NewPipeExtractor/extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/YoutubeParsingHelper.java

1196 lines
52 KiB
Java
Raw Normal View History

package org.schabi.newpipe.extractor.services.youtube;
2017-03-01 18:47:52 +01:00
import com.grack.nanojson.JsonArray;
import com.grack.nanojson.JsonBuilder;
import com.grack.nanojson.JsonObject;
import com.grack.nanojson.JsonParser;
import com.grack.nanojson.JsonParserException;
import com.grack.nanojson.JsonWriter;
import org.schabi.newpipe.extractor.MetaInfo;
import org.schabi.newpipe.extractor.Page;
import org.schabi.newpipe.extractor.downloader.Response;
import org.schabi.newpipe.extractor.exceptions.*;
import org.schabi.newpipe.extractor.localization.ContentCountry;
2020-02-29 16:55:07 +01:00
import org.schabi.newpipe.extractor.localization.Localization;
import org.schabi.newpipe.extractor.stream.Description;
import org.schabi.newpipe.extractor.utils.JsonUtils;
import org.schabi.newpipe.extractor.utils.Parser;
import org.schabi.newpipe.extractor.utils.Utils;
2017-03-01 18:47:52 +01:00
2020-02-29 16:42:04 +01:00
import java.io.IOException;
2020-02-27 17:39:23 +01:00
import java.io.UnsupportedEncodingException;
import java.net.MalformedURLException;
import java.net.URL;
2020-02-27 17:39:23 +01:00
import java.net.URLDecoder;
import java.time.LocalDate;
import java.time.OffsetDateTime;
import java.time.ZoneOffset;
import java.time.format.DateTimeParseException;
2021-04-07 12:25:59 +02:00
import java.util.*;
2021-01-17 18:48:16 +01:00
import javax.annotation.Nonnull;
import javax.annotation.Nullable;
2020-02-26 15:22:59 +01:00
import static org.schabi.newpipe.extractor.NewPipe.getDownloader;
import static org.schabi.newpipe.extractor.utils.Utils.EMPTY_STRING;
import static org.schabi.newpipe.extractor.utils.Utils.HTTP;
import static org.schabi.newpipe.extractor.utils.Utils.HTTPS;
import static org.schabi.newpipe.extractor.utils.Utils.UTF_8;
import static org.schabi.newpipe.extractor.utils.Utils.isNullOrEmpty;
/*
2017-03-01 18:47:52 +01:00
* Created by Christian Schabesberger on 02.03.16.
*
* Copyright (C) Christian Schabesberger 2016 <chris.schabesberger@mailbox.org>
* YoutubeParsingHelper.java is part of NewPipe.
*
* NewPipe is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* NewPipe is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with NewPipe. If not, see <http://www.gnu.org/licenses/>.
*/
public class YoutubeParsingHelper {
private YoutubeParsingHelper() {
}
public static final String YOUTUBEI_V1_URL = "https://www.youtube.com/youtubei/v1/";
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
private static final String HARDCODED_CLIENT_VERSION = "2.20210728.00.00";
private static final String HARDCODED_KEY = "AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8";
private static final String MOBILE_YOUTUBE_KEY = "AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w";
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
private static final String MOBILE_YOUTUBE_CLIENT_VERSION = "16.29.38";
2020-02-26 15:22:59 +01:00
private static String clientVersion;
2020-07-26 12:00:56 +02:00
private static String key;
private static final String[] HARDCODED_YOUTUBE_MUSIC_KEY =
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
{"AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30", "67", "1.20210726.00.01"};
private static String[] youtubeMusicKey;
2020-03-17 11:33:39 +01:00
private static boolean keyAndVersionExtracted = false;
@SuppressWarnings("OptionalUsedAsFieldOrParameterType")
private static Optional<Boolean> hardcodedClientVersionAndKeyValid = Optional.empty();
private static Random numberGenerator = new Random();
2021-04-07 12:25:59 +02:00
/**
* <code>PENDING+</code> means that the user did not yet submit their choices.
* Therefore, YouTube & Google should not track the user, because they did not give consent.
* The three digits at the end can be random, but are required.
*/
2021-04-09 11:51:54 +02:00
private static final String CONSENT_COOKIE_VALUE = "PENDING+";
2021-04-07 12:25:59 +02:00
/**
* Youtube <code>CONSENT</code> cookie. Should prevent redirect to consent.youtube.com
*/
2021-04-09 11:51:54 +02:00
private static final String CONSENT_COOKIE = "CONSENT=" + CONSENT_COOKIE_VALUE;
2021-04-07 12:25:59 +02:00
private static final String FEED_BASE_CHANNEL_ID =
"https://www.youtube.com/feeds/videos.xml?channel_id=";
private static final String FEED_BASE_USER = "https://www.youtube.com/feeds/videos.xml?user=";
private static boolean isGoogleURL(String url) {
url = extractCachedUrlIfNeeded(url);
try {
final URL u = new URL(url);
final String host = u.getHost();
2021-06-24 18:39:16 +02:00
return host.startsWith("google.")
|| host.startsWith("m.google.")
