303 lines
11 KiB
C#
303 lines
11 KiB
C#
using System.Text.RegularExpressions;
|
||
using HtmlAgilityPack;
|
||
using Microsoft.Extensions.Logging;
|
||
using WebImageViewer.Class.Common;
|
||
using WebImageViewer.Class.Models;
|
||
|
||
namespace WebImageViewer.Class.Scrapers;
|
||
|
||
/// <summary>
|
||
/// aurahentai.com 固有のスクレイパークラス
|
||
/// </summary>
|
||
public partial class AuraHentaiScraper : BaseScraper
|
||
{
|
||
private const string BaseDomain = "aurahentai.com";
|
||
private const string BaseUrl = "https://aurahentai.com";
|
||
private const string ImageBaseUrl = "https://img.aurahentai.com";
|
||
|
||
public override string SiteName => "aurahentai";
|
||
public override string SiteKey => "aurahentai";
|
||
public override IReadOnlyList<string> Aliases => new[] { "aura", "aurahentai.com", "gallery" };
|
||
|
||
public AuraHentaiScraper(HttpHelper httpHelper, ILogger<AuraHentaiScraper> logger)
|
||
: base(httpHelper, logger)
|
||
{
|
||
}
|
||
|
||
/// <inheritdoc/>
|
||
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;
|
||
}
|
||
|
||
// gallery パスが含まれているか(例: /gallery/013597)
|
||
if (trimmed.Contains("gallery", StringComparison.OrdinalIgnoreCase))
|
||
{
|
||
return true;
|
||
}
|
||
|
||
|
||
// 数字のみ(例: 013597 や 13597)の場合
|
||
if (Regex.IsMatch(trimmed, @"^\d{4,8}$"))
|
||
{
|
||
return true;
|
||
}
|
||
|
||
return false;
|
||
}
|
||
|
||
/// <inheritdoc/>
|
||
public override async Task<GalleryItem?> ScrapeAsync(string idOrUrl, string? category = null, CancellationToken cancellationToken = default)
|
||
{
|
||
var (targetUrl, normalizedId) = ResolveUrlAndId(idOrUrl);
|
||
|
||
Logger.LogInformation("[{SiteName}] スクレイピング開始 - ID: {Id}, URL: {Url}", SiteName, normalizedId, targetUrl);
|
||
|
||
// aurahentai.com は Cloudflare で保護されているため FlareSolverr を使用
|
||
var doc = await LoadHtmlDocumentAsync(targetUrl, useFlareSolverr: true, cancellationToken);
|
||
if (doc == null || doc.DocumentNode == null)
|
||
{
|
||
Logger.LogWarning("[{SiteName}] ドキュメントの読み込みに失敗しました: {Url}", SiteName, targetUrl);
|
||
return null;
|
||
}
|
||
|
||
// タイトルの抽出(Age-gateのAdults onlyを除外し、メインタイトルを取得)
|
||
var titleNode = doc.DocumentNode.SelectSingleNode("//main//h1[not(@id='age-gate-title')]") ??
|
||
doc.DocumentNode.SelectSingleNode("//h1[not(@id='age-gate-title')]") ??
