Merge pull request #3 from OUCC/feat/#1-2

#1-2 ScraperSelectorを追加
OUCC · Mar 2, 2024 · d2b3d40 · d2b3d40
2 parents c9dbc54 + 7320070
commit d2b3d40
Show file tree

Hide file tree

Showing 12 changed files with 178 additions and 128 deletions.
diff --git a/Epub/KoeBook.Epub/Contracts/Services/IFileExtensionService.cs b/Epub/KoeBook.Epub/Contracts/Services/IFileExtensionService.cs
@@ -0,0 +1,6 @@
+namespace KoeBook.Epub.Contracts.Services;
+
+public interface IFileExtensionService
+{
+    public string GetImagesMediaType(string fileName);
+}
diff --git a/Epub/KoeBook.Epub/Contracts/Services/IScraperSelectorService.cs b/Epub/KoeBook.Epub/Contracts/Services/IScraperSelectorService.cs
@@ -0,0 +1,11 @@
+using KoeBook.Epub.Models;
+
+namespace KoeBook.Epub.Contracts.Services;
+
+/// <summary>
+/// スクレイピングを行い、EpubDocumentを作成します。
+/// </summary>
+public interface IScraperSelectorService
+{
+    public ValueTask<EpubDocument> ScrapingAsync(string url, string coverFillePath, string tempDirectory, Guid id, CancellationToken ct);
+}
diff --git a/Epub/KoeBook.Epub/Contracts/Services/IScrapingService.cs b/Epub/KoeBook.Epub/Contracts/Services/IScrapingService.cs
@@ -2,7 +2,7 @@
 
 namespace KoeBook.Epub.Contracts.Services;
 
-public interface IScrapingService
+public interface IScrapingService : IScraperSelectorService
 {
-    public Task<EpubDocument> ScrapingAsync(string url, string coverFillePath, string imageDirectory, Guid id, CancellationToken ct);
+    public bool IsMatchSite(Uri url);
 }
diff --git a/Epub/KoeBook.Epub/Models/EpubDocument.cs b/Epub/KoeBook.Epub/Models/EpubDocument.cs
@@ -37,6 +37,33 @@ public class EpubDocument(string title, string author, string coverFilePath, Gui
     ];
     public List<Chapter> Chapters { get; set; } = [];
 
+    internal void EnsureChapter()
+    {
+        if (Chapters.Count == 0)
+            Chapters.Add(new Chapter() { Title = null });
+    }
+
+    internal void EnsureSection(int chapterIndex)
+    {
+        EnsureChapter();
+
+        if (Chapters[chapterIndex].Sections.Count == 0)
+        {
+            if (Chapters[chapterIndex].Title != null)
+                Chapters[chapterIndex].Sections.Add(new Section(Chapters[chapterIndex].Title!));
+            else
+                Chapters[chapterIndex].Sections.Add(new Section(Title));
+        }
+    }
+
+    internal void EnsureParagraph(int chapterIndex, int sectionIndex)
+    {
+        EnsureSection(chapterIndex);
+
+        if (Chapters[chapterIndex].Sections[sectionIndex].Elements.Count == 0)
+            Chapters[chapterIndex].Sections[sectionIndex].Elements.Add(new Paragraph());
+    }
+
     public string CreateNavXhtml()
     {
         var builder = new StringBuilder($"""

