feat: implement multiple new job crawlers and add corresponding test suites

2026-08-25 00:12:52 -07:00
parent 0de312edbd
commit 55d4f4a0d9
26 changed files with 4045 additions and 15 deletions
@@ -0,0 +1,229 @@
+package linkedin
+
+import (
+	"context"
+	"crypto/sha256"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"regexp"
+	"strings"
+	"time"
+
+	"github.com/hireme/hireme/internal/crawler"
+	"github.com/hireme/hireme/internal/logger"
+)
+
+type LinkedInCrawler struct {
+	client *http.Client
+}
+
+func NewLinkedInCrawler() *LinkedInCrawler {
+	return NewLinkedInCrawlerWithClient(&http.Client{
+		Timeout: 30 * time.Second,
+	})
+}
+
+func NewLinkedInCrawlerWithClient(client *http.Client) *LinkedInCrawler {
+	return &LinkedInCrawler{
+		client: client,
+	}
+}
+
+func (c *LinkedInCrawler) CanHandle(platform string) bool {
+	p := strings.ToLower(platform)
+	return p == "linkedin" || strings.Contains(p, "linkedin.com/jobs")
+}
+
+func (c *LinkedInCrawler) DiscoverJobs(ctx context.Context, targetURL string) ([]crawler.DiscoveredJob, error) {
+	seen := make(map[string]bool)
+	var discovered []crawler.DiscoveredJob
+
+	searchAPI := targetURL
+	if !strings.Contains(searchAPI, "jobs-guest/jobs/api/seeMoreJobPostings/search") {
+		u, err := url.Parse(targetURL)
+		if err == nil && u.Host != "" {
+			q := u.Query()
+			searchAPI = fmt.Sprintf("https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search?%s", q.Encode())
+		} else {
+			searchAPI = fmt.Sprintf("https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search?keywords=%s&location=Seattle%%2C+Washington%%2C+United+States", url.QueryEscape(targetURL))
+		}
+	}
+
+	pageSize := 25
+	reID := regexp.MustCompile(`(?:urn:li:jobPosting:(\d+)|/jobs/view/(?:[a-z0-9\-]+-)?(\d+))`)
+
+	for page := 0; page < 4; page++ {
+		start := page * pageSize
+		reqURL := searchAPI
+		if strings.Contains(reqURL, "?") {
+			reqURL = fmt.Sprintf("%s&start=%d", searchAPI, start)
+		} else {
+			reqURL = fmt.Sprintf("%s?start=%d", searchAPI, start)
+		}
+
+		body, err := c.fetchURL(ctx, reqURL)
+		if err != nil {
+			if page == 0 {
+				return nil, err
+			}
+			break
+		}
+
+		matches := reID.FindAllStringSubmatch(body, -1)
+		if len(matches) == 0 {
+			break
+		}
+
+		newFound := false
+		for _, m := range matches {
+			jobID := m[1]
+			if jobID == "" && len(m) >= 3 {
+				jobID = m[2]
+			}
+			if jobID == "" {
+				continue
+			}
+
+			applyURL := fmt.Sprintf("https://www.linkedin.com/jobs/view/%s", jobID)
+			if !seen[applyURL] {
+				seen[applyURL] = true
+				newFound = true
+				discovered = append(discovered, crawler.DiscoveredJob{
+					CanonicalID: fmt.Sprintf("linkedin:%s", jobID),
+					ApplyURL:    applyURL,
+				})
+			}
+		}
+
+		if !newFound || len(matches) < pageSize {
+			break
+		}
+	}
+
+	logger.Log.Infow("LinkedIn discovery finished", "found_count", len(discovered))
+	return discovered, nil
+}
+
+func (c *LinkedInCrawler) FetchJobDetail(ctx context.Context, applyURL string) (*crawler.ExtractedJob, error) {
+	jobID := parseLinkedInJobID(applyURL)
+	if jobID == "" {
+		jobID = fmt.Sprintf("%x", sha256.Sum256([]byte(applyURL)))[:16]
+	}
+
+	// LinkedIn guest API endpoint for job details:
+	guestDetailURL := fmt.Sprintf("https://www.linkedin.com/jobs-guest/jobs/api/jobPosting/%s", jobID)
+	body, err := c.fetchURL(ctx, guestDetailURL)
