Is Youtube change page structure again?

[OtakuEmperor]

After #315, It seems like work a few days, but this week, the regular expression string functNamePattern = @"\""signature"",\s?([a-zA-Z0-9\$]+)\("; in Decipherer.cs got empty string Did anyone get the same situation?

[LinPolly]

https://github.com/flagbug/YoutubeExtractor/issues/313#issuecomment-419663362

[ghost]

#313 (comment)

Don't work

[OtakuEmperor]

I don't know others how to fix it. In my situation, I can't find out the function name. And I mix two solution.

313 and https://github.com/i3arnon/libvideo/commit/85dba890c50221d7fd4a27f37a60b5ac316cf601#diff-064c53a6b75fe468598d6396ede2b27a

I change string functNamePattern = @"\""signature"",\s?([a-zA-Z0-9\$]+)\("; to string functNamePattern = @"(\w+)&&(\w+)\.set\(\w+,(\w+)\(\1\)\);return\s+\2";

and change

string funcPattern = @"(?!h\.)" + @funcName + @"=function\(\w+\)\{.*?\}"; //Escape funcName string to string funcPattern = @"(?!h\.)" + @funcName + @"(\w+)\s*=\s*function\(\s*(\w+)\s*\)\s*{\s*\2\s*=\s*\2\.split\(\""\""\)\s*;(.+)return\s*\2\.join\(\""\""\)\s*}\s*;"; //Escape funcName string

And it works for me now, hope it will help you too. Hope Youtube won't change the structure.

[sameer]

According to youtube-dl, these are the new regexes that need to be added.

[kelvinRosa]

@sameer did you know where to update this?

[sameer]

@sameer did you know where to update this?

No but I can look.

[anilgit90]

@kelvinRosa Just look at the pull request "Changes to DownloadUrlResolver #333". I hope I have covered the changes done by youtube-dl.

[Makzz90]

My solution. 1) DownloadUrlResolver: ` public static class DownloadUrlResolver { private const string RateBypassFlag = "ratebypass"; private const string SignatureQuery = "sig";//signature

    /// <summary>
    /// Decrypts the signature in the <see cref="VideoInfo.DownloadUrl" /> property and sets it
    /// to the decrypted URL. Use this method, if you have decryptSignature in the <see
    /// cref="GetDownloadUrls" /> method set to false.
    /// </summary>
    /// <param name="videoInfo">The video info which's downlaod URL should be decrypted.</param>
    /// <exception cref="YoutubeParseException">
    /// There was an error while deciphering the signature.
    /// </exception>
    public static void DecryptDownloadUrl(VideoInfo videoInfo)
    {
        Regex regex = new Regex("&"+SignatureQuery+"=(.+)&");
        Match match = regex.Match(videoInfo.DownloadUrl);
        if(match.Success)
        {
            string encryptedSignature = match.Groups[1].Value;
            string decrypted;

            try
            {
                decrypted = GetDecipheredSignature(videoInfo.HtmlPlayerVersion, encryptedSignature);
            }

            catch (Exception ex)
            {
                throw new YoutubeParseException("Could not decipher signature", ex);
            }

            videoInfo.DownloadUrl = regex.Replace(videoInfo.DownloadUrl, "&"+SignatureQuery+"="+ decrypted + "&");
            videoInfo.RequiresDecryption = false;
        }
    }

