AZEN-SGG/Kavita
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Simplify Folder Watcher (#1484)

Browse source 
* Refactored Library Watcher to use Hangfire under the hood.

* Support .kavitaignore at root level.

* Refactored a lot of the library watching code to process faster and handle when FileSystemWatcher runs out of internal buffer space. It's still not perfect, but good enough for basic use.

* Make folder watching as experimental and default it to off by default.

* Revert #1479

* Tweaked the messaging for OPDS to remove a note about download role.

Moved some code closer to where it's used.

* Cleaned up how the events widget reports

* Fixed a null issue when deleting series in the UI

* Cleaned up some debug code

* Added more information for when we skip a scan

* Cleaned up some logging messages in CoverGen tasks

* More log message tweaks

* Added some debug to help identify a rare issue

* Fixed a bug where save bookmarks as webp could get reset to false when saving other server settings

* Updated some documentation on library watcher.

* Make LibraryWatcher fire every 5 mins
This commit is contained in:
Joseph Milazzo 2022-08-28 15:20:46 -05:00 • committed by GitHub
parent b64ed6df8d
commit b07aaf1eb5
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
19 changed files with 187 additions and 259 deletions
Download patch file Download diff file Expand all files Collapse all files

170
API/Services/Tasks/Scanner/Parser/DefaultParser.cs Normal file
View file

@ -0,0 +1,170 @@
using System.IO;
using System.Linq;
using API.Entities.Enums;
using API.Services;
namespace API.Parser;
public interface IDefaultParser
{
ParserInfo Parse(string filePath, string rootPath, LibraryType type = LibraryType.Manga);
void ParseFromFallbackFolders(string filePath, string rootPath, LibraryType type, ref ParserInfo ret);
}
/// <summary>
/// This is an implementation of the Parser that is the basis for everything
/// </summary>
public class DefaultParser : IDefaultParser
{
private readonly IDirectoryService _directoryService;
public DefaultParser(IDirectoryService directoryService)
{
_directoryService = directoryService;
}
/// <summary>
/// Parses information out of a file path. Will fallback to using directory name if Series couldn't be parsed
/// from filename.
/// </summary>
/// <param name="filePath"></param>
/// <param name="rootPath">Root folder</param>
/// <param name="type">Defaults to Manga. Allows different Regex to be used for parsing.</param>
/// <returns><see cref="ParserInfo"/> or null if Series was empty</returns>
public ParserInfo Parse(string filePath, string rootPath, LibraryType type = LibraryType.Manga)
{
var fileName = _directoryService.FileSystem.Path.GetFileNameWithoutExtension(filePath);
ParserInfo ret;
if (Parser.IsEpub(filePath))
{
ret = new ParserInfo()
{
Chapters = Parser.ParseChapter(fileName) ?? Parser.ParseComicChapter(fileName),
Series = Parser.ParseSeries(fileName) ?? Parser.ParseComicSeries(fileName),
Volumes = Parser.ParseVolume(fileName) ?? Parser.ParseComicVolume(fileName),
Filename = Path.GetFileName(filePath),
Format = Parser.ParseFormat(filePath),
FullFilePath = filePath
};
}
else
{
ret = new ParserInfo()
{
Chapters = type == LibraryType.Comic ? Parser.ParseComicChapter(fileName) : Parser.ParseChapter(fileName),
Series = type == LibraryType.Comic ? Parser.ParseComicSeries(fileName) : Parser.ParseSeries(fileName),
Volumes = type == LibraryType.Comic ? Parser.ParseComicVolume(fileName) : Parser.ParseVolume(fileName),
Filename = Path.GetFileName(filePath),
Format = Parser.ParseFormat(filePath),
Title = Path.GetFileNameWithoutExtension(fileName),
FullFilePath = filePath
};
}
if (Parser.IsImage(filePath) && Parser.IsCoverImage(filePath)) return null;
