Parsing a french carpooling site
This script will examine the first 100th results of a French carpooling website to find the busiest hour to make a trip.
For that, the script must be run regularly. The first time, it creates a file for the current day. The next time it will be run, the values will be read from the file, updated and wrote back to the file.
If the first time the script is read, the trip is already full, we have missing data. As a convention, it will stay as "Complet.
We parse the file twice: once to have a list of ids and once to have the availability. We cannot access the children of "a" so we have to do 2 passes. This supposes the document is well constructed so an id always match the availability. It also supposes it does not change.