|
||
doc.DocumentNode.SelectSingleNode("//title");
|
||
var title = CleanText(titleNode?.InnerText) ?? normalizedId;
|
||
|
||
// 余分なサイト名・サフィックスを除去(例: "タイトル ? 100% No Ads | AuraHentai" -> "タイトル")
|
||
if (title.Contains(" ? "))
|
||
{
|
||
title = title.Split(" ? ")[0].Trim();
|
||
}
|
||
else if (title.Contains(" | "))
|
||
{
|
||
title = title.Split(" | ")[0].Trim();
|
||
}
|
||
|
||
var galleryItem = new GalleryItem
|
||
{
|
||
Id = normalizedId,
|
||
Title = string.IsNullOrEmpty(title) ? normalizedId : title,
|
||
Category = "gallery",
|
||
SourceSite = SiteName,
|
||
SourceUrl = targetUrl,
|
||
ThumbnailUrl = $"{ImageBaseUrl}/w575/g/{normalizedId}/cover.webp",
|
||
FetchedAt = DateTime.UtcNow
|
||
};
|
||
|
||
// メタデータ(作者、サークル、タグ、ページ数等)の抽出
|
||
var pageCount = ExtractMetadata(doc, galleryItem);
|
||
|
||
// 画像一覧の抽出
|
||
ExtractImages(doc, galleryItem, normalizedId, pageCount);
|
||
|
||
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;
|
||
}
|
||
|
||
/// <summary>
|
||
/// IDまたはURLから完全なリクエストURLと正規化されたIDを解決します。
|
||
/// </summary>
|
||
private static (string TargetUrl, string NormalizedId) ResolveUrlAndId(string idOrUrl)
|
||
{
|
||
var trimmed = idOrUrl.Trim();
|
||
string normalizedId;
|
||
|
||
if (trimmed.StartsWith("http://", StringComparison.OrdinalIgnoreCase) ||
|
||
trimmed.StartsWith("https://", StringComparison.OrdinalIgnoreCase))
|
||
{
|
||
var uri = new Uri(trimmed);
|
||
var segments = uri.AbsolutePath.Split('/', StringSplitOptions.RemoveEmptyEntries);
|
||
if (segments.Length >= 2 && segments[0].Equals("gallery", StringComparison.OrdinalIgnoreCase))
|
||
{
|
||
normalizedId = segments[1];
|
||
}
|
||
else if (segments.Length >= 1)
|
||
{
|
||
normalizedId = segments[^1];
|
||
}
|
||
else
|
||
{
|
||
normalizedId = trimmed;
|
||
}
|
||
}
|
||
else if (trimmed.Contains('/'))
|
||
{
|
||
var parts = trimmed.Split('/', StringSplitOptions.RemoveEmptyEntries);
|
||
normalizedId = parts[^1];
|
||
}
|
||
else
|
||
{
|
||
normalizedId = trimmed;
|
||
}
|
||
|
||
// 6桁のゼロ埋め形式に対応(例: "13597" -> "013597"、数字のみかつ6桁未満の場合)
|
||
if (int.TryParse(normalizedId, out var num) && normalizedId.Length < 6)
|
||
{
|
||
normalizedId = num.ToString("D6");
|
||
}
|
||
|
||
var fullUrl = $"{BaseUrl}/gallery/{normalizedId}";
|
||
return (fullUrl, normalizedId);
|
||
}
|
||
|
||
/// <summary>
|
||
/// dlタグからメタデータ(タグ、作者、サークル、カテゴリ、総ページ数)を抽出します。
|
||
/// </summary>
|
||
private static int ExtractMetadata(HtmlDocument doc, GalleryItem item)
|
||
{
|
||
int detectedPageCount = 0;
|
||
var dlNode = doc.DocumentNode.SelectSingleNode("//main//dl");
|
||
if (dlNode == null) return detectedPageCount;
|
||
|
||
var dtNodes = dlNode.SelectNodes(".//dt");
|
||
if (dtNodes == null) return detectedPageCount;
|
||
|
||
foreach (var dt in dtNodes)
|
||
{
|
||
var label = CleanText(dt.InnerText).ToLowerInvariant();
|
||
var dd = dt.ParentNode?.SelectSingleNode(".//dd");
|
||
if (dd == null) continue;
|
||
|
||
|
||
var links = dd.SelectNodes(".//a");
|
||
var values = links != null
|
||
? links.Select(a => CleanText(a.InnerText)).Where(s => !string.IsNullOrEmpty(s)).ToList()
|
||
: new List<string> { CleanText(dd.InnerText) };
|
||
|
||
switch (label)
|
||
{
|
||
case "artists":
|
||
case "artist":
|
||
item.Author = string.Join(", ", values);
|
||
break;
|
||
case "groups":
|
||
case "group":
|
||
case "circles":
|
||
case "circle":
|
||
item.Circle = string.Join(", ", values);
|
||
break;
|
||
case "category":
|
||
if (values.Count > 0) item.Category = values[0];
|
||
break;
|
||
case "tags":
|
||
case "characters":
|
||
case "character":
|
||
case "parodies":
|
||
case "parody":
|
||
item.Tags.AddRange(values);
|
||
break;
|
||
case "pages":
|
||
var rawPages = CleanText(dd.InnerText);
|
||
var match = Regex.Match(rawPages, @"\d+");
|
||
if (match.Success && int.TryParse(match.Value, out var count))
|
||
{
|
||
detectedPageCount = count;
|
||
}
|
||
break;
|
||
}
|
||
}
|
||
|
||
return detectedPageCount;
|
||
}
|
||
|
||
/// <summary>
|
||
/// 画像リストを構築します。
|
||
/// </summary>
|
||
private static void ExtractImages(HtmlDocument doc, GalleryItem item, string galleryId, int pageCount)
|
||
{
|
||
// ページグリッド内のサムネイル画像から抽出
|
||
var imgNodes = doc.DocumentNode.SelectNodes("//div[contains(@class, 'page-grid')]//img") ??