if (Parser.IsImage(filePath))
{
// Reset Chapters, Volumes, and Series as images are not good to parse information out of. Better to use folders.
ret.Volumes = Parser.DefaultVolume;
ret.Chapters = Parser.DefaultChapter;
ret.Series = string.Empty;
}
if (ret.Series == string.Empty || Parser.IsImage(filePath))
{
// Try to parse information out of each folder all the way to rootPath
ParseFromFallbackFolders(filePath, rootPath, type, ref ret);
}
var edition = Parser.ParseEdition(fileName);
if (!string.IsNullOrEmpty(edition))
{
ret.Series = Parser.CleanTitle(ret.Series.Replace(edition, ""), type is LibraryType.Comic);
ret.Edition = edition;
}
var isSpecial = type == LibraryType.Comic ? Parser.ParseComicSpecial(fileName) : Parser.ParseMangaSpecial(fileName);
// We must ensure that we can only parse a special out. As some files will have v20 c171-180+Omake and that
// could cause a problem as Omake is a special term, but there is valid volume/chapter information.
if (ret.Chapters == Parser.DefaultChapter && ret.Volumes == Parser.DefaultVolume && !string.IsNullOrEmpty(isSpecial))
{
ret.IsSpecial = true;
ParseFromFallbackFolders(filePath, rootPath, type, ref ret); // NOTE: This can cause some complications, we should try to be a bit less aggressive to fallback to folder
}
// If we are a special with marker, we need to ensure we use the correct series name. we can do this by falling back to Folder name
if (Parser.HasSpecialMarker(fileName))
{
ret.IsSpecial = true;
ret.Chapters = Parser.DefaultChapter;
ret.Volumes = Parser.DefaultVolume;
ParseFromFallbackFolders(filePath, rootPath, type, ref ret);
}
if (string.IsNullOrEmpty(ret.Series))
{
ret.Series = Parser.CleanTitle(fileName, type is LibraryType.Comic);
}
// Pdfs may have .pdf in the series name, remove that
if (Parser.IsPdf(filePath) && ret.Series.ToLower().EndsWith(".pdf"))
{
ret.Series = ret.Series.Substring(0, ret.Series.Length - ".pdf".Length);
}
return ret.Series == string.Empty ? null : ret;
}
/// <summary>
/// Fills out <see cref="ParserInfo"/> by trying to parse volume, chapters, and series from folders
/// </summary>
/// <param name="filePath"></param>
/// <param name="rootPath"></param>
/// <param name="type"></param>
/// <param name="ret">Expects a non-null ParserInfo which this method will populate</param>
public void ParseFromFallbackFolders(string filePath, string rootPath, LibraryType type, ref ParserInfo ret)
{
var fallbackFolders = _directoryService.GetFoldersTillRoot(rootPath, filePath).ToList();
for (var i = 0; i < fallbackFolders.Count; i++)
{
var folder = fallbackFolders[i];
if (!string.IsNullOrEmpty(Parser.ParseMangaSpecial(folder))) continue;
var parsedVolume = type is LibraryType.Manga ? Parser.ParseVolume(folder) : Parser.ParseComicVolume(folder);
var parsedChapter = type is LibraryType.Manga ? Parser.ParseChapter(folder) : Parser.ParseComicChapter(folder);
if (!parsedVolume.Equals(Parser.DefaultVolume) || !parsedChapter.Equals(Parser.DefaultChapter))
{
if ((string.IsNullOrEmpty(ret.Volumes) || ret.Volumes.Equals(Parser.DefaultVolume)) && !parsedVolume.Equals(Parser.DefaultVolume))
{
ret.Volumes = parsedVolume;
}
if ((string.IsNullOrEmpty(ret.Chapters) || ret.Chapters.Equals(Parser.DefaultChapter)) && !parsedChapter.Equals(Parser.DefaultChapter))
{
ret.Chapters = parsedChapter;
}
}
// Generally users group in series folders. Let's try to parse series from the top folder
if (!folder.Equals(ret.Series) && i == fallbackFolders.Count - 1)
{
var series = Parser.ParseSeries(folder);
if (string.IsNullOrEmpty(series))
{
ret.Series = Parser.CleanTitle(folder, type is LibraryType.Comic);
break;
}
if (!string.IsNullOrEmpty(series) && (string.IsNullOrEmpty(ret.Series) || !folder.Contains(ret.Series)))
{
ret.Series = series;
break;
}
}
}
}
}

