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

Reimplement YoutubeChannelInfoItemExtractor

Browse source
This commit is contained in:
wb9688 2020-02-23 18:27:28 +01:00 committed by TobiGr
parent 3187116a63
commit 8aea4d445b
2 changed files with 31 additions and 57 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

86
extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/extractors/YoutubeChannelInfoItemExtractor.java
View file

@ -1,13 +1,12 @@
package org.schabi.newpipe.extractor.services.youtube.extractors;
import org.jsoup.nodes.Element;
import com.grack.nanojson.JsonObject;
import org.schabi.newpipe.extractor.channel.ChannelInfoItemExtractor;
import org.schabi.newpipe.extractor.exceptions.ParsingException;
import org.schabi.newpipe.extractor.services.youtube.linkHandler.YoutubeChannelLinkHandlerFactory;
import org.schabi.newpipe.extractor.utils.Utils;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
/*
* Created by Christian Schabesberger on 12.02.17.
*
@ -29,87 +28,62 @@ import java.util.regex.Pattern;
*/
public class YoutubeChannelInfoItemExtractor implements ChannelInfoItemExtractor {
private final Element el;
private JsonObject channelInfoItem;
public YoutubeChannelInfoItemExtractor(Element el) {
this.el = el;
public YoutubeChannelInfoItemExtractor(JsonObject channelInfoItem) {
this.channelInfoItem = channelInfoItem;
}
@Override
public String getThumbnailUrl() throws ParsingException {
Element img = el.select("span[class*=\"yt-thumb-simple\"]").first()
.select("img").first();
String url = img.attr("abs:src");
if (url.contains("gif")) {
url = img.attr("abs:data-thumb");
try {
return channelInfoItem.getObject("thumbnails").getArray("thumbnails").getObject(0).getString("url");
} catch (Exception e) {
throw new ParsingException("Could not get thumbnail url", e);
}
return url;
}
@Override
public String getName() throws ParsingException {
return el.select("a[class*=\"yt-uix-tile-link\"]").first()
.text();
try {
return channelInfoItem.getObject("title").getString("simpleText");
} catch (Exception e) {
throw new ParsingException("Could not get name", e);
}
}
@Override
public String getUrl() throws ParsingException {
try {
String buttonTrackingUrl = el.select("button[class*=\"yt-uix-button\"]").first()
.attr("abs:data-href");
Pattern channelIdPattern = Pattern.compile("(?:.*?)\\%252Fchannel\\%252F([A-Za-z0-9\\-\\_]+)(?:.*)");
Matcher match = channelIdPattern.matcher(buttonTrackingUrl);
if (match.matches()) {
return YoutubeChannelExtractor.CHANNEL_URL_BASE + match.group(1);
}
} catch(Exception ignored) {}
// fallback method for channels without "Subscribe" button (or just in case yt changes things)
// provides an url with "/user/NAME", inconsistent with stream and channel extractor: tests will fail
try {
return el.select("a[class*=\"yt-uix-tile-link\"]").first()
.attr("abs:href");
String id = "channel/" + channelInfoItem.getString("channelId"); // Does prepending 'channel/' always work?
return YoutubeChannelLinkHandlerFactory.getInstance().getUrl(id);
} catch (Exception e) {
throw new ParsingException("Could not get channel url", e);
throw new ParsingException("Could not get url", e);
}
}
@Override
public long getSubscriberCount() throws ParsingException {
final Element subsEl = el.select("span[class*=\"yt-subscriber-count\"]").first();
if (subsEl != null) {
try {
return Long.parseLong(Utils.removeNonDigitCharacters(subsEl.text()));
} catch (NumberFormatException e) {
throw new ParsingException("Could not get subscriber count", e);
}
} else {
// If the element is null, the channel have the subscriber count disabled
return -1;
}
public long getSubscriberCount() {
// TODO: get subscriber count, it's in subscriberCountText.simpleText as a string like "103M subscribers"
return -1;
}
@Override
public long getStreamCount() throws ParsingException {
Element metaEl = el.select("ul[class*=\"yt-lockup-meta-info\"]").first();
if (metaEl == null) {
return 0;
} else {
return Long.parseLong(Utils.removeNonDigitCharacters(metaEl.text()));
try {
return Long.parseLong(Utils.removeNonDigitCharacters(channelInfoItem.getObject("videoCountText")
.getArray("runs").getObject(0).getString("text")));
} catch (Exception e) {
throw new ParsingException("Could not get name", e);
}
}
@Override
public String getDescription() throws ParsingException {
Element desEl = el.select("div[class*=\"yt-lockup-description\"]").first();
if (desEl == null) {
return "";
} else {
return desEl.text();
try {
return channelInfoItem.getObject("descriptionSnippet").getArray("runs").getObject(0).getString("text");
} catch (Exception e) {
throw new ParsingException("Could not get description url", e);
}
}
}

2
extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/extractors/YoutubeSearchExtractor.java
View file

@ -130,7 +130,7 @@ public class YoutubeSearchExtractor extends SearchExtractor {
} else if (((JsonObject) item).getObject("videoRenderer") != null) {
collector.commit(new YoutubeStreamInfoItemExtractor(((JsonObject) item).getObject("videoRenderer"), timeAgoParser));
} else if (((JsonObject) item).getObject("channelRenderer") != null) {
// collector.commit(new YoutubeChannelInfoItemExtractor(((JsonObject) item).getObject("channelRenderer")));
collector.commit(new YoutubeChannelInfoItemExtractor(((JsonObject) item).getObject("channelRenderer")));
} else if (((JsonObject) item).getObject("playlistRenderer") != null) {
// collector.commit(new YoutubePlaylistInfoItemExtractor(((JsonObject) item).getObject("playlistRenderer")));
}