Files
hamkadr/src/JobsMedical.Web/Program.cs
T
soroush.asadi cea27c8684
CI/CD / CI · dotnet build (push) Successful in 53s
CI/CD / Deploy · hamkadr (push) Successful in 1m12s
[Ingest] Route scraping through an optional V2Ray/Xray proxy (Telegram in Iran)
Telegram and some sources are filtered in Iran. .NET cannot speak vmess/vless/trojan, so add an Xray sidecar (compose service 'xray', behind the 'proxy' profile) that converts the admin's config into a local SOCKS5 proxy (xray:10808). New ScrapeHttpClients provider builds a proxied or direct HttpClient (WebProxy supports socks5/socks4/http) cached per proxy URL; all five ingestion sources (Telegram/Bale/Divar/Medjobs/Websites) now use it. Admin settings gain IngestProxyEnabled + IngestProxyUrl (migration; UI under sources). Added deploy/xray/config.json template + README with vmess/vless/trojan examples.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-04 17:53:17 +03:30

305 lines
14 KiB
C#
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
using System.Text.Encodings.Web;
using System.Text.Unicode;
using JobsMedical.Web.Data;
using JobsMedical.Web.Models;
using JobsMedical.Web.Services;
using Microsoft.AspNetCore.Authentication.Cookies;
using Microsoft.AspNetCore.HttpOverrides;
using Microsoft.AspNetCore.Mvc;
using Microsoft.EntityFrameworkCore;
var builder = WebApplication.CreateBuilder(args);
// Add services to the container.
builder.Services.AddRazorPages();
// Interest tracking + recommendation engine.
builder.Services.AddHttpContextAccessor();
builder.Services.AddMemoryCache();
builder.Services.AddScoped<VisitorContext>();
builder.Services.AddScoped<InterestService>();
builder.Services.AddScoped<RecommendationService>();
builder.Services.AddHttpClient("sms");
builder.Services.AddSingleton<ISmsSender, KavenegarSmsSender>();
builder.Services.AddScoped<OtpService>();
builder.Services.AddSingleton<CaptchaService>();
builder.Services.AddScoped<SubmissionGuard>();
builder.Services.AddScoped<NotificationService>();
builder.Services.AddScoped<PushNotifier>();
builder.Services.AddSingleton<NotificationHub>(); // in-memory SSE broker (live in-app notifications)
// Listing parser: heuristic now; swap for an LLM-backed IListingParser later.
builder.Services.AddSingleton<IListingParser, HeuristicListingParser>();
// Scrape/ingestion engine: pluggable sources → dedupe → parse → validate → (AI audit) → publish/queue.
builder.Services.AddHttpClient("scrape", c =>
{
c.Timeout = TimeSpan.FromSeconds(20);
c.DefaultRequestHeaders.UserAgent.ParseAdd("Mozilla/5.0 (compatible; HamkadrBot/1.0)");
});
builder.Services.AddHttpClient("ai");
// Proxy-aware client provider for ingestion (routes through Xray/V2Ray SOCKS proxy when set).
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.ScrapeHttpClients>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.ListingValidator>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IAiAuditor,
JobsMedical.Web.Services.Scraping.OpenAiCompatibleAuditor>();
builder.Services.AddScoped<JobsMedical.Web.Services.Scraping.SettingsService>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.SampleListingSource>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.TelegramListingSource>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.BaleListingSource>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.DivarListingSource>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.MedjobsListingSource>();
builder.Services.AddSingleton<JobsMedical.Web.Services.Scraping.IListingSource,
JobsMedical.Web.Services.Scraping.WebsiteListingSource>();
builder.Services.AddScoped<JobsMedical.Web.Services.Scraping.ListingArchiver>();
builder.Services.AddScoped<JobsMedical.Web.Services.Scraping.IngestionService>();
builder.Services.AddHostedService<JobsMedical.Web.Services.Scraping.IngestionWorker>();
// Phone-OTP cookie auth.
builder.Services.AddAuthentication(CookieAuthenticationDefaults.AuthenticationScheme)
.AddCookie(o =>
{
o.LoginPath = "/Account/Login";
o.AccessDeniedPath = "/Account/Login";
o.ExpireTimeSpan = TimeSpan.FromDays(30);
o.SlidingExpiration = true;
});
// Emit Persian/Arabic characters directly in HTML instead of \u-style entities.
