29 Comments

1. 

   Simon Ghislain

    /  May 12, 2018

   Hello again,

   I just left a comment on another of your post, but I guess this one is closer to the bojective i’m trying to achieve…

   Btw i have again same kind of error when i run the code (entire python code snippet):
   AttributeError: module ‘urllib’ has no attribute ‘urlopen’

   I’m completely new to Google map API (never used it), but there seem to be a bunch of parameters that can be accessed from a itineray:
   https://developers.google.com/maps/documentation/directions/intro#Waypoints

   (with MAPS Direction API, sorry it’s french version of the site)

   Do you think the process of collecting all points of the road (as well as directions –to be looking ahead on the road, not sideways–) can be done only with the API, without ArcGIS?

   Oh god, I really don’t know how to put all this together…I even don’t know yet how I can use that Google Maps API, I never used JSON i don’t know what is it yet 😀

   Anyway, thank you again, i’m very glad to discover someone has already done something similar to what i wish to do…

   Going back to analyse your code…
   Hoping to hear from you soon!

   Reply
   • 

     apwheele

      /  May 12, 2018

     It sounds like you have most of the ingredients. Using the directions API you can get JSON for the route. At that point you just need to write code to interpolate along the path at regular intervals. That will be some work though to figure that out with the spherical coordinates. It might be easier to project the coordinates and use shapely, http://shapely.readthedocs.io/en/stable/manual.html.

     To generate the route foregoing the Google directions API you might want to check out OSMnx, http://geoffboeing.com/2016/11/osmnx-python-street-networks/.

     Reply
     • 

       Simon Ghislain

        /  May 12, 2018

       Thank you for your answer…
       I found someone doing exactly that, but in Java:
       https://stackoverflow.com/questions/47784512/plotting-coordinates-on-route-in-gmap-google-maps-android-api

       I would like to do it in Python, as I already “took in hands” your code and it’s working…
       I installed googlemaps python API…but i don’t know where to start…

       could you give me an example of like 5 lines of codes to use googlemaps API to generate itinerary from A to B, and how to access the .routes, .legs, .steps, etc…information?

       Again, thank you so much for any help, you save me

     • 

       apwheele

        /  May 12, 2018

       I don’t have code to do your exact task, so you will need to strike out some on your own. The order of tasks will be something like:

       – use begin & end locations to query a route via the google directions api.
       – take that route in spherical coordinates, project it into any suitable local projection (can use pyproj)
       – use the shapely library to interpolate points at regular intervals along the route
       – calculate the local orientation of the road from those interpolated points
       – reproject the interpolated points back to spherical coordinates
       – grab the streetview images at those interpolated points given your preferred orientation of the camera (which you can figure out via knowing the direction of the road)
       – stitch the resulting images into a GIF

       You should add a small inset map that shows a point flowing along the route where the images are for as well ;).

       Pretty sure that is doable, but will take some work.

     • 

       Simon Ghislain

        /  May 12, 2018

       Thank you very much, I will try that, and I keep you updated on my full code once it is working!
       Hope i’ll be back soon…
       Thank you again

     • 

       Simon Ghislain

        /  May 12, 2018

       I DID IT!!
       I F***** MANAGED TO DO IT!
       (except orientation sector)

       Here is my brute code:

       import json
       import urllib.request
       import urllib.parse
       import os
       from pyproj import Proj
       import math
       myloc = r”C:\Gmap\01″
       key = “&key=” + “”

       def distance_cart(p1, p2):
       return( math.sqrt((p1[0]-p2[0])*(p1[0]-p2[0]) + (p1[1]-p2[1])*(p1[1]-p2[1])))

       def GetStreet(Add,num, SaveLoc):
       base = “https://maps.googleapis.com/maps/api/streetview?size=1200×800&location=”
       MyUrl = base + urllib.parse.quote_plus(Add) + “&heading=” + str(00.00) + key #added url encoding
       fi = str(num) + “.jpg”
       urllib.request.urlretrieve(MyUrl, os.path.join(SaveLoc,fi))

       def decode_polyline(polyline_str):
       index, lat, lng = 0, 0, 0
       coordinates = []
       changes = {‘latitude’: 0, ‘longitude’: 0}

