diff options
author | Robey Pointer <robey@lag.net> | 2004-06-10 18:02:13 +0000 |
---|---|---|
committer | Robey Pointer <robey@lag.net> | 2004-06-10 18:02:13 +0000 |
commit | f0ba3c482eb937a450cac99b578a6197ed54101b (patch) | |
tree | 39667c10c66edd74e283ebe42e3289ff59044b29 | |
parent | cba104ce3f83ef7844ae9598502a974a36ce8682 (diff) |
[project @ Arch-1:robey@lag.net--2003-public%secsh--dev--1.0--patch-59]
speed up parts of BufferedFile
BufferedFile uses cStringIO for the write buffer now (i don't actually notice
any speed difference so this might revert later) and the default buffer size
has been upped from 1KB to 8KB.
when scanning for linefeeds (when writing to a line-buffered file), only scan
the newly-written bytes, since we know all the previously buffered data is
linefeed-free. this was the #1 slowdown on the 1MB-file unit test.
also, limit the buffering on line-buffered files to whatever the default
buffer size is. there's no reason to buffer 1MB waiting for a linefeed.
-rw-r--r-- | paramiko/file.py | 34 |
1 files changed, 21 insertions, 13 deletions
diff --git a/paramiko/file.py b/paramiko/file.py index 3b9812a8..b6b86691 100644 --- a/paramiko/file.py +++ b/paramiko/file.py @@ -15,13 +15,15 @@ # details. # # You should have received a copy of the GNU Lesser General Public License -# along with Foobar; if not, write to the Free Software Foundation, Inc., +# along with Paramiko; if not, write to the Free Software Foundation, Inc., # 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. """ BufferedFile. """ +from cStringIO import StringIO + _FLAG_READ = 0x1 _FLAG_WRITE = 0x2 _FLAG_APPEND = 0x4 @@ -36,7 +38,7 @@ class BufferedFile (object): simpler stream """ - _DEFAULT_BUFSIZE = 1024 + _DEFAULT_BUFSIZE = 8192 SEEK_SET = 0 SEEK_CUR = 1 @@ -45,7 +47,8 @@ class BufferedFile (object): def __init__(self): self._flags = 0 self._bufsize = self._DEFAULT_BUFSIZE - self._wbuffer = self._rbuffer = '' + self._wbuffer = StringIO() + self._rbuffer = '' self._at_trailing_cr = False self._closed = False # pos - position within the file, according to the user @@ -80,8 +83,8 @@ class BufferedFile (object): Write out any data in the write buffer. This may do nothing if write buffering is not turned on. """ - self._write_all(self._wbuffer) - self._wbuffer = '' + self._write_all(self._wbuffer.getvalue()) + self._wbuffer = StringIO() return def next(self): @@ -300,16 +303,21 @@ class BufferedFile (object): if not (self._flags & _FLAG_BUFFERED): self._write_all(data) return - self._wbuffer += data + self._wbuffer.write(data) if self._flags & _FLAG_LINE_BUFFERED: - last_newline_pos = self._wbuffer.rfind('\n') + # only scan the new data for linefeed, to avoid wasting time. + last_newline_pos = data.rfind('\n') if last_newline_pos >= 0: - self._write_all(self._wbuffer[:last_newline_pos + 1]) - self._wbuffer = self._wbuffer[last_newline_pos+1:] - else: - if len(self._wbuffer) >= self._bufsize: - self._write_all(self._wbuffer) - self._wbuffer = '' + wbuf = self._wbuffer.getvalue() + last_newline_pos += len(wbuf) - len(data) + self._write_all(wbuf[:last_newline_pos + 1]) + self._wbuffer = StringIO() + self._wbuffer.write(wbuf[last_newline_pos+1:]) + return + # even if we're line buffering, if the buffer has grown past the + # buffer size, force a flush. + if len(self._wbuffer.getvalue()) >= self._bufsize: + self.flush() return def writelines(self, sequence): |