使用 Openweather API 获取特定日期的天气 - 解析 JSON 响应

Question

I am trying to build a simple weather forecasting app using Android studio and Java. I followed some instructions here ( https://www.androdocs.com/java/creating-an-android-weather-app-using-java.html ) to get up and running and this works. However, I can only get the current weather. Openweather forecasting API calls seem to be for 5 days. That's ok, but how do I get the weather - say temp and wind speed - for a specific date specified by the user that's within the next 5 days?

Below is a sample JSON response (shortened). Even if I could pull out the info around a specific date at 12pm, and get the temp and wind speed for that date, that would be enough. How do I parse this JSON response to get the temp and wind speed for a specific date? Thanks so much... sorry I'm a beginner...

{"cod":"200","message":0,"cnt":40,"list":[{"dt":1574283600,"main":{"temp":281.75,"temp_min":281.68,"temp_max":281.75,"pressure":995,"sea_level":995,"grnd_level":980,"humidity":93,"temp_kf":0.07},"weather":[{"id":501,"main":"Rain","description":"moderate rain","icon":"10n"}],"clouds":{"all":100},"wind":{"speed":4.82,"deg":147},"rain":{"3h":5.38},"sys":{"pod":"n"},"dt_txt":"2019-11-20 21:00:00"},{"dt":1574294400,"main":{"temp":281.79,"temp_min":281.74,"temp_max":281.79,"pressure":995,"sea_level":995,"grnd_level":980,"humidity":91,"temp_kf":0.05},"weather":[{"id":500,"main":"Rain","description":"light rain","icon":"10n"}],"clouds":{"all":100},"wind":{"speed":5.55,"deg":140},"rain":{"3h":1.75},"sys":{"pod":"n"},"dt_txt":"2019-11-21 00:00:00"},{"dt":1574305200,"main":{"temp":279.48,"temp_min":279.44,"temp_max":279.48,"pressure":994,"sea_level":994,"grnd_level":980,"humidity":95,"temp_kf":0.04},"weather":[{"id":500,"main":"Rain","description":"light rain","icon":"10n"}],"clouds":{"all":100}, "wind":{"speed":2.37,"deg":155},"rain":{"3h":0.94},"sys":{"pod":"n"},"dt_txt":"2019-11-21 03:00:00"},{"dt":1574316000,"main":{"temp":278.56,"temp_min":278.54,"temp_max":278.56,"pressure":995,"sea_level":995,"grnd_level":980,"humidity":94,"temp_kf":0.02},"weather":[{"id":500,"main":"Rain","description":"light rain","icon":"10n"}],"clouds":{"all":100},"wind":{"speed":1.73,"deg":128},"rain":{"3h":0.06},"sys":{"pod":"n"},"dt_txt":"2019-11-21 06:00:00"},{"dt":1574326800,"main":{"temp":279.19,"temp_min":279.19,"temp_max":279.19,"pressure":995,"sea_level":995,"grnd_level":981,"humidity":95,"temp_kf":0},"weather":[{"id":804,"main":"Clouds","description":"overcast clouds","icon":"04d"}],"clouds":{"all":100},"wind":{"speed":1.79,"deg":104},"sys":{"pod":"d"},"dt_txt":"2019-11-21 09:00:00"},{"dt":1574337600,"main":{"temp":282.2,"temp_min":282.2,"temp_max":282.2,"pressure":995,"sea_level":995,"grnd_level":980,"humidity":85,"temp_kf":0},"weather":[{"id":500,"main":"Rain","description":"light rain"," icon":"10d"}],"clouds":{"all":100},"wind":{"speed":2.78,"deg":129},"rain":{"3h":0.19},"sys":{"pod":"d"},"dt_txt":"2019-11-21 12:00:00"}

Answer 1

JSON-Simple

Here is an example app using the JSON-Simple library to parse the JSON data downloaded from OpenWeatherMap.org .

package work.basil.example;

import org.json.simple.JSONArray;
import org.json.simple.JSONObject;
import org.json.simple.parser.JSONParser;
import org.json.simple.parser.ParseException;

import java.io.BufferedReader;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStreamReader;
import java.net.URL;
import java.net.URLConnection;
import java.time.Instant;
import java.time.ZoneId;
import java.time.ZonedDateTime;

public class Weather
{

    public static void main ( String[] args )
    {

        Weather app = new Weather();
        app.demo();
    }

    private void demo ( )
    {
        //Creating a JSONParser object
        JSONParser jsonParser = new JSONParser();
        try
        {
            // Download JSON.
            String yourKey = "b6907d289e10d714a6e88b30761fae22";
            URL url = new URL( "https://samples.openweathermap.org/data/2.5/forecast/hourly?zip=79843&appid=b6907d289e10d714a6e88b30761fae22" + yourKey ); // 79843 = US postal Zip Code for Marfa, Texas.
            URLConnection conn = url.openConnection();
            BufferedReader reader = new BufferedReader( new InputStreamReader( conn.getInputStream() ) );


            // Parse JSON
            JSONObject jsonObject = ( JSONObject ) jsonParser.parse( reader );
            System.out.println( "jsonObject = " + jsonObject );