diff --git a/Epub/KoeBook.Epub/ScrapingHelper.cs b/Epub/KoeBook.Epub/ScrapingHelper.cs
diff --git a/Epub/KoeBook.Epub/Services/FileExtensionService.cs b/Epub/KoeBook.Epub/Services/FileExtensionService.cs
@@ -0,0 +1,19 @@
+using KoeBook.Epub.Contracts.Services;
+
+namespace KoeBook.Epub.Services;
+
+public class FileExtensionService : IFileExtensionService
+{
+    public string GetImagesMediaType(string fileName)
+    {
+        return Path.GetExtension(fileName) switch
+        {
+            ".gif" => "image/gif",
+            ".jpg" or ".jpeg" => "image/jpeg",
+            ".png" => "image/png",
+            ".svg" => "image/svg+xml",
+            ".webp" => "image/webp",
+            _ => string.Empty,
+        };
+    }
+}
diff --git a/Epub/KoeBook.Epub/Services/ScraperSelectorService.cs b/Epub/KoeBook.Epub/Services/ScraperSelectorService.cs
@@ -0,0 +1,23 @@
+using System.Collections.Immutable;
+using KoeBook.Epub.Contracts.Services;
+using KoeBook.Epub.Models;
+
+namespace KoeBook.Epub.Services;
+
+public class ScraperSelectorService(IEnumerable<IScrapingService> scrapingServices) : IScraperSelectorService
+{
+    private readonly ImmutableArray<IScrapingService> _scrapingServices = scrapingServices.ToImmutableArray();
+
+    public async ValueTask<EpubDocument> ScrapingAsync(string url, string coverFillePath, string tempDirectory, Guid id, CancellationToken ct)
+    {
+        var uri = new Uri(url);
+
+        foreach (var service in _scrapingServices)
+        {
+            if (service.IsMatchSite(uri))
+                return await service.ScrapingAsync(url, coverFillePath, tempDirectory, id, ct);
+        }
+
+        throw new ArgumentException("対応するURLではありません");
+    }
+}
diff --git a/Epub/KoeBook.Epub/Services/ScrapingAozora.cs → ...ok.Epub/Services/ScrapingAozoraService.cs b/Epub/KoeBook.Epub/Services/ScrapingAozora.cs → ...ok.Epub/Services/ScrapingAozoraService.cs
@@ -4,21 +4,25 @@
 using AngleSharp.Io;
 using KoeBook.Epub.Contracts.Services;
 using KoeBook.Epub.Models;
-using static KoeBook.Epub.ScrapingHelper;
+using static KoeBook.Epub.Utility.ScrapingHelper;
 
 
 namespace KoeBook.Epub.Services
 {
-    public partial class ScrapingAozora : IScrapingService
+    public partial class ScrapingAozoraService : IScrapingService
     {
-        private int chapterNum;
-        private int sectionNum;
-        private bool chapterExist = false;
-        private bool sectionExist = false;
-
+        public bool IsMatchSite(Uri uri)
+        {
+            return uri.Host == "www.aozora.gr.jp";
+        }
 
