Hooked up the API layer to be able to extract images from PDF again for Tachiyomi explicitly (#1686)

API/Controllers/ReaderController.cs

@@ -87,16 +87,17 @@ public class ReaderController : BaseApiController
     /// Returns an image for a given chapter. Will perform bounding checks
     /// </summary>
     /// <remarks>This will cache the chapter images for reading</remarks>
-    /// <param name="chapterId"></param>
+    /// <param name="chapterId">Chapter Id</param>
-    /// <param name="page"></param>
+    /// <param name="page">Page in question</param>
+    /// <param name="extractPdf">Should Kavita extract pdf into images. Defaults to false.</param>
     /// <returns></returns>
     [HttpGet("image")]
     [ResponseCache(CacheProfileName = ResponseCacheProfiles.Hour)]
     [AllowAnonymous]
-    public async Task<ActionResult> GetImage(int chapterId, int page)
+    public async Task<ActionResult> GetImage(int chapterId, int page, bool extractPdf = false)
     {
         if (page < 0) page = 0;
-        var chapter = await _cacheService.Ensure(chapterId);
+        var chapter = await _cacheService.Ensure(chapterId, extractPdf);
         if (chapter == null) return BadRequest("There was an issue finding image file for reading");
 
         try
@@ -155,12 +156,13 @@ public class ReaderController : BaseApiController
     /// Returns various information about a Chapter. Side effect: This will cache the chapter images for reading.
     /// </summary>
     /// <param name="chapterId"></param>
+    /// <param name="extractPdf">Should Kavita extract pdf into images. Defaults to false.</param>
     /// <returns></returns>
     [HttpGet("chapter-info")]
-    public async Task<ActionResult<ChapterInfoDto>> GetChapterInfo(int chapterId)
+    public async Task<ActionResult<ChapterInfoDto>> GetChapterInfo(int chapterId, bool extractPdf = false)
     {
         if (chapterId <= 0) return null; // This can happen occasionally from UI, we should just ignore
-        var chapter = await _cacheService.Ensure(chapterId);
+        var chapter = await _cacheService.Ensure(chapterId, extractPdf);
         if (chapter == null) return BadRequest("Could not find Chapter");
 
         var dto = await _unitOfWork.ChapterRepository.GetChapterInfoDtoAsync(chapterId);

API/Services/BookService.cs

@@ -49,9 +49,9 @@ public interface IBookService
     /// <summary>
     /// Extracts a PDF file's pages as images to an target directory
     /// </summary>
+    /// <remarks>This method relies on Docnet which has explict patches from Kavita for ARM support. This should only be used with Tachiyomi</remarks>
     /// <param name="fileFilePath"></param>
     /// <param name="targetDirectory">Where the files will be extracted to. If doesn't exist, will be created.</param>
-    [Obsolete("This method of reading is no longer supported. Please use native pdf reader")]
     void ExtractPdfImages(string fileFilePath, string targetDirectory);
 
     Task<string> ScopePage(HtmlDocument doc, EpubBookRef book, string apiBase, HtmlNode body, Dictionary<string, int> mappings, int page);

API/Services/CacheService.cs

@@ -20,8 +20,9 @@ public interface ICacheService
     /// cache operations (except cleanup).
     /// </summary>
     /// <param name="chapterId"></param>
+    /// <param name="extractPdfToImages">Extracts a PDF into images for a different reading experience</param>
     /// <returns>Chapter for the passed chapterId. Side-effect from ensuring cache.</returns>
-    Task<Chapter> Ensure(int chapterId);
+    Task<Chapter> Ensure(int chapterId, bool extractPdfToImages = false);
     /// <summary>
     /// Clears cache directory of all volumes. This can be invoked from deleting a library or a series.
     /// </summary>
@@ -31,7 +32,7 @@ public interface ICacheService
     string GetCachedPagePath(Chapter chapter, int page);
     string GetCachedBookmarkPagePath(int seriesId, int page);
     string GetCachedFile(Chapter chapter);
-    public void ExtractChapterFiles(string extractPath, IReadOnlyList<MangaFile> files);
+    public void ExtractChapterFiles(string extractPath, IReadOnlyList<MangaFile> files, bool extractPdfImages = false);
     Task<int> CacheBookmarkForSeries(int userId, int seriesId);
     void CleanupBookmarkCache(int seriesId);
 }
@@ -95,7 +96,7 @@ public class CacheService : ICacheService
     /// </summary>
     /// <param name="chapterId"></param>
     /// <returns>This will always return the Chapter for the chapterId</returns>
-    public async Task<Chapter> Ensure(int chapterId)
+    public async Task<Chapter> Ensure(int chapterId, bool extractPdfToImages = false)
     {
         _directoryService.ExistOrCreate(_directoryService.CacheDirectory);
         var chapter = await _unitOfWork.ChapterRepository.GetChapterAsync(chapterId);
@@ -103,7 +104,7 @@ public class CacheService : ICacheService
 
