CSV reader and unique ids

[Mike P]

Hi All,

I'm trying to use the CSV module to read in some data and then use a
hashable method (as there are millions of records) to find unique ids
and push these out to another file,

can anyone advise? Below is the code so far
fin = open(CSV_INPUT, "rb")
fout = open(CSV_OUTPUT, "wb")
reader = csv.reader(fin, delimiter=chr(254))
writer = csv.writer(fout)

headerList = reader.next()
UID = {}

#For help
#print headerList
# ['Time', 'User-ID', 'IP']

try:
for row in reader[1]:
UID[row] = 1
else:
List= UID.keys()
writer.writerows(List)
fin.close()
fout.close()

Mike

[Tim Golden]

Mike P wrote:

I'm trying to use the CSV module to read in some data and then use a
hashable method (as there are millions of records) to find unique ids
and push these out to another file,

You could either zip with a counter or use the uuid module,
depending on just how unique you want your ids to be.

<code>
import os, sys
import csv
import itertools
import uuid

stuff = "the quick brown fox jumps over the lazy dog".split ()

f = open ("output.csv", "wb")
writer = csv.writer (f)

#
# Style 1 - numeric counter
#
writer.writerows (zip (itertools.count (), stuff))

#
# Style 2 - uuid
#
writer.writerows ((uuid.uuid1 (), s) for s in stuff)

f.close ()
os.startfile ("output.csv")

</code>

TJG

[python]

Anyone have any benchmarks on the difference in performance between 32
and 64 bit versions of Python for specific categories of operation, eg.
math, file, string, etc. operations?

My question is OS neutral so feel free to share your experience with
either Windows or Linux OS's.

Thank you,
Malcolm