    /// <summary>
    /// Gets a list of <see cref="VideoInfo" />s for the specified URL.
    /// </summary>
    /// <param name="videoUrl">The URL of the YouTube video.</param>
    /// <param name="decryptSignature">
    /// A value indicating whether the video signatures should be decrypted or not. Decrypting
    /// consists of a HTTP request for each <see cref="VideoInfo" />, so you may want to set
    /// this to false and call <see cref="DecryptDownloadUrl" /> on your selected <see
    /// cref="VideoInfo" /> later.
    /// </param>
    /// <returns>A list of <see cref="VideoInfo" />s that can be used to download the video.</returns>
    /// <exception cref="ArgumentNullException">
    /// The <paramref name="videoUrl" /> parameter is <c>null</c>.
    /// </exception>
    /// <exception cref="ArgumentException">
    /// The <paramref name="videoUrl" /> parameter is not a valid YouTube URL.
    /// </exception>
    /// <exception cref="VideoNotAvailableException">The video is not available.</exception>
    /// <exception cref="WebException">
    /// An error occurred while downloading the YouTube page html.
    /// </exception>
    /// <exception cref="YoutubeParseException">The Youtube page could not be parsed.</exception>
    public static IEnumerable<VideoInfo> GetDownloadUrls(string videoUrl, bool decryptSignature = true)
    {
        if (videoUrl == null)
            throw new ArgumentNullException("videoUrl");

        bool isYoutubeUrl = TryNormalizeYoutubeUrl(videoUrl, out videoUrl);

        if (!isYoutubeUrl)
        {
            throw new ArgumentException("URL is not a valid youtube URL!");
        }

        try
        {
            var json = LoadJson(videoUrl);

            string videoTitle = GetVideoTitle(json);

            IEnumerable<ExtractionInfo> downloadUrls = ExtractDownloadUrls(json);

            IEnumerable<VideoInfo> infos = GetVideoInfos(downloadUrls, videoTitle).ToList();

            string htmlPlayerVersion = GetHtml5PlayerVersion(json);

            foreach (VideoInfo info in infos)
            {
                info.HtmlPlayerVersion = htmlPlayerVersion;

                if (decryptSignature && info.RequiresDecryption)
                {
                    DecryptDownloadUrl(info);
                }
            }

            return infos;
        }

        catch (Exception ex)
        {
            if (ex is WebException || ex is VideoNotAvailableException)
            {
                throw;
            }

            ThrowYoutubeParseException(ex, videoUrl);
        }

        return null; // Will never happen, but the compiler requires it
    }

if PORTABLE

    public static System.Threading.Tasks.Task<IEnumerable<VideoInfo>> GetDownloadUrlsAsync(string videoUrl, bool decryptSignature = true)
    {
        return System.Threading.Tasks.Task.Run(() => GetDownloadUrls(videoUrl, decryptSignature));
    }

endif

    /// <summary>
    /// Normalizes the given YouTube URL to the format http://youtube.com/watch?v={youtube-id}
    /// and returns whether the normalization was successful or not.
    /// </summary>
    /// <param name="url">The YouTube URL to normalize.</param>
    /// <param name="normalizedUrl">The normalized YouTube URL.</param>
    /// <returns>
    /// <c>true</c>, if the normalization was successful; <c>false</c>, if the URL is invalid.
    /// </returns>
    public static bool TryNormalizeYoutubeUrl(string url, out string normalizedUrl)
    {
        url = url.Trim();

        url = url.Replace("youtu.be/", "youtube.com/watch?v=");
        url = url.Replace("www.youtube", "youtube");
        url = url.Replace("youtube.com/embed/", "youtube.com/watch?v=");

        if (url.Contains("/v/"))
        {
            url = "http://youtube.com" + new Uri(url).AbsolutePath.Replace("/v/", "/watch?v=");
        }

        url = url.Replace("/watch#", "/watch?");

        IDictionary<string, string> query = HttpHelper.ParseQueryString(url);

        string v;

        if (!query.TryGetValue("v", out v))
        {
            normalizedUrl = null;
            return false;
        }

        normalizedUrl = "http://youtube.com/watch?v=" + v;

        return true;
    }
    /*
    private static IEnumerable<ExtractionInfo> ExtractDownloadUrls(JObject json)
    {
        string[] splitByUrls = GetStreamMap(json).Split(',');
        string[] adaptiveFmtSplitByUrls = GetAdaptiveStreamMap(json).Split(',');
        splitByUrls = splitByUrls.Concat(adaptiveFmtSplitByUrls).ToArray();