2021-04-26 11:49:47 +02:00
|| host.startsWith("www.google.");
} catch (final MalformedURLException e) {
return false;
}
}
public static boolean isYoutubeURL(@Nonnull final URL url) {
final String host = url.getHost();
2021-06-24 18:39:16 +02:00
return host.equalsIgnoreCase("youtube.com")
|| host.equalsIgnoreCase("www.youtube.com")
|| host.equalsIgnoreCase("m.youtube.com")
|| host.equalsIgnoreCase("music.youtube.com");
}
public static boolean isYoutubeServiceURL(@Nonnull final URL url) {
final String host = url.getHost();
return host.equalsIgnoreCase("www.youtube-nocookie.com")
|| host.equalsIgnoreCase("youtu.be");
}
public static boolean isHooktubeURL(@Nonnull final URL url) {
final String host = url.getHost();
return host.equalsIgnoreCase("hooktube.com");
}
public static boolean isInvidioURL(@Nonnull final URL url) {
final String host = url.getHost();
2020-07-02 21:31:05 +02:00
return host.equalsIgnoreCase("invidio.us")
|| host.equalsIgnoreCase("dev.invidio.us")
|| host.equalsIgnoreCase("www.invidio.us")
2021-01-22 19:20:22 +01:00
|| host.equalsIgnoreCase("redirect.invidious.io")
2020-07-02 21:31:05 +02:00
|| host.equalsIgnoreCase("invidious.snopyta.org")
|| host.equalsIgnoreCase("yewtu.be")
|| host.equalsIgnoreCase("tube.connect.cafe")
2021-08-12 10:06:41 +02:00
|| host.equalsIgnoreCase("tubus.eduvid.org")
|| host.equalsIgnoreCase("invidious.kavin.rocks")
2021-06-23 14:12:03 +02:00
|| host.equalsIgnoreCase("invidious-us.kavin.rocks")
|| host.equalsIgnoreCase("piped.kavin.rocks")
|| host.equalsIgnoreCase("invidious.site")
|| host.equalsIgnoreCase("vid.mint.lgbt")
|| host.equalsIgnoreCase("invidiou.site")
2021-01-22 19:20:22 +01:00
|| host.equalsIgnoreCase("invidious.fdn.fr")
|| host.equalsIgnoreCase("invidious.048596.xyz")
|| host.equalsIgnoreCase("invidious.zee.li")
|| host.equalsIgnoreCase("vid.puffyan.us")
2021-06-23 14:12:03 +02:00
|| host.equalsIgnoreCase("ytprivate.com")
|| host.equalsIgnoreCase("invidious.namazso.eu")
|| host.equalsIgnoreCase("invidious.silkky.cloud")
|| host.equalsIgnoreCase("invidious.exonip.de")
|| host.equalsIgnoreCase("inv.riverside.rocks")
|| host.equalsIgnoreCase("invidious.blamefran.net")
|| host.equalsIgnoreCase("invidious.moomoo.me")
|| host.equalsIgnoreCase("ytb.trom.tf")
|| host.equalsIgnoreCase("yt.cyberhost.uk")
|| host.equalsIgnoreCase("y.com.cm");
}
/**
2020-07-02 21:31:05 +02:00
* Parses the duration string of the video expecting ":" or "." as separators
2021-02-07 22:12:22 +01:00
*
* @return the duration in seconds
2020-07-02 21:31:05 +02:00
* @throws ParsingException when more than 3 separators are found
*/
public static int parseDurationString(@Nonnull final String input)
2017-03-01 18:47:52 +01:00
throws ParsingException, NumberFormatException {
// If time separator : is not detected, try . instead
2018-09-09 14:01:39 +02:00
final String[] splitInput = input.contains(":")
? input.split(":")
: input.split("\\.");
2017-03-01 18:47:52 +01:00
String days = "0";
String hours = "0";
String minutes = "0";
2018-09-09 14:01:39 +02:00
final String seconds;
2017-03-01 18:47:52 +01:00
switch (splitInput.length) {
2017-03-01 18:47:52 +01:00
case 4:
days = splitInput[0];
hours = splitInput[1];
minutes = splitInput[2];
seconds = splitInput[3];
break;
case 3:
hours = splitInput[0];
minutes = splitInput[1];
seconds = splitInput[2];
break;
case 2:
minutes = splitInput[0];
seconds = splitInput[1];
break;
case 1:
seconds = splitInput[0];
break;
default:
throw new ParsingException("Error duration string with unknown format: " + input);
}
2020-05-30 17:20:54 +02:00
return ((Integer.parseInt(Utils.removeNonDigitCharacters(days)) * 24
+ Integer.parseInt(Utils.removeNonDigitCharacters(hours))) * 60
+ Integer.parseInt(Utils.removeNonDigitCharacters(minutes))) * 60
+ Integer.parseInt(Utils.removeNonDigitCharacters(seconds));
2017-03-01 18:47:52 +01:00
}
@Nonnull
public static String getFeedUrlFrom(@Nonnull final String channelIdOrUser) {
if (channelIdOrUser.startsWith("user/")) {
return FEED_BASE_USER + channelIdOrUser.replace("user/", "");
} else if (channelIdOrUser.startsWith("channel/")) {
return FEED_BASE_CHANNEL_ID + channelIdOrUser.replace("channel/", "");
} else {
return FEED_BASE_CHANNEL_ID + channelIdOrUser;
}
}
public static OffsetDateTime parseDateFrom(final String textualUploadDate)
throws ParsingException {
try {
return OffsetDateTime.parse(textualUploadDate);
} catch (final DateTimeParseException e) {
try {
return LocalDate.parse(textualUploadDate).atStartOfDay().atOffset(ZoneOffset.UTC);
} catch (final DateTimeParseException e1) {
throw new ParsingException("Could not parse date: \"" + textualUploadDate + "\"",
e1);
}
}
}
2020-02-02 18:15:47 +01:00
/**
* Checks if the given playlist id is a YouTube Mix (auto-generated playlist)
* Ids from a YouTube Mix start with "RD"
2021-02-07 22:12:22 +01:00
*
2021-06-24 18:39:16 +02:00
* @param playlistId the playlist id
* @return Whether given id belongs to a YouTube Mix
2020-02-02 18:15:47 +01:00
*/
public static boolean isYoutubeMixId(@Nonnull final String playlistId) {
return playlistId.startsWith("RD") && !isYoutubeMusicMixId(playlistId);
}
/**
* Checks if the given playlist id is a YouTube Music Mix (auto-generated playlist)
* Ids from a YouTube Music Mix start with "RDAMVM" or "RDCLAK"
2021-02-07 22:12:22 +01:00
*
* @param playlistId the playlist id
* @return Whether given id belongs to a YouTube Music Mix
*/
public static boolean isYoutubeMusicMixId(@Nonnull final String playlistId) {
return playlistId.startsWith("RDAMVM") || playlistId.startsWith("RDCLAK");
}
2021-02-07 22:12:22 +01:00
/**
* Checks if the given playlist id is a YouTube Channel Mix (auto-generated playlist)
* Ids from a YouTube channel Mix start with "RDCM"
2021-02-07 22:12:22 +01:00
*
* @return Whether given id belongs to a YouTube Channel Mix
*/
public static boolean isYoutubeChannelMixId(@Nonnull final String playlistId) {
return playlistId.startsWith("RDCM");
}
/**
* Extracts the video id from the playlist id for Mixes.
2021-02-07 22:12:22 +01:00
*
* @throws ParsingException If the playlistId is a Channel Mix or not a mix.
