Tranga-Website/Tranga/Connector.cs

390 lines
19 KiB
C#
Raw Normal View History

using System.IO.Compression;
using System.Net;
2023-06-01 18:28:58 +02:00
using System.Runtime.InteropServices;
2023-06-03 22:25:24 +02:00
using System.Text.RegularExpressions;
2023-05-20 00:19:04 +02:00
using System.Xml.Linq;
using Logging;
using Tranga.TrangaTasks;
2023-06-01 18:28:58 +02:00
using static System.IO.UnixFileMode;
namespace Tranga;
/// <summary>
/// Base-Class for all Connectors
/// Provides some methods to be used by all Connectors, as well as a DownloadClient
/// </summary>
public abstract class Connector
{
protected TrangaSettings settings { get; }
protected DownloadClient downloadClient { get; init; } = null!;
protected readonly Logger? logger;
2023-05-25 14:23:33 +02:00
protected Connector(TrangaSettings settings, Logger? logger = null)
{
this.settings = settings;
this.logger = logger;
if (!Directory.Exists(settings.coverImageCache))
Directory.CreateDirectory(settings.coverImageCache);
}
public abstract string name { get; } //Name of the Connector (e.g. Website)
/// <summary>
/// Returns all Publications with the given string.
/// If the string is empty or null, returns all Publication of the Connector
/// </summary>
/// <param name="publicationTitle">Search-Query</param>
/// <returns>Publications matching the query</returns>
public abstract Publication[] GetPublications(string publicationTitle = "");
/// <summary>
/// Returns all Chapters of the publication in the provided language.
/// If the language is empty or null, returns all Chapters in all Languages.
/// </summary>
/// <param name="publication">Publication to get Chapters for</param>
/// <param name="language">Language of the Chapters</param>
/// <returns>Array of Chapters matching Publication and Language</returns>
public abstract Chapter[] GetChapters(Publication publication, string language = "");
2023-06-09 11:06:18 +02:00
public Chapter[] SelectChapters(Publication publication, string searchTerm, string? language = null)
2023-06-09 11:06:18 +02:00
{
Chapter[] availableChapters = this.GetChapters(publication, language??"en");
Regex volumeRegex = new ("((v(ol)*(olume)*)+ *([0-9]+(-[0-9]+)?){1})", RegexOptions.IgnoreCase);
Regex chapterRegex = new ("((c(h)*(hapter)*)+ *([0-9]+(-[0-9]+)?){1})", RegexOptions.IgnoreCase);
Regex singleResultRegex = new("([0-9]+)", RegexOptions.IgnoreCase);
Regex rangeResultRegex = new("([0-9]+(-[0-9]+))", RegexOptions.IgnoreCase);
Regex allRegex = new("a(ll)?", RegexOptions.IgnoreCase);
2023-06-09 11:06:18 +02:00
if (volumeRegex.IsMatch(searchTerm) && chapterRegex.IsMatch(searchTerm))
{
string volume = singleResultRegex.Match(volumeRegex.Match(searchTerm).Value).Value;
string chapter = singleResultRegex.Match(chapterRegex.Match(searchTerm).Value).Value;
return availableChapters.Where(aCh => aCh.volumeNumber is not null && aCh.chapterNumber is not null &&
aCh.volumeNumber.Equals(volume, StringComparison.InvariantCultureIgnoreCase) &&
aCh.chapterNumber.Equals(chapter, StringComparison.InvariantCultureIgnoreCase))
.ToArray();
}
else if (volumeRegex.IsMatch(searchTerm))
{
string volume = volumeRegex.Match(searchTerm).Value;
if (rangeResultRegex.IsMatch(volume))
{
string range = rangeResultRegex.Match(volume).Value;
int start = Convert.ToInt32(range.Split('-')[0]);
int end = Convert.ToInt32(range.Split('-')[1]);
return availableChapters.Where(aCh => aCh.volumeNumber is not null &&
Convert.ToInt32(aCh.volumeNumber) >= start &&
Convert.ToInt32(aCh.volumeNumber) <= end).ToArray();
}
2023-06-11 19:17:03 +02:00
else if (singleResultRegex.IsMatch(volume))
{
string volumeNumber = singleResultRegex.Match(volume).Value;
2023-06-09 11:06:18 +02:00
return availableChapters.Where(aCh =>