        foreach (string s in splitByUrls)
        {
            IDictionary<string, string> queries = HttpHelper.ParseQueryString(s);
            string url;

            bool requiresDecryption = false;

            if (queries.ContainsKey("s") || queries.ContainsKey("sig"))
            {
                requiresDecryption = queries.ContainsKey("s");
                string signature = queries.ContainsKey("s") ? queries["s"] : queries["sig"];

                url = string.Format("{0}&{1}={2}", queries["url"], SignatureQuery, signature);

                string fallbackHost = queries.ContainsKey("fallback_host") ? "&fallback_host=" + queries["fallback_host"] : String.Empty;

                url += fallbackHost;
            }

            else
            {
                url = queries["url"];
            }

            url = HttpHelper.UrlDecode(url);
            url = HttpHelper.UrlDecode(url);

            IDictionary<string, string> parameters = HttpHelper.ParseQueryString(url);
            if (!parameters.ContainsKey(RateBypassFlag))
                url += string.Format("&{0}={1}", RateBypassFlag, "yes");

            yield return new ExtractionInfo { RequiresDecryption = requiresDecryption, Uri = new Uri(url) };
        }
    }
    */
    //
    private static IEnumerable<ExtractionInfo> ExtractDownloadUrls(JObject json)
    {
        var info = new List<ExtractionInfo>();

        var formats = GetStreamMap(json);
        var adaptiveFormats = GetAdaptiveStreamMap(json);

        ExtractInfo(info, formats);
        ExtractInfo(info, adaptiveFormats);

        return info;
    }
    private static void ExtractInfo(List<ExtractionInfo> info, JToken items)
    {
        if (items != null)
        {
            foreach (var item in items)
            {
                string url = "";
                bool requiresDecryption = false;

                if (item["cipher"] != null )
                {
                    string s = item["cipher"].ToString();
                    IDictionary<string, string> queries = HttpHelper.ParseQueryString(s);

                    if (queries.ContainsKey("s") || queries.ContainsKey("sig"))
                    {
                        requiresDecryption = queries.ContainsKey("s");
                        string signature = queries.ContainsKey("s") ? queries["s"] : queries["sig"];

                        url = string.Format("{0}&{1}={2}", queries["url"], SignatureQuery, signature);

                        string fallbackHost = queries.ContainsKey("fallback_host") ? "&fallback_host=" + queries["fallback_host"] : String.Empty;

                        url += fallbackHost;
                    }
                }
                else
                {
                    url = item["url"].ToString();
                }

                url = HttpHelper.UrlDecode(url);

                IDictionary<string, string> parameters = HttpHelper.ParseQueryString(url);
                if (!parameters.ContainsKey(RateBypassFlag))
                    url += string.Format("&{0}={1}", RateBypassFlag, "yes");

                info.Add( new ExtractionInfo { RequiresDecryption = requiresDecryption, Uri = new Uri(url) });
            }
        }
    }
    //
    /*
    private static string GetAdaptiveStreamMap(JObject json)
    {
        JToken streamMap = json["args"]["adaptive_fmts"];

        // bugfix: adaptive_fmts is missing in some videos, use url_encoded_fmt_stream_map instead
        if (streamMap == null)
        {
          streamMap = json["args"]["url_encoded_fmt_stream_map"];
        }

        return streamMap.ToString();
    }
    */
    private static JToken GetAdaptiveStreamMap(JObject json)
    {
        JToken adaptiveFormat = null;
        JToken streamMap = json["args"]["player_response"];

        string streamMapString = streamMap == null ? null : streamMap.ToString();

        if (streamMapString != null)
        {
            JObject playerResponse = JObject.Parse(streamMapString);
            var temp = playerResponse?["streamingData"]?["adaptiveFormats"];

            adaptiveFormat = temp;
        }

        return adaptiveFormat;
    }
    private static string GetDecipheredSignature(string htmlPlayerVersion, string signature)
    {
        return Decipherer.DecipherWithVersion(signature, htmlPlayerVersion);
    }
    /*
    private static string GetHtml5PlayerVersion(JObject json)
    {
        var regex = new Regex(@"player(.+?).js");

