Implement pagination in YoutubeSearchExtractor

extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/extractors/YoutubeSearchExtractor.java

@@ -5,9 +5,7 @@ import com.grack.nanojson.JsonObject;
 import com.grack.nanojson.JsonParser;
 import com.grack.nanojson.JsonParserException;
 
-import org.jsoup.Jsoup;
 import org.jsoup.nodes.Document;
-import org.jsoup.nodes.Element;
 import org.schabi.newpipe.extractor.InfoItem;
 import org.schabi.newpipe.extractor.StreamingService;
 import org.schabi.newpipe.extractor.downloader.Downloader;
@@ -19,12 +17,12 @@ import org.schabi.newpipe.extractor.localization.TimeAgoParser;
 import org.schabi.newpipe.extractor.search.InfoItemsSearchCollector;
 import org.schabi.newpipe.extractor.search.SearchExtractor;
 import org.schabi.newpipe.extractor.services.youtube.linkHandler.YoutubeParsingHelper;
-import org.schabi.newpipe.extractor.utils.Parser;
 
 import java.io.IOException;
-import java.io.UnsupportedEncodingException;
+import java.util.Collections;
-import java.net.MalformedURLException;
+import java.util.HashMap;
-import java.net.URL;
+import java.util.List;
+import java.util.Map;
 
 import javax.annotation.Nonnull;
 
@@ -73,58 +71,70 @@ public class YoutubeSearchExtractor extends SearchExtractor {
 
     @Override
     public String getSearchSuggestion() {
-        final Element el = doc.select("div[class*=\"spell-correction\"]").first();
+        JsonObject showingResultsForRenderer = initialData.getObject("contents")
-        if (el != null) {
+                .getObject("twoColumnSearchResultsRenderer").getObject("primaryContents")
-            return el.select("a").first().text();
+                .getObject("sectionListRenderer").getArray("contents").getObject(0)
-        } else {
+                .getObject("itemSectionRenderer").getArray("contents").getObject(0)
+                .getObject("showingResultsForRenderer");
+        if (showingResultsForRenderer == null) {
             return "";
+        } else {
+            return showingResultsForRenderer.getObject("correctedQuery").getArray("runs")
+                    .getObject(0).getString("text");
         }
     }
 
     @Nonnull
     @Override
     public InfoItemsPage<InfoItem> getInitialPage() throws ExtractionException {
-        return new InfoItemsPage<>(collectItems(doc), getNextPageUrl());
+        InfoItemsSearchCollector collector = getInfoItemSearchCollector();
+        JsonArray videos = initialData.getObject("contents").getObject("twoColumnSearchResultsRenderer")
+                .getObject("primaryContents").getObject("sectionListRenderer").getArray("contents")
+                .getObject(0).getObject("itemSectionRenderer").getArray("contents");
+
+        collectStreamsFrom(collector, videos);
+        return new InfoItemsPage<>(collector, getNextPageUrl());
     }
 
     @Override
     public String getNextPageUrl() throws ExtractionException {
-        return getUrl() + "&page=" + 2;
+        return getNextPageUrlFrom(initialData.getObject("contents").getObject("twoColumnSearchResultsRenderer")
+                .getObject("primaryContents").getObject("sectionListRenderer").getArray("contents")
+                .getObject(0).getObject("itemSectionRenderer").getArray("continuations"));
     }
 
     @Override
     public InfoItemsPage<InfoItem> getPage(String pageUrl) throws IOException, ExtractionException {
-        // TODO: Get extracting next pages working
+        if (pageUrl == null || pageUrl.isEmpty()) {
-        final String response = getDownloader().get(pageUrl, getExtractorLocalization()).responseBody();
+            throw new ExtractionException(new IllegalArgumentException("Page url is empty or null"));
-        doc = Jsoup.parse(response, pageUrl);
-
-        return new InfoItemsPage<>(collectItems(doc), getNextPageUrlFromCurrentUrl(pageUrl));
         }
 
-    private String getNextPageUrlFromCurrentUrl(String currentUrl)
-            throws MalformedURLException, UnsupportedEncodingException {
-        final int pageNr = Integer.parseInt(
-                Parser.compatParseMap(
-                        new URL(currentUrl)
-                                .getQuery())
-                        .get("page"));
-
-        return currentUrl.replace("&page=" + pageNr,
-                "&page=" + Integer.toString(pageNr + 1));
-    }
-
-    private InfoItemsSearchCollector collectItems(Document doc) throws NothingFoundException, ParsingException {
         InfoItemsSearchCollector collector = getInfoItemSearchCollector();
+        JsonArray ajaxJson;
+        try {
+            Map<String, List<String>> headers = new HashMap<>();
+            headers.put("X-YouTube-Client-Name", Collections.singletonList("1"));
+            headers.put("X-YouTube-Client-Version", Collections.singletonList("2.20200221.03.00")); // TODO: Automatically get YouTube client version somehow
+            final String response = getDownloader().get(pageUrl, headers, getExtractorLocalization()).responseBody();
+            ajaxJson = JsonParser.array().from(response);
+        } catch (JsonParserException pe) {
+            throw new ParsingException("Could not parse json data for next streams", pe);
+        }
+
+        JsonObject itemSectionRenderer = ajaxJson.getObject(1).getObject("response")
+                .getObject("continuationContents").getObject("itemSectionContinuation");
+
+        collectStreamsFrom(collector, itemSectionRenderer.getArray("contents"));
+
+        return new InfoItemsPage<>(collector, getNextPageUrlFrom(itemSectionRenderer.getArray("continuations")));
+    }
+
+    private void collectStreamsFrom(InfoItemsSearchCollector collector, JsonArray videos) throws NothingFoundException, ParsingException {
         collector.reset();
 
         final TimeAgoParser timeAgoParser = getTimeAgoParser();
 
-        if (initialData == null) initialData = YoutubeParsingHelper.getInitialData(doc.toString());
+        for (Object item : videos) {
-        JsonArray list = initialData.getObject("contents").getObject("twoColumnSearchResultsRenderer")
-                .getObject("primaryContents").getObject("sectionListRenderer").getArray("contents")
-                .getObject(0).getObject("itemSectionRenderer").getArray("contents");
-
-        for (Object item : list) {
             if (((JsonObject) item).getObject("backgroundPromoRenderer") != null) {
                 throw new NothingFoundException(((JsonObject) item).getObject("backgroundPromoRenderer")
                         .getObject("bodyText").getArray("runs").getObject(0).getString("text"));
@@ -136,7 +146,17 @@ public class YoutubeSearchExtractor extends SearchExtractor {
                 collector.commit(new YoutubePlaylistInfoItemExtractor(((JsonObject) item).getObject("playlistRenderer")));
             }
         }
-        return collector;
     }
 
+    private String getNextPageUrlFrom(JsonArray continuations) throws ParsingException {
+        if (continuations == null) {
+            return "";
+        }
+
+        JsonObject nextContinuationData = continuations.getObject(0).getObject("nextContinuationData");
+        String continuation = nextContinuationData.getString("continuation");
+        String clickTrackingParams = nextContinuationData.getString("clickTrackingParams");
+        return getUrl() + "&pbj=1&ctoken=" + continuation + "&continuation=" + continuation
+                + "&itct=" + clickTrackingParams;
+    }
 }

extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/linkHandler/YoutubeSearchQueryHandlerFactory.java

@@ -24,13 +24,13 @@ public class YoutubeSearchQueryHandlerFactory extends SearchQueryHandlerFactory
     public String getUrl(String searchString, List<String> contentFilters, String sortFilter) throws ParsingException {
         try {
             final String url = "https://www.youtube.com/results"
-                    + "?q=" + URLEncoder.encode(searchString, CHARSET_UTF_8);
+                    + "?search_query=" + URLEncoder.encode(searchString, CHARSET_UTF_8);
 
             if (contentFilters.size() > 0) {
                 switch (contentFilters.get(0)) {
-                    case VIDEOS: return url + "&sp=EgIQAVAU";
+                    case VIDEOS: return url + "&sp=EgIQAQ%253D%253D";
-                    case CHANNELS: return url + "&sp=EgIQAlAU";
+                    case CHANNELS: return url + "&sp=EgIQAg%253D%253D";
-                    case PLAYLISTS: return url + "&sp=EgIQA1AU";
+                    case PLAYLISTS: return url + "&sp=EgIQAw%253D%253D";
                     case ALL:
                     default:
                 }