Files
myAi/Jobs/cv-search-job/Services/HtmlJobSearcher.cs
T
claude 16bb195cb5 Add XML doc to all service interfaces and implementations (#26)
- Update CLAUDE.md: replace incorrect 'no XML doc on internal code' rule
  with the correct convention (XML doc on all public methods and
  non-trivial private/protected helpers)
- Restore /// <summary> on FileDownloadController private helpers
  (HandleRangeRequest, StreamRangeAsync)
- Add full XML doc to all service contracts:
  ICaptchaVerifier, IEmailSender, ICvMatcherService, IJobTextExtractor,
  IJobTokenService, IDocumentClassifier, IRagService, ITextChunker,
  ITextExtractor, IEmailTemplateService, ITemplateService
- Add /// <summary> and /// <inheritdoc /> to all concrete service classes
  and their methods: RecaptchaVerifier, EmailApiEmailSender,
  SmtpEmailDispatcher, CvMatcherService, JobTextExtractor, JobTokenService,
  RagService, DocumentClassifier, TextChunker, TextExtractor,
  HtmlJobSearcher, CvSearchEmailSender, CvSearchJobTask,
  EmailTemplateService, DbTemplateService

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-28 09:17:42 +03:00

102 lines
4.0 KiB
C#

using System.Text.RegularExpressions;
using System.Web;
using CvMatcher.Models.Settings;
using Microsoft.Extensions.Logging;
namespace CvSearchJob.Services;
/// <summary>
/// Config-driven HTML scraper that fetches a provider's job listing page and extracts matching job URLs.
/// Uses a two-stage anchor filter: href must contain the provider's link pattern, and anchor text must
/// contain at least one CV keyword.
/// </summary>
public sealed class HtmlJobSearcher
{
private readonly HttpClient _http;
private readonly ILogger<HtmlJobSearcher> _logger;
public HtmlJobSearcher(HttpClient http, ILogger<HtmlJobSearcher> logger)
{
_http = http;
_logger = logger;
_http.Timeout = TimeSpan.FromSeconds(20);
_http.DefaultRequestHeaders.UserAgent.ParseAdd("Mozilla/5.0 (compatible; MyAi.ro CV-Search/1.0)");
}
/// <summary>
/// Fetches the provider's search result page for the combined initial + CV keywords, parses all anchor
/// tags, applies the two-stage filter, and returns up to <see cref="JobProviderConfig.MaxResults"/> absolute URLs.
/// Returns an empty list when the HTTP request fails rather than throwing.
/// </summary>
/// <param name="provider">Provider configuration including search URL template, link filter, and result cap.</param>
/// <param name="cvKeywords">Keywords extracted from the user's CV to inject into the search query.</param>
/// <param name="ct">Cancellation token.</param>
/// <returns>Deduplicated list of absolute job page URLs (query string stripped).</returns>
public async Task<IReadOnlyList<string>> SearchJobUrlsAsync(
JobProviderConfig provider,
IReadOnlyList<string> cvKeywords,
CancellationToken ct)
{
var allKeywords = provider.InitialKeywords
.Concat(cvKeywords)
.Where(k => !string.IsNullOrWhiteSpace(k))
.Distinct(StringComparer.OrdinalIgnoreCase)
.ToList();
if (allKeywords.Count == 0)
return [];
var keywordsEncoded = HttpUtility.UrlEncode(string.Join(" ", allKeywords));
var searchUrl = provider.SearchUrlTemplate.Replace("{keywords}", keywordsEncoded);
string html;
try
{
html = await _http.GetStringAsync(searchUrl, ct);
}
catch (Exception ex)
{
_logger.LogWarning(ex, "Failed to fetch search results from {Provider} at {Url}", provider.Name, searchUrl);
return [];
}
var baseUri = new Uri(searchUrl);
var results = new List<string>();
var seen = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
// Match all anchor tags capturing href and inner text
var anchorPattern = new Regex(@"<a[^>]+href=[""']([^""']+)[""'][^>]*>(.*?)</a>",
RegexOptions.IgnoreCase | RegexOptions.Singleline);
foreach (Match match in anchorPattern.Matches(html))
{
if (results.Count >= provider.MaxResults) break;
var href = match.Groups[1].Value.Trim();
var anchorText = Regex.Replace(match.Groups[2].Value, "<[^>]+>", " ").Trim();
if (!href.Contains(provider.JobLinkContains, StringComparison.OrdinalIgnoreCase))
continue;
// Stage 2: anchor text must contain at least one CV keyword
if (!cvKeywords.Any(k => anchorText.Contains(k, StringComparison.OrdinalIgnoreCase)))
continue;
// Make absolute URL
if (!Uri.TryCreate(href, UriKind.Absolute, out var absoluteUri))
{
if (!Uri.TryCreate(baseUri, href, out absoluteUri))
continue;
}
// Strip query string and fragment so different tracking variants of the same URL collapse to one.
var url = absoluteUri.GetLeftPart(UriPartial.Path);
if (seen.Add(url))
results.Add(url);
}
_logger.LogInformation("Provider {Provider}: found {Count} job URLs", provider.Name, results.Count);
return results;
}
}