+	if err != nil {
+		body, err = c.fetchURL(ctx, applyURL)
+		if err != nil {
+			return nil, err
+		}
+	}
+
+	title := extractRegex(body, `<h2[^>]*class="[^"]*top-card-layout__title[^"]*"[^>]*>(.*?)</h2>`)
+	if title == "" {
+		title = extractRegex(body, `<h1[^>]*>(.*?)</h1>`)
+	}
+	if title == "" {
+		title = "Software Engineer"
+	}
+
+	company := extractRegex(body, `<a[^>]*class="[^"]*topcard__org-name-link[^"]*"[^>]*>(.*?)</a>`)
+	if company == "" {
+		company = extractRegex(body, `<span[^>]*class="[^"]*topcard__flavor[^"]*"[^>]*>(.*?)</span>`)
+	}
+	if company == "" {
+		company = "LinkedIn"
+	}
+
+	location := extractRegex(body, `<span[^>]*class="[^"]*topcard__flavor--bullet[^"]*"[^>]*>(.*?)</span>`)
+	if location == "" {
+		location = "Seattle, WA, USA"
+	}
+
+	desc := extractRegex(body, `<div[^>]*class="[^"]*show-more-less-html__markup[^"]*"[^>]*>([\s\S]*?)</div>`)
+	if desc == "" {
+		desc = body
+	}
+
+	return &crawler.ExtractedJob{
+		CanonicalID: fmt.Sprintf("linkedin:%s", jobID),
+		Title:       cleanHTML(title),
+		Company:     cleanHTML(company),
+		Location:    cleanHTML(location),
+		ApplyURL:    applyURL,
+		RawContent:  cleanHTML(desc),
+	}, nil
+}
+
+func (c *LinkedInCrawler) fetchURL(ctx context.Context, urlStr string) (string, error) {
+	req, err := http.NewRequestWithContext(ctx, "GET", urlStr, nil)
+	if err != nil {
+		return "", fmt.Errorf("failed to create request: %w", err)
+	}
+
+	req.Header.Set("User-Agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36")
+	req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8")
+	req.Header.Set("Accept-Language", "en-US,en;q=0.9")
+
+	resp, err := c.client.Do(req)
+	if err != nil {
+		return "", fmt.Errorf("HTTP GET failed: %w", err)
+	}
+	defer func() { _ = resp.Body.Close() }()
+
+	if resp.StatusCode != http.StatusOK {
+		return "", fmt.Errorf("HTTP GET returned status %d for %s", resp.StatusCode, urlStr)
+	}
+
+	body, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return "", fmt.Errorf("failed to read response: %w", err)
+	}
+
+	return string(body), nil
+}
+
+func parseLinkedInJobID(u string) string {
+	re := regexp.MustCompile(`/jobs/view/(?:[a-z0-9\-]+-)?(\d+)`)
+	m := re.FindStringSubmatch(u)
+	if len(m) >= 2 {
+		return m[1]
+	}
+	re2 := regexp.MustCompile(`/jobPosting/(\d+)`)
+	m2 := re2.FindStringSubmatch(u)
+	if len(m2) >= 2 {
+		return m2[1]
+	}
+	return ""
+}
+
+func extractRegex(s, pattern string) string {
+	re := regexp.MustCompile(pattern)
+	m := re.FindStringSubmatch(s)
+	if len(m) >= 2 {
+		return m[1]
+	}
+	return ""
+}
+
+func cleanHTML(s string) string {
+	s = strings.ReplaceAll(s, "<b>", "**")
+	s = strings.ReplaceAll(s, "</b>", "**")
+	s = strings.ReplaceAll(s, "<strong>", "**")
+	s = strings.ReplaceAll(s, "</strong>", "**")
+	s = strings.ReplaceAll(s, "<br>", "\n")
+	s = strings.ReplaceAll(s, "<br/>", "\n")
+	s = strings.ReplaceAll(s, "<br />", "\n")
+	s = strings.ReplaceAll(s, "<li>", "- ")
+	s = strings.ReplaceAll(s, "</li>", "\n")
+	s = strings.ReplaceAll(s, "<ul>", "\n")
+	s = strings.ReplaceAll(s, "</ul>", "\n")
+	s = strings.ReplaceAll(s, "<p>", "\n")
+	s = strings.ReplaceAll(s, "</p>", "\n")
+	reTags := regexp.MustCompile(`<[^>]*>`)
+	s = reTags.ReplaceAllString(s, "")
+	return strings.TrimSpace(s)
+}
@@ -0,0 +1,104 @@
+package linkedin
+
+import (
+	"bytes"
