I am trying to get addresses from a web page for mapping purposes. I know I could copy and paste the addresses into an Excel file and geocode them but I was wondering if there was any faster way to get the address from a web page and create a point location file?
-
It all depends on how the addresses are formatted on said web page, but yes parsing addressing into a .csv and then geocoding is quite possible.– MLowryJan 23, 2012 at 19:15
-
2It sounds like you are talking about web scraping. If that is the case, which isn't really a GIS question, consider asking the question over at StackOverflow, or searching their tags screen-scraping and web-scraping.– RyanKDaltonJan 23, 2012 at 19:41
-
as above, but it's fairly easy to write a small app that looks at a specific part of a web page and disemintaes the information.– HairyJan 24, 2012 at 12:15
Add a comment
|
1 Answer
This should get you started. Python and the BeautifulSoup module to the rescue. The code below will print out a list of the 26 addresses on that webpage. I used Firebug in Firefox to look at the page source, which told me that the cell width was 37%. I gambled that those cells were maybe the only ones at 37% width, and was right. You should be able to feed the list of addresses you get into a online geocoder and get point locations.
import BeautifulSoup as bs
import urllib2
url = 'http://www.phillypal.com/pal_locations.php'
response = urllib2.urlopen(url)
html = response.read()
soup = bs.BeautifulSoup(html)
addresses = soup.findAll('td', {'width':'37%'})
print len(addresses)
for address in addresses:
print address.find(text=True)
answered Jan 24, 2012 at 19:36