       while index < len(polyline_str):
       for unit in ['latitude', 'longitude']:
       shift, result = 0, 0

       while True:
       byte = ord(polyline_str[index]) – 63
       index+=1
       result |= (byte & 0x1f) <= 0x20:
       break

       if (result & 1):
       changes[unit] = ~(result >> 1)
       else:
       changes[unit] = (result >> 1)

       lat += changes[‘latitude’]
       lng += changes[‘longitude’]

       coordinates.append((lat / 100000.0, lng / 100000.0))

       return coordinates

       start = “Ottignies, Belgique”
       finish = “Namur, Belgique”

       url = ‘http://maps.googleapis.com/maps/api/directions/json?%s’ % urllib.parse.urlencode((
       (‘origin’, start),
       (‘destination’, finish)
       ))
       ur = urllib.request.urlopen(url)
       result = json.load(ur)
       coord=[]
       for i in range (0, len (result[‘routes’][0][‘legs’][0][‘steps’])):
       points1 = result[‘routes’][0][‘legs’][0][‘steps’][i][‘polyline’][‘points’]
       if (points1 is not None):
       coord.append(decode_polyline(points1))

       cc=[]
       cce=[]
       for i in range (0, len(coord)):
       for j in range(0, len(coord[i])):
       cc.append(coord[i][j])
       lat=coord[i][j][0]
       long=coord[i][j][1]
       coord_f = str(lat) + “,” + str(long)

       p = Proj(proj=’utm’, zone=10, ellps=’WGS84′) # use kwargs
       x, y = p(lat, long)
       cce.append([x,y])

       mess= ‘x=%9.3f’ % cce[i][0]

       #print(mess)

       #GetStreet(Add=coord_f,SaveLoc=myloc)

       print(len(cc))
       print(len(cce))

       res=[]
       p0=cce[0]
       res.append(p0)
       temp=0
       prev=p0
       distance_max=500
       for i in range (1, len(cce)):
       temp = temp + distance_cart(cce[i], prev)
       if (temp>distance_max):
       res.append(cce[i])
       temp=0
       prev=cce[i]

       print(len(res))

       final=[]
       for i in range (0, len(res)):
       p = Proj(proj=’utm’, zone=10, ellps=’WGS84′)
       lat, long = p(res[i][0], res[i][1], inverse=True)
       coord_f = str(lat) + “,” + str(long)
       GetStreet(Add=coord_f, num=i, SaveLoc=myloc)

       It saves all the images..it’s working perfectly! withtout shapely : a simply sqrt to compute distances…

2. 

   Simon Ghislain

    /  May 12, 2018

   But it seems i’ve been banned now because i didnt use the key while devugging…
   How can I know I’m banned?

   At run, I get a cascade of error starting from urlretrieve() to
   request.py”, line 650, in http_error_default
   raise HTTPError(req.full_url, code, msg, hdrs, fp)
   urllib.error.HTTPError: HTTP Error 403: Forbidden

   Error 403 forbidden?
   I worked 2min ago, i changed nothing… :-s

   Reply
   • 

     apwheele

      /  May 12, 2018

     Nice, yeah just set up a new api key, make sure it works, and then you should be good to go. Sometimes with the google APIs it will let me get ~100 requests before it bans me, but some of the api’s don’t work at all without a key.

     Reply
     • 

       Simon Ghislain

        /  May 12, 2018

       Yeah definitely looks like i’m banned…
       I tried with a valid key (i know it’s valid because i made a request using googlemaps module and i worked)

       The line triggering the errors is in GetStreet() function:
       urllib.request.urlretrieve(MyUrl, os.path.join(SaveLoc,fi))

       It Triggers all the way to HTTP Error 403: Forbidden

       How long generally are you banned?
       Are there methods to “reset” the ~100 requests quota without key?

       I have several API keys in Google api console (credentials tab)…without restrictions and enabled for API Directions (but not for Google street view…and i can’t find it)

       Again, thank you so much for your help, I would never have been able to achieve that without you…

3. 

