Improve CV rewrite flow and parser accuracy

2026-04-01 11:30:37 +02:00
parent f402213526
commit f22c6791a7
9 changed files with 581 additions and 55 deletions
@@ -0,0 +1,139 @@
+using System.Reflection;
+using System.Text.Json;
+using JobTrackerApi.Controllers;
+using JobTrackerApi.Models;
+using JobTrackerApi.Services;
+using JobTrackerApi.Tests.TestSupport;
+using Microsoft.AspNetCore.Http;
+using Microsoft.AspNetCore.Identity;
+using Microsoft.AspNetCore.Mvc;
+using Microsoft.Extensions.Configuration;
+using Microsoft.Extensions.Hosting;
+using Moq;
+using Xunit;
+
+namespace JobTrackerApi.Tests;
+
+public sealed class CvCorpusHarnessTests
+{
+    private static readonly string CorpusRoot = "/home/pi/cvs";
+
+    [Fact]
+    public async Task Local_cv_corpus_harness_produces_repeatable_parse_report_when_available()
+    {
+        if (!Directory.Exists(CorpusRoot)) return;
+
+        var files = Directory.EnumerateFiles(CorpusRoot, "*.*", SearchOption.TopDirectoryOnly)
+            .Where(path => path.EndsWith(".pdf", StringComparison.OrdinalIgnoreCase)
+                || path.EndsWith(".docx", StringComparison.OrdinalIgnoreCase)
+                || path.EndsWith(".txt", StringComparison.OrdinalIgnoreCase)
+                || path.EndsWith(".md", StringComparison.OrdinalIgnoreCase))
+            .OrderBy(path => path, StringComparer.OrdinalIgnoreCase)
+            .Take(8)
+            .ToList();
+
+        if (files.Count == 0) return;
+
+        var user = new ApplicationUser { Id = "user-1", ProfileCvText = "seed" };
+        var userManager = TestHostFactory.CreateUserManager();
+        userManager.Setup(x => x.GetUserAsync(It.IsAny<System.Security.Claims.ClaimsPrincipal>())).ReturnsAsync(user);
+        userManager.Setup(x => x.UpdateAsync(It.IsAny<ApplicationUser>())).ReturnsAsync(IdentityResult.Success);
+
+        var aiService = new Mock<ISummarizerService>();
+        aiService.Setup(x => x.SummarizeSectionAsync(It.Is<string>(instruction => instruction.Contains("Extract this CV into structured JSON", StringComparison.Ordinal)), It.IsAny<string>(), 3200, 900)).ReturnsAsync(string.Empty);
+        aiService.Setup(x => x.SummarizeSectionAsync(It.Is<string>(instruction => instruction.Contains("Reconstruct this CV text extracted from a PDF", StringComparison.Ordinal)), It.IsAny<string>(), 2800, 900)).ReturnsAsync((string _, string text, int _, int __) => text);
+
+        await using var db = TestHostFactory.CreateInMemoryDb();
+        var paths = CreatePaths();
+        var controller = new ProfileCvController(userManager.Object, aiService.Object, db, paths, NoOpCvAiClassifier.Instance)
+        {
+            ControllerContext = new ControllerContext { HttpContext = new DefaultHttpContext() }
+        };
+
+        var extractMethod = typeof(ProfileCvController).GetMethod("ExtractTextAsync", BindingFlags.NonPublic | BindingFlags.Static);
+        var buildMethod = typeof(ProfileCvController).GetMethod("BuildStructuredCvAsync", BindingFlags.NonPublic | BindingFlags.Instance);
+        Assert.NotNull(extractMethod);
+        Assert.NotNull(buildMethod);
+
+        var report = new List<object>();
+        foreach (var path in files)
+        {
+            await using var stream = File.OpenRead(path);
+            var fileName = Path.GetFileName(path);
+            var formFile = new FormFile(stream, 0, stream.Length, "file", fileName)
+            {
+                Headers = new HeaderDictionary(),
+                ContentType = GuessContentType(path)
+            };
+
+            var extension = Path.GetExtension(path);
+            var extractTask = (Task<string>)extractMethod!.Invoke(null, new object[] { formFile, extension })!;
+            var text = await extractTask;
+            Assert.False(string.IsNullOrWhiteSpace(text));
+
+            var buildTask = (Task<StructuredCvProfile>)buildMethod!.Invoke(controller, new object[] { text, CancellationToken.None })!;
+            var structured = await buildTask;
+            Assert.NotNull(structured);
+
+            report.Add(new
+            {
+                file = fileName,
+                characters = text.Length,
+                contactLocation = structured.Contact.Location,
+                firstJob = structured.Jobs.FirstOrDefault()?.Title,
+                firstJobLocation = structured.Jobs.FirstOrDefault()?.Location,
+                firstEducation = structured.Education.FirstOrDefault()?.Qualification,
+                firstEducationLocation = structured.Education.FirstOrDefault()?.Location,
+                suspiciousLocations = structured.Jobs.Select(job => job.Location)
+                    .Concat(structured.Education.Select(education => education.Location))
+                    .Append(structured.Contact.Location)
+                    .Where(value => !string.IsNullOrWhiteSpace(value))
+                    .Where(LooksSuspiciousLocation)
+                    .ToList()
+            });
+        }
+
+        var reportPath = Path.Combine(Path.GetTempPath(), $"jobtracker-cv-corpus-{DateTime.UtcNow:yyyyMMddHHmmss}.json");
+        await File.WriteAllTextAsync(reportPath, JsonSerializer.Serialize(report, new JsonSerializerOptions { WriteIndented = true }));
+
+        Assert.True(report.Count > 0);
+    }
+
+    private static bool LooksSuspiciousLocation(string? value)
+    {
+        if (string.IsNullOrWhiteSpace(value)) return false;
+        return value.Contains("Python", StringComparison.OrdinalIgnoreCase)
+            || value.Contains("Ruby", StringComparison.OrdinalIgnoreCase)
+            || value.Contains(" S A L E S ", StringComparison.OrdinalIgnoreCase)
+            || value.Any(char.IsDigit);
+    }
+
+    private static string GuessContentType(string path)
+    {
+        return Path.GetExtension(path).ToLowerInvariant() switch
+        {
+            ".pdf" => "application/pdf",
+            ".docx" => "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+            ".md" => "text/markdown",
+            _ => "text/plain"
+        };
+    }
+
+    private static AppPaths CreatePaths()
+    {
+        var tempRoot = Path.Combine(Path.GetTempPath(), $"jobtracker-cv-corpus-{Guid.NewGuid():N}");
+        Directory.CreateDirectory(tempRoot);
+
+        var config = new ConfigurationBuilder()
+            .AddInMemoryCollection(new Dictionary<string, string?>
+            {
+                ["Data:Root"] = tempRoot,
+                ["Data:CvArtifactsRoot"] = Path.Combine(tempRoot, "CvArtifacts")
+            })
+            .Build();
+
+        var env = new Mock<IHostEnvironment>();
+        env.SetupGet(x => x.ContentRootPath).Returns(tempRoot);
+        return new AppPaths(config, env.Object);
+    }
+}