-        public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath, string imageDirectory, Guid id, CancellationToken ct)
+        public async ValueTask<EpubDocument> ScrapingAsync(string url, string coverFilePath, string imageDirectory, Guid id, CancellationToken ct)
         {
+            var chapterNum = 0;
+            var sectionNum = 0;
+            var chapterExist = false;
+            var sectionExist = false;
+
             var config = Configuration.Default.WithDefaultLoader();
             using var context = BrowsingContext.New(config);
             var doc = await context.OpenAsync(url, ct).ConfigureAwait(false);
@@ -61,7 +65,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                         }
                         if ((MidashiId - previousMidashiId) == 10)
                         {
-                            checkChapter(document);
+                            document.EnsureChapter();
                             document.Chapters[^1].Sections.Add(new Section(TextProcess(midashi)));
                             sectionExist = true;
                         }
@@ -97,7 +101,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                 {
                     if (previous == true)
                     {
-                        checkSection(document, chapterNum);
+                        document.EnsureSection(chapterNum);
                         document.Chapters[chapterNum].Sections[sectionNum].Elements.Add(new Paragraph());
                     }
                 }
@@ -156,12 +160,12 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                             {
                                 if (sectionExist)
                                 {
-                                    checkChapter(document);
+                                    document.EnsureChapter();
                                     document.Chapters[^1].Sections.Insert(0, new Section("___"));
                                 }
                                 sectionNum++;
                             }
-                            checkParagraph(document, chapterNum, sectionNum);
+                            document.EnsureParagraph(chapterNum, sectionNum);
                             if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                             {
                                 paragraph.Text += TextProcess(midashi);
@@ -183,7 +187,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                         if (element.ClassName == "caption")
                         {
                             // https://www.aozora.gr.jp/annotation/graphics.html#:~:text=%3Cdiv%20class%3D%22caption%22%3E を処理するための部分
-                            checkParagraph(document, chapterNum, sectionNum);
+                            document.EnsureParagraph(chapterNum, sectionNum);
                             if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                             {
                                 var split = SplitBrace(TextProcess(element));
@@ -216,12 +220,12 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                             {
                                 if (sectionExist)
                                 {
-                                    checkChapter(document);
+                                    document.EnsureChapter();
                                     document.Chapters[^1].Sections.Insert(0, new Section("___"));
                                 }
                                 sectionNum++;
                             }
-                            checkParagraph(document, chapterNum, sectionNum);
+                            document.EnsureParagraph(chapterNum, sectionNum);
                             if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                             {
                                 foreach (var splitText in SplitBrace(TextProcess(element)))
@@ -253,7 +257,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                         {
                             if (sectionExist)
                             {
-                                checkChapter(document);
+                                document.EnsureChapter();
                                 document.Chapters[^1].Sections.Insert(0, new Section("___"));
                             }
                             sectionNum++;
@@ -274,7 +278,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                                     await response.Content.CopyToAsync(ms, ct).ConfigureAwait(false);
                                     var filePass = System.IO.Path.Combine(imageDirectory, FileUrlToFileName().Replace(img.Source, "$1"));
                                     File.WriteAllBytes(filePass, ms.ToArray());
-                                    checkSection(document, chapterNum);
+                                    document.EnsureSection(chapterNum);
                                     if (document.Chapters[chapterNum].Sections[sectionNum].Elements.Count > 1)
                                     {
                                         document.Chapters[chapterNum].Sections[sectionNum].Elements.Insert(document.Chapters[chapterNum].Sections[sectionNum].Elements.Count - 1, new Picture(filePass));
@@ -283,7 +287,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                             }
                             if (img.AlternativeText != null)
                             {
-                                checkParagraph(document, chapterNum, sectionNum);
+                                document.EnsureParagraph(chapterNum, sectionNum);
                                 if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                                 {
                                     paragraph.Text += TextReplace(img.AlternativeText);
@@ -332,7 +336,7 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                             case "［＃ページの左右中央］":
                                 break;
                             default:
-                                checkParagraph(document, chapterNum, sectionNum);
+                                document.EnsureParagraph(chapterNum, sectionNum);
                                 if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                                 {
                                     foreach (var splitText in SplitBrace(TextProcess(element)))
@@ -362,12 +366,12 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                         {
                             if (sectionExist)
                             {
-                                checkChapter(document);
+                                document.EnsureChapter();
                                 document.Chapters[^1].Sections.Insert(0, new Section("___"));
                             }
                             sectionNum++;
                         }
-                        checkParagraph(document, chapterNum, sectionNum);
+                        document.EnsureParagraph(chapterNum, sectionNum);
                         if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                         {
                             var split = SplitBrace(TextProcess(element));
@@ -402,12 +406,12 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                     {
                         if (sectionExist)
                         {
-                            checkChapter(document);
+                            document.EnsureChapter();
                             document.Chapters[^1].Sections.Insert(0, new Section("___"));
                         }
                         sectionNum++;
                     }
-                    checkParagraph(document, chapterNum, sectionNum);
+                    document.EnsureParagraph(chapterNum, sectionNum);
                     if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                     {
                         paragraph.Text += TextProcess(element);
@@ -450,12 +454,12 @@ public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath,
                             {
                                 if (sectionExist)
                                 {
-                                    checkChapter(document);
+                                    document.EnsureChapter();
                                     document.Chapters[^1].Sections.Insert(0, new Section("___"));
                                 }
                                 sectionNum++;
                             }
-                            checkParagraph(document, chapterNum, sectionNum);
+                            document.EnsureParagraph(chapterNum, sectionNum);
                             if ((document.Chapters[chapterNum].Sections[sectionNum].Elements[^1] is Paragraph paragraph))
                             {
                                 var split = SplitBrace(TextReplace(nextNode.Text()));

diff --git a/Epub/KoeBook.Epub/Services/ScrapingNarou.cs → ...Book.Epub/Services/ScrapingNaroService.cs b/Epub/KoeBook.Epub/Services/ScrapingNarou.cs → ...Book.Epub/Services/ScrapingNaroService.cs
@@ -5,20 +5,20 @@
 using AngleSharp.Io;
 using KoeBook.Epub.Contracts.Services;
 using KoeBook.Epub.Models;
-using static KoeBook.Epub.ScrapingHelper;
+using static KoeBook.Epub.Utility.ScrapingHelper;
 
 namespace KoeBook.Epub.Services
 {
-    public partial class ScrapingNarouService : IScrapingService
+    public partial class ScrapingNaroService(IHttpClientFactory httpClientFactory) : IScrapingService
     {
-        public ScrapingNarouService(IHttpClientFactory httpClientFactory)
+        private readonly IHttpClientFactory _httpCliantFactory = httpClientFactory;
+
+        public bool IsMatchSite(Uri uri)
         {
-            _httpCliantFactory = httpClientFactory;
+            return uri.Host == "ncode.syosetu.com";
         }
 
-        private readonly IHttpClientFactory _httpCliantFactory;
-
-        public async Task<EpubDocument> ScrapingAsync(string url, string coverFilePath, string imageDirectory, Guid id, CancellationToken ct)
+        public async ValueTask<EpubDocument> ScrapingAsync(string url, string coverFilePath, string imageDirectory, Guid id, CancellationToken ct)
         {
             var config = Configuration.Default.WithDefaultLoader();
             using var context = BrowsingContext.New(config);