         if (_directoryService.Exists(extractPath)) return chapter;
         var files = chapter?.Files.ToList();
-        ExtractChapterFiles(extractPath, files);
+        ExtractChapterFiles(extractPath, files, extractPdfToImages);
 
         return  chapter;
     }
@@ -114,8 +115,9 @@ public class CacheService : ICacheService
     /// </summary>
     /// <param name="extractPath"></param>
     /// <param name="files"></param>
+    /// <param name="extractPdfImages">Defaults to false, if true, will extract the images from the PDF renderer and not move the pdf file</param>
     /// <returns></returns>
-    public void ExtractChapterFiles(string extractPath, IReadOnlyList<MangaFile> files)
+    public void ExtractChapterFiles(string extractPath, IReadOnlyList<MangaFile> files, bool extractPdfImages = false)
     {
         var removeNonImages = true;
         var fileCount = files.Count;
@@ -143,12 +145,17 @@ public class CacheService : ICacheService
                 case MangaFormat.Epub:
                 case MangaFormat.Pdf:
                 {
-                    removeNonImages = false;
                     if (!_directoryService.FileSystem.File.Exists(files[0].FilePath))
                     {
                         _logger.LogError("{File} does not exist on disk", files[0].FilePath);
                         throw new KavitaException($"{files[0].FilePath} does not exist on disk");
                     }
+                    if (extractPdfImages)
+                    {
+                        _readingItemService.Extract(file.FilePath, Path.Join(extractPath, extraPath), file.Format);
+                        break;
+                    }
+                    removeNonImages = false;
 
                     _directoryService.ExistOrCreate(extractPath);
                     _directoryService.CopyFileToDirectory(files[0].FilePath, extractPath);

API/Services/ReadingItemService.cs

@@ -199,6 +199,8 @@ public class ReadingItemService : IReadingItemService
                 _imageService.ExtractImages(fileFilePath, targetDirectory, imageCount);
                 break;
             case MangaFormat.Pdf:
+                _bookService.ExtractPdfImages(fileFilePath, targetDirectory);
+                break;
             case MangaFormat.Unknown:
             case MangaFormat.Epub:
                 break;

openapi.json

@@ -7,7 +7,7 @@
       "name": "GPL-3.0",
       "url": "https://github.com/Kareadita/Kavita/blob/develop/LICENSE"
     },
-    "version": "0.6.1.7"
+    "version": "0.6.1.8"
   },
   "servers": [
     {
@@ -3411,7 +3411,7 @@
           {
             "name": "chapterId",
             "in": "query",
-            "description": "",
+            "description": "Chapter Id",
             "schema": {
               "type": "integer",
               "format": "int32"
@@ -3420,11 +3420,20 @@
           {
             "name": "page",
             "in": "query",
-            "description": "",
+            "description": "Page in question",
             "schema": {
               "type": "integer",
               "format": "int32"
             }
+          },
+          {
+            "name": "extractPdf",
+            "in": "query",
+            "description": "Should Kavita extract pdf into images. Defaults to false.",
+            "schema": {
+              "type": "boolean",
+              "default": false
+            }
           }
         ],
         "responses": {
@@ -3491,6 +3500,15 @@
               "type": "integer",
               "format": "int32"
             }
+          },
+          {
+            "name": "extractPdf",
+            "in": "query",
+            "description": "Should Kavita extract pdf into images. Defaults to false.",
+            "schema": {
+              "type": "boolean",
+              "default": false
+            }
           }
         ],
         "responses": {