How Nest Watcher analyzes nests

When configuring Nest Watcher, it might be helpful to understand how it operates.

• Get values needed for analyzation

  • If enabled, get the last nest migration:

    • Check this for regular migrations
    • See here if there’s an event start/end between that time and right now If yes, use that.
  • Go through your areas.json and save it together with the appropiate settings in a list, to later iterate over
  • If enabled check this for current event mons and save them as a list
  • Get Nesting species from here and subtract the event mon list from it
  • Connect to the DB and, if meganests are enabled, query the most common nest mon in your db and remove it from the nest list
• For each area, do:

  • Get the OSM data file, if not existent, query the data from overpass-turbo (taking into account rate-limiting)

    • Overpass only allows for bbox-querying, if parks are outside of the given geofence, they’re later being filtered out
  • Get area data. If not existent, use an empty dict
  • Delete nests within the area from the database
  • Save all nodes, ways and relations from osm in seperate lists
  • If less_queries is enabled, query all mons and spawns from the area
  • Go throgh all relations, generate polygons from them and save used ways in an extra list from the way-list. Then generate polygons from the ways. (The generating of polygons is pretty complicated and I wouldn’t be able to easily explain it here)
  • Go through the area data and connect parks by merging the polygons and then removing the to-be-connected park from the list

  • Go through each park now and do:

    • If there was an error with the geometry calculation, skip
    • If the park polygon is not within the area polygon, skip
    • If the park ID a way and part of a relation, skip
    • If enabled, query Pokestops from the park polygon
    • Get Spawnpoints (if less_queries is enabled, go through the prequeried list, else grab them straight from the database)
    • If no Spawns/Stops exist, skip
    • If Spawns + Stops < min_spawnpoints (settings), skip
    • Get the most common mon and its count where the ID is in the generated nesting list and which spawned on the queried spawns/stops. (if less_queries, get them from the pregenerated list, else get them from the db)
    • Calculate mon_avg: (mon_amount / scan_hours_per_day)) * (24.00 / hours_since_change)
    • Calculate mon_ratio: mon_avg / (stops + spawns)
    • Now see if those values are higher than the filter configured in settings
    • Get name and center lat/lon of the park. If not existent in area_data, try to get the name from the OSM data and calculate the center using the polylabel algorithm for normal polygons. Or the simple centroid for multipolygons
    • Save the park details in GeoJSON format
    • Put the nest into the database
    • Save the park in a list
  • Convert the list into area_data format (sorted by mon_avg) and append data of nests that haven’t been found this run, then save the area_data
• Frontend features
  • Save the GeoJSON of all found parks according to the config
  • Send nests to Discord

---