Getting street,city and country by reverse geocoding using google
Asked Answered
A

4

13

I'm trying to getting the $street, $city and $country string from google json. It works for my home address : http://maps.googleapis.com/maps/api/geocode/json?latlng=52.108662,6.307370&sensor=true

$url = "http://maps.googleapis.com/maps/api/geocode/json?latlng=".$lat.",".$lng."&sensor=true";
    $data = @file_get_contents($url);
    $jsondata = json_decode($data,true);
    if(is_array($jsondata) && $jsondata['status'] == "OK")
    {
          $city = $jsondata['results']['0']['address_components']['2']['long_name'];
          $country = $jsondata['results']['0']['address_components']['5']['long_name'];
          $street = $jsondata['results']['0']['address_components']['1']['long_name'];
    }

But for a different address with more data in the arrays like this example: http://maps.googleapis.com/maps/api/geocode/json?latlng=52.154184,6.199592&sensor=true it doesn't work, because there is more data in the json array and it makes the province the country.

How can I select the type that I need (long_name)?

  • for street : long_name where "types" : [ "route" ]
  • for city : long_name where "types" : [ "locality", "political" ]
  • for country : long_name where "types" : [ "country", "political" ]

Example output from the geocode JSON:

{
   "results" : [
      {
         "address_components" : [
            {
               "long_name" : "89",
               "short_name" : "89",
               "types" : [ "street_number" ]
            },
            {
               "long_name" : "Wieck De",
               "short_name" : "Wieck De",
               "types" : [ "establishment" ]
            },
            {
               "long_name" : "Industrieweg",
               "short_name" : "Industrieweg",
               "types" : [ "route" ]
            },
            {
               "long_name" : "Zutphen",
               "short_name" : "Zutphen",
               "types" : [ "locality", "political" ]
            },
            {
               "long_name" : "Zutphen",
               "short_name" : "Zutphen",
               "types" : [ "administrative_area_level_2", "political" ]
            },
            {
               "long_name" : "Gelderland",
               "short_name" : "GE",
               "types" : [ "administrative_area_level_1", "political" ]
            },
            {
               "long_name" : "Nederland",
               "short_name" : "NL",
               "types" : [ "country", "political" ]
            },
            {
               "long_name" : "7202 CA",
               "short_name" : "7202 CA",
               "types" : [ "postal_code" ]
            }

I think I fixed it myself, hereby my code:

// street
foreach ($jsondata["results"] as $result) {
    foreach ($result["address_components"] as $address) {
        if (in_array("route", $address["types"])) {
            $street = $address["long_name"];
        }
    }
}
// city
foreach ($jsondata["results"] as $result) {
    foreach ($result["address_components"] as $address) {
        if (in_array("locality", $address["types"])) {
            $city = $address["long_name"];
        }
    }
}
// country
foreach ($jsondata["results"] as $result) {
    foreach ($result["address_components"] as $address) {
        if (in_array("country", $address["types"])) {
            $country = $address["long_name"];
        }
    }
}
Aegyptus answered 11/8, 2013 at 15:0 Comment(0)
A
10

You could convert the data to the associative array and work with it like

 $data = array();
 foreach($jsondata['results']['0']['address_components'] as $element){
     $data[ implode(' ',$element['types']) ] = $element['long_name'];
 }
 print_r($data);

 echo 'route: ' . $data['route'] . "\n";
 echo 'country: ' . $data['country political'];
Anglophile answered 11/8, 2013 at 15:12 Comment(0)
B
3

Your code is perfectly good, but wouldn't it be better to use a switch inside 1 foreach instead of repeated foreach loops? Here is how I parse the exact same array :

  $location = array();

  foreach ($result['address_components'] as $component) {

    switch ($component['types']) {
      case in_array('street_number', $component['types']):
        $location['street_number'] = $component['long_name'];
        break;
      case in_array('route', $component['types']):
        $location['street'] = $component['long_name'];
        break;
      case in_array('sublocality', $component['types']):
        $location['sublocality'] = $component['long_name'];
        break;
      case in_array('locality', $component['types']):
        $location['locality'] = $component['long_name'];
        break;
      case in_array('administrative_area_level_2', $component['types']):
        $location['admin_2'] = $component['long_name'];
        break;
      case in_array('administrative_area_level_1', $component['types']):
        $location['admin_1'] = $component['long_name'];
        break;
      case in_array('postal_code', $component['types']):
        $location['postal_code'] = $component['long_name'];
        break;
      case in_array('country', $component['types']):
        $location['country'] = $component['long_name'];
        break;
    }

  }
Behlke answered 13/2, 2014 at 9:59 Comment(1)
This won't work because the first component is the most reliable, just make the cases, if loops and it will be fine.Later
E
2

If you use Postal Code to find the address, as i have recently generated street,city, country using Google MAP API the code is:

$search_code = urlencode($postcode);
        $url = 'http://maps.googleapis.com/maps/api/geocode/json?address=' . $search_code . '&sensor=false';
        $json = json_decode(file_get_contents($url));
        if($json->results == []){
            return '';
        }
        $lat = $json->results[0]->geometry->location->lat;
        $lng = $json->results[0]->geometry->location->lng;

        //Now build the actual lookup
        $address_url = 'http://maps.googleapis.com/maps/api/geocode/json?latlng=' . $lat . ',' . $lng . '&sensor=false';
        $address_json = json_decode(file_get_contents($address_url));

        $address_data = $address_json->results[0]->address_components;
        //return $address_data = $address_json->results[0]->formatted_address;

        $street = str_replace('Dr', 'Drive', $address_data[1]->long_name);
        $town = $address_data[2]->long_name;
        $county = $address_data[3]->long_name;

        return $street.', '. $town. ', '.$county;
Extremity answered 18/3, 2015 at 6:16 Comment(0)
C
0

Looks like the job for a set-parser like JMESpath http://jmespath.org/

Given the array

{
  "locations": [
    {"name": "Seattle", "state": "WA"},
    {"name": "New York", "state": "NY"},
    {"name": "Bellevue", "state": "WA"},
    {"name": "Olympia", "state": "WA"}
  ]
}

A JMESPath of:

locations[?state == 'WA'].name | sort(@) | {WashingtonCities: join(', ', @)}

yields

{
  "WashingtonCities": "Bellevue, Olympia, Seattle"
}

You'd have to rewrite for your case, but you get the idea how powerful this language is. You can use composer to install a JMESPath implementation for PHP

Coimbra answered 4/5, 2018 at 22:35 Comment(0)

© 2022 - 2024 — McMap. All rights reserved.