4 # gcget -- screen scrape Geocaching.com's annoying web interface
5 # aka SHOW ME THE CACHE!!!
7 # Copyright 2007, Evan Battaglia
8 # Distributed under the terms of the GPL v2.
11 # requires module mechanize
14 # DEFAULT USERNAME AND PASSWORD: THESE ARE OVERWRITTEN BY COMMAND-LINE OPTIONS
20 # this has some extra args in:
21 # gcget lat,lon maxnumgcs [maxdist] [threshold]
22 # threshold -- if find more than this # of geocaches, don't get ANY,
23 # instead give warning and quit
30 This program is free software, distributed under the terms of the GNU GPL v2.
32 Usage: gcget [-u username] [-p password] lat,lon maxnumberofgcs [maxdistance] [threshold]
33 Downloads up to maxnumberofgcs at a distance of up to maxdistance from lat,lon.
34 If we number of geocaches within maxdistance is above threshold, don't download any
35 geocaches, just give a warning and quit.
37 If username and password are not given, will use default values hard-coded in script.
43 # PARSE OPTIONS: USERNAME AND PASSWORD
48 opts, args = getopt.gnu_getopt(sys.argv[1:], "u:p:d", ["help"])
49 except getopt.GetoptError:
50 # print help information and exit:
61 if o == "--help" or o == "-h":
71 #########################
72 #ll = args[0].split(",")
75 #The following line replaced the previous 3 lines.
76 lat, lon = args[0].split(",")
84 threshold = int(args[3])
88 # rounds up to multiples of 20. 20
89 n = int((int(args[1])+19)/20)
92 from mechanize import Browser
95 # get magic number for "Next" button.
96 # this is normally 16 (link hidden is $ctl16), unless there are less than 10 pages of results,
97 # in which case it will be less (e.g. 09 for 3 pages of results)
98 def getmagicnumber(b):
99 for i in range(16,0,-1):
100 if re.compile("pgrBottom.ctl%02d" % i).search(b.response().get_data()):
105 b.open("http://geocaching.com/seek/")
106 b.follow_link(text="Log in")
108 # The Username and Password fields on the Login form changed
109 b["ctl00$ContentBody$myUsername"] = USER
110 b["ctl00$ContentBody$myPassword"] = PASS
113 magicnumber = 0 # the ctl number of Next. get only once
115 try: b.select_form("form4")
118 b.select_form("form4")
120 b.select_form("form4")
121 print >> sys.stderr, "Invalid username/password"
123 f=open("gcget.badlogin.html","w")
124 f.write(b.response().get_data())
126 print >> sys.stderr, "Dumping last HTML page recieved into gcget.badlogin.html"
129 b["origin_lat"] = lat
130 b["origin_long"] = lon
134 thresholdre = re.compile("Total Records: <b>([0-9]*)</b>")
135 m = thresholdre.search(b.response().get_data())
137 if int(m.group(1)) > threshold:
138 sys.stderr.write("THRESHOLD %d > %d\n" % (int(m.group(1)), threshold))
141 records = int(m.group(1))
142 sys.stderr.write("ok found %d, getting min(%d,%d) gcs\n" % (int(m.group(1)), int(records), int(args[1])))
144 print "can't find total records"
148 # (records+19)/20 is the max pages
149 for ii in range(min(n,(records+19)/20)):
152 b['CID'] = [i.name for i in b.find_control('CID').items]
157 # only print one header, start of xml file
158 lines = b.response().get_data().split("\n")
160 print "\n".join(lines[0:2])
163 print "\n".join(lines[2:-1])
168 sys.stderr.write("i")
174 magicnumber = getmagicnumber(b)
176 # print "couldn't find magic number!" # why does this happen?
180 [f for f in b.forms()][0].new_control("hidden", "ctl00$ContentBody$pgrBottom$ctl%02d" % magicnumber, {})
183 sys.stderr.write("\n")
188 # f=open("delmeNOW","w")
189 # f.write(b.response().get_data())