| |
@@ -6,12 +6,12 @@
|
| |
# start on week 156, which is the end of 2014, because first-seen dates start in the end of 2012 because
|
| |
# that's when fedmsg starts. So the old-vs-mid-vs-new metrics are only really meaningful starting two years
|
| |
# after that. (Actually, push back to 158 because the data looks ugly with low points before that.)
|
| |
- STARTWEEK=158
|
| |
+ STARTWEEK=209
|
| |
|
| |
ENDWEEK=$(( ( $(date +'%s') - $(date -ud '2012-01-01 0:0:0' +'%s') )/60/60/24/7 -1 ))
|
| |
|
| |
- python new-and-old-users-report.py --csvh $(( $STARTWEEK - 1 )) | tee data/contributor-count.csv
|
| |
+ python3 new-and-old-users-report.py --csvh $(( $STARTWEEK - 1 )) | tee data/contributor-count.csv
|
| |
for week in $(seq $STARTWEEK $ENDWEEK ); do
|
| |
- python new-and-old-users-report.py --csv $week | tee -a data/contributor-count.csv
|
| |
+ python3 new-and-old-users-report.py --csv $week | tee -a data/contributor-count.csv
|
| |
done
|
| |
|
| |
Issue resolved: #19
Issue name: How to handle "old school" users in the beginning?
The approach used: The graph will start from 2016
Changed the start week from 158 to 209 ( start of 2016)