diff --git a/WebImageViewer/Class/Renderers/HtmlRenderer.cs b/WebImageViewer/Class/Renderers/HtmlRenderer.cs index 1da7900..5ab51a5 100644 --- a/WebImageViewer/Class/Renderers/HtmlRenderer.cs +++ b/WebImageViewer/Class/Renderers/HtmlRenderer.cs @@ -1350,6 +1350,7 @@ public static class HtmlRenderer + @@ -1357,7 +1358,7 @@ public static class HtmlRenderer
- +
@@ -1415,6 +1416,16 @@ public static class HtmlRenderer
開く →
+ +
+
+ HENTAIPAW + /hentaipaw/2299095 +
+
時間停止と催眠と透明人間で水着の担任の先生をプールの授業中に犯しまくる (50ページ)
+
+
開く →
+
diff --git a/WebImageViewer/Class/Scrapers/HentaiPawScraper.cs b/WebImageViewer/Class/Scrapers/HentaiPawScraper.cs new file mode 100644 index 0000000..2c6f339 --- /dev/null +++ b/WebImageViewer/Class/Scrapers/HentaiPawScraper.cs @@ -0,0 +1,335 @@ +using System.Text.RegularExpressions; +using HtmlAgilityPack; +using Microsoft.Extensions.Logging; +using WebImageViewer.Class.Common; +using WebImageViewer.Class.Models; + +namespace WebImageViewer.Class.Scrapers; + +/// +/// ja.hentaipaw.com(エロモフ)固有のスクレイパークラス +/// +public partial class HentaiPawScraper : BaseScraper +{ + private const string BaseDomain = "hentaipaw.com"; + private const string BaseUrl = "https://ja.hentaipaw.com"; + + public override string SiteName => "hentaipaw"; + public override string SiteKey => "hentaipaw"; + public override IReadOnlyList Aliases => new[] { "hentaipaw.com", "ja.hentaipaw.com", "eromofu", "エロモフ" }; + + public HentaiPawScraper(HttpHelper httpHelper, ILogger logger) + : base(httpHelper, logger) + { + } + + /// + public override bool CanHandle(string urlOrPath) + { + if (string.IsNullOrWhiteSpace(urlOrPath)) + { + return false; + } + + var trimmed = urlOrPath.Trim(); + + // サイトキーまたはエイリアスと完全一致 + if (string.Equals(trimmed, SiteKey, StringComparison.OrdinalIgnoreCase) || + Aliases.Any(a => string.Equals(trimmed, a, StringComparison.OrdinalIgnoreCase))) + { + return true; + } + + // ドメイン名が含まれているか + if (trimmed.Contains(BaseDomain, StringComparison.OrdinalIgnoreCase)) + { + return true; + } + + // エロモフ(eromofu)が含まれているか + if (trimmed.Contains("eromofu", StringComparison.OrdinalIgnoreCase)) + { + return true; + } + + // hentaipaw パス(例: hentaipaw/2299095)が含まれているか + if (trimmed.StartsWith("hentaipaw/", StringComparison.OrdinalIgnoreCase)) + { + return true; + } + + return false; + } + + /// + public override async Task ScrapeAsync(string idOrUrl, string? category = null, CancellationToken cancellationToken = default) + { + var normalizedId = ResolveArticleId(idOrUrl); + if (string.IsNullOrEmpty(normalizedId)) + { + Logger.LogWarning("[{SiteName}] 有効な記事IDを抽出できませんでした: {Input}", SiteName, idOrUrl); + return null; + } + + var articleUrl = $"{BaseUrl}/articles/{normalizedId}"; + var viewerUrl = $"{BaseUrl}/viewer?articleId={normalizedId}&page=1"; + + Logger.LogInformation("[{SiteName}] スクレイピング開始 - ID: {Id}, 記事URL: {ArticleUrl}", SiteName, normalizedId, articleUrl); + + // 1. ビュワーページと記事詳細ページを取得(Cloudflare保護下のためFlareSolverrを使用) + var viewerDoc = await LoadHtmlDocumentAsync(viewerUrl, useFlareSolverr: true, cancellationToken); + var articleDoc = await LoadHtmlDocumentAsync(articleUrl, useFlareSolverr: true, cancellationToken); + + if (viewerDoc == null && articleDoc == null) + { + Logger.LogWarning("[{SiteName}] ビュワーページおよび記事詳細ページの取得に両方失敗しました: ID={Id}", SiteName, normalizedId); + return