Rich Shepard wrote:
On Tue, 22 Nov 2011, Ethan Furman wrote:
They are closed automatically.
Ah, so.
It's not good practice to do this way, though better would be to assign
the open files to their own name, then you could 'name.close()' when you
were done.
Ethan,
That's what I thought you did in the lines
reader = csv.reader(open(filename,"r"),
delimiter="|",
quotechar="'")
output = csv.writer(open("out.txt","w"),
delimiter="|",
quotechar="'")
but reader.close() and output.close() generated python errors.
What's happening there is the two files are being opened, and then those
open file objects are being handed into the csv.reader and csv.writer
functions, and the result of the two csv.* calls are being assigned to
'reader' and 'output'. Basically, the actual file objects are anonymous.
When reader and output go out of scope they are destroyed, and at that
point the only references to the two open file objects are lost, and
then the open file objects are closed by Python.
Like I said, it's better to do it explicity. ;)
~Ethan~
_______________________________________________
Portland mailing list
[email protected]
http://mail.python.org/mailman/listinfo/portland