aCh.volumeNumber is not null &&
2023-06-11 19:17:03 +02:00
aCh.volumeNumber.Equals(volumeNumber, StringComparison.InvariantCultureIgnoreCase)).ToArray();
}
2023-06-09 11:06:18 +02:00
}
else if (chapterRegex.IsMatch(searchTerm))
{
string chapter = volumeRegex.Match(searchTerm).Value;
if (rangeResultRegex.IsMatch(chapter))
{
string range = rangeResultRegex.Match(chapter).Value;
int start = Convert.ToInt32(range.Split('-')[0]);
int end = Convert.ToInt32(range.Split('-')[1]);
return availableChapters.Where(aCh => aCh.chapterNumber is not null &&
Convert.ToInt32(aCh.chapterNumber) >= start &&
Convert.ToInt32(aCh.chapterNumber) <= end).ToArray();
}
2023-06-11 19:17:03 +02:00
else if (singleResultRegex.IsMatch(chapter))
{
string chapterNumber = singleResultRegex.Match(chapter).Value;
2023-06-09 11:06:18 +02:00
return availableChapters.Where(aCh =>
aCh.chapterNumber is not null &&
2023-06-11 19:17:03 +02:00
aCh.chapterNumber.Equals(chapterNumber, StringComparison.InvariantCultureIgnoreCase)).ToArray();
}
2023-06-09 11:06:18 +02:00
}
else
{
if (rangeResultRegex.IsMatch(searchTerm))
{
int start = Convert.ToInt32(searchTerm.Split('-')[0]);
int end = Convert.ToInt32(searchTerm.Split('-')[1]);
return availableChapters[start..(end + 1)];
}
else if(singleResultRegex.IsMatch(searchTerm))
return new [] { availableChapters[Convert.ToInt32(searchTerm)] };
else if (allRegex.IsMatch(searchTerm))
return availableChapters;
2023-06-09 11:06:18 +02:00
}
return Array.Empty<Chapter>();
}
/// <summary>
/// Retrieves the Chapter (+Images) from the website.
2023-05-31 20:39:23 +02:00
/// Should later call DownloadChapterImages to retrieve the individual Images of the Chapter and create .cbz archive.
/// </summary>
/// <param name="publication">Publication that contains Chapter</param>
/// <param name="chapter">Chapter with Images to retrieve</param>
/// <param name="parentTask">Will be used for progress-tracking</param>
/// <param name="cancellationToken"></param>
public abstract HttpStatusCode DownloadChapter(Publication publication, Chapter chapter, DownloadChapterTask parentTask, CancellationToken? cancellationToken = null);
2023-05-26 14:07:11 +02:00
/// <summary>
/// Copies the already downloaded cover from cache to downloadLocation
/// </summary>
/// <param name="publication">Publication to retrieve Cover for</param>
2023-05-26 14:07:11 +02:00
/// <param name="settings">TrangaSettings</param>
public void CopyCoverFromCacheToDownloadLocation(Publication publication, TrangaSettings settings)
{
2023-06-11 19:05:08 +02:00
logger?.WriteLine(this.GetType().ToString(), $"Cloning cover {publication.sortName} -> {publication.internalId}");
//Check if Publication already has a Folder and cover
string publicationFolder = publication.CreatePublicationFolder(settings.downloadLocation);
DirectoryInfo dirInfo = new (publicationFolder);
2023-06-11 19:05:08 +02:00
if (dirInfo.EnumerateFiles().Any(info => info.Name.Contains("cover", StringComparison.InvariantCultureIgnoreCase)))
{
logger?.WriteLine(this.GetType().ToString(), $"Cover exists {publication.sortName}");
return;
}
string fileInCache = Path.Join(settings.coverImageCache, publication.coverFileNameInCache);
string newFilePath = Path.Join(publicationFolder, $"cover.{Path.GetFileName(fileInCache).Split('.')[^1]}" );
logger?.WriteLine(this.GetType().ToString(), $"Cloning cover {fileInCache} -> {newFilePath}");
File.Copy(fileInCache, newFilePath, true);
2023-06-01 22:32:11 +02:00
if(RuntimeInformation.IsOSPlatform(OSPlatform.Linux))
File.SetUnixFileMode(newFilePath, GroupRead | GroupWrite | OtherRead | OtherWrite | UserRead | UserWrite);
}
2023-05-20 00:19:04 +02:00
2023-05-20 15:05:41 +02:00
/// <summary>
/// Creates a string containing XML of publication and chapter.