   Simon Ghislain

    /  May 12, 2018

   Oh…sorry…nevermind my last message, I just found the Street View API, i just had to get an API key for that particular API, and now it works again 🙂

   Reply
4. 

   Simon Ghislain

    /  May 12, 2018

   Here is my final workingg code (with correct orientation+saving all images into a .gif and a .mp5 video):

   import json
   import urllib.request
   import urllib.parse
   import googlemaps
   import os
   from pyproj import Proj
   import math
   import imageio
   import sys
   import moviepy.editor as mp

   myloc = r”C:\Gmap\01″
   key = “&key=” + “API_KEY”
   res_dir=[]
   filesn=[]

   def distance_cart(p1, p2):
   return( math.sqrt((p1[0]-p2[0])*(p1[0]-p2[0]) + (p1[1]-p2[1])*(p1[1]-p2[1])))

   def dir_cart(p1, p2):
   if p1[0]-p2[0] == 0:
   return math.degrees(math.pi/2.0)
   else:
   return( math.degrees(math.atan((p1[1]-p2[1]) / (p1[0]-p2[0]))))

   def create_gif(filenames, duration):
   images = []
   for filename in filenames:
   images.append(imageio.imread(filename))
   output_file = ‘Gif-1.gif’
   imageio.mimsave(output_file, images, duration=duration)

   def GetStreet(Add,num, SaveLoc):
   base = “https://maps.googleapis.com/maps/api/streetview?size=1200×800&location=”
   MyUrl = base + urllib.parse.quote_plus(Add) + “&heading=” + str(res_dir[num]+180.00) + key
   print(“processing ” + str(num) + “//” + str(len(res_dir)))
   fi = str(num) + “.jpg”
   filesn.append(str(os.path.join(SaveLoc, fi)))
   urllib.request.urlretrieve(MyUrl, os.path.join(SaveLoc,fi))

   def decode_polyline(polyline_str):
   index, lat, lng = 0, 0, 0
   coordinates = []
   changes = {‘latitude’: 0, ‘longitude’: 0}

   while index < len(polyline_str):
   for unit in ['latitude', 'longitude']:
   shift, result = 0, 0

   while True:
   byte = ord(polyline_str[index]) – 63
   index+=1
   result |= (byte & 0x1f) <= 0x20:
   break

   if (result & 1):
   changes[unit] = ~(result >> 1)
   else:
   changes[unit] = (result >> 1)

   lat += changes[‘latitude’]
   lng += changes[‘longitude’]

   coordinates.append((lat / 100000.0, lng / 100000.0))

   return coordinates

   itin=[]
   #create a list of checkpoints for the entire itinerary:
   itin.append(“Bruxelles, Belgique”)
   itin.append(“Ottignies, Belgique”)
   itin.append(“Namur, Belgique”)

   cc=[]
   cce=[]
   print(“len itin: ” + str(len(itin)))

   for k in range(1, len(itin)):

   url = ‘http://maps.googleapis.com/maps/api/directions/json?%s’ % urllib.parse.urlencode((
   (‘origin’, itin[k-1]),
   (‘destination’, itin[k])
   ))
   ur = urllib.request.urlopen(url)
   result = json.load(ur)
   coord=[]
   for i in range (0, len (result[‘routes’][0][‘legs’][0][‘steps’])):
   points1 = result[‘routes’][0][‘legs’][0][‘steps’][i][‘polyline’][‘points’]
   if (points1 is not None):
   coord.append(decode_polyline(points1))

   for i in range (0, len(coord)):
   for j in range(0, len(coord[i])):
   cc.append(coord[i][j])
   lat=coord[i][j][0]
   long=coord[i][j][1]

   p = Proj(proj=’utm’, zone=10, ellps=’WGS84′) # use kwargs
   x, y = p(lat, long)
   cce.append([x,y])

   print(“len cce ” + str(k) + “:” + str(len(cce)))
   print(“len cce tot: ” + str(len(cce)))
   res=[]

   p0=cce[0]
   res.append(p0)
   temp=0
   prev=p0
   distance_max=250
   for i in range (1, len(cce)):
   temp = temp + distance_cart(cce[i], prev)
   if (temp>distance_max):
   res.append(cce[i])
   res_dir.append(dir_cart(cce[i], prev))
   temp=0
   prev=cce[i]
   res.append(cce[len(cce)-1])
   res_dir.append(dir_cart(cce[len(cce)-1], prev))
   print(len(res))
   print(“resdir: ” + str(len(res_dir)))