        string js = json["assets"]["js"].ToString();

        return regex.Match(js).Result("$1");
    }
    */
    private static string GetHtml5PlayerVersion(JObject json)
    {
        var regex = new Regex(@"player[-|_](.+?).js");

        string js = json["assets"]["js"].ToString();

        return regex.Match(js).Result("$1");
    }
    /*
    private static string GetStreamMap(JObject json)
    {
        JToken streamMap = json["args"]["url_encoded_fmt_stream_map"];

        string streamMapString = streamMap == null ? null : streamMap.ToString();

        if (streamMapString == null || streamMapString.Contains("been+removed"))
        {
            throw new VideoNotAvailableException("Video is removed or has an age restriction.");
        }

        return streamMapString;
    }
    */
    private static JToken GetStreamMap(JObject json)
    {
        JToken streamMap = json["args"]["player_response"];

        string streamMapString = streamMap == null ? null : streamMap.ToString();

        if (streamMapString == null || streamMapString.Contains("been+removed"))
        {
            throw new Exception("Video is removed or has an age restriction.");
        }

        JObject playerResponse = JObject.Parse(streamMapString);
        JToken temp = playerResponse?["streamingData"]?["adaptiveFormats"];
        return temp;
    }

    private static IEnumerable<VideoInfo> GetVideoInfos(IEnumerable<ExtractionInfo> extractionInfos, string videoTitle)
    {
        var downLoadInfos = new List<VideoInfo>();

        foreach (ExtractionInfo extractionInfo in extractionInfos)
        {
            string itag = HttpHelper.ParseQueryString(extractionInfo.Uri.Query)["itag"];

            int formatCode = int.Parse(itag);

            VideoInfo info = VideoInfo.Defaults.SingleOrDefault(videoInfo => videoInfo.FormatCode == formatCode);

            if (info != null)
            {
                info = new VideoInfo(info)
                {
                    DownloadUrl = extractionInfo.Uri.ToString(),
                    Title = videoTitle,
                    RequiresDecryption = extractionInfo.RequiresDecryption
                };
            }

            else
            {
                info = new VideoInfo(formatCode)
                {
                    DownloadUrl = extractionInfo.Uri.ToString()
                };
            }

            downLoadInfos.Add(info);
        }

        return downLoadInfos;
    }

    private static string GetVideoTitle(JObject json)
    {
        JToken title = json["args"]["title"];

        return title == null ? String.Empty : title.ToString();
    }

    private static bool IsVideoUnavailable(string pageSource)
    {
        const string unavailableContainer = "<div id=\"watch-player-unavailable\">";

        return pageSource.Contains(unavailableContainer);
    }

    private static JObject LoadJson(string url)
    {
        string pageSource = HttpHelper.DownloadString(url);

        if (IsVideoUnavailable(pageSource))
        {
            throw new VideoNotAvailableException();
        }

        var dataRegex = new Regex(@"ytplayer\.config\s*=\s*(\{.+?\});", RegexOptions.Multiline);

        string extractedJson = dataRegex.Match(pageSource).Result("$1");

        return JObject.Parse(extractedJson);
    }

    private static void ThrowYoutubeParseException(Exception innerException, string videoUrl)
    {
        throw new YoutubeParseException("Could not parse the Youtube page for URL " + videoUrl + "\n" +
                                        "This may be due to a change of the Youtube page structure.\n" +
                                        "Please report this bug at www.github.com/flagbug/YoutubeExtractor/issues", innerException);
    }

    private class ExtractionInfo
    {
        public bool RequiresDecryption { get; set; }

        public Uri Uri { get; set; }
    }
}

}`

2) Decipherer: `internal static class Decipherer {

    public static string DecipherWithVersion(string cipher, string cipherVersion)
    {
        string jsUrl = string.Format("http://s.ytimg.com/yts/jsbin/player_{0}.js", cipherVersion);//_
        string js = HttpHelper.DownloadString(jsUrl);