            JSONArray list = ( JSONArray ) jsonObject.get( "list" );
            System.out.println( "list = " + list );

            // Loop through each item
            for ( Object o : list )
            {
                JSONObject forecast = ( JSONObject ) o;

                Long dt = ( Long ) forecast.get( "dt" );          // Parse text into a number of whole seconds.
                Instant instant = Instant.ofEpochSecond( dt );    // Parse the count of whole seconds since 1970-01-01T00:00Z into a `Instant` object, representing a moment in UTC with a resolution of nanoseconds.
                ZoneId z = ZoneId.of( "America/Chicago" );        // Specify a time zone using a real `Continent/Region` time zone name. Never use 2-4 letter pseudo-zones such as `PDT`, `CST`, `IST`, etc.
                ZonedDateTime zdt = instant.atZone( z );          // Adjust from a moment in UTC to the wall-clock used by the people of a particular region (a time zone). Same moment, same point on the timeline, different wall-clock time.
                LocalTime lt = zdt.toLocalTime() ;
                // … compare with lt.equals( LocalTime.NOON ) to find the data sample you desire. 
                System.out.println( "dt : " + dt );
                System.out.println( "instant : " + instant );
                System.out.println( "zdt : " + zdt );

                JSONObject main = ( JSONObject ) forecast.get( "main" );
                System.out.println( "main = " + main );


                Double temp = ( Double ) main.get( "temp" );  // Better to use BigDecimal instead of Double for accuracy. But I do not know how to get the JSON-Simple library to parse the original string input as a BigDecimal.
                System.out.println( "temp = " + temp );

                JSONObject wind = ( JSONObject ) forecast.get( "wind" );
                System.out.println( "wind = " + wind );

                System.out.println( "BASIL - wind.getCLass: " + wind.getClass() );
                Double speed = ( Double ) wind.get( "speed" );
                System.out.println( "speed = " + speed );

                System.out.println( "\n" );
            }
        }
        catch ( FileNotFoundException e )
        {
            e.printStackTrace();
        }
        catch ( IOException e )
        {
            e.printStackTrace();
        }
        catch ( ParseException e )
        {
            e.printStackTrace();
        }
    }
}

Decimal separator

Notice that this code blows up when encountering a wind speed data point lacking the decimal separator. The publisher of that data should be writing, for example, 1.0 rather than 1 for consistency. If they had done so, the library would have parsed 1.0 as a Double rather than parsing 1 as Long .

JSON-Simple 1 now defunct

Also this code uses the original version 1 of JSON-Simple, now defunct. This project was forked, producing dramatically different versions 2 & 3.

See this page Parsing decimal numbers, some of which lack a decimal separator, in JSON data using JSON-Simple (Java) for details about the parsing-decimal problem and about links to the forked project .

Not for production use

So while I would not recommend this code for production use , it may help you along the way. For real work, consider the later version 3 of JSON-Simple or any of the several other JSON-processing libraries available for Java.

See the example data at this URL . To make it readable, use your text-editor or IDE to reformat the JSON data.

Sample output:

dt : 1553709600
instant : 2019-03-27T18:00:00Z
zdt : 2019-03-27T13:00-05:00[America/Chicago]
main = {"temp":286.44,"temp_min":286.258,"grnd_level":1002.193,"temp_kf":0.18,"humidity":100,"pressure":1015.82,"sea_level":1015.82,"temp_max":286.44}
temp = 286.44
wind = {"deg":202.816,"speed":5.51}
speed = 5.51


dt : 1553713200
instant : 2019-03-27T19:00:00Z
zdt : 2019-03-27T14:00-05:00[America/Chicago]
main = {"temp":286.43,"temp_min":286.3,"grnd_level":1002.667,"temp_kf":0.13,"humidity":100,"pressure":1016.183,"sea_level":1016.183,"temp_max":286.43}
temp = 286.43
wind = {"deg":206.141,"speed":4.84}
speed = 4.84

Answer 2

You can serialize the response JSON string to POJOs by any one of the most popular JSON libraries such as Jackson or Gson , then retrieve fields you want of the object whose date field equals given date. BTW, your JSON string is invalid, ]} is missing at the end of it.

POJOs

@JsonIgnoreProperties(ignoreUnknown = true)
class Response {
    List<Weather> list;

    //general getters and setters
}

@JsonIgnoreProperties(ignoreUnknown = true)
class Weather {
    JsonNode main;
    JsonNode wind;

    @JsonProperty("dt_txt")
    String dtTxt;

    //general getters and setters
}

Use @JsonIgnoreProperties (provided by Jackson) to ignore those fields you don't care while serialization.

Code snippet

ObjectMapper mapper = new ObjectMapper();
Response response = mapper.readValue(jsonStr, Response.class);

String givenDate = "2019-11-21 12:00:00";
response.getList().forEach(e -> {
    if (givenDate.equals(e.getDtTxt())) {
        System.out.println("temp: " + e.getMain().get("temp").asText());
        System.out.println("wind speed:" + e.getWind().get("speed").asText());
    }
});

Console output

temp: 282.2
wind speed:2.78