#!/bin/ksh
#verify change data
#check if any one has too many fields
for i in *.log1.gz; do gunzip < $i ; done | gawk -F\; '{if (NF != 12) print $0;}'

for i in *.log1.gz; do gunzip < $i ; done | perl collectComments.perl | gzip > changes.gz

##############################################
#now inspect comments to see if there is a pattern on how to extract 
# a) bugs, 
# b) code submitters
# c) code reviewers
# d) super code reviewers
############################
#explore varios ways of submitting
gunzip < changes.gz |  perl getNeighbors.perl submitted  | sort | uniq -c | sort -n > submitexamples

gzcat changes.gz | grep -i "$i"
"Thanks to" (793), "submitted by" (165),  "contributed by" (187), 
"contribution from" (12), "contributions from" (2)


gzcat changes.gz | perl getNeighbors1.perl '\D\d\d\d+\D'

gzcat changes.gz | perl findbug.perl

