如何用 C# 和 HttpClient 识别 DataDome 验证码

dotnet new console -o datadome-demo
cd datadome-demo
dotnet add package CapSkip

using System.Net;
using System.Text.Json;
using System.Text.RegularExpressions;
using CapSkip;

static class BlockPage
{
    // Build the challenge address from a DataDome block page, or return null.
    public static string? CaptchaUrl(HttpStatusCode status, string html, string pageUrl)
    {
        if (status != HttpStatusCode.Forbidden && status != HttpStatusCode.Unauthorized) return null;
        var found = Regex.Match(html, @"var\s+dd\s*=\s*(\{[^}]*\})");
        if (!found.Success) return null;

        using var doc = JsonDocument.Parse(found.Groups[1].Value.Replace('\'', '"'));
        var dd = doc.RootElement;
        string? Get(string name) => dd.TryGetProperty(name, out var v)
            ? (v.ValueKind == JsonValueKind.String ? v.GetString() : v.GetRawText())
            : null;

        var query = new List<(string Key, string? Value)>
        {
            ("initialCid", Get("cid")), ("hash", Get("hsh")), ("cid", Get("cookie")),
        };
        if (!string.IsNullOrEmpty(Get("t"))) query.Add(("t", Get("t")));  // fe challenge, bv banned
        query.Add(("referer", pageUrl));
        query.Add(("s", Get("s")));
        foreach (var key in new[] { "e", "b" })
            if (!string.IsNullOrEmpty(Get(key))) query.Add((key, Get(key)));
        query.Add(("dm", "cd"));

        var path = Get("rt") == "i" ? "interstitial" : "captcha";
        var qs = string.Join("&", query.Select(p => p.Key + "=" + Uri.EscapeDataString(p.Value ?? "")));
        return $"https://{Get("host") ?? "geo.captcha-delivery.com"}/{path}/?{qs}";
    }
}

static class Solve
{
    // Local mode. In Server mode, use your server's address instead.
    static readonly CapSkipClient Solver = new(host: "127.0.0.1", port: 8080);

    public static Task<SolveResult> Run(string captchaUrl, string pageUrl, string? proxyUrl)
    {
        var options = new Dictionary<string, object?>();
        if (proxyUrl != null)
        {
            var i = proxyUrl.IndexOf("://", StringComparison.Ordinal);
            options["proxy"] = new Proxy(proxyUrl[..i].ToUpperInvariant(), proxyUrl[(i + 3)..]);
        }
        // No userAgent option: CapSkip uses its own browser's and reports it.
        return Solver.DataDomeAsync(captchaUrl, pageUrl, options);
    }
}

static class Web
{
    static HttpClient Client(string? proxyUrl)
    {
        // UseCookies is off, so the Cookie header set by hand goes out as it is.
        var handler = new HttpClientHandler { UseCookies = false };
        if (proxyUrl != null)
        {
            var uri = new Uri(proxyUrl);
            var proxy = new WebProxy($"{uri.Scheme}://{uri.Host}:{uri.Port}");
            if (uri.UserInfo.Length > 0)
            {
                var login = uri.UserInfo.Split(':', 2);
                proxy.Credentials = new NetworkCredential(
                    Uri.UnescapeDataString(login[0]),
                    login.Length > 1 ? Uri.UnescapeDataString(login[1]) : "");
            }
            handler.Proxy = proxy;
        }
        return new HttpClient(handler) { Timeout = TimeSpan.FromSeconds(30) };
    }

    // Returns the status, the body and a replacement datadome cookie if the site sent one.
    public static async Task<(HttpStatusCode Status, string Body, string? Fresh)> Get(
        string url, string userAgent, string? cookie, string? proxyUrl)
    {
        using var http = Client(proxyUrl);
        using var request = new HttpRequestMessage(HttpMethod.Get, url);
        request.Headers.TryAddWithoutValidation("User-Agent", userAgent);
        if (cookie != null) request.Headers.TryAddWithoutValidation("Cookie", cookie);
        using var response = await http.SendAsync(request);

        string? fresh = null;
        if (response.Headers.TryGetValues("Set-Cookie", out var lines))
            fresh = lines.Select(l => Regex.Match(l, "^datadome=[^;]+"))
                         .FirstOrDefault(m => m.Success)?.Value;
        return (response.StatusCode, await response.Content.ReadAsStringAsync(), fresh);
    }
}

static class Program
{
    const string PageUrl = "https://www.example.com/";  // a page behind DataDome
    const string Browser = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 " +
                           "(KHTML, like Gecko) Chrome/156.0.0.0 Safari/537.36";
    // The proxy your requests use, e.g. "http://user:[email protected]:8080". Null means
    // they leave from the PC that runs CapSkip.
    static readonly string? ProxyUrl = null;

    static async Task<int> Main()
    {
        for (var attempt = 1; attempt <= 3; attempt++)
        {
            var (status, body, _) = await Web.Get(PageUrl, Browser, null, ProxyUrl);  // fresh block page
            var captchaUrl = BlockPage.CaptchaUrl(status, body, PageUrl);
            if (captchaUrl is null)
            {
                Console.WriteLine($"{(int)status}: no DataDome block page on this response");
                return 0;
            }

            SolveResult result;
            try { result = await Solve.Run(captchaUrl, PageUrl, ProxyUrl); }
            catch (ApiException ex)
            {
                Console.WriteLine($"attempt {attempt}: {ex.Message}");  // the next try gets a new page
                continue;
            }

            // The cookie works only from the address and User-Agent that earned it.
            var via = ProxyUrl ?? (result.Proxy is null ? null : "http://" + result.Proxy);
            var cookie = result.Cookie;
            var (code, page, fresh) = await Web.Get(PageUrl, result.UserAgent!, cookie, via);
            if (code != HttpStatusCode.OK)
            {
                Console.WriteLine($"attempt {attempt}: HTTP {(int)code} with the cookie");
                continue;
            }
            cookie = fresh ?? cookie;  // DataDome may rotate the cookie; keep the newest
            Console.WriteLine($"{(int)code} {page[..Math.Min(40, page.Length)]}");
            var (again, _, _) = await Web.Get(PageUrl, result.UserAgent!, cookie, via);  // reuse it
            Console.WriteLine((int)again);
            return 0;
        }
        Console.WriteLine("DataDome not passed in 3 attempts");
        return 1;
    }
}