*/
@Nonnull
public static String extractVideoIdFromMixId(@Nonnull final String playlistId)
throws ParsingException {
if (playlistId.startsWith("RDMM")) { // My Mix
return playlistId.substring(4);
} else if (isYoutubeMusicMixId(playlistId)) { // starts with "RDAMVM" or "RDCLAK"
return playlistId.substring(6);
} else if (isYoutubeChannelMixId(playlistId)) { // starts with "RMCM"
// Channel mix are build with RMCM{channelId}, so videoId can't be determined
throw new ParsingException("Video id could not be determined from mix id: "
+ playlistId);
} else if (isYoutubeMixId(playlistId)) { // normal mix, starts with "RD"
return playlistId.substring(2);
} else { // not a mix
throw new ParsingException("Video id could not be determined from mix id: "
+ playlistId);
}
}
public static JsonObject getInitialData(final String html) throws ParsingException {
try {
2020-10-16 20:27:40 +02:00
try {
final String initialData = Parser.matchGroup1(
"window\\[\"ytInitialData\"\\]\\s*=\\s*(\\{.*?\\});", html);
2020-10-16 20:27:40 +02:00
return JsonParser.object().from(initialData);
} catch (final Parser.RegexException e) {
final String initialData = Parser.matchGroup1(
"var\\s*ytInitialData\\s*=\\s*(\\{.*?\\});", html);
2020-10-16 20:27:40 +02:00
return JsonParser.object().from(initialData);
}
} catch (final JsonParserException | Parser.RegexException e) {
throw new ParsingException("Could not get ytInitialData", e);
}
}
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
public static boolean areHardcodedClientVersionAndKeyValid()
2021-06-24 18:39:16 +02:00
throws IOException, ExtractionException {
if (hardcodedClientVersionAndKeyValid.isPresent()) {
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
return hardcodedClientVersionAndKeyValid.get();
}
// @formatter:off
final byte[] body = JsonWriter.string()
.object()
.object("context")
.object("client")
2021-06-24 18:39:16 +02:00
.value("hl", "en-GB")
.value("gl", "GB")
2021-06-24 18:39:16 +02:00
.value("clientName", "WEB")
.value("clientVersion", HARDCODED_CLIENT_VERSION)
.end()
2021-06-24 18:39:16 +02:00
.object("user")
.value("lockedSafetyMode", false)
.end()
.value("fetchLiveState", true)
.end()
.end().done().getBytes(UTF_8);
// @formatter:on
final Map<String, List<String>> headers = new HashMap<>();
headers.put("X-YouTube-Client-Name", Collections.singletonList("1"));
headers.put("X-YouTube-Client-Version",
Collections.singletonList(HARDCODED_CLIENT_VERSION));
2020-02-26 15:22:59 +01:00
// This endpoint is fetched by the YouTube website to get the items of its main menu and is
// pretty lightweight (around 30kB)
final Response response = getDownloader().post(YOUTUBEI_V1_URL + "guide?key="
+ HARDCODED_KEY, headers, body);
final String responseBody = response.responseBody();
final int responseCode = response.responseCode();
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
hardcodedClientVersionAndKeyValid = Optional.of(responseBody.length() > 5000
&& responseCode == 200); // Ensure to have a valid response
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
return hardcodedClientVersionAndKeyValid.get();
}
private static void extractClientVersionAndKey() throws IOException, ExtractionException {
// Don't extract the client version and the InnerTube key if it has been already extracted
if (keyAndVersionExtracted) return;
// Don't provide a search term in order to have a smaller response
final String url = "https://www.youtube.com/results?search_query=&ucbcb=1";
final Map<String, List<String>> headers = new HashMap<>();
addCookieHeader(headers);
final String html = getDownloader().get(url, headers).responseBody();
2020-07-26 12:00:56 +02:00
final JsonObject initialData = getInitialData(html);
final JsonArray serviceTrackingParams = initialData.getObject("responseContext")
.getArray("serviceTrackingParams");
String shortClientVersion = null;
// Try to get version from initial data first
2020-07-26 12:00:56 +02:00
for (final Object service : serviceTrackingParams) {
final JsonObject s = (JsonObject) service;
if (s.getString("service").equals("CSI")) {
2020-07-26 12:00:56 +02:00
final JsonArray params = s.getArray("params");
for (final Object param : params) {
final JsonObject p = (JsonObject) param;
final String key = p.getString("key");
if (key != null && key.equals("cver")) {
2020-07-26 12:00:56 +02:00
clientVersion = p.getString("value");
}
}
} else if (s.getString("service").equals("ECATCHER")) {
// Fallback to get a shortened client version which does not contain the last two
// digits
2020-07-26 12:00:56 +02:00
final JsonArray params = s.getArray("params");
for (final Object param : params) {
final JsonObject p = (JsonObject) param;
final String key = p.getString("key");
if (key != null && key.equals("client.version")) {
shortClientVersion = p.getString("value");
}
}
}
}
String contextClientVersion;
2020-07-26 12:00:56 +02:00
final String[] patterns = {
"INNERTUBE_CONTEXT_CLIENT_VERSION\":\"([0-9\\.]+?)\"",
"innertube_context_client_version\":\"([0-9\\.]+?)\"",
"client.version=([0-9\\.]+)"
};
2020-07-26 12:00:56 +02:00
for (final String pattern : patterns) {
try {
contextClientVersion = Parser.matchGroup1(pattern, html);
2020-04-15 18:49:58 +02:00
if (!isNullOrEmpty(contextClientVersion)) {
2020-07-26 12:00:56 +02:00
clientVersion = contextClientVersion;
break;
2020-02-26 15:22:59 +01:00
}
} catch (final Parser.RegexException ignored) {
2021-02-07 22:12:22 +01:00
}
}
if (!isNullOrEmpty(clientVersion) && !isNullOrEmpty(shortClientVersion)) {
2020-07-26 12:00:56 +02:00
clientVersion = shortClientVersion;
}
try {
key = Parser.matchGroup1("INNERTUBE_API_KEY\":\"([0-9a-zA-Z_-]+?)\"", html);
2021-06-05 18:17:26 +02:00
} catch (final Parser.RegexException e1) {
2020-07-26 12:00:56 +02:00
try {
key = Parser.matchGroup1("innertubeApiKey\":\"([0-9a-zA-Z_-]+?)\"", html);
2021-06-05 18:17:26 +02:00
} catch (final Parser.RegexException e2) {
throw new ParsingException("Could not extract client version and key");
2021-02-07 22:12:22 +01:00
}
}
2021-06-05 18:17:26 +02:00
keyAndVersionExtracted = true;
2020-07-26 12:00:56 +02:00
}
/**
* Get the client version
*/
public static String getClientVersion() throws IOException, ExtractionException {
if (!isNullOrEmpty(clientVersion)) return clientVersion;
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
if (areHardcodedClientVersionAndKeyValid()) {
return clientVersion = HARDCODED_CLIENT_VERSION;
}
2020-07-26 12:00:56 +02:00
extractClientVersionAndKey();
2020-07-26 12:00:56 +02:00
return clientVersion;
}
/**
* Get the key
*/
public static String getKey() throws IOException, ExtractionException {
if (!isNullOrEmpty(key)) return key;
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
if (areHardcodedClientVersionAndKeyValid()) {
return key = HARDCODED_KEY;
}
extractClientVersionAndKey();
2020-07-26 12:00:56 +02:00
return key;
}
2020-02-27 17:39:23 +01:00
/**
* <p>
* <b>Only use in tests.</b>
* </p>
*
* <p>
* Quick-and-dirty solution to reset global state in between test classes.
