First Commit

2026-03-21 11:55:27 +01:00
commit 2e8a29b4d0
1757 changed files with 166084 additions and 0 deletions
@@ -0,0 +1,48 @@
+using System;
+
+namespace JobTrackerApi.Services.JobImport.Plugins;
+
+public sealed class FinnPlugin : IJobSitePlugin
+{
+    public bool CanHandle(string url) => url.Contains("finn.no", StringComparison.OrdinalIgnoreCase);
+
+    public JobImportResult Parse(string html, string url)
+    {
+        var meta = HtmlExtract.ReadMeta(html);
+        var title = meta.TryGetValue("og:title", out var t) ? t : HtmlExtract.ReadTitle(html);
+        var desc = meta.TryGetValue("og:description", out var d) ? d : null;
+        var company = ExtractCompanyFromTitle(title);
+
+        return new JobImportResult
+        {
+            SourceUrl = url,
+            Title = CleanTitle(title),
+            Company = company,
+            Location = meta.TryGetValue("job:location", out var loc) ? loc : null,
+            Description = HtmlExtract.ToPlainText(desc),
+            Parser = "finn",
+            Success = !string.IsNullOrWhiteSpace(title) && !string.IsNullOrWhiteSpace(desc),
+        };
+    }
+
+    private static string? CleanTitle(string? title)
+    {
+        if (string.IsNullOrWhiteSpace(title)) return null;
+        // FINN often appends " - FINN.no" etc.
+        var s = title.Replace(" - FINN.no", "", StringComparison.OrdinalIgnoreCase).Trim();
+        return s.Length == 0 ? title : s;
+    }
+
+    private static string? ExtractCompanyFromTitle(string? title)
+    {
+        if (string.IsNullOrWhiteSpace(title)) return null;
+        // Common pattern: "Role hos Company" / "Role - Company"
+        var s = title;
+        var idx = s.LastIndexOf(" - ", StringComparison.Ordinal);
+        if (idx > 0 && idx < s.Length - 3) return s[(idx + 3)..].Trim();
+        idx = s.LastIndexOf(" hos ", StringComparison.OrdinalIgnoreCase);
+        if (idx > 0 && idx < s.Length - 5) return s[(idx + 5)..].Trim();
+        return null;
+    }
+}
+
@@ -0,0 +1,25 @@
+using System;
+
+namespace JobTrackerApi.Services.JobImport.Plugins;
+
+public sealed class JobbnorgePlugin : IJobSitePlugin
+{
+    public bool CanHandle(string url) => url.Contains("jobbnorge.no", StringComparison.OrdinalIgnoreCase);
+
+    public JobImportResult Parse(string html, string url)
+    {
+        var meta = HtmlExtract.ReadMeta(html);
+        var title = meta.TryGetValue("og:title", out var t) ? t : HtmlExtract.ReadTitle(html);
+        var desc = meta.TryGetValue("og:description", out var d) ? d : null;
+
+        return new JobImportResult
+        {
+            SourceUrl = url,
+            Title = title,
+            Description = HtmlExtract.ToPlainText(desc),
+            Parser = "jobbnorge",
+            Success = !string.IsNullOrWhiteSpace(title) && !string.IsNullOrWhiteSpace(desc),
+        };
+    }
+}
+
@@ -0,0 +1,27 @@
+using System;
+
+namespace JobTrackerApi.Services.JobImport.Plugins;
+
+public sealed class LinkedInPlugin : IJobSitePlugin
+{
+    public bool CanHandle(string url) => url.Contains("linkedin.com/jobs", StringComparison.OrdinalIgnoreCase);
+
+    public JobImportResult Parse(string html, string url)
+    {
+        // LinkedIn heavily relies on JS; meta tags are often the best available without a headless browser.
+        var meta = HtmlExtract.ReadMeta(html);
+        var title = meta.TryGetValue("og:title", out var t) ? t : HtmlExtract.ReadTitle(html);
+        var desc = meta.TryGetValue("og:description", out var d) ? d : null;
+
+        return new JobImportResult
+        {
+            SourceUrl = url,
+            Title = title,
+            Company = meta.TryGetValue("og:site_name", out var sn) ? sn : null,
+            Description = HtmlExtract.ToPlainText(desc),
+            Parser = "linkedin",
+            Success = !string.IsNullOrWhiteSpace(title) && !string.IsNullOrWhiteSpace(desc),
+        };
+    }
+}
+
@@ -0,0 +1,29 @@
+using System;
+
+namespace JobTrackerApi.Services.JobImport.Plugins;
+
+public sealed class NavPlugin : IJobSitePlugin
+{
+    public bool CanHandle(string url)
+        => url.Contains("arbeidsplassen.nav.no", StringComparison.OrdinalIgnoreCase) ||
+           url.Contains("nav.no", StringComparison.OrdinalIgnoreCase);
+
+    public JobImportResult Parse(string html, string url)
+    {
+        var meta = HtmlExtract.ReadMeta(html);
+        var title = meta.TryGetValue("og:title", out var t) ? t : HtmlExtract.ReadTitle(html);
+        var desc = meta.TryGetValue("og:description", out var d) ? d : null;
+        var siteName = meta.TryGetValue("og:site_name", out var sn) ? sn : null;
+
+        return new JobImportResult
+        {
+            SourceUrl = url,
+            Title = title,
+            Company = siteName, // better than nothing; universal parser often gets this anyway.
+            Description = HtmlExtract.ToPlainText(desc),
+            Parser = "nav",
+            Success = !string.IsNullOrWhiteSpace(title) && !string.IsNullOrWhiteSpace(desc),
+        };
+    }
+}
+