Coverage counts cannot be more granular than the mbtile dimension

[jothirnadh]

trafficstars

While working on our hackday project using OSM-coverage processor we came across with this issue which is discrbed below.

In our project we wanted to calculate the length of the roads per district in Bangkok metropolitan area. There are a total of 50 districts in Bangkok metropolitan area but the coverage processor returned only 25 results when we ran it across the city. On research we found that the reason for this result is the size of the mbtiles is much larger than some of the districts. It is only giving the statistics for entire mbtile for the first district it came across and skipping the rest of them. This shows the issue with the granularity of coverage counts as we run it for smaller areas.