* </p>
* <p>
* This is needed for the mocks because in order to reach that state a network request has to
* be made. If the global state is not reset and the RecordingDownloader is used,
* then only the first test class has that request recorded. Meaning running the other
* tests with mocks will fail, because the mock is missing.
* </p>
*/
2021-01-17 18:48:16 +01:00
public static void resetClientVersionAndKey() {
clientVersion = null;
key = null;
}
/**
* <p>
* <b>Only use in tests.</b>
* </p>
*/
public static void setNumberGenerator(final Random random) {
numberGenerator = random;
}
public static boolean isHardcodedYoutubeMusicKeyValid() throws IOException,
ReCaptchaException {
final String url =
"https://music.youtube.com/youtubei/v1/music/get_search_suggestions?alt=json&key="
+ HARDCODED_YOUTUBE_MUSIC_KEY[0];
2020-03-20 11:05:19 +01:00
// @formatter:off
byte[] json = JsonWriter.string()
.object()
.object("context")
.object("client")
.value("clientName", "WEB_REMIX")
.value("clientVersion", HARDCODED_YOUTUBE_MUSIC_KEY[2])
.value("hl", "en-GB")
2020-03-20 11:05:19 +01:00
.value("gl", "GB")
.array("experimentIds").end()
2021-06-24 18:39:16 +02:00
.value("experimentsToken", EMPTY_STRING)
2020-03-20 11:05:19 +01:00
.object("locationInfo").end()
.object("musicAppInfo").end()
.end()
.object("capabilities").end()
.object("request")
.array("internalExperimentFlags").end()
.object("sessionIndex").end()
.end()
.object("activePlayers").end()
.object("user")
.value("enableSafetyMode", false)
.end()
.end()
.value("input", "")
2021-02-07 22:12:22 +01:00
.end().done().getBytes(UTF_8);
2020-03-20 11:05:19 +01:00
// @formatter:on
final Map<String, List<String>> headers = new HashMap<>();
headers.put("X-YouTube-Client-Name", Collections.singletonList(
HARDCODED_YOUTUBE_MUSIC_KEY[1]));
headers.put("X-YouTube-Client-Version", Collections.singletonList(
HARDCODED_YOUTUBE_MUSIC_KEY[2]));
2020-03-20 11:05:19 +01:00
headers.put("Origin", Collections.singletonList("https://music.youtube.com"));
headers.put("Referer", Collections.singletonList("music.youtube.com"));
2020-03-20 11:05:19 +01:00
headers.put("Content-Type", Collections.singletonList("application/json"));
final Response response = getDownloader().post(url, headers, json);
// Ensure to have a valid response
return response.responseBody().length() > 500 && response.responseCode() == 200;
2020-03-20 11:05:19 +01:00
}
public static String[] getYoutubeMusicKey() throws IOException, ReCaptchaException,
Parser.RegexException {
if (youtubeMusicKey != null && youtubeMusicKey.length == 3) return youtubeMusicKey;
if (isHardcodedYoutubeMusicKeyValid()) {
return youtubeMusicKey = HARDCODED_YOUTUBE_MUSIC_KEY;
}
2020-03-17 11:33:39 +01:00
final String url = "https://music.youtube.com/";
final Map<String, List<String>> headers = new HashMap<>();
addCookieHeader(headers);
final String html = getDownloader().get(url, headers).responseBody();
2020-03-17 11:33:39 +01:00
2020-03-20 11:05:19 +01:00
String key;
try {
key = Parser.matchGroup1("INNERTUBE_API_KEY\":\"([0-9a-zA-Z_-]+?)\"", html);
} catch (final Parser.RegexException e) {
2020-03-20 11:05:19 +01:00
key = Parser.matchGroup1("innertube_api_key\":\"([0-9a-zA-Z_-]+?)\"", html);
}
final String clientName = Parser.matchGroup1("INNERTUBE_CONTEXT_CLIENT_NAME\":([0-9]+?),",
html);
2020-03-20 11:05:19 +01:00
String clientVersion;
try {
clientVersion = Parser.matchGroup1(
"INNERTUBE_CONTEXT_CLIENT_VERSION\":\"([0-9\\.]+?)\"", html);
} catch (final Parser.RegexException e) {
2020-03-20 11:05:19 +01:00
try {
clientVersion = Parser.matchGroup1(
"INNERTUBE_CLIENT_VERSION\":\"([0-9\\.]+?)\"", html);
} catch (final Parser.RegexException ee) {
clientVersion = Parser.matchGroup1(
"innertube_context_client_version\":\"([0-9\\.]+?)\"", html);
2020-03-20 11:05:19 +01:00
}
}
2020-03-17 11:33:39 +01:00
return youtubeMusicKey = new String[]{key, clientName, clientVersion};
2020-03-17 11:33:39 +01:00
}
@Nullable
public static String getUrlFromNavigationEndpoint(@Nonnull final JsonObject navigationEndpoint)
throws ParsingException {
2020-04-16 16:08:14 +02:00
if (navigationEndpoint.has("urlEndpoint")) {
2020-02-27 17:39:23 +01:00
String internUrl = navigationEndpoint.getObject("urlEndpoint").getString("url");
2021-02-12 22:22:11 +01:00
if (internUrl.startsWith("https://www.youtube.com/redirect?")) {
// remove https://www.youtube.com part to fall in the next if block
2021-02-12 22:22:11 +01:00
internUrl = internUrl.substring(23);
}
2020-02-27 17:39:23 +01:00
if (internUrl.startsWith("/redirect?")) {
// q parameter can be the first parameter
internUrl = internUrl.substring(10);
String[] params = internUrl.split("&");
for (String param : params) {
if (param.split("=")[0].equals("q")) {
String url;
try {
2021-02-07 22:12:22 +01:00
url = URLDecoder.decode(param.split("=")[1], UTF_8);
} catch (final UnsupportedEncodingException e) {
