README.rst

Data Queues

dataqs (Data Queues) is a simple Django app to download, process, and import spatial data into GeoServer/GeoNode.

Quick start

1. Add "dataqs" to your INSTALLED_APPS setting like this:

   INSTALLED_APPS = (
       ...
       'dataqs',
       'dataqs.forecastio',
       'dataqs.gfms',
       'dataqs.airnow',
       'dataqs.wqp',
       'dataqs.aqicn',
       #etc
   )
   

2. In your settings.py or local_settings.py file, add a CELERYBEAT_SCHEDULE setting to specify when Celery should run data_queues tasks:

   from celery.schedules import crontab
   CELERYBEAT_SCHEDULE = {
       'gfms': {
           'task': 'dataqs.gfms.tasks.gfms_task',
           'schedule': crontab(minute='3'),
           'args': ()
       },
       'forecast_io': {
           'task': 'dataqs.forecastio.tasks.forecast_io_task',
           'schedule': crontab(minute='1'),
           'args': ()
       },
           'task': 'dataqs.aqicn.tasks.aqicn_task',
           'schedule': crontab(hour='*/6', minute='0'),
           'args': ([],)
       },
   }
   

3. Also add the following settings:

   #Location of GeoServer data directory
   GS_DATA_DIR = '/usr/share/geoserver/data'
   
   #Directory where temporary dataqs geoprocessing files should be downloaded
   GS_TMP_DIR = GS_DATA_DIR + '/tmp'
   
   #AirNow API username:password
   #(sign up for a free account at http://airnowapi.org/account/request/)
   AIRNOW_ACCOUNT = 'your_airnow_username:your_airnow_password'
   
   #NASA GPM FTP ACCOUNT
   #(sign up at http://registration.pps.eosdis.nasa.gov/registration/)
   GPM_ACCOUNT = 'your_gpm_email_account'
   
   #HIFLD: Dictionary of layers to process in the form of:
   HIFLD_LAYERS = [
       {
       'name': 'US State Boundaries',
       'url': 'https://hifld-dhs-gii.opendata.arcgis.com/datasets/718791120f6549708cb642dac6ff0dbf_0.geojson',
       'table': 'state_boundaries',
       'sld': 'polygon'
   },
   {
       'name': 'Cities and Towns NTAD',
       'url': 'https://hifld-dhs-gii.opendata.arcgis.com/datasets/6a1e404a10754e59bac4bfa50db3f487_0.geojson',
       'table': 'cities_towns',
       'sld': 'point'
   },
   {
       'name': 'Roads and Railroad Tunnels',
       'url': 'https://hifld-dhs-gii.opendata.arcgis.com/datasets/2f6abb736360437ba363e0a1210b4d36_0.geojson',
       'table': 'roads_tunnels',
       'sld': 'line'
   }
   ]
   
   #Time to wait before updating Geoserver mosaic (keep at 0 unless Geoserver
   #is on a different server. In that case, there will need to be an automated
   #rsync between GS_TMP_DIR where celery is running and
   #GS_DATA_DIR where GeoServer is running.
   RSYNC_WAIT_TIME = 0
   

4. In order to run the spei processor, the following must be installed:

   sudo apt-get install netcdf-bin
   sudo apt-get install cdo