        //Find "C" in this: var A = B.sig||C (B.s)
        //string functNamePattern = @"\""signature"",\s?([a-zA-Z0-9\$]+)\("; //Regex Formed To Find Word or DollarSign
        string functNamePattern = @"(\w+)=function\(\w+\){(\w+)=\2\.split\(\x22{2}\);.*?return\s+\2\.join\(\x22{2}\)}";

        var funcName = Regex.Match(js, functNamePattern).Groups[1].Value;

        if(string.IsNullOrEmpty(funcName))
            throw new NotImplementedException("Couldn't find signature function.");

        if (funcName.Contains("$"))
            funcName = "\\" + funcName; //Due To Dollar Sign Introduction, Need To Escape

        string funcPattern = @"(?!h\.)" + @funcName + @"=function\(\w+\)\{.*?\}"; //Escape funcName string
        var funcBody = Regex.Match(js, funcPattern, RegexOptions.Singleline).Value; //Entire sig function
        var lines = funcBody.Split(';'); //Each line in sig function

        string functionIdentifier = "";
        string operations = "";

        foreach (var line in lines) //Matches the funcBody with each cipher method. Only runs till all three are defined.
        {
            Match m;

            //
            // Get the name of the function called in this statement
            functionIdentifier = Regex.Match(line, @"\w+(?:.|\[)(\""?\w+(?:\"")?)\]?\(").Groups[1].Value;
            if (string.IsNullOrWhiteSpace(functionIdentifier))
                continue;
            //

            string reReverse = string.Format(@"{0}:\bfunction\b\(\w+\)", functionIdentifier); //Regex for reverse (one parameter)
            string reSlice = string.Format(@"{0}:\bfunction\b\([a],b\).(\breturn\b)?.?\w+\.", functionIdentifier); //Regex for slice (return or not)
            string reSwap = string.Format(@"{0}:\bfunction\b\(\w+\,\w\).\bvar\b.\bc=a\b", functionIdentifier); //Regex for the char swap.

            if ((m=Regex.Match(js, reSlice)).Success)
            {
                if ((m = Regex.Match(line, @"\(\w+,(?<index>\d+)\)")).Success)
                    operations += "s" + m.Groups["index"].Value + " "; //operation is a slice
            }
            else if ((m=Regex.Match(js, reSwap)).Success)
            {
                if ((m = Regex.Match(line, @"\(\w+,(?<index>\d+)\)")).Success)
                    operations += "w" + m.Groups["index"].Value + " "; //operation is a swap (w)
            }
            else if((m = Regex.Match(js, reReverse)).Success)
            {
                operations += "r "; //operation is a reverse
            }
        }

        operations = operations.Trim();

        return DecipherWithOperations(cipher, operations);
    }

    private static string ApplyOperation(string cipher, string op)
    {
        switch (op[0])
        {
            case 'r'://reverse
                return new string(cipher.ToCharArray().Reverse().ToArray());

            case 'w'://Swap
                {
                    int index = GetOpIndex(op);
                    return SwapFirstChar(cipher, index);
                }

            case 's'://Slice
                {
                    int index = GetOpIndex(op);
                    return cipher.Substring(index);
                }

            default:
                throw new NotImplementedException("Couldn't find cipher operation.");
        }
    }

    private static string DecipherWithOperations(string cipher, string operations)
    {
        return operations.Split(new[] { " " }, StringSplitOptions.RemoveEmptyEntries).Aggregate(cipher, ApplyOperation);
    }

    private static int GetOpIndex(string op)
    {
        string parsed = new Regex(@".(\d+)").Match(op).Result("$1");
        int index = Int32.Parse(parsed);

        return index;
    }

    private static string SwapFirstChar(string cipher, int index)
    {
        var builder = new StringBuilder(cipher);
        builder[0] = cipher[index];
        builder[index] = cipher[0];

        return builder.ToString();
    }
}`

based on YoutubeExplode.