##!/usr/bin/env python

# xmlweather.cgi is a python cgi script that parses National Weather Service XML weather data and displays it in a web page.
# Set the url variable with the weather station of your choice.  I include this in webpages using server-side includes with the
# following statement: <!--#include FILE="cgi-bin/xmlweather.cgi"-->  Notice also that you'll need to create a directory for the image
# cache and a file for the xml cache.  Both need to be writable by apache.
# Enjoy.  http://www.braggtown.com

## Import Python modules
import cgi, cgitb; cgitb.enable()
from urllib import urlopen, urlretrieve
from xml.dom import minidom
from sys import exit
import os
from os.path import exists, join, split
from time import time

## Set url variable.  Get url from http://www.nws.noaa.gov/data/current_obs/
url = "http://www.nws.noaa.gov/data/current_obs/KRDU.xml"
## Set cache variable to True to use local cache, False to always retrieve newest file
useCache = True
## Set variable to location of locally cached xml file
cache = "/home/jtuttle/public_html/weather_cache.xml"
## Set image directory.  This reduces network latency by utilizing local image directory.  Set to False if no local images.
imgDir = "/home/jtuttle/public_html/images"
imgDirUrl = "http://www.prairienet.org/~jtuttle/images/weather_icons"

# Acquire, parse, and normalize xml from NWS.  Return xml tree object
def getWebXML(url):
	try:			
		content = urlopen(url).read()
		xml = minidom.parseString(content)
		xml.normalize()	
		return 	xml
	except IOError:
		print "Content-Type: text/html\n"
		print " No valid data at ", url
		exit()

# Acquire, parse, and normalize xml from local cached file.  Return xml tree object
def getCacheXML(cache):
	xml = minidom.parse(cache)
	xml.normalize()
	return xml

# Parse out selected elements and populate variables.  Return variable dictionary
def getElements(xml):
	root = xml.getElementsByTagName('current_observation')
	imgroot = xml.getElementsByTagName('image')
	location = (root[0].getElementsByTagName('location')[0]).childNodes[0].data.encode('iso-8859-1')
	weather = (root[0].getElementsByTagName('weather')[0]).childNodes[0].data.encode('iso-8859-1')
	temperature_string = (root[0].getElementsByTagName('temperature_string')[0]).childNodes[0].data.encode('iso-8859-1')
	relative_humidity = (root[0].getElementsByTagName('relative_humidity')[0]).childNodes[0].data.encode('iso-8859-1')
	wind_string = (root[0].getElementsByTagName('wind_string')[0]).childNodes[0].data.encode('iso-8859-1')
	heat_index_string = (root[0].getElementsByTagName('heat_index_string')[0]).childNodes[0].data.encode('iso-8859-1')
	observation_time = (root[0].getElementsByTagName('observation_time')[0]).childNodes[0].data.encode('iso-8859-1')
	credit = (root[0].getElementsByTagName('credit')[0]).childNodes[0].data.encode('iso-8859-1')
	credit_URL = (root[0].getElementsByTagName('credit_URL')[0]).childNodes[0].data.encode('iso-8859-1')
	img_url = (imgroot[0].getElementsByTagName('url')[0]).childNodes[0].data.encode('iso-8859-1')
	if (imgDir) and exists(imgDir):
		img_url = cacheIcons(imgDir, img_url, imgDirUrl)	
	img_title = (imgroot[0].getElementsByTagName('title')[0]).childNodes[0].data.encode('iso-8859-1')
	img_link = (imgroot[0].getElementsByTagName('link')[0]).childNodes[0].data.encode('iso-8859-1')
	icon_url_base = (root[0].getElementsByTagName('icon_url_base')[0]).childNodes[0].data.encode('iso-8859-1') 
	icon_url_name = (root[0].getElementsByTagName('icon_url_name')[0]).childNodes[0].data.encode('iso-8859-1')
	icon_url = join(icon_url_base, icon_url_name)
	if (imgDir) and exists(imgDir):
		icon_url = cacheIcons(imgDir, icon_url, imgDirUrl)	

	elDict = {'location': location, 'weather': weather, 'temperature_string': temperature_string, 'relative_humidity': relative_humidity,\
	'wind_string': wind_string, 'heat_index_string': heat_index_string, 'observation_time': observation_time, 'credit': credit, \
	'credit_URL': credit_URL, 'img_url': img_url, 'img_title': img_title, 'img_link': img_link, 'icon_url': icon_url} 

	return elDict

# Determine if local cache of icon exists.  If not, copy to image cache directory
def cacheIcons(imgDir, url, imgDirUrl):
	path, icon = split(url)
	if not exists(join(imgDir,icon)):
		urlretrieve(url, join(imgDir, icon))
	return join(imgDirUrl, icon)	

# Format results and xhtml in verbose fashion
def formatDataVerbose(elDict, url):
	print "Content-Type: text/html\n"
	print """<img src="%s" border="0" alt="weather icon" align="left" width="32" height="32" /><br />\n""" \
	% (elDict["icon_url"])
	print "<h2>Local conditions at %s : %s </h2><br />\n" % (elDict["location"], elDict["observation_time"])
	print "The weather is %s and the temperature is %s." % (elDict["weather"].lower(), elDict["temperature_string"])
	print "The heat index is %s, the relative humidity is %s, and the wind is %s.<br />\n" % (elDict["heat_index_string"].lower(), \
	elDict["relative_humidity"].lower(), elDict["wind_string"].lower())
	print """<br /><a href="%s" title="%s"><img src="%s" alt="%s" border="0" /></a>""" % (elDict["img_link"], elDict["img_title"], \
	elDict["img_url"], elDict["img_title"])
	print "<!-- data source: %s -->" % (url)

# Format results and xhtml in succint fashion
def formatDataSuccinct(elDict, url):
	print "Content-Type: text/html\n"
	print """<img src="%s" border="0" alt="weather icon" width="32" height="32" /><br /><br />""" \
	% (elDict["icon_url"])
	print "<strong>Local conditions at %s : %s </strong><br />\n" % (elDict["location"], elDict["observation_time"])
	print "Current conditions: ", elDict["weather"], "<br />"
	print "Temperature: ", elDict["temperature_string"], "<br />"
	print "Relative Humidity: ", elDict["relative_humidity"], "&#37;<br />"
	print "Heat Index: ", elDict["heat_index_string"], "<br />"
	print "Wind: ", elDict["wind_string"], "<br />\n"
	print """<br /><a href="%s" title="%s"><img src="%s" alt="%s" border="0" /></a><br />""" % (elDict["img_link"], elDict["img_title"], \
	elDict["img_url"], elDict["img_title"])
	print "<!-- data source: %s -->" % (url)

# Determine age of cached file. If older than 1 hour, retrieve new data
def checkCacheAge(url, cache):
	cache_age = os.path.getmtime(cache)
	if ((time() - cache_age)/60 > 30):
   		content = urlopen(url)
   		f = open(cache, 'w')
   		f.write(content.read())
  		content.close()
   		f.close()


## main logic
if (useCache) and (exists(cache)):
	checkCacheAge(url, cache)
	xml = getCacheXML(cache)

else:	
	xml = getWebXML(url)

elDict = getElements(xml)
#formatDataVerbose(elDict, url)
formatDataSuccinct(elDict, url)