null; + } + + // 2. タイトルの抽出(記事詳細ページのh1優先、フォールバックでビュワーページのtitleタグ) + string title = string.Empty; + if (articleDoc?.DocumentNode != null) + { + var h1 = articleDoc.DocumentNode.SelectSingleNode("//h1"); + if (h1 != null) + { + title = CleanText(h1.InnerText); + } + } + + if (string.IsNullOrWhiteSpace(title) && viewerDoc?.DocumentNode != null) + { + var titleNode = viewerDoc.DocumentNode.SelectSingleNode("//title"); + title = CleanTitle(CleanText(titleNode?.InnerText)); + } + + if (string.IsNullOrWhiteSpace(title)) + { + title = $"エロモフ作品 #{normalizedId}"; + } + + var galleryItem = new GalleryItem + { + Id = normalizedId, + Title = title, + Category = category ?? "doujinshi", + SourceSite = SiteName, + SourceUrl = articleUrl, + ThumbnailUrl = $"https://cdn.imagedeliveries.com/{normalizedId}/thumbnails/cover.webp", + FetchedAt = DateTime.UtcNow + }; + + // 3. メタデータ(作者、サークル、タグ、原作、キャラクター等)の抽出 + if (articleDoc?.DocumentNode != null) + { + ExtractMetadata(articleDoc, galleryItem); + } + + // 4. 画像一覧の抽出(ビュワーページの高解像度スライド優先、フォールバックでサムネイル) + ExtractImages(viewerDoc, articleDoc, galleryItem, normalizedId); + + Logger.LogInformation("[{SiteName}] スクレイピング完了: 『{Title}』, 画像数: {Count}枚", + SiteName, galleryItem.Title, galleryItem.Images.Count); + + return galleryItem; + } + + /// + /// 入力文字列(URLまたはID)から記事ID(数字列)を抽出・正規化します。 + /// + private static string ResolveArticleId(string idOrUrl) + { + var trimmed = idOrUrl.Trim(); + + // 1. URL形式(例: .../articles/2299095, .../viewer?articleId=2299095) + var matchViewer = Regex.Match(trimmed, @"[?&]articleId=(?\d+)", RegexOptions.IgnoreCase); + if (matchViewer.Success) + { + return matchViewer.Groups["id"].Value; + } + + var matchArticles = Regex.Match(trimmed, @"/articles/(?\d+)", RegexOptions.IgnoreCase); + if (matchArticles.Success) + { + return matchArticles.Groups["id"].Value; + } + + // 2. hentaipaw/2299095 形式 + var matchSitePath = Regex.Match(trimmed, @"(?:hentaipaw/)?(?\d+)", RegexOptions.IgnoreCase); + if (matchSitePath.Success) + { + return matchSitePath.Groups["id"].Value; + } + + // 3. 数字のみ + var matchDigits = Regex.Match(trimmed, @"^\d+$"); + if (matchDigits.Success) + { + return matchDigits.Value; + } + + return trimmed.TrimEnd('/').Split('/').Last(); + } + + /// + /// タイトル文字列からサイト名サフィックス(例: " - エロモフ" 等)を除去します。 + /// + private static string CleanTitle(string? rawTitle) + { + if (string.IsNullOrWhiteSpace(rawTitle)) + { + return string.Empty; + } + + var title = rawTitle.Trim(); + if (title.Contains(" - エロモフ")) + { + title = title.Split(" - エロモフ")[0].Trim(); + } + else if (title.Contains(" | エロモフ")) + { + title = title.Split(" | エロモフ")[0].Trim(); + } + else if (title.Contains(" - hentaipaw")) + { + title = title.Split(" - hentaipaw")[0].Trim(); + } + + return title; + } + + /// + /// 記事詳細ページのHTMLから作者、サークル、タグ、カテゴリー等のメタ情報を抽出します。 + /// + private static void ExtractMetadata(HtmlDocument doc, GalleryItem item) + { + // #article-tag-information 内のタグ群 + var tagContainer = doc.DocumentNode.SelectSingleNode("//*[@id='article-tag-information']"); + if (tagContainer == null) + { + return; + } + + var rows = tagContainer.SelectNodes(".//div[contains(@class, 'flex') and contains(@class, 'flex-wrap')]"); + if (rows == null) + { + return; + } + + foreach (var row in rows) + { + var h3 = row.SelectSingleNode(".