Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-11-04 07:31:38 +00:00
Code Issues Projects Releases Packages Wiki Activity

Revert gzip readline performance patch #1281707 until a more generic performance improvement can be found

Browse source
This commit is contained in:
Bob Ippolito 2006-05-22 15:22:46 +00:00
parent d72aab5e31
commit b97597316b
1 changed files with 19 additions and 27 deletions
Download patch file Download diff file Expand all files Collapse all files

46
Lib/gzip.py
View file

@ -107,7 +107,6 @@ def __init__(self, filename=None, mode=None,
self.extrabuf = "" self.extrabuf = ""
self.extrasize = 0 self.extrasize = 0
self.filename = filename self.filename = filename
self.min_readsize = 64 # Starts small, scales exponentially
elif mode[0:1] == 'w' or mode[0:1] == 'a': elif mode[0:1] == 'w' or mode[0:1] == 'a':
self.mode = WRITE self.mode = WRITE
@ -382,39 +381,32 @@ def seek(self, offset):
self.read(count % 1024) self.read(count % 1024)
def readline(self, size=-1): def readline(self, size=-1):
if size < 0: if size < 0: size = sys.maxint
size = sys.maxint # Line can be as long as maxint bufs = []
readsize = self.min_readsize # Read from file in small chunks readsize = min(100, size) # Read from the file in small chunks
else:
readsize = size # Only read in as much as specified
bufs = ""
while True: while True:
if size == 0: return bufs # Return line (reached max len) if size == 0:
return "".join(bufs) # Return resulting line
c = self.read(readsize) c = self.read(readsize)
i = c.find('\n') i = c.find('\n')
if size is not None:
# We set i=size to break out of the loop under two
# conditions: 1) there's no newline, and the chunk is
# larger than size, or 2) there is a newline, but the
# resulting line would be longer than 'size'.
if i==-1 and len(c) > size: i=size-1
elif size <= i: i = size -1
# If there is a newline, or the string is empty
if i >= 0 or c == '': if i >= 0 or c == '':
if size <= i: i = size - 1 # Another larger than size check bufs.append(c[:i+1]) # Add portion of last chunk
self._unread(c[i+1:]) # Push back rest of chunk
return ''.join(bufs) # Return resulting line
self._unread(c[i+1:]) # Push back rest of chunk # Append chunk to list, decrease 'size',
bufs.append(c)
return bufs + c[:i+1] # Stored line, plus new segment size = size - len(c)
readsize = min(size, readsize * 2)
# If there is no newline
else:
if len(c) > size: i = size - 1 # If lineis larger than size
bufs = bufs + c
size = size - len(c)
readsize = min(size, int(readsize * 1.1))
# Optimize future readline() calls
if readsize > self.min_readsize:
self.min_readsize = readsize
def readlines(self, sizehint=0): def readlines(self, sizehint=0):
# Negative numbers result in reading all the lines # Negative numbers result in reading all the lines