2020-02-27 17:39:23 +01:00
return null;
}
return url;
}
}
} else if (internUrl.startsWith("http")) {
return internUrl;
} else if (internUrl.startsWith("/channel") || internUrl.startsWith("/user")
|| internUrl.startsWith("/watch")) {
2021-02-12 22:22:11 +01:00
return "https://www.youtube.com" + internUrl;
2020-02-27 17:39:23 +01:00
}
2020-04-16 16:08:14 +02:00
} else if (navigationEndpoint.has("browseEndpoint")) {
final JsonObject browseEndpoint = navigationEndpoint.getObject("browseEndpoint");
final String canonicalBaseUrl = browseEndpoint.getString("canonicalBaseUrl");
final String browseId = browseEndpoint.getString("browseId");
// All channel ids are prefixed with UC
if (browseId != null && browseId.startsWith("UC")) {
return "https://www.youtube.com/channel/" + browseId;
}
2020-04-15 18:49:58 +02:00
if (!isNullOrEmpty(canonicalBaseUrl)) {
return "https://www.youtube.com" + canonicalBaseUrl;
}
throw new ParsingException("canonicalBaseUrl is null and browseId is not a channel (\""
+ browseEndpoint + "\")");
2020-04-16 16:08:14 +02:00
} else if (navigationEndpoint.has("watchEndpoint")) {
2020-02-27 17:39:23 +01:00
StringBuilder url = new StringBuilder();
url.append("https://www.youtube.com/watch?v=").append(navigationEndpoint
.getObject("watchEndpoint").getString("videoId"));
if (navigationEndpoint.getObject("watchEndpoint").has("playlistId")) {
url.append("&list=").append(navigationEndpoint.getObject("watchEndpoint")
.getString("playlistId"));
}
if (navigationEndpoint.getObject("watchEndpoint").has("startTimeSeconds")) {
url.append("&amp;t=").append(navigationEndpoint.getObject("watchEndpoint")
.getInt("startTimeSeconds"));
}
2020-02-27 17:39:23 +01:00
return url.toString();
2020-04-16 16:08:14 +02:00
} else if (navigationEndpoint.has("watchPlaylistEndpoint")) {
2020-03-17 11:33:39 +01:00
return "https://www.youtube.com/playlist?list=" +
navigationEndpoint.getObject("watchPlaylistEndpoint").getString("playlistId");
2020-02-27 17:39:23 +01:00
}
return null;
}
2020-04-20 14:27:33 +02:00
/**
* Get the text from a JSON object that has either a simpleText or a runs array.
2021-02-07 22:12:22 +01:00
*
2020-04-20 14:27:33 +02:00
* @param textObject JSON object to get the text from
* @param html whether to return HTML, by parsing the navigationEndpoint
* @return text in the JSON object or {@code null}
2020-04-20 14:27:33 +02:00
*/
@Nullable
public static String getTextFromObject(final JsonObject textObject, final boolean html)
throws ParsingException {
if (isNullOrEmpty(textObject)) return null;
2020-02-27 17:39:23 +01:00
if (textObject.has("simpleText")) return textObject.getString("simpleText");
if (textObject.getArray("runs").isEmpty()) return null;
final StringBuilder textBuilder = new StringBuilder();
for (final Object textPart : textObject.getArray("runs")) {
2020-02-27 17:39:23 +01:00
String text = ((JsonObject) textPart).getString("text");
2020-04-16 16:08:14 +02:00
if (html && ((JsonObject) textPart).has("navigationEndpoint")) {
String url = getUrlFromNavigationEndpoint(((JsonObject) textPart)
.getObject("navigationEndpoint"));
2020-04-15 18:49:58 +02:00
if (!isNullOrEmpty(url)) {
textBuilder.append("<a href=\"").append(url).append("\">").append(text)
.append("</a>");
2020-02-27 17:39:23 +01:00
continue;
}
}
textBuilder.append(text);
}
String text = textBuilder.toString();
if (html) {
text = text.replaceAll("\\n", "<br>");
text = text.replaceAll(" ", " &nbsp;");
}
return text;
}
@Nullable
public static String getTextFromObject(final JsonObject textObject) throws ParsingException {
2020-02-27 17:39:23 +01:00
return getTextFromObject(textObject, false);
}
2021-03-24 09:04:43 +01:00
@Nullable
public static String getTextAtKey(@Nonnull final JsonObject jsonObject, final String key)
2021-03-24 09:04:43 +01:00
throws ParsingException {
if (jsonObject.isString(key)) {
return jsonObject.getString(key);
} else {
return getTextFromObject(jsonObject.getObject(key));
}
}
public static String fixThumbnailUrl(@Nonnull String thumbnailUrl) {
if (thumbnailUrl.startsWith("//")) {
thumbnailUrl = thumbnailUrl.substring(2);
}
if (thumbnailUrl.startsWith(HTTP)) {
thumbnailUrl = Utils.replaceHttpWithHttps(thumbnailUrl);
} else if (!thumbnailUrl.startsWith(HTTPS)) {
thumbnailUrl = "https://" + thumbnailUrl;
}
return thumbnailUrl;
}
2020-02-29 16:42:04 +01:00
@Nonnull
public static String getValidJsonResponseBody(@Nonnull final Response response)
throws ParsingException, MalformedURLException {
if (response.responseCode() == 404) {
throw new ContentNotAvailableException("Not found"
+ " (\"" + response.responseCode() + " " + response.responseMessage() + "\")");
}
final String responseBody = response.responseBody();
if (responseBody.length() < 50) { // Ensure to have a valid response
2020-02-29 16:42:04 +01:00
throw new ParsingException("JSON response is too short");
}
// Check if the request was redirected to the error page.