1089
API/Services/Tasks/Scanner/Parser/Parser.cs Normal file
View file

File diff suppressed because it is too large Load diff

100
API/Services/Tasks/Scanner/Parser/ParserInfo.cs Normal file
View file

@ -0,0 +1,100 @@
using API.Data.Metadata;
using API.Entities.Enums;
namespace API.Parser
{
/// <summary>
/// This represents all parsed information from a single file
/// </summary>
public class ParserInfo
{
/// <summary>
/// Represents the parsed chapters from a file. By default, will be 0 which means nothing could be parsed.
/// <remarks>The chapters can only be a single float or a range of float ie) 1-2. Mainly floats should be multiples of 0.5 representing specials</remarks>
/// </summary>
public string Chapters { get; set; } = "";
/// <summary>
/// Represents the parsed series from the file or folder
/// </summary>
public string Series { get; set; } = string.Empty;
/// <summary>
/// This can be filled in from ComicInfo.xml/Epub during scanning. Will update the SortName field on <see cref="Entities.Series"/>
/// </summary>
public string SeriesSort { get; set; } = string.Empty;
/// <summary>
/// This can be filled in from ComicInfo.xml/Epub during scanning. Will update the LocalizedName field on <see cref="Entities.Series"/>
/// </summary>
public string LocalizedSeries { get; set; } = string.Empty;
/// <summary>
/// Represents the parsed volumes from a file. By default, will be 0 which means that nothing could be parsed.
/// If Volumes is 0 and Chapters is 0, the file is a special. If Chapters is non-zero, then no volume could be parsed.
/// <example>Beastars Vol 3-4 will map to "3-4"</example>
/// <remarks>The volumes can only be a single int or a range of ints ie) 1-2. Float based volumes are not supported.</remarks>
/// </summary>
public string Volumes { get; set; } = "";
/// <summary>
/// Filename of the underlying file
/// <example>Beastars v01 (digital).cbz</example>
/// </summary>
public string Filename { get; init; } = "";
/// <summary>
/// Full filepath of the underlying file
/// <example>C:/Manga/Beastars v01 (digital).cbz</example>
/// </summary>
public string FullFilePath { get; set; } = "";
/// <summary>
/// <see cref="MangaFormat"/> that represents the type of the file
/// <remarks>Mainly used to show in the UI and so caching service knows how to cache for reading.</remarks>
/// </summary>
public MangaFormat Format { get; set; } = MangaFormat.Unknown;
/// <summary>
/// This can potentially story things like "Omnibus, Color, Full Contact Edition, Extra, Final, etc"
/// </summary>
/// <remarks>Not Used in Database</remarks>
public string Edition { get; set; } = "";
/// <summary>
/// If the file contains no volume/chapter information or contains Special Keywords <see cref="Parser.MangaSpecialRegex"/>
/// </summary>
public bool IsSpecial { get; set; }
/// <summary>
/// Used for specials or books, stores what the UI should show.
/// <remarks>Manga does not use this field</remarks>
/// </summary>
public string Title { get; set; } = string.Empty;
/// <summary>
/// If the ParserInfo has the IsSpecial tag or both volumes and chapters are default aka 0
/// </summary>
/// <returns></returns>
public bool IsSpecialInfo()
{
return (IsSpecial || (Volumes == "0" && Chapters == "0"));
}
/// <summary>
/// This will contain any EXTRA comicInfo information parsed from the epub or archive. If there is an archive with comicInfo.xml AND it contains
/// series, volume information, that will override what we parsed.
/// </summary>
public ComicInfo ComicInfo { get; set; }
/// <summary>
/// Merges non empty/null properties from info2 into this entity.
/// </summary>
/// <remarks>This does not merge ComicInfo as they should always be the same</remarks>
/// <param name="info2"></param>
public void Merge(ParserInfo info2)
{
if (info2 == null) return;
Chapters = string.IsNullOrEmpty(Chapters) || Chapters == "0" ? info2.Chapters: Chapters;
Volumes = string.IsNullOrEmpty(Volumes) || Volumes == "0" ? info2.Volumes : Volumes;
Edition = string.IsNullOrEmpty(Edition) ? info2.Edition : Edition;
Title = string.IsNullOrEmpty(Title) ? info2.Title : Title;
Series = string.IsNullOrEmpty(Series) ? info2.Series : Series;
IsSpecial = IsSpecial || info2.IsSpecial;
}
}
}