Hi!
Please excuse me if this i common knowledge, or if I've one again
re-implemented something
that turned out to be in the standard library, but I think I came up
with something rather neat.
I'm writing a lot of programs that call external programs, and as much
as I love subproces.Popen, I do get tired of writing
rather wordy things like:
args = ['cmd', 'arg1', 'arg2', 'etc']
p = subprocess.Popen(args, stdin=subprocess.PIPE)
for line in p:
line = line.decode('latin1')
do_something_with(line)
Bleh.
so last week I had enough, and I sat down and wrote a few functions.
No I can do:
from lib import inpipe
for line in inpipe(args, encoding='latin1'):
do_something_with(line)
and
from lib import outpipe
with outpipe(args, addnl=True) as write:
write('Python is awesome!')
write('')
Some code from a program I'm writing:
args = ['metaflac',
'--show-total-samples',
'--show-sample-rate',
file]
for line in stripped(inpipe(args)):
(name, value) = line.split('=')
# ... process tag
Now, that is a lot more readable than what I had before!
The library has a lot of other things in it as well, and is available here:
https://github.com/olemb/lib
I love Python!
--
Ole Martin,
http://nerdly.info/ole/
--
http://mail.python.org/mailman/listinfo/python-list