final URL latestUrl = new URL(response.latestUrl());
if (latestUrl.getHost().equalsIgnoreCase("www.youtube.com")) {
final String path = latestUrl.getPath();
if (path.equalsIgnoreCase("/oops") || path.equalsIgnoreCase("/error")) {
throw new ContentNotAvailableException("Content unavailable");
}
}
final String responseContentType = response.getHeader("Content-Type");
if (responseContentType != null
&& responseContentType.toLowerCase().contains("text/html")) {
throw new ParsingException("Got HTML document, expected JSON response"
+ " (latest url was: \"" + response.latestUrl() + "\")");
}
return responseBody;
}
public static Response getResponse(final String url, final Localization localization)
throws IOException, ExtractionException {
final Map<String, List<String>> headers = new HashMap<>();
2021-04-07 12:25:59 +02:00
addYouTubeHeaders(headers);
final Response response = getDownloader().get(url, headers, localization);
getValidJsonResponseBody(response);
return response;
}
public static JsonObject getJsonPostResponse(final String endpoint,
final byte[] body,
final Localization localization)
throws IOException, ExtractionException {
final Map<String, List<String>> headers = new HashMap<>();
addClientInfoHeaders(headers);
headers.put("Content-Type", Collections.singletonList("application/json"));
final Response response = getDownloader().post(YOUTUBEI_V1_URL + endpoint + "?key="
+ getKey(), headers, body, localization);
return JsonUtils.toJsonObject(getValidJsonResponseBody(response));
}
public static JsonObject getJsonMobilePostResponse(final String endpoint,
final byte[] body,
@Nonnull final ContentCountry
contentCountry,
final Localization localization)
throws IOException, ExtractionException {
final Map<String, List<String>> headers = new HashMap<>();
headers.put("Content-Type", Collections.singletonList("application/json"));
// Spoofing an Android 11 device with the hardcoded version of the Android app
headers.put("User-Agent", Collections.singletonList("com.google.android.youtube/"
+ MOBILE_YOUTUBE_CLIENT_VERSION + "Linux; U; Android 11; "
+ contentCountry.getCountryCode() + ") gzip"));
headers.put("x-goog-api-format-version", Collections.singletonList("2"));
final Response response = getDownloader().post(
"https://youtubei.googleapis.com/youtubei/v1/" + endpoint + "?key="
+ MOBILE_YOUTUBE_KEY, headers, body, localization);
return JsonUtils.toJsonObject(getValidJsonResponseBody(response));
}
public static JsonArray getJsonResponse(final String url, final Localization localization)
throws IOException, ExtractionException {
Map<String, List<String>> headers = new HashMap<>();
2021-04-07 12:25:59 +02:00
addYouTubeHeaders(headers);
final Response response = getDownloader().get(url, headers, localization);
return JsonUtils.toJsonArray(getValidJsonResponseBody(response));
}
public static JsonArray getJsonResponse(@Nonnull final Page page,
final Localization localization)
throws IOException, ExtractionException {
final Map<String, List<String>> headers = new HashMap<>();
2021-04-07 12:25:59 +02:00
addYouTubeHeaders(headers);
final Response response = getDownloader().get(page.getUrl(), headers, localization);
return JsonUtils.toJsonArray(getValidJsonResponseBody(response));
}
@Nonnull
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
public static JsonBuilder<JsonObject> prepareDesktopJsonBuilder(
@Nonnull final Localization localization,
@Nonnull final ContentCountry contentCountry)
throws IOException, ExtractionException {
// @formatter:off
return JsonObject.builder()
.object("context")
.object("client")
.value("hl", localization.getLocalizationCode())
.value("gl", contentCountry.getCountryCode())
2021-06-24 18:39:16 +02:00
.value("clientName", "WEB")
.value("clientVersion", getClientVersion())
.end()
.object("user")
// TO DO: provide a way to enable restricted mode with:
// .value("enableSafetyMode", boolean)
2021-06-24 18:39:16 +02:00
.value("lockedSafetyMode", false)
.end()
.end();
// @formatter:on
}
@Nonnull
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
public static JsonBuilder<JsonObject> prepareAndroidMobileJsonBuilder(
@Nonnull final Localization localization,
@Nonnull final ContentCountry contentCountry) {
// @formatter:off
return JsonObject.builder()
.object("context")
.object("client")
.value("clientName", "ANDROID")
.value("clientVersion", MOBILE_YOUTUBE_CLIENT_VERSION)
.value("hl", localization.getLocalizationCode())
.value("gl", contentCountry.getCountryCode())
.end()
2021-06-24 18:39:16 +02:00
.object("user")
// TO DO: provide a way to enable restricted mode with:
// .value("enableSafetyMode", boolean)
.value("lockedSafetyMode", false)
.end()
.end();
// @formatter:on
}
Rebase + some code improvements + fix extraction of age-restricted videos + update clients version Here is now the requests which will be made by the `onFetchPage` method of `YoutubeStreamExtractor`: - the desktop API is fetched. If there is no streaming data, the desktop player API with the embed client screen will be fetched (and also the player code), then the Android mobile API. - if there is no streaming data, a `ContentNotAvailableException` will be thrown by using the message provided in playability status If the video is age restricted, a request to the next endpoint of the desktop player with the embed client screen will be sent. Otherwise, the next endpoint will be fetched normally, if the content is available. If the video is not age-restricted, a request to the player endpoint of the Android mobile API will be made. We can get more streams by using the Android mobile API but some streams may be not available on this API, so the streaming data of the Android mobile API will be first used to get itags and then the streaming data of the desktop internal API will be used. If the parsing of the Android mobile API went wrong, only the streams of the desktop API will be used. Other code changes: - `prepareJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareDesktopJsonBuilder` - `prepareMobileJsonBuilder` in `YoutubeParsingHelper` was renamed to `prepareAndroidMobileJsonBuilder` - two new methods in `YoutubeParsingHelper` were added: `prepareDesktopEmbedVideoJsonBuilder` and `prepareAndroidMobileEmbedVideoJsonBuilder` - `createPlayerBodyWithSts` is now public and was moved to `YoutubeParsingHelper` - a new method in `YoutubeJavaScriptExtractor` was added: `resetJavaScriptCode`, which was needed for the method `resetDebofuscationCode` of `YoutubeStreamExtractor` - `areHardcodedClientVersionAndKeyValid` in `YoutubeParsingHelper` returns now a `boolean` instead of an `Optional<Boolean>` - the `fetchVideoInfoPage` method of `YoutubeStreamExtractor` was removed because YouTube returns now 404 for every client with the `get_video_info` page - some unused objects and some warnings in `YoutubeStreamExtractor` were removed and fixed Co-authored-by: TiA4f8R <74829229+TiA4f8R@users.noreply.github.com>