   for i in range (0, len(res)-1):
   #for i in range(0, 10):
   p = Proj(proj=’utm’, zone=10, ellps=’WGS84′)
   lat, long = p(res[i][0], res[i][1], inverse=True)
   coord_f = str(lat) + “,” + str(long)
   GetStreet(Add=coord_f, num=i, SaveLoc=myloc)

   create_gif(filesn, .2)

   clip = mp.VideoFileClip(“Gif-1.gif”)
   clip.write_videofile(“gifvideo.mp4”)

   Reply
   • 

     Oswaldo Reategui

      /  May 13, 2020

     Hi dear Simon,
     Could you please provide me the python file for your project? I tried to replicate this code but a lot of errors popped up.
     Thank you.

     Reply
5. 

   Valerio De Luca

    /  January 4, 2019

   NameError: global name ‘mid’ is not defined

   Reply
6. 

   hieund198

    /  May 13, 2019

   So sorry, I don’t know the place “mid” in your code define
   MyUrl = base + mid + end
   fi = File + “.jpg”
   MetaUrl = base + r”/metadata” + size + end

   Could you tell me what’s the “mid”?

   Thanks

   Reply
   • 

     apwheele

      /  May 13, 2019

     You are right, that is a typo. “Mid” should be “size”, I will update the code snippet.

     Reply
     • 

       hieund198

        /  May 13, 2019

       Thank you, I got run your code. It’s helpful for me

7. 

   Fatin Abdilah

    /  January 6, 2021

   Dear Andrew, I’m so glad I found this post. I’m unsure if you’re familiar with Treepedia (https://github.com/mittrees/Treepedia_Public). The second script of Treepedia calls for the Google Street API to collect metadatas containing panoIDs of point locations (shp) (created in its first script) to be fed into its third and final script. However, since Treepedia was created before Google updated their API urls, I’m having trouble calling the API. Could you please take a quick look at the script. I’m at my wits end. I suspect the problem is the urlAddress line.

   # This function is used to collect the metadata of the GSV panoramas based on the sample point shapefile

   # Copyright(C) Xiaojiang Li, Ian Seiferling, Marwa Abdulhai, Senseable City Lab, MIT

   def GSVpanoMetadataCollector(samplesFeatureClass,num,ouputTextFolder):
   ”’
   This function is used to call the Google API url to collect the metadata of
   Google Street View Panoramas. The input of the function is the shpfile of the create sample site, the output
   is the generate panoinfo matrics stored in the text file

   Parameters:
   samplesFeatureClass: the shapefile of the create sample sites
   num: the number of sites proced every time
   ouputTextFolder: the output folder for the panoinfo

   ”’

   import urllib.request
   import google_streetview.api
   import xmltodict
   import io
   import ogr, osr
   import time
   import os,os.path
   import json

   if not os.path.exists(ouputTextFolder):
   os.makedirs(ouputTextFolder)

   driver = ogr.GetDriverByName(‘ESRI Shapefile’)

   # change the projection of shapefile to the WGS84
   dataset = driver.Open(samplesFeatureClass)
   layer = dataset.GetLayer()

   sourceProj = layer.GetSpatialRef()
   targetProj = osr.SpatialReference()
   targetProj.ImportFromEPSG(4326)
   transform = osr.CoordinateTransformation(sourceProj, targetProj)

   # loop all the features in the featureclass
   feature = layer.GetNextFeature()
   featureNum = layer.GetFeatureCount()
   batch = featureNum//num

   for b in range(batch):
   # for each batch process num GSV site
   start = b*num
   end = (b+1)*num
   if end > featureNum:
   end = featureNum

   ouputTextFile = ‘Pnt_start%s_end%s.txt’%(start,end)
   ouputGSVinfoFile = os.path.join(ouputTextFolder,ouputTextFile)