//h3"); + var label = CleanText(h3?.InnerText).Replace(":", "").Replace("", "").Trim(); + + var valueNodes = row.SelectNodes(".//a//div | .//a | .//p"); + var values = valueNodes != null + ? valueNodes.Select(v => CleanText(v.InnerText)) + .Where(v => !string.IsNullOrEmpty(v) && !string.Equals(v, "N/A", StringComparison.OrdinalIgnoreCase)) + .Distinct() + .ToList() + : new List(); + + if (values.Count == 0) + { + continue; + } + + if (label.Contains("作者") || label.Contains("作家")) + { + item.Author = string.Join(", ", values); + } + else if (label.Contains("サークル")) + { + item.Circle = string.Join(", ", values); + } + else if (label.Contains("カテゴリー")) + { + item.Category = values[0]; + } + else if (label.Contains("タグ") || label.Contains("キャラクター") || label.Contains("原作")) + { + item.Tags.AddRange(values); + } + } + + // タグの重複を除外 + item.Tags = item.Tags.Distinct().ToList(); + } + + /// + /// 画像リストを構築します。ビュワーページ内の高解像度スライド画像を優先し、フォールバックでサムネイルを使用します。 + /// + private static void ExtractImages(HtmlDocument? viewerDoc, HtmlDocument? articleDoc, GalleryItem item, string articleId) + { + var collectedImages = new List(); + + // 1. ビュワーページの HTML から高解像度画像(slides)を抽出 + if (viewerDoc?.DocumentNode != null) + { + var rawHtml = viewerDoc.DocumentNode.OuterHtml; + var pattern = $@"https://cdn\.imagedeliveries\.com/{Regex.Escape(articleId)}/[a-f0-9]{{32,64}}/(?\d+)\.webp"; + var matches = Regex.Matches(rawHtml, pattern, RegexOptions.IgnoreCase); + + var pageMap = new Dictionary(); + foreach (Match match in matches) + { + if (int.TryParse(match.Groups["page"].Value, out var pageNum)) + { + if (!pageMap.ContainsKey(pageNum)) + { + pageMap[pageNum] = match.Value; + } + } + } + + if (pageMap.Count > 0) + { + foreach (var kvp in pageMap.OrderBy(k => k.Key)) + { + int page = kvp.Key; + string highResUrl = kvp.Value; + string thumbUrl = $"https://cdn.imagedeliveries.com/{articleId}/thumbnails/{page}.webp"; + + collectedImages.Add(new ImageInfo + { + PageNumber = page, + ImageUrl = highResUrl, + ThumbnailUrl = thumbUrl, + AltText = $"{item.Title} - {page}ページ" + }); + } + } + } + + // 2. フォールバック: ビュワーページから取得できなかった場合、記事詳細ページのサムネイル一覧から復元 + if (collectedImages.Count == 0 && articleDoc?.DocumentNode != null) + { + var rawHtml = articleDoc.DocumentNode.OuterHtml; + var thumbPattern = $@"https://cdn\.imagedeliveries\.com/{Regex.Escape(articleId)}/thumbnails/(?\d+)\.webp"; + var matches = Regex.Matches(rawHtml, thumbPattern, RegexOptions.IgnoreCase); + + var pages = new SortedSet(); + foreach (Match match in matches) + { + if (int.TryParse(match.Groups["page"].Value, out var pageNum)) + { + pages.Add(pageNum); + } + } + + foreach (var page in pages) + { + string thumbUrl = $"https://cdn.imagedeliveries.com/{articleId}/thumbnails/{page}.webp"; + collectedImages.Add(new ImageInfo + { + PageNumber = page, + ImageUrl = thumbUrl, + ThumbnailUrl = thumbUrl, + AltText = $"{item.Title} - {page}ページ" + }); + } + } + + item.Images = collectedImages; + } +} diff --git a/WebImageViewer/Program.cs b/WebImageViewer/Program.cs index 0770d1c..5f2301f 100644 --- a/WebImageViewer/Program.cs +++ b/WebImageViewer/Program.cs @@ -37,6 +37,8 @@ builder.Services.AddScoped(); builder.Services.AddScoped(); builder.Services.AddScoped(); builder.Services.AddScoped(); +builder.Services.AddScoped(); +builder.Services.AddScoped(); // ギャラリー統括管理サービスの登録 builder.Services.AddScoped();