2021-07-28 23:55:09 +02:00
@Nonnull
public static JsonBuilder<JsonObject> prepareDesktopEmbedVideoJsonBuilder(
@Nonnull final Localization localization,
@Nonnull final ContentCountry contentCountry,
@Nonnull final String videoId) throws IOException, ExtractionException {
// @formatter:off
return JsonObject.builder()
.object("context")
.object("client")
.value("hl", localization.getLocalizationCode())
.value("gl", contentCountry.getCountryCode())
.value("clientName", "WEB")
.value("clientVersion", getClientVersion())
.value("clientScreen", "EMBED")
.end()
.object("thirdParty")
.value("embedUrl", "https://www.youtube.com/watch?v=" + videoId)
.end()
.object("user")
// TO DO: provide a way to enable restricted mode with:
// .value("enableSafetyMode", boolean)
.value("lockedSafetyMode", false)
.end()
.end()
.value("videoId", videoId);
// @formatter:on
}
@Nonnull
public static JsonBuilder<JsonObject> prepareAndroidMobileEmbedVideoJsonBuilder(
@Nonnull final Localization localization,
@Nonnull final ContentCountry contentCountry,
@Nonnull final String videoId) {
// @formatter:off
return JsonObject.builder()
.object("context")
.object("client")
.value("clientName", "ANDROID")
.value("clientVersion", MOBILE_YOUTUBE_CLIENT_VERSION)
.value("clientScreen", "EMBED")
.value("hl", localization.getLocalizationCode())
.value("gl", contentCountry.getCountryCode())
.end()
.object("thirdParty")
.value("embedUrl", "https://www.youtube.com/watch?v=" + videoId)
.end()
.object("user")
// TO DO: provide a way to enable restricted mode with:
// .value("enableSafetyMode", boolean)
.value("lockedSafetyMode", false)
.end()
.end()
.value("videoId", videoId);
// @formatter:on
}
@Nonnull
public static byte[] createPlayerBodyWithSts(final Localization localization,
final ContentCountry contentCountry,
final String videoId,
final boolean withThirdParty,
@Nullable final String sts)
throws IOException, ExtractionException {
if (withThirdParty) {
// @formatter:off
return JsonWriter.string(prepareDesktopEmbedVideoJsonBuilder(localization, contentCountry, videoId)
.object("playbackContext")
.object("contentPlaybackContext")
.value("signatureTimestamp", sts)
.end()
.end()
.done())
.getBytes(UTF_8);
// @formatter:on
} else {
// @formatter:off
return JsonWriter.string(prepareDesktopJsonBuilder(localization, contentCountry)
.value("videoId", videoId)
.object("playbackContext")
.object("contentPlaybackContext")
.value("signatureTimestamp", sts)
.end()
.end()
.done())
.getBytes(UTF_8);
// @formatter:on
}
}
2021-04-07 12:25:59 +02:00
/**
* Add required headers and cookies to an existing headers Map.
* @see #addClientInfoHeaders(Map)
* @see #addCookieHeader(Map)
*/
public static void addYouTubeHeaders(final Map<String, List<String>> headers)
throws IOException, ExtractionException {
addClientInfoHeaders(headers);
addCookieHeader(headers);
}
/**
* Add the <code>X-YouTube-Client-Name</code>, <code>X-YouTube-Client-Version</code>,
* <code>Origin</code>, and <code>Referer</code> headers.
2021-04-07 12:25:59 +02:00
* @param headers The headers which should be completed
*/
public static void addClientInfoHeaders(@Nonnull final Map<String, List<String>> headers)
2021-04-07 12:25:59 +02:00
throws IOException, ExtractionException {
2021-06-24 18:39:16 +02:00
headers.computeIfAbsent("Origin", k -> Collections.singletonList(
"https://www.youtube.com"));
headers.computeIfAbsent("Referer", k -> Collections.singletonList(
"https://www.youtube.com"));
headers.computeIfAbsent("X-YouTube-Client-Name", k -> Collections.singletonList("1"));
2021-04-07 12:25:59 +02:00
if (headers.get("X-YouTube-Client-Version") == null) {
headers.put("X-YouTube-Client-Version", Collections.singletonList(getClientVersion()));
}
}
/**
* Add the <code>CONSENT</code> cookie to prevent redirect to <code>consent.youtube.com</code>
* @see #CONSENT_COOKIE
* @param headers the headers which should be completed
*/
public static void addCookieHeader(@Nonnull final Map<String, List<String>> headers) {
2021-04-07 12:25:59 +02:00
if (headers.get("Cookie") == null) {
headers.put("Cookie", Arrays.asList(generateConsentCookie()));
2021-04-07 12:25:59 +02:00
} else {
headers.get("Cookie").add(generateConsentCookie());
2021-04-07 12:25:59 +02:00
}
}
@Nonnull
public static String generateConsentCookie() {
2021-07-03 13:14:22 +02:00
final int statusCode = 100 + numberGenerator.nextInt(900);
return CONSENT_COOKIE + statusCode;
}
public static String extractCookieValue(final String cookieName,
@Nonnull final Response response) {
2021-04-07 12:25:59 +02:00
final List<String> cookies = response.responseHeaders().get("set-cookie");
int startIndex;
String result = "";
for (final String cookie : cookies) {
startIndex = cookie.indexOf(cookieName);
if (startIndex != -1) {
result = cookie.substring(startIndex + cookieName.length() + "=".length(),
cookie.indexOf(";", startIndex));
}
}
return result;
}
/**
* Shared alert detection function, multiple endpoints return the error similarly structured.
* <p>
* Will check if the object has an alert of the type "ERROR".
* </p>
*
* @param initialData the object which will be checked if an alert is present
* @throws ContentNotAvailableException if an alert is detected
*/
public static void defaultAlertsCheck(@Nonnull final JsonObject initialData)
throws ParsingException {
final JsonArray alerts = initialData.getArray("alerts");
2020-04-15 18:49:58 +02:00
if (!isNullOrEmpty(alerts)) {
final JsonObject alertRenderer = alerts.getObject(0).getObject("alertRenderer");
2020-05-03 10:28:45 +02:00
final String alertText = getTextFromObject(alertRenderer.getObject("text"));
final String alertType = alertRenderer.getString("type", EMPTY_STRING);
if (alertType.equalsIgnoreCase("ERROR")) {
if (alertText != null && alertText.contains("This account has been terminated")) {
if (alertText.contains("violation") || alertText.contains("violating")
|| alertText.contains("infringement")) {
// Possible error messages:
// "This account has been terminated for a violation of YouTube's Terms of Service."
2021-03-30 10:14:33 +02:00
// "This account has been terminated due to multiple or severe violations of YouTube's policy prohibiting hate speech."
// "This account has been terminated due to multiple or severe violations of YouTube's policy prohibiting content designed to harass, bully or threaten."
// "This account has been terminated due to multiple or severe violations of YouTube's policy against spam, deceptive practices and misleading content or other Terms of Service violations."