/// See ComicInfo.xml
/// </summary>
/// <returns>XML-string</returns>
internal static string GetComicInfoXmlString(Publication publication, Chapter chapter, Logger? logger)
2023-05-20 00:19:04 +02:00
{
2023-05-26 15:09:26 +02:00
logger?.WriteLine("Connector", $"Creating ComicInfo.Xml for {publication.sortName} {publication.internalId} {chapter.volumeNumber}-{chapter.chapterNumber}");
2023-05-20 00:19:04 +02:00
XElement comicInfo = new XElement("ComicInfo",
new XElement("Tags", string.Join(',',publication.tags)),
new XElement("LanguageISO", publication.originalLanguage),
new XElement("Title", chapter.name),
2023-06-10 14:05:23 +02:00
new XElement("Writer", string.Join(',', publication.authors)),
new XElement("Volume", chapter.volumeNumber),
2023-06-01 22:06:10 +02:00
new XElement("Number", chapter.chapterNumber)
2023-05-20 00:19:04 +02:00
);
return comicInfo.ToString();
}
2023-05-20 15:05:41 +02:00
/// <summary>
/// Checks if a chapter-archive is already present
/// </summary>
/// <returns>true if chapter is present</returns>
public bool CheckChapterIsDownloaded(Publication publication, Chapter chapter)
{
2023-06-03 22:25:24 +02:00
string newFilePath = GetArchiveFilePath(publication, chapter);
if (!Directory.Exists(Path.Join(settings.downloadLocation, publication.folderName)))
return false;
FileInfo[] archives = new DirectoryInfo(Path.Join(settings.downloadLocation, publication.folderName)).GetFiles();
Regex infoRex = new(@"(Vol.[0-9]*)?Ch.[0-9]+");
Regex chapterInfoRex = new(@"Ch.[0-9]+");
Regex chapterRex = new(@"[0-9]+");
2023-06-24 20:46:35 +02:00
if (File.Exists(newFilePath))
return true;
if (archives.FirstOrDefault(archive => chapterRex.Match(chapterInfoRex.Match(infoRex.Match(archive.Name).Value).Value).Value == chapter.chapterNumber) is { } path)
2023-06-24 20:46:35 +02:00
{
logger?.WriteLine(this.GetType().ToString(), "Move existing Chapter to new name.");
File.Move(path.FullName, newFilePath);
return true;
}
return false;
}
2023-05-20 15:05:41 +02:00
/// <summary>
/// Creates full file path of chapter-archive
/// </summary>
/// <returns>Filepath</returns>
protected string GetArchiveFilePath(Publication publication, Chapter chapter)
{
return Path.Join(settings.downloadLocation, publication.folderName, $"{publication.folderName} - {chapter.fileName}.cbz");
}
2023-05-22 18:15:24 +02:00
2023-05-19 20:22:13 +02:00
/// <summary>
/// Downloads Image from URL and saves it to the given path(incl. fileName)
/// </summary>
/// <param name="imageUrl"></param>
/// <param name="fullPath"></param>
2023-05-31 21:18:41 +02:00
/// <param name="requestType">RequestType for Rate-Limit</param>
/// <param name="referrer">referrer used in html request header</param>
private HttpStatusCode DownloadImage(string imageUrl, string fullPath, byte requestType, string? referrer = null)
{
2023-06-01 13:13:53 +02:00
DownloadClient.RequestResult requestResult = downloadClient.MakeRequest(imageUrl, requestType, referrer);
if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300 || requestResult.result == Stream.Null)
return requestResult.statusCode;
byte[] buffer = new byte[requestResult.result.Length];
requestResult.result.ReadExactly(buffer, 0, buffer.Length);
File.WriteAllBytes(fullPath, buffer);
return requestResult.statusCode;
}
2023-05-22 18:15:24 +02:00
2023-05-19 20:22:13 +02:00
/// <summary>
/// Downloads all Images from URLs, Compresses to zip(cbz) and saves.