builder.Services.AddSingleton(HtmlEncoder.Create(
UnicodeRanges.BasicLatin, UnicodeRanges.Arabic, UnicodeRanges.ArabicSupplement,
UnicodeRanges.ArabicExtendedA, UnicodeRanges.GeneralPunctuation));
builder.Services.AddDbContext<AppDbContext>(opt =>
opt.UseNpgsql(builder.Configuration.GetConnectionString("Default")));
var app = builder.Build();
// Apply migrations + seed on startup (fine for MVP single-instance deploy).
using (var scope = app.Services.CreateScope())
{
var db = scope.ServiceProvider.GetRequiredService<AppDbContext>();
db.Database.Migrate();
await SeedData.SeedReferenceAsync(db); // cities/roles/districts always
// Demo board in Development, or whenever the admin has turned Demo Mode on.
var st = await scope.ServiceProvider
.GetRequiredService<JobsMedical.Web.Services.Scraping.SettingsService>().GetAsync();
if (app.Environment.IsDevelopment() || st.DemoMode) await SeedData.SeedDemoAsync(db);
// Archive any listings that went stale while the app was down.
await scope.ServiceProvider
.GetRequiredService<JobsMedical.Web.Services.Scraping.ListingArchiver>()
.ArchiveStaleAsync();
}
// Configure the HTTP request pipeline.
if (!app.Environment.IsDevelopment())
{
app.UseExceptionHandler("/Error");
// The default HSTS value is 30 days. You may want to change this for production scenarios, see https://aka.ms/aspnetcore-hsts.
app.UseHsts();
}
// Behind nginx (TLS terminated upstream): trust X-Forwarded-Proto/For so the app knows it's
// HTTPS — required for correct secure cookies and to avoid HTTPS-redirect loops.
var forwardedOptions = new ForwardedHeadersOptions
{
ForwardedHeaders = ForwardedHeaders.XForwardedFor | ForwardedHeaders.XForwardedProto
};
forwardedOptions.KnownIPNetworks.Clear(); // only nginx can reach the container's bound port
forwardedOptions.KnownProxies.Clear();
app.UseForwardedHeaders(forwardedOptions);
app.UseHttpsRedirection();
app.UseRouting();
// Assign every visitor a stable cookie id so we can track interest from the first visit.
app.UseMiddleware<VisitorCookieMiddleware>();
app.UseAuthentication();
app.UseAuthorization();
app.MapStaticAssets();
app.MapRazorPages()
.WithStaticAssets();
// Lightweight liveness probe for the deploy health-wait loop (and uptime checks).
app.MapGet("/healthz", () => Results.Text("ok"));
// ---- PWA: web manifest + service worker (served from root for full scope) ----
app.MapGet("/manifest.webmanifest", () => Results.Content("""
{
"name": "همکادر — شیفت و استخدام کادر درمان",
"short_name": "همکادر",
"lang": "fa", "dir": "rtl",
"start_url": "/", "scope": "/",
"display": "standalone", "orientation": "portrait",
"background_color": "#f4f7f9", "theme_color": "#0e8f8a",
"icons": [
{ "src": "/icons/icon-192.png", "sizes": "192x192", "type": "image/png", "purpose": "any maskable" },
{ "src": "/icons/icon-512.png", "sizes": "512x512", "type": "image/png", "purpose": "any maskable" }
],
"shortcuts": [
{ "name": "شیفت‌ها", "url": "/Shifts" },
{ "name": "استخدام", "url": "/Jobs" }
]
}
""", "application/manifest+json"));
// Store a browser's push subscription (from the "enable notifications" flow).
app.MapPost("/push/subscribe", async (PushSubscriptionDto dto, AppDbContext db, VisitorContext vc) =>
{
if (string.IsNullOrWhiteSpace(dto.Endpoint) || dto.Keys?.P256dh is null || dto.Keys?.Auth is null)
return Results.BadRequest();
if (!await db.WebPushSubscriptions.AnyAsync(s => s.Endpoint == dto.Endpoint))
{
db.WebPushSubscriptions.Add(new WebPushSubscription
{
Endpoint = dto.Endpoint, P256dh = dto.Keys.P256dh, Auth = dto.Keys.Auth, VisitorId = vc.VisitorId,
});
await db.SaveChangesAsync();
}
return Results.Ok();
});
// Live notification stream (Server-Sent Events). Runs over our own origin, so it reaches
// users in Iran (unlike Web Push, which goes via the browser's blocked push service).