// "This account has been terminated due to multiple or severe violations of YouTube's policy on nudity or sexual content."
// "This account has been terminated for violating YouTube's Community Guidelines."
// "This account has been terminated because we received multiple third-party claims of copyright infringement regarding material that the user posted."
2021-03-30 10:14:33 +02:00
// "This account has been terminated because it is linked to an account that received multiple third-party claims of copyright infringement."
throw new AccountTerminatedException(alertText, AccountTerminatedException.Reason.VIOLATION);
} else {
throw new AccountTerminatedException(alertText);
}
}
throw new ContentNotAvailableException("Got error: \"" + alertText + "\"");
}
}
}
@Nonnull
public static List<MetaInfo> getMetaInfo(@Nonnull final JsonArray contents)
throws ParsingException {
final List<MetaInfo> metaInfo = new ArrayList<>();
for (final Object content : contents) {
final JsonObject resultObject = (JsonObject) content;
if (resultObject.has("itemSectionRenderer")) {
for (final Object sectionContentObject :
resultObject.getObject("itemSectionRenderer").getArray("contents")) {
final JsonObject sectionContent = (JsonObject) sectionContentObject;
if (sectionContent.has("infoPanelContentRenderer")) {
metaInfo.add(getInfoPanelContent(sectionContent
.getObject("infoPanelContentRenderer")));
}
if (sectionContent.has("clarificationRenderer")) {
metaInfo.add(getClarificationRendererContent(sectionContent
.getObject("clarificationRenderer")
));
}
}
}
}
return metaInfo;
}
@Nonnull
private static MetaInfo getInfoPanelContent(@Nonnull final JsonObject infoPanelContentRenderer)
throws ParsingException {
final MetaInfo metaInfo = new MetaInfo();
final StringBuilder sb = new StringBuilder();
for (final Object paragraph : infoPanelContentRenderer.getArray("paragraphs")) {
if (sb.length() != 0) {
sb.append("<br>");
}
sb.append(YoutubeParsingHelper.getTextFromObject((JsonObject) paragraph));
}
metaInfo.setContent(new Description(sb.toString(), Description.HTML));
if (infoPanelContentRenderer.has("sourceEndpoint")) {
final String metaInfoLinkUrl = YoutubeParsingHelper.getUrlFromNavigationEndpoint(
infoPanelContentRenderer.getObject("sourceEndpoint"));
try {
metaInfo.addUrl(new URL(Objects.requireNonNull(extractCachedUrlIfNeeded(
metaInfoLinkUrl))));
} catch (final NullPointerException | MalformedURLException e) {
throw new ParsingException("Could not get metadata info URL", e);
}
final String metaInfoLinkText = YoutubeParsingHelper.getTextFromObject(
infoPanelContentRenderer.getObject("inlineSource"));
if (isNullOrEmpty(metaInfoLinkText)) {
throw new ParsingException("Could not get metadata info link text.");
}
metaInfo.addUrlText(metaInfoLinkText);
}
return metaInfo;
}
@Nonnull
private static MetaInfo getClarificationRendererContent(@Nonnull final JsonObject clarificationRenderer)
throws ParsingException {
final MetaInfo metaInfo = new MetaInfo();
final String title = YoutubeParsingHelper.getTextFromObject(clarificationRenderer
.getObject("contentTitle"));
final String text = YoutubeParsingHelper.getTextFromObject(clarificationRenderer
.getObject("text"));
2021-02-07 22:12:22 +01:00
if (title == null || text == null) {
throw new ParsingException("Could not extract clarification renderer content");
}
metaInfo.setTitle(title);
metaInfo.setContent(new Description(text, Description.PLAIN_TEXT));
if (clarificationRenderer.has("actionButton")) {
final JsonObject actionButton = clarificationRenderer.getObject("actionButton")
.getObject("buttonRenderer");
try {
final String url = YoutubeParsingHelper.getUrlFromNavigationEndpoint(actionButton
.getObject("command"));
metaInfo.addUrl(new URL(Objects.requireNonNull(extractCachedUrlIfNeeded(url))));
} catch (final NullPointerException | MalformedURLException e) {
throw new ParsingException("Could not get metadata info URL", e);
}
final String metaInfoLinkText = YoutubeParsingHelper.getTextFromObject(
actionButton.getObject("text"));
if (isNullOrEmpty(metaInfoLinkText)) {
throw new ParsingException("Could not get metadata info link text.");
}
metaInfo.addUrlText(metaInfoLinkText);
}
if (clarificationRenderer.has("secondaryEndpoint") && clarificationRenderer
.has("secondarySource")) {
final String url = getUrlFromNavigationEndpoint(clarificationRenderer
.getObject("secondaryEndpoint"));
// Ignore Google URLs, because those point to a Google search about "Covid-19"
if (url != null && !isGoogleURL(url)) {
try {
metaInfo.addUrl(new URL(url));
final String description = getTextFromObject(clarificationRenderer
.getObject("secondarySource"));
metaInfo.addUrlText(description == null ? url : description);
} catch (final MalformedURLException e) {
throw new ParsingException("Could not get metadata info secondary URL", e);
}
}
}
return metaInfo;
}
/**
* Sometimes, YouTube provides URLs which use Google's cache. They look like
* {@code https://webcache.googleusercontent.com/search?q=cache:CACHED_URL}
2021-02-07 22:12:22 +01:00
*
* @param url the URL which might refer to the Google's webcache
* @return the URL which is referring to the original site
*/
public static String extractCachedUrlIfNeeded(final String url) {
if (url == null) {
return null;
}
if (url.contains("webcache.googleusercontent.com")) {
return url.split("cache:")[1];
}
return url;
}
public static boolean isVerified(final JsonArray badges) {
if (Utils.isNullOrEmpty(badges)) {
return false;
}
for (Object badge : badges) {
final String style = ((JsonObject) badge).getObject("metadataBadgeRenderer")
.getString("style");
if (style != null && (style.equals("BADGE_STYLE_TYPE_VERIFIED")
|| style.equals("BADGE_STYLE_TYPE_VERIFIED_ARTIST"))) {
return true;
}
}
return false;
}
@Nonnull
public static String unescapeDocument(@Nonnull final String doc) {
return doc
.replaceAll("\\\\x22", "\"")
.replaceAll("\\\\x7b", "{")
.replaceAll("\\\\x7d", "}")
.replaceAll("\\\\x5b", "[")
.replaceAll("\\\\x5d", "]");
}
2017-03-01 18:47:52 +01:00
}