From 36c3928fdc7d246fda50d48c108fb6116c315229 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Kristj=C3=A1n=20Valur=20J=C3=B3nsson?= Date: Tue, 2 Jun 2009 13:14:08 +0000 Subject: [PATCH] http://bugs.python.org/issue6117 Fix O(n**2) performance problem in socket._fileobject --- Lib/socket.py | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/Lib/socket.py b/Lib/socket.py index e694587ea8e..dd0f327a3a3 100644 --- a/Lib/socket.py +++ b/Lib/socket.py @@ -235,7 +235,7 @@ class _fileobject(object): __slots__ = ["mode", "bufsize", "softspace", # "closed" is a property, see below - "_sock", "_rbufsize", "_wbufsize", "_rbuf", "_wbuf", + "_sock", "_rbufsize", "_wbufsize", "_rbuf", "_wbuf", "_wbuf_len", "_close"] def __init__(self, sock, mode='rb', bufsize=-1, close=False): @@ -261,6 +261,7 @@ class _fileobject(object): # realloc()ed down much smaller than their original allocation. self._rbuf = StringIO() self._wbuf = [] # A list of strings + self._wbuf_len = 0 self._close = close def _getclosed(self): @@ -287,6 +288,7 @@ class _fileobject(object): if self._wbuf: buffer = "".join(self._wbuf) self._wbuf = [] + self._wbuf_len = 0 self._sock.sendall(buffer) def fileno(self): @@ -297,25 +299,22 @@ class _fileobject(object): if not data: return self._wbuf.append(data) + self._wbuf_len += len(data) if (self._wbufsize == 0 or self._wbufsize == 1 and '\n' in data or - self._get_wbuf_len() >= self._wbufsize): + self._wbuf_len >= self._wbufsize): self.flush() def writelines(self, list): # XXX We could do better here for very long lists # XXX Should really reject non-string non-buffers - self._wbuf.extend(filter(None, map(str, list))) + lines = filter(None, map(str, list)) + self._wbuf_len += sum(map(len, lines)) + self._wbuf.extend(lines) if (self._wbufsize <= 1 or - self._get_wbuf_len() >= self._wbufsize): + self._wbuf_len >= self._wbufsize): self.flush() - def _get_wbuf_len(self): - buf_len = 0 - for x in self._wbuf: - buf_len += len(x) - return buf_len - def read(self, size=-1): # Use max, disallow tiny reads in a loop as they are very inefficient. # We never leave read() with any leftover data from a new recv() call