DocumentService/DocumentService.Tests.Live/API/PdfAttachmentLiveTests.cs

using DocumentService.Application.Common.DTOs;
using FluentAssertions;

namespace DocumentService.Tests.Live.API;

/// <summary>
/// Live tests for PDF attachment endpoints (check + extract).
/// Uses PdfAttachmentClient — routes managed by the client.
