collector.TwitterCollector.java Source code

Introduction

Here is the source code for collector.TwitterCollector.java

Source

/*
 * To change this license header, choose License Headers in Project Properties.
 * To change this template file, choose Tools | Templates
 * and open the template in the editor.
 */
package collector;

import java.util.ArrayList;
import java.util.LinkedHashSet;
import java.util.List;
import model.Tweet;
import model.TwitterUser;
import twitter4j.Paging;
import twitter4j.Query;
import twitter4j.QueryResult;
import twitter4j.Status;
import twitter4j.Twitter;
import twitter4j.TwitterException;
import twitter4j.TwitterFactory;
import twitter4j.conf.ConfigurationBuilder;

/**
 *
 * @author fabio
 */
public class TwitterCollector {

    private final Twitter twitter;

    public TwitterCollector(List<String> credentials) {

        ConfigurationBuilder cb = new ConfigurationBuilder();
        String OAuthConsumerKey = credentials.get(0);
        String OAuthConsumerSecret = credentials.get(1);
        String OAuthAccessToken = credentials.get(2);
        String OAuthAccessTokenSecret = credentials.get(3);

        cb.setDebugEnabled(true).setOAuthConsumerKey(OAuthConsumerKey).setOAuthConsumerSecret(OAuthConsumerSecret)
                .setOAuthAccessToken(OAuthAccessToken).setOAuthAccessTokenSecret(OAuthAccessTokenSecret);
        //                .setJSONStoreEnabled(true);

        TwitterFactory tf = new TwitterFactory(cb.build());
        twitter = tf.getInstance();

    }

    public LinkedHashSet<Tweet> search(String queryExpression, int maxResults) {
        Query query = new Query(queryExpression);
        int numberOfTweets = maxResults;//512;
        long lastID = Long.MAX_VALUE;
        query = query.lang("pt");
        List<Status> tweets = new ArrayList<>();
        boolean finish = false;
        while ((tweets.size() < numberOfTweets) && !finish) {
            System.out.print(".");
            if (numberOfTweets - tweets.size() > 100) {//100) {
                query.setCount(100);//100);
            } else {
                query.setCount(numberOfTweets - tweets.size());
            }
            try {
                QueryResult result = twitter.search(query);
                List<Status> resultList = result.getTweets();
                if (resultList == null || resultList.isEmpty()) {
                    finish = true;
                    System.out.println("no foram encontrados mais tweets");
                } else {
                    tweets.addAll(resultList);
                    for (Status t : tweets) {
                        if (t.getId() < lastID) {
                            lastID = t.getId();
                        }
                    }
                }
            } catch (TwitterException ex) {
                System.err.println(ex.getMessage());
            }
            query.setMaxId(lastID - 1);
        }
        LinkedHashSet<Tweet> out = new LinkedHashSet<>();
        for (Status status : tweets) {
            if (!status.getText().startsWith("RT")) {
                TwitterUser user;
                user = new TwitterUser().addID(status.getUser().getId()).addName(status.getUser().getName())
                        .addLocation(status.getUser().getLocation()).addDateSignin(status.getUser().getCreatedAt())
                        .addCountTweets(status.getUser().getStatusesCount())
                        .addCountFavorites(status.getUser().getFavouritesCount())
                        .addCountFriends(status.getUser().getFriendsCount())
                        .addCountFollowers(status.getUser().getFollowersCount());
                Tweet tweet = new Tweet().addUser(user).addText(status.getText()).addID(status.getId())
                        .addDate(status.getCreatedAt())
                        .addLatitude(status.getGeoLocation() != null ? status.getGeoLocation().getLatitude()
                                : Double.MAX_VALUE)
                        .addLongitude(status.getGeoLocation() != null ? status.getGeoLocation().getLongitude()
                                : Double.MAX_VALUE);
                out.add(tweet);
            }
        }
        return out;
    }

    public LinkedHashSet<Tweet> userSearchData(String userName, int maxResults) {

        LinkedHashSet<Tweet> out = new LinkedHashSet<>();
        Paging paging = new Paging(1, 180);
        int numberOfTweets = maxResults;//512;
        long lastID = Long.MAX_VALUE;
        ArrayList<Status> status = new ArrayList<>();
        while (status.size() < numberOfTweets) {
            if (numberOfTweets - status.size() > 180) {//100) {
                paging.setCount(180);//100);
            } else {
                paging.setCount(numberOfTweets - status.size());
            }
            try {
                List<Status> timeLine = twitter.getUserTimeline(userName, paging);
                status.addAll(timeLine);
                for (Status t : status) {
                    if (t.getId() < lastID) {
                        lastID = t.getId();
                    }
                }
            } catch (TwitterException ex) {
                System.err.println(ex.getMessage());
            }
            paging.setMaxId(lastID - 1);
        }

        //armazenar os atributos interessantes a analise dos tweets
        int qtdretweet = 0;
        for (Status sta : status) {
            String text = sta.getText();
            if (!sta.isRetweet() && !sta.isRetweeted() && !text.startsWith("RT")) { //&& !TweetMediaDetect.detect(text)) {
                TwitterUser user;
                user = new TwitterUser().addID(sta.getUser().getId()).addName(sta.getUser().getName())
                        .addLocation(sta.getUser().getLocation()).addDateSignin(sta.getUser().getCreatedAt())
                        .addCountTweets(sta.getUser().getStatusesCount())
                        .addCountFavorites(sta.getUser().getFavouritesCount())
                        .addCountFriends(sta.getUser().getFriendsCount())
                        .addCountFollowers(sta.getUser().getFollowersCount());
                Tweet tweet = new Tweet().addUser(user).addText(sta.getText()).addID(sta.getId())
                        .addDate(sta.getCreatedAt())
                        .addLatitude(sta.getGeoLocation() != null ? sta.getGeoLocation().getLatitude()
                                : Double.MAX_VALUE)
                        .addLongitude(sta.getGeoLocation() != null ? sta.getGeoLocation().getLongitude()
                                : Double.MAX_VALUE);
                out.add(tweet);
            } else {
                qtdretweet++;
            }
        }

        return out;
    }
}