amo-stats-analyze.py

#!/usr/bin/env python

import csv
import re
import sys

################################################################################

# Good reference: http://www.get-firefox.eu/default_025.html
app_ids={
	'{136c295a-4a5a-41cf-bf24-5cee526720d5}': 'Nvu',
	'{9b0e-13a3a9e97384}': 'Firefox', # Somebody's buggy submission
	#'{b58db1ce-a21c-4fcd-abef-68f559f273d7}': '',
	'{ec8030f7-c20a-464f-9b0e-13a3a9e96384}': 'Firefox', # Somebody's buggy submission
	'{ec8030f7-c20a-4>4f-9b0e-13a3a9e97384}': 'Firefox', # Somebody's buggy submission
	'ec8030f7-g20a-464f-9b0e-13a3a9e97384': 'Firefox', # Somebody's buggy submission
	'{ee53ece0-255c-4cc6-8a7e-81a8b6e5ba2c}': 'Spicebird',
	'songbird@songbirdnest.com': 'Songbird',
	'prism@developer.mozilla.org': 'Prism',
	'{a23983c0-fd0e-11dc-95ff-0800200c9a66}': 'Fennec',
	'{3db10fab-e461-4c80-8b97-957ad5f8ea47}': 'Netscape',
	'{718e30fb-e89b-41dd-9da7-e25a45638b28}': 'Sunbird',
	'{92650c4d-4b8e-4d2a-b7eb-24ecf4f6b63a}': 'SeaMonkey',
	'{3550f703-e582-4d05-9a08-453d09bdfdc6}': 'Thunderbird',
	'{a463f10c-3994-11da-9945-000d60ca027b}': 'Flock',
	'{ec8030f7-c20a-464f-9b0e-13a3a9e97384}': 'Firefox',
}

################################################################################

# Read from stdin.
statReader=csv.reader(sys.stdin)

# Find the column names.
while True:
	cols=statReader.next()
	if cols[0][0:8]=='# Fields':
		# Extract just the column names from this line.
		cols=cols[0][11:-1]
		# And split!
		cols=cols.split(';')

		# Stop reading headers.
		break

# Group the columns into buckets.
buckets=['x', 'x']
for col in cols[2:]:
	try:
		(app, version)=col.split('/', 2)
	except ValueError:
		(app, version)=('Unknown', '0')
	
	# If we know what app this is, get its name instead of ID.
	if app in app_ids:
		app=app_ids[app]
	else:
		app='Unknown'
		version='0'

	# Trim the version
	vnum=re.search('^(\d)(\.\d)?', version)
	if vnum and vnum.group(1) and vnum.group(2):
		version=vnum.group(1)+vnum.group(2)
	else:
		version='0'

	# Super-bucket-ize.
	if 'Firefox'==app:
		if float(version)<1.0:
			version='<1.0'
		elif float(version)<2.0:
			version='1.5'
		elif float(version)<3.0:
			version='2.0'
		elif float(version)<3.5:
			version='3.0'
		else:
			version='>3.5'
#	elif 'Thunderbird'==app:
#		if float(version)>=3.0:
#			version='3.0'
#	elif app in ['Flock', 'SeaMonkey', 'Songbird', 'Spicebird', 'Sunbird']:
#		version='All'
	elif 'Unknown'!=app:
		# Super-super-bucket-ize.
		app='Other'

	bucket=app+'/'+version
	if app in ['Other', 'Unknown']: bucket=app

	buckets.append(bucket)

# Figure out the ordered set of headers to produce.
headers=list(set(buckets[3:]))
headers.sort()
headers=['date']+headers

# Print out these header names.
statWriter=csv.writer(sys.stdout)
statWriter.writerow(headers)
# Start a csvwriter to hold these columns.
statWriter=csv.DictWriter(sys.stdout, headers, restval='0')

for row in statReader:
	data={'date':row[0]}

	for i in range(3, len(row)):
		data.setdefault(buckets[i], 0)
		data[buckets[i]]+=int(row[i])

	statWriter.writerow(data)