/// </summary>
/// <param name="imageUrls">List of URLs to download Images from</param>
/// <param name="saveArchiveFilePath">Full path to save archive to (without file ending .cbz)</param>
/// <param name="parentTask">Used for progress tracking</param>
2023-05-20 01:06:12 +02:00
/// <param name="comicInfoPath">Path of the generate Chapter ComicInfo.xml, if it was generated</param>
2023-05-22 18:15:24 +02:00
/// <param name="requestType">RequestType for RateLimits</param>
/// <param name="referrer">Used in http request header</param>
/// <param name="cancellationToken"></param>
protected HttpStatusCode DownloadChapterImages(string[] imageUrls, string saveArchiveFilePath, byte requestType, DownloadChapterTask parentTask, string? comicInfoPath = null, string? referrer = null, CancellationToken? cancellationToken = null)
{
if (cancellationToken?.IsCancellationRequested ?? false)
return HttpStatusCode.RequestTimeout;
2023-05-26 15:09:26 +02:00
logger?.WriteLine("Connector", $"Downloading Images for {saveArchiveFilePath}");
2023-05-19 20:22:13 +02:00
//Check if Publication Directory already exists
2023-05-22 17:09:47 +02:00
string directoryPath = Path.GetDirectoryName(saveArchiveFilePath)!;
if (!Directory.Exists(directoryPath))
Directory.CreateDirectory(directoryPath);
2023-05-22 17:09:47 +02:00
if (File.Exists(saveArchiveFilePath)) //Don't download twice.
return HttpStatusCode.OK;
2023-05-19 20:22:13 +02:00
//Create a temporary folder to store images
string tempFolder = Directory.CreateTempSubdirectory().FullName;
int chapter = 0;
2023-05-19 20:22:13 +02:00
//Download all Images to temporary Folder
foreach (string imageUrl in imageUrls)
{
string[] split = imageUrl.Split('.');
string extension = split[^1];
logger?.WriteLine("Connector", $"Downloading Image {chapter + 1:000}/{imageUrls.Length:000} {parentTask.publication.sortName} {parentTask.publication.internalId} Vol.{parentTask.chapter.volumeNumber} Ch.{parentTask.chapter.chapterNumber} {parentTask.progress:P2}");
HttpStatusCode status = DownloadImage(imageUrl, Path.Join(tempFolder, $"{chapter++}.{extension}"), requestType, referrer);
if ((int)status < 200 || (int)status >= 300)
return status;
parentTask.IncrementProgress(1.0 / imageUrls.Length);
if (cancellationToken?.IsCancellationRequested ?? false)
return HttpStatusCode.RequestTimeout;
}
2023-05-20 00:19:04 +02:00
if(comicInfoPath is not null)
File.Copy(comicInfoPath, Path.Join(tempFolder, "ComicInfo.xml"));
2023-05-26 15:09:26 +02:00
logger?.WriteLine("Connector", $"Creating archive {saveArchiveFilePath}");
2023-05-19 20:22:13 +02:00
//ZIP-it and ship-it
2023-05-22 17:09:47 +02:00
ZipFile.CreateFromDirectory(tempFolder, saveArchiveFilePath);
2023-06-01 18:28:58 +02:00
if(RuntimeInformation.IsOSPlatform(OSPlatform.Linux))
File.SetUnixFileMode(saveArchiveFilePath, GroupRead | GroupWrite | OtherRead | OtherWrite | UserRead | UserWrite);
Directory.Delete(tempFolder, true); //Cleanup
return HttpStatusCode.OK;
}
2023-06-01 13:13:53 +02:00
protected string SaveCoverImageToCache(string url, byte requestType)
{
string[] split = url.Split('/');
string filename = split[^1];
string saveImagePath = Path.Join(settings.coverImageCache, filename);
2023-06-01 13:13:53 +02:00
if (File.Exists(saveImagePath))
return filename;
DownloadClient.RequestResult coverResult = downloadClient.MakeRequest(url, requestType);
using MemoryStream ms = new();
coverResult.result.CopyTo(ms);
