These are set of scripts used to count the release downloads from apache http stat logs.
1.Where to download the http stat logs from
The logs are located in people.apache.org under x1/logarchive/people/<year>/<month>/??.bz2
I usually download them on a folder structure like : <year>/<month>
2.Get ready to process the scripts
Unzip the log files
>bunzip2 ??.bz2
Concatenate together all the logs from a given month
>ls -1 | xargs -i -t cat {} >> ../<year>-<month>.log
3.Running the scripts
all-report.sh - runs all the scripts
>./all-reports.sh
???-report.sh - runs reports for a given type of release/project (e.g cpp, sca, sdo, das)
>./sca.report.sh
4.Adding a new month to be processed
Edit the ???-report.sh and add a new line for the new month as appropriated
>./sca-099.sh 2007/2007-09.log >> sca-downloads.log
5.Adding a new project/release
Create a new script that filters the specific project/release
>total=$(cat $1 | grep tuscany-das-1.0-incubating-beta1 | awk '{ print $2 }' | uniq | wc -l)
echo $1 $total
The update the other scripts to properly call this to process a given month
d134d725ec