// The browser keeps this open while the tab/PWA is alive; the client updates the bell,
// shows a toast, and fires a local OS notification (no push server) when permission is on.
app.MapGet("/notifications/stream", async (HttpContext ctx, NotificationHub hub) =>
{
var claim = ctx.User.FindFirst(System.Security.Claims.ClaimTypes.NameIdentifier)?.Value;
if (!int.TryParse(claim, out var uid)) { ctx.Response.StatusCode = 401; return; }
ctx.Response.Headers.ContentType = "text/event-stream";
ctx.Response.Headers.CacheControl = "no-cache";
ctx.Response.Headers.Connection = "keep-alive";
ctx.Response.Headers["X-Accel-Buffering"] = "no"; // tell nginx not to buffer the stream
ctx.Features.Get<Microsoft.AspNetCore.Http.Features.IHttpResponseBodyFeature>()?.DisableBuffering();
var (reader, unsubscribe) = hub.Subscribe(uid);
var ct = ctx.RequestAborted;
try
{
await ctx.Response.WriteAsync(": connected\n\n", ct);
await ctx.Response.Body.FlushAsync(ct);
while (!ct.IsCancellationRequested)
{
var readTask = reader.WaitToReadAsync(ct).AsTask();
var keepAlive = Task.Delay(TimeSpan.FromSeconds(25), ct);
if (await Task.WhenAny(readTask, keepAlive) == keepAlive)
{
await ctx.Response.WriteAsync(": ping\n\n", ct); // comment line keeps the connection warm
await ctx.Response.Body.FlushAsync(ct);
continue;
}
if (!await readTask) break;
while (reader.TryRead(out var notice))
{
var json = System.Text.Json.JsonSerializer.Serialize(notice);
await ctx.Response.WriteAsync($"event: notice\ndata: {json}\n\n", ct);
await ctx.Response.Body.FlushAsync(ct);
}
}
}
catch (OperationCanceledException) { /* client disconnected — normal */ }
finally { unsubscribe(); }
}).RequireAuthorization();
// Serve a facility verification document — only the facility owner or an admin may read it.
app.MapGet("/facility-doc/{id:int}", async (int id, HttpContext ctx, AppDbContext db) =>
{
var doc = await db.FacilityDocuments.Include(d => d.Facility).FirstOrDefaultAsync(d => d.Id == id);
if (doc is null) return Results.NotFound();
var isAdmin = ctx.User.IsInRole("Admin");
var uid = int.TryParse(ctx.User.FindFirst(System.Security.Claims.ClaimTypes.NameIdentifier)?.Value, out var n) ? n : (int?)null;
if (!isAdmin && doc.Facility.OwnerUserId != uid) return Results.Forbid();
return Results.File(doc.Data, doc.ContentType, doc.FileName);
}).RequireAuthorization();
// User-submitted report against a listing (abuse/fake/wrong info).