   # skip over those existing txt files
   if os.path.exists(ouputGSVinfoFile):
   continue

   time.sleep(1)

   with open(ouputGSVinfoFile, ‘w’) as panoInfoText:
   # process num feature each time
   for i in range(start, end):
   feature = layer.GetFeature(i)
   geom = feature.GetGeometryRef()

   # trasform the current projection of input shapefile to WGS84
   #WGS84 is Earth centered, earth fixed terrestrial ref system
   geom.Transform(transform)
   lon = geom.GetX()
   lat = geom.GetY()
   key = r’????????????????’ #Input Your Key here

   # get the meta data of panoramas
   # previous link by the code authors: http://maps.google.com/cbk?output=xml&ll=%s,%s
   urlAddress = r’https://maps.googleapis.com/maps/api/streetview’%(lat,lon)

   time.sleep(0.05)
   # the output result of the meta data is a xml object
   metaDataxml = urllib2.urlopen(urlAddress)
   metaData = metaDataxml.read()

   data = xmltodict.parse(metaData)

   # in case there is not panorama in the site, therefore, continue
   if data[‘panorama’]==None:
   continue
   else:
   panoInfo = data[‘panorama’][‘data_properties’]

   # get the meta data of the panorama
   panoDate = panoInfo.items()[4][1]
   panoId = panoInfo.items()[5][1]
   panoLat = panoInfo.items()[8][1]
   panoLon = panoInfo.items()[9][1]

   # print ‘The coordinate (%s,%s), panoId is: %s, panoDate is: %s’%(panoLon,panoLat,panoId, panoDate)
   # lineTxt = ‘panoID: %s panoDate: %s longitude: %s latitude: %s\n’%(panoId, panoDate, panoLon, panoLat)
   # panoInfoText.write(lineTxt)
   print(‘The coordinate (%s,%s), panoId is: %s, panoDate is: %s’ % (pano_lon, pano_lat, pano_id,
   pano_date))
   line_txt = ‘panoID: %s panoDate: %s longitude: %s latitude: %s\n’ % (pano_id, pano_date, pano_lon,
   pano_lat)
   panoInfoText.write(line_txt)

   panoInfoText.close()

   # ————Main Function ——————-
   if __name__ == “__main__”:
   import os, os.path

   root = ‘C:\\Users\\fatin\\AppData\Local\Programs\Python\Python36\Scripts\Treepedia\Data’
   inputShp = os.path.join(root,’GeorgeTown20.shp’)
   outputTxt = root

   GSVpanoMetadataCollector(inputShp,1000,outputTxt)

   Thanks so much.

   Reply
   • 

     apwheele

      /  January 6, 2021

     This will take alittle bit of work, but should be doable. (When pasting comments don’t paste your key! I have deleted it out.)

     You need to get the panoid from the metadata api, https://developers.google.com/maps/documentation/streetview/metadata?hl=en, so the url line should be :

     urlAddress = f’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key={your_key}’

     This api only returns results in json, so you will need to update the code the parses the xml to parse the json like my post. After that it should be OK with the newer API.

     Reply
     • 

       Fatin Abdilah

        /  January 6, 2021

       Thanks Andrew for the help. I tried to parse the json, unfortunately the log said:

       Traceback (most recent call last):
       File “C:\Users\fatin\AppData\Local\Programs\Python\Python36\Scripts\Treepedia\metadataCollector1.py”, line 124, in
       GSVpanoMetadataCollector(inputShp,1000,outputTxt)
       File “C:\Users\fatin\AppData\Local\Programs\Python\Python36\Scripts\Treepedia\metadataCollector1.py”, line 77, in GSVpanoMetadataCollector
       urlAddress = r’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key=**’ % (lat,lon)
       TypeError: not all arguments converted during string formatting

       **what I edited**

       time.sleep(0.05)
       # the output result of the meta data is a xml object
       response = urllib2.urlopen(urlAddress)
       jsonRaw = response.read()
       jsonData = json.loads(jsonRaw)
       if jsonData[‘status’] == “OK”:
       if ‘date’ in jsonData:
       return (jsonData[‘date’],jsonData[‘pano_id’]) #sometimes it does not have a date!
       else:
       return (None,jsonData[‘pano_id’])
       else:
       return (None,None)

       # in case there is not panorama in the site, therefore, continue
       if jsonData[‘panorama’]==None:
       continue
       else:
       panoInfo = jsonData[‘panorama’][‘data_properties’]

       *****

       I’m obviously new to python so I’m kind of lost at the moment. Am I going the right direction though?

