I’ve operated a [Weather Underground](http://www.wunderground.com/) [Personal Weather Station](http://www.wunderground.com/weatherstation/about.asp) (PWS) [[KMEBERWI7](http://www.wunderground.com/personal-weather-station/dashboard?ID=KMEBERWI7#history)] off-and-on (hardware issues notwithstanding) for as long as I can remember, and I thought it was about time to finally do an Alfred↔PWS mashup. My personal requirements were quite modest:

– 5 reading history (including most current)
– Ability to copy all the current day’s readings as CSV
– Quickly get to my PWS data w/o a bookmark

[Alfred](http://www.alfredapp.com/) makes all this possible via customized workflows that support many scripting environments, including Python. Here’s a quick preview before going into the details:

It’s a fairly simple workflow:

– Grab today’s “raw” data (and clean it up)
– Select the last 5 entries
– Connect a URL action and use the full CSV as clipboard contents for any copy action

The full Python code is below and [on github](https://github.com/hrbrmstr/alfred-pws), and you can hit that github link or [Packal](http://www.packal.org/workflow/pws-history) for the compiled workflow. It’s been tested on Mavericks, but more eyes are always welcome.

Customizing the workflow for your own/favorite PWS is as simple as changing the `station` variable.

There’s plenty of room for improvement, including

– performing a background download of the PWS data
– using a sparkline graph as the icon
– customizing which data fields are returned
– providing commands to get/set your/favorite PWS
– providing options for the “copy” return type (currently CSV, but potentially XML or JSON)

Don’t hesitate to post issues or pull requests and check back for updates (as I’m sure some of those improvements will be making their way into the workflow).

import re
import csv
import sys
import datetime
from lib import requests
from workflow import Workflow, web
from StringIO import StringIO
 
# retrieve today's history for station "X"
 
def get_wx_data(station):
 
  tdy = datetime.datetime.today()
 
  # construct the URL for "today"
  url = 'http://www.wunderground.com/weatherstation/WXDailyHistory.asp?ID=%s&day=%d&month=%d&year=%d&graphspan=day&format=1' % (station, tdy.day, tdy.month, tdy.year)
 
  r = web.get(url) # get the data
 
  r.raise_for_status() # report any errors
 
  return(re.sub("\n\n", "\n", re.sub("^\n|<br>", "", r.text))) # clean up the output & pass it back to main control
 
 
# main workflow control
 
def main(wf):
 
  station = "KMEBERWI7" # change to use your own/favorite station
 
  resp = get_wx_data(station)
 
  # only want last 5 readings, change this to whatever you want
  max = 5
 
  i=0
  for row in reversed(list(csv.reader(StringIO(resp)))):
    wf.add_item(title=row[0] + " | " + row[1] + "F | " + row[3] + "in | " + row[8] + "%", 
                subtitle=station, # so you know where you're getting data from
                arg=station, # passed as query to URL action - http://www.wunderground.com/personal-weather-station/dashboard?ID={query}#history
                valid=True, # it can be opened in the browser
                icon="/System/Library/CoreServices/CoreTypes.bundle/Contents/Resources/ToolbarInfo.icns", # info icon
                copytext=resp) # get the whole CSV file in a copy
    if (i==max): break
    i += 1
 
  # output to alfred
 
  wf.send_feedback()
 
if __name__ == u"__main__":
  wf = Workflow()
  sys.exit(wf.run(main))