I've seen several tail implementations and they've seemed overly complicated, or overengineered, for my purposes.
Here's an implementation I'm using in my production code, it's _ONLY_ for following an open stream, line by line, and sleeps for a second while waiting for activity. No heuristics, no configurability, it's simple and --in my opinion-- clean enough to understand in a single look.
Enjoy!!
Python, 50 lines
Download
Copy to clipboard1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 | #!/usr/bin/python2.7
# vim:ai:et:ts=4:sw=4:wm=0:encoding=utf8:fileencoding=utf8
"""
tail.py
=======
Exports 'follow', which yields linewise the content being added
to a text file, similar to unix' "tail -f" functionality.
As a demonstration, you can run this script directly with an
argument, the filename you wish to follow. I use it to follow
my httpd logs:
$ python2.7 tail.py /var/log/httpd/access_log
....
"""
__all__ = ('follow',)
import time
def follow(stream):
"Follow the live contents of a text file."
line = ''
for block in iter(lambda:stream.read(1024), None):
if '\n' in block:
# Only enter this block if we have at least one line to yield.
# The +[''] part is to catch the corner case of when a block
# ends in a newline, in which case it would repeat a line.
for line in (line+block).splitlines(True)+['']:
if line.endswith('\n'):
yield line
# When exiting the for loop, 'line' has any remaninig text.
elif not block:
# Wait for data.
time.sleep(1.0)
# The End.
if __name__ == '__main__':
# As a simple demonstration, run it with the filename to tail.
import sys
with open(sys.argv[1], 'rt') as following:
following.seek(-64, 2)
try:
for line in follow(following):
sys.stdout.write(line)
except KeyboardInterrupt:
pass
# Fin.
|