File.WriteAllBytes(saveImagePath, ms.ToArray());
logger?.WriteLine(this.GetType().ToString(), $"Saving image to {saveImagePath}");
return filename;
}
protected class DownloadClient
{
private static readonly HttpClient Client = new()
{
Timeout = TimeSpan.FromSeconds(60)
};
2023-05-22 18:15:24 +02:00
private readonly Dictionary<byte, DateTime> _lastExecutedRateLimit;
private readonly Dictionary<byte, TimeSpan> _rateLimit;
// ReSharper disable once InconsistentNaming
private readonly Logger? logger;
2023-05-22 18:15:24 +02:00
2023-05-19 20:22:13 +02:00
/// <summary>
/// Creates a httpClient
/// </summary>
2023-05-22 18:15:24 +02:00
/// <param name="rateLimitRequestsPerMinute">Rate limits for requests. byte is RequestType, int maximum requests per minute for RequestType</param>
2023-05-31 21:18:41 +02:00
/// <param name="logger"></param>
public DownloadClient(Dictionary<byte, int> rateLimitRequestsPerMinute, Logger? logger)
{
this.logger = logger;
2023-05-22 18:15:24 +02:00
_lastExecutedRateLimit = new();
_rateLimit = new();
2023-05-22 18:15:24 +02:00
foreach(KeyValuePair<byte, int> limit in rateLimitRequestsPerMinute)
_rateLimit.Add(limit.Key, TimeSpan.FromMinutes(1).Divide(limit.Value));
}
2023-05-22 18:15:24 +02:00
2023-05-19 20:22:13 +02:00
/// <summary>
/// Request Webpage
/// </summary>
/// <param name="url"></param>
2023-05-22 18:15:24 +02:00
/// <param name="requestType">For RateLimits: Same Endpoints use same type</param>
/// <param name="referrer">Used in http request header</param>
2023-05-19 20:22:13 +02:00
/// <returns>RequestResult with StatusCode and Stream of received data</returns>
2023-06-01 13:13:53 +02:00
public RequestResult MakeRequest(string url, byte requestType, string? referrer = null)
{
if (_rateLimit.TryGetValue(requestType, out TimeSpan value))
2023-05-22 18:15:24 +02:00
_lastExecutedRateLimit.TryAdd(requestType, DateTime.Now.Subtract(value));
else
{
logger?.WriteLine(this.GetType().ToString(), "RequestType not configured for rate-limit.");
return new RequestResult(HttpStatusCode.NotAcceptable, Stream.Null);
}
TimeSpan rateLimitTimeout = _rateLimit[requestType]
.Subtract(DateTime.Now.Subtract(_lastExecutedRateLimit[requestType]));
2023-05-22 18:15:24 +02:00
2023-05-22 21:41:11 +02:00
if(rateLimitTimeout > TimeSpan.Zero)
Thread.Sleep(rateLimitTimeout);
HttpResponseMessage? response = null;
while (response is null)
{
try
{
HttpRequestMessage requestMessage = new(HttpMethod.Get, url);
2023-06-01 13:13:53 +02:00
if(referrer is not null)
requestMessage.Headers.Referrer = new Uri(referrer);
_lastExecutedRateLimit[requestType] = DateTime.Now;
response = Client.Send(requestMessage);
}
catch (HttpRequestException e)
{
2023-05-22 21:44:52 +02:00
logger?.WriteLine(this.GetType().ToString(), e.Message);
2023-06-01 21:16:57 +02:00
logger?.WriteLine(this.GetType().ToString(), $"Waiting {_rateLimit[requestType] * 2}... Retrying.");
Thread.Sleep(_rateLimit[requestType] * 2);
}
}
if (!response.IsSuccessStatusCode)
{
logger?.WriteLine(this.GetType().ToString(), $"Request-Error {response.StatusCode}: {response.ReasonPhrase}");
return new RequestResult(response.StatusCode, Stream.Null);
}
return new RequestResult(response.StatusCode, response.Content.ReadAsStream());
}
public struct RequestResult
{
public HttpStatusCode statusCode { get; }
public Stream result { get; }
public RequestResult(HttpStatusCode statusCode, Stream result)
{
this.statusCode = statusCode;
this.result = result;
}
}
}
}