app.MapPost("/report", async (HttpContext ctx, AppDbContext db, VisitorContext vc,
[FromForm] string targetType, [FromForm] int targetId, [FromForm] string reason,
[FromForm] string? label, [FromForm] string? returnUrl) =>
{
if (!string.IsNullOrWhiteSpace(reason) && Enum.TryParse<ReportTargetType>(targetType, true, out var tt))
{
int? uid = ctx.User?.FindFirst(System.Security.Claims.ClaimTypes.NameIdentifier) is { } c
&& int.TryParse(c.Value, out var n) ? n : null;
db.Reports.Add(new Report
{
TargetType = tt, TargetId = targetId, TargetLabel = label,
Reason = reason.Trim()[..Math.Min(reason.Trim().Length, 500)],
ReporterUserId = uid, ReporterVisitorId = vc.VisitorId,
});
await db.SaveChangesAsync();
}
return Results.Redirect((string.IsNullOrWhiteSpace(returnUrl) ? "/" : returnUrl) + "?reported=1");
}).DisableAntiforgery();
app.MapGet("/sw.js", () => Results.Content("""
const CACHE = 'hamkadr-v1';
self.addEventListener('install', e => { self.skipWaiting(); e.waitUntil(caches.open(CACHE).then(c => c.addAll(['/']))); });
self.addEventListener('activate', e => { e.waitUntil(caches.keys().then(ks => Promise.all(ks.filter(k => k !== CACHE).map(k => caches.delete(k))))); self.clients.claim(); });
self.addEventListener('fetch', e => {
const req = e.request;
if (req.method !== 'GET' || new URL(req.url).origin !== location.origin) return;
e.respondWith(fetch(req).then(res => { const copy = res.clone(); caches.open(CACHE).then(c => c.put(req, copy)); return res; })
.catch(() => caches.match(req).then(m => m || caches.match('/'))));
});
self.addEventListener('push', e => {
let d = { title: 'همکادر', body: 'فرصت جدید برای شما', url: '/' };
try { if (e.data) d = Object.assign(d, e.data.json()); } catch (_) { if (e.data) d.body = e.data.text(); }
e.waitUntil(self.registration.showNotification(d.title, { body: d.body, icon: '/icons/icon-192.png', badge: '/icons/icon-192.png', dir: 'rtl', lang: 'fa', data: { url: d.url } }));
});
self.addEventListener('notificationclick', e => {
e.notification.close();
const url = (e.notification.data && e.notification.data.url) || '/';
e.waitUntil(clients.matchAll({ type: 'window' }).then(cl => { for (const c of cl) { if ('focus' in c) { c.navigate(url); return c.focus(); } } return clients.openWindow(url); }));
});
""", "text/javascript"));
// ---- SEO: robots.txt + dynamic sitemap.xml (so Google indexes every live shift/job page) ----
app.MapGet("/robots.txt", (HttpContext ctx) =>
{
var b = $"{ctx.Request.Scheme}://{ctx.Request.Host}";
return Results.Text($"User-agent: *\nAllow: /\nDisallow: /Admin\nDisallow: /Employer\nSitemap: {b}/sitemap.xml\n", "text/plain");
});
app.MapGet("/sitemap.xml", async (HttpContext ctx, AppDbContext db) =>
{
var b = $"{ctx.Request.Scheme}://{ctx.Request.Host}";
var today = DateOnly.FromDateTime(DateTime.UtcNow);
var jobCutoff = JobsMedical.Web.Services.Scraping.ListingPolicy.JobCutoffUtc;
var sb = new System.Text.StringBuilder();
sb.Append("<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");
sb.Append("<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\">\n");
void Url(string loc, DateTime? mod = null, string freq = "daily")
{
sb.Append(" <url><loc>").Append(System.Security.SecurityElement.Escape(loc)).Append("</loc>");
if (mod is not null) sb.Append("<lastmod>").Append(mod.Value.ToString("yyyy-MM-dd")).Append("</lastmod>");
sb.Append("<changefreq>").Append(freq).Append("</changefreq></url>\n");
}
foreach (var p in new[] { "", "/Shifts", "/Jobs", "/Calendar", "/Facilities" })
Url($"{b}{p}", DateTime.UtcNow, p == "" ? "daily" : "hourly");
foreach (var s in await db.Shifts.Where(s => s.Status == ShiftStatus.Open && s.Date >= today)
.Select(s => new { s.Id, s.CreatedAt }).ToListAsync())
Url($"{b}/Shifts/Details/{s.Id}", s.CreatedAt, "daily");
foreach (var j in await db.JobOpenings.Where(j => j.Status == ShiftStatus.Open && j.CreatedAt >= jobCutoff)
.Select(j => new { j.Id, j.CreatedAt }).ToListAsync())
Url($"{b}/Jobs/Details/{j.Id}", j.CreatedAt, "weekly");
sb.Append("</urlset>");
return Results.Content(sb.ToString(), "application/xml");
});
app.Run();