Merge pull request #1 from hexmind/master

thanks hexmind

.gitignore

@@ -0,0 +1,3 @@
+.idea
+*.html
+maps/*

bookmarkstokml.py

@@ -1,10 +1,6 @@
 # -*- coding: utf-8 -*-
 """
-Go to Google Bookmarks: https://www.google.com/bookmarks/
-
-On the bottom left, click "Export bookmarks": https://www.google.com/bookmarks/bookmarks.html?hl=en
-
-After downloading the html file, run this script on it to generate a KML.
+See readme.md
 
 """
 
@@ -13,81 +9,60 @@ import simplekml
 
 from urllib import FancyURLopener
 
-import os
-import random
 import re
 import sys
 import time
 
-# filename = r'GoogleBookmarks.html'
-filename = sys.argv[1]
+coords_in_content = re.compile('\/@(\d+\.\d+),(\d+\.\d+),')
+mobile_agent = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.73 Safari/537.36'
 
+filename = r'GoogleBookmarks.html'
+if(len(sys.argv) > 1):
+    filename = sys.argv[1]
+
+print 'opening ' + filename
 with open(filename) as bookmarks_file:
     data = bookmarks_file.read()
 
-# kml = simplekml.Kml()
-
-# Hacky and doesn't work for all of the stars:
-lat_re = re.compile('markers:[^\]]*latlng[^}]*lat:([^,]*)')
-lon_re = re.compile('markers:[^\]]*latlng[^}]*lng:([^}]*)')
-coords_in_url = re.compile('\?q=(-?\d{,3}\.\d*),\s*(-?\d{,3}\.\d*)')
-
 doc = document_fromstring(data)
 
 class Browser(FancyURLopener):
-    user_agents = [
-        'Mozilla/5.0 (Windows; U; Windows NT 5.1; it; rv:1.8.1.11) Gecko/20071127 Firefox/2.0.0.11',
-        'Opera/9.25 (Windows NT 5.1; U; en)',
-        'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET CLR 1.1.4322; .NET CLR 2.0.50727)',
-        'Mozilla/5.0 (compatible; Konqueror/3.5; Linux) KHTML/3.5.5 (like Gecko) (Kubuntu)',
-        'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.12) Gecko/20070731 Ubuntu/dapper-security Firefox/1.5.0.12',
-        'Lynx/2.8.5rel.1 libwww-FM/2.14 SSL-MM/1.4.1 GNUTLS/1.2.9'
-    ]
+    version = mobile_agent
 
-    version = random.choice(user_agents)
-
-for label in doc.body.iterfind('dl/dl/h3'):
+for label in doc.body.iterfind('dl/dt/h3'):
     labelName = label.text_content()
-    #print labelName
 
     kml = simplekml.Kml()
     kml.document.name = labelName
 
-    for element, attribute, url, pos in label.getnext().iterlinks():
+    for element, attribute, url, pos in label.getparent().getnext().iterlinks():
         if 'maps.google' in url:
+            print
             description = element.text or ''
-            print description.encode('UTF8')
-            print "URL: {0}".format(url)
-
+            print  'GET {0} {1}'.format(url, description.encode('UTF8'))
             browser = Browser()
 
-            if coords_in_url.search(url):
-                # Coordinates are in URL itself
-                latitude = coords_in_url.search(url).groups()[0]
-                longitude = coords_in_url.search(url).groups()[1]
-            else:
-                # Load map and find coordinates in source of page
-                sock = False
-
-                while not sock:
-                    try:
-                        sock = browser.open(url.replace(' ','+'))
-                    except Exception, e:
-                        print 'Connection problem:'
-                        print repr(e)
-                        print 'Retrying randomly between 15 and 60 seconds.'
-                        time.sleep(random.randint(15, 60))
-
-                content = sock.read()
-                sock.close()
-                time.sleep(random.randint(15, 60)) # Don't annoy server
+            # Load map and find coordinates in source of page
+            sock = False
+            while not sock:
                 try:
-                    latitude = lat_re.findall(content)[0]
-                    longitude = lon_re.findall(content)[0]
-                except IndexError:
-                    print '[Coordinates not found]'
-                    print
-                    continue
+                    sock = browser.open(url.replace(' ','+'))
+                except Exception, e:
+                    print 'Connection problem:' + repr(e)
+                    print 'Retrying randomly between 15 and 60 seconds.'
+                    time.sleep(random.randint(15, 60))
+
+            content = sock.read()
+            sock.close()
+
+            try:
+                coords = coords_in_content.search(content)
+                latitude = coords.groups()[0]
+                longitude = coords.groups()[1]
+
+            except (AttributeError, IndexError):
+                print '[Coordinates not found] ' + str(coords) + ' Try to update "mobile_agent"'
+                continue
 
             print latitude, longitude
             try:
@@ -95,6 +70,7 @@ for label in doc.body.iterfind('dl/dl/h3'):
                              coords=[(float(longitude), float(latitude))])
             except ValueError:
                 print '[Invalid coordinates]'
-            print
 
-    kml.save("./maps/" + labelName + ".kml")
+    output = './maps/' + labelName + '.kml'
+    print 'saving results to ' + output
+    kml.save(output)

readme.md

@@ -1,9 +1,12 @@
-To export your Google Maps starred locations:
+V2016.04.17
 
 Go to Google Bookmarks: https://www.google.com/bookmarks/
 
 On the bottom left, click "Export bookmarks": https://www.google.com/bookmarks/bookmarks.html?hl=en
 
+Install script dependencies:
+pip install simplekml
+
 After downloading the html file, run this script on it to generate a KML file per bookmark label.
 
-It's hacky and doesn't work on all of them, but it kinda works.
+It's hacky and doesn't work on all of them, but it kinda works.