     • 

       apwheele

        /  January 6, 2021

       So for this line:

       urlAddress = r’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key=**’ % (lat,lon)

       That will not work, as you need to pass in the key now directly. It won’t return any results without a key. Your code still just pipes in the lat/lon, and does not include the key. So I use a formatted string (your not using an old V2.? version of python are you?)

       urlAddress = f’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key={key}’

       If you are using an older version, it will need to be something like:

       url_base = r’https://maps.googleapis.com/maps/api/streetview/metadata?location=’
       urlAddress = url_base + str(lat) + ‘,’ + str(lon) + ‘&key=’ + key

       or use some other type of string substitution, see

       String substitution in Python

       or

       String substitution in Python

     • 

       Fatin Abdilah

        /  January 6, 2021

       Dear Andrew,

       Yes I did include the API key, but for the sake of commenting, I changed it to the asterisks. I should’ve noted that sorry bout that. I’m currently using python 3.6 but I’ve yet to try (and tweak it for) python 2.7.

       Like I mentioned the log reported:
       *****
       Traceback (most recent call last):
       File “C:\Users\fatin\AppData\Local\Programs\Python\Python36\Scripts\Treepedia\metadataCollector1.py”, line 124, in
       GSVpanoMetadataCollector(inputShp,1000,outputTxt)
       File “C:\Users\fatin\AppData\Local\Programs\Python\Python36\Scripts\Treepedia\metadataCollector1.py”, line 77, in GSVpanoMetadataCollector
       urlAddress = r’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key=**’ % (lat,lon)
       TypeError: not all arguments converted during string formatting

       *****
       Did I write the json parse codes correctly?

       **what I edited**

       time.sleep(0.05)
       # the output result of the meta data is a xml object
       response = urllib2.urlopen(urlAddress)
       jsonRaw = response.read()
       jsonData = json.loads(jsonRaw)
       if jsonData[‘status’] == “OK”:
       if ‘date’ in jsonData:
       return (jsonData[‘date’],jsonData[‘pano_id’]) #sometimes it does not have a date!
       else:
       return (None,jsonData[‘pano_id’])
       else:
       return (None,None)

       # in case there is not panorama in the site, therefore, continue
       if jsonData[‘panorama’]==None:
       continue
       else:
       panoInfo = jsonData[‘panorama’][‘data_properties’]

       *****

     • 

       apwheele

        /  January 6, 2021

       That error has to do with the line:

       urlAddress = r’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key=**’ % (lat,lon)

       Which is wrong as written. Again I think it is easier to just do f string substitution (since you are on python 3 it is available):

       urlAddress = f’https://maps.googleapis.com/maps/api/streetview/metadata?location={lat},{lon}&key=**’

       You should have enough though with my post to figure out the Json on your own, it is a pretty simple return nested dictionary, and it looks like you have the error condition as well.

   • 

     apwheele

      /  January 6, 2021

     You may also be interested in this post, https://andrewpwheeler.com/2020/10/24/using-the-google-vision-and-streetview-api-to-explore-hotspots/. I use the Google Vision API to get labels for items in the image, which includes several green space related items.

     Reply
     • 

       Fatin Abdilah

        /  January 7, 2021

       Dear Andrew,

       I managed to get an updated version of the script. However, I haven’t been able to execute it correctly using bash. The author mentioned an example to call the inputshp, outdir, key variable:

       $ python metadataCollector.py –inputshp=’C:/Users/senseablecity/Dropbox (MIT)/ResearchProj/treepedia/cities-proj/Oakland/OaklandSlowStreets/SlowStreets_points/SS_20m.shp’ –outdir=’C:/Users/senseablecity/Dropbox (MIT)/ResearchProj/treepedia/cities-proj/Oakland/OaklandSlowStreets/SlowStreets_points’ –key=’***’

       However the log mentioned that it couldn’t recognise my C:/Users/… directory. Could you show me the correct way to call data for the argparse with path directories?