+	"context"
+	"io"
+	"net/http"
+	"strings"
+	"testing"
+
+	"github.com/hireme/hireme/internal/crawler/crawlertest"
+)
+
+type roundTripFunc func(req *http.Request) *http.Response
+
+func (f roundTripFunc) RoundTrip(req *http.Request) (*http.Response, error) {
+	return f(req), nil
+}
+
+func TestLinkedInCrawler_DiscoverAndFetch(t *testing.T) {
+	mockClient := &http.Client{
+		Transport: roundTripFunc(func(r *http.Request) *http.Response {
+			if strings.Contains(r.URL.Path, "seeMoreJobPostings") {
+				body := `<html>
+					<li data-row="1">
+						<div class="base-card" data-entity-urn="urn:li:jobPosting:3920192831">
+							<a class="base-card__full-link" href="https://www.linkedin.com/jobs/view/software-engineer-infra-at-linkedin-3920192831"></a>
+						</div>
+					</li>
+				</html>`
+				return &http.Response{
+					StatusCode: http.StatusOK,
+					Body:       io.NopCloser(bytes.NewBufferString(body)),
+					Header:     make(http.Header),
+				}
+			}
+
+			if strings.Contains(r.URL.Path, "3920192831") {
+				body := `<html>
+					<h2 class="top-card-layout__title">Staff Software Engineer - Distributed Graph</h2>
+					<a class="topcard__org-name-link">LinkedIn</a>
+					<span class="topcard__flavor--bullet">Bellevue, WA</span>
+					<div class="show-more-less-html__markup">
+						<p>LinkedIn graph systems scale to billions of edge connections.</p>
+					</div>
+				</html>`
+				return &http.Response{
+					StatusCode: http.StatusOK,
+					Body:       io.NopCloser(bytes.NewBufferString(body)),
+					Header:     make(http.Header),
+				}
+			}
+
+			return &http.Response{
+				StatusCode: http.StatusNotFound,
+				Body:       io.NopCloser(bytes.NewBufferString("not found")),
+				Header:     make(http.Header),
+			}
+		}),
+	}
+
+	c := NewLinkedInCrawlerWithClient(mockClient)
+	if !c.CanHandle("linkedin") {
+		t.Errorf("CanHandle should support linkedin")
+	}
+
+	discovered, err := c.DiscoverJobs(context.Background(), "https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search?keywords=Software+Engineer")
+	if err != nil {
+		t.Fatalf("DiscoverJobs failed: %v", err)
+	}
+
+	if len(discovered) != 1 {
+		t.Fatalf("expected 1 discovered job, got %d", len(discovered))
+	}
+
+	if discovered[0].CanonicalID != "linkedin:3920192831" {
+		t.Errorf("unexpected canonical ID: %s", discovered[0].CanonicalID)
+	}
+
+	extracted, err := c.FetchJobDetail(context.Background(), "https://www.linkedin.com/jobs/view/3920192831")
+	if err != nil {
+		t.Fatalf("FetchJobDetail failed: %v", err)
+	}
+
+	if extracted.Title != "Staff Software Engineer - Distributed Graph" {
+		t.Errorf("unexpected title: %s", extracted.Title)
+	}
+	if extracted.Company != "LinkedIn" {
+		t.Errorf("unexpected company: %s", extracted.Company)
+	}
+	if extracted.Location != "Bellevue, WA" {
+		t.Errorf("unexpected location: %s", extracted.Location)
+	}
+}
+
+func TestLinkedInCrawler_LiveIntegration(t *testing.T) {
+	crawlertest.RunLiveIntegration(t, crawlertest.LiveTestConfig{
+		PlatformName:            "LinkedIn",
+		Crawler:                 NewLinkedInCrawler(),
+		TargetURL:               "https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search?keywords=Software+Engineer&location=Seattle%2C+Washington%2C+United+States",
+		ExpectedCanonicalPrefix: "linkedin:",
+		ExpectedCompany:         "",
+	})
+}