|
||
doc.DocumentNode.SelectNodes("//main//img[contains(@src, '/g/')]");
|
||
|
||
var collectedImages = new List<ImageInfo>();
|
||
|
||
if (imgNodes != null)
|
||
{
|
||
int pageNum = 1;
|
||
foreach (var img in imgNodes)
|
||
{
|
||
var src = img.GetAttributeValue("src", null) ??
|
||
img.GetAttributeValue("data-src", null);
|
||
|
||
if (string.IsNullOrWhiteSpace(src) || src.Contains("cover.webp") || src.Contains("logo"))
|
||
{
|
||
continue;
|
||
}
|
||
|
||
// サムネイルURL(例: https://img.aurahentai.com/thumb/g/013597/001.webp)
|
||
var thumbUrl = src.StartsWith("http") ? src : $"{ImageBaseUrl}{src}";
|
||
|
||
// 本文の高解像度URL(例: https://img.aurahentai.com/g/013597/001.webp)
|
||
var fullImageUrl = thumbUrl.Replace("/thumb/g/", "/g/");
|
||
|
||
var alt = img.GetAttributeValue("alt", null);
|
||
|
||
collectedImages.Add(new ImageInfo
|
||
{
|
||
PageNumber = pageNum++,
|
||
ImageUrl = fullImageUrl,
|
||
ThumbnailUrl = thumbUrl,
|
||
AltText = CleanText(alt) ?? $"{item.Title} {pageNum}ページ"
|
||
});
|
||
}
|
||
}
|
||
|
||
// HTMLから取得できなかった場合、またはページ数情報がある場合は連番URLを補完生成
|
||
if (collectedImages.Count == 0 && pageCount > 0)
|
||
{
|
||
for (int i = 1; i <= pageCount; i++)
|
||
{
|
||
var pageStr = i.ToString("D3");
|
||
collectedImages.Add(new ImageInfo
|
||
{
|
||
PageNumber = i,
|
||
ImageUrl = $"{ImageBaseUrl}/g/{galleryId}/{pageStr}.webp",
|
||
ThumbnailUrl = $"{ImageBaseUrl}/thumb/g/{galleryId}/{pageStr}.webp",
|
||
AltText = $"{item.Title} {i}ページ"
|
||
});
|
||
}
|
||
}
|
||
else if (pageCount > collectedImages.Count)
|
||
{
|
||
// グリッドが折りたたまれて一部しかHTMLにない場合、総ページ数まで補完
|
||
int existingCount = collectedImages.Count;
|
||
for (int i = existingCount + 1; i <= pageCount; i++)
|
||
{
|
||
var pageStr = i.ToString("D3");
|
||
collectedImages.Add(new ImageInfo
|
||
{
|
||
PageNumber = i,
|
||
ImageUrl = $"{ImageBaseUrl}/g/{galleryId}/{pageStr}.webp",
|
||
ThumbnailUrl = $"{ImageBaseUrl}/thumb/g/{galleryId}/{pageStr}.webp",
|
||
AltText = $"{item.Title} {i}ページ"
|
||
});
|
||
}
|
||
}
|
||
|
||
item.Images.AddRange(collectedImages);
|
||
}
|
||
}
|