       **** the updated script****

       def GSVpanoMetadataCollector(samplesFeatureClass,num,ouputTextFolder,key):
       ”’
       This function is used to call the Google API url to collect the metadata of
       Google Street View Panoramas. The input of the function is the shpfile of the create sample site, the output
       is the generate panoinfo matrics stored in the text file

       Parameters:
       samplesFeatureClass: the shapefile of the create sample sites
       num: the number of sites proced every time
       ouputTextFolder: the output folder for the panoinfo

       ”’
       import urllib,urllib2
       #import urllib.request
       #import google_streetview.api
       import xmltodict
       import ogr, osr
       import time
       import os,os.path
       import sys
       import json

       if not os.path.exists(ouputTextFolder):
       os.makedirs(ouputTextFolder)

       driver = ogr.GetDriverByName(‘ESRI Shapefile’)

       # change the projection of shapefile to the WGS84
       dataset = driver.Open(samplesFeatureClass)
       layer = dataset.GetLayer()

       #sourceProj = layer.GetSpatialRef()
       #targetProj = osr.SpatialReference()
       #targetProj.ImportFromEPSG(4326)
       #transform = osr.CoordinateTransformation(sourceProj, targetProj)

       # loop all the features in the featureclass
       feature = layer.GetNextFeature()
       featureNum = layer.GetFeatureCount()
       #batch = featureNum//num
       batch = int(featureNum/num + 0.5)

       print(batch)
       for b in range(batch):
       # for each batch process num GSV site
       start = b*num
       geom.Transform(transform)
       lon = geom.GetX()
       lat = geom.GetY()

       #get the metadata of panoramas
       urlAddress = r’https://maps.googleapis.com/maps/api/streetview/metadata?size=600×300&location=%s,%s&heading=-45&pitch=42&fov=110&key=%s’%(lon, lat, key)
       time.sleep(0.1)

       time.sleep(0.05)
       #the output result of the metadata is a xml object
       #using different url reading method in python 2 and python 3
       if sys.version_info[0] ==2:
       request = urllib.request.Request(urlAddress)
       metaData = urllib.request.urlopen(request).read()

       data = json.loads(metaData)
       panoDate = data[‘date’]
       panoId = data[‘pano_id’]
       panoLat = data[‘location’][‘lat’]
       panoLon = data[‘location’][‘lng’]

       lineTxt = ‘panoID: %s panoDate: %s longitude: %s latitude: %s\n’%(panoId, panoDate, panoLon, panoLat)
       panoInfoText.write(lineTxt)

       panoInfoText.close()

       # ————Main Function ——————-
       if __name__ == “__main__”:
       import os, os.path
       import sys
       import argparse

       parser = argparse.ArgumentParser(
       description = “parameters”
       )

       parser.add_argument(
       “–inputshp”,
       required = True,
       type = str,
       help = “the path of the shapefile”
       )

       parser.add_argument(
       “–outdir”,
       default = “mosaic-mrt.tif”,
       help = “the output dir for the meta text file”,
       type = str,
       )

       parser.add_argument(
       “–key”,
       default = “”,
       help = “Google Street View key”,
       type = str,
       )

       args = parser.parse_args()
       inputshp = args.inputshp
       outdir = args.outdir
       key = args.key

       print(inputshp, outdir, key)
       GSVpanoMetadataCollector(inputshp, 1000, outdir, key)

       ****end of script***

     • 

       apwheele

        /  January 7, 2021

       In the command line don’t do the equal signs:

       python script.py –inputshp ‘C:/Users/whatever’

       Your script it doesn’t look like to me it actually writes to the results to the specified output directory.

8. 

   eddie

    /  October 11, 2021

   I have same issue in treepedia.
   Thanks to you, I’ve figured out what the cause is, but haven’t been able to solve it yet.
   did you ever solve…??

   Reply

1. Using the Google Vision and Streetview API to Explore Hotspots | Andrew Wheeler
2. Using google places API in criminology research? | Andrew Wheeler
3. GUI Tool to download Google Streetview Imagery | Andrew Wheeler