(mapping, key, newval, hash) == -1) break; Py_CLEAR(newval); }