CypherpunkSamurai/NewPipeExtractor
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

find playerUrl in another place when assetsPattern regex fails

Browse source
This commit is contained in:
bopol 2020-10-27 13:48:58 +01:00
parent db0ef83d6b
commit 6dc5ab4015
1 changed files with 26 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

25
extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/extractors/YoutubeStreamExtractor.java
View file

@ -3,6 +3,10 @@ package org.schabi.newpipe.extractor.services.youtube.extractors;
import com.grack.nanojson.JsonArray; import com.grack.nanojson.JsonArray;
import com.grack.nanojson.JsonObject; import com.grack.nanojson.JsonObject;
import com.grack.nanojson.JsonParser; import com.grack.nanojson.JsonParser;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import org.mozilla.javascript.Context; import org.mozilla.javascript.Context;
import org.mozilla.javascript.Function; import org.mozilla.javascript.Function;
import org.mozilla.javascript.ScriptableObject; import org.mozilla.javascript.ScriptableObject;
@ -740,11 +744,30 @@ public class YoutubeStreamExtractor extends StreamExtractor {
final String embedPageContent = downloader.get(embedUrl, getExtractorLocalization()).responseBody(); final String embedPageContent = downloader.get(embedUrl, getExtractorLocalization()).responseBody();
// Get player url // Get player url
String playerUrl = null;
try {
final String assetsPattern = "\"assets\":.+?\"js\":\\s*(\"[^\"]+\")"; final String assetsPattern = "\"assets\":.+?\"js\":\\s*(\"[^\"]+\")";
String playerUrl = Parser.matchGroup1(assetsPattern, embedPageContent) playerUrl = Parser.matchGroup1(assetsPattern, embedPageContent)
.replace("\\", "").replace("\"", ""); .replace("\\", "").replace("\"", "");
} catch (Parser.RegexException ex) {
// playerUrl is still available in the file, just somewhere else
final Document doc = Jsoup.parse(embedPageContent);
final Elements elems = doc.select("script").attr("name", "player_ias/base");
for (Element elem : elems) {
if (elem.attr("src").contains("base.js")) {
playerUrl = elem.attr("src");
}
}
if (playerUrl == null) {
throw new ParsingException("Could not get playerUrl");
}
}
if (playerUrl.startsWith("//")) { if (playerUrl.startsWith("//")) {
playerUrl = HTTPS + playerUrl; playerUrl = HTTPS + playerUrl;
} else if (playerUrl.startsWith("/")) {
playerUrl = HTTPS + "//youtube.com" + playerUrl;
} }
try { try {