#! /usr/bin/env python

# This script aggregates many files containing Bro script coverage information
# into a single file and reports the overall coverage information. Usage:
#
#   coverage-calc <quoted glob of filenames> <output file> <ignored script dir>
#
# The last argument is used to ignore Bro scripts that are part of the test
# suite itself as those should not count towards the coverage calculation.

import os
import sys
import glob

stats = {}
inputglob = sys.argv[1]
outputfile = sys.argv[2]
ignoredir = os.path.abspath(sys.argv[3])

for filename in glob.glob(inputglob):
    with open(filename, 'r') as f:
        for line in f.read().splitlines():
            parts = line.split("\t")
            exec_count = int(parts[0])
            location = os.path.normpath(parts[1])
            # ignore scripts that don't appear to be part of Bro distribution
            if location.startswith(ignoredir) or not location.startswith("/"):
                continue
            desc = parts[2]
            key = location + desc
            if key in stats:
                stats[key][0] += exec_count
            else:
                stats[key] = [exec_count, location, desc]

with open(outputfile, 'w') as f:
    for k in sorted(stats, key=lambda i: stats[i][1]):
        f.write("%s\t%s\t%s\n" % (stats[k][0], stats[k][1], stats[k][2]))

num_covered = 0
for k in stats:
    if stats[k][0] > 0:
        num_covered += 1

if len(stats) > 0:
    print "%s/%s (%.1f%%) Bro script statements covered." % (num_covered, len(stats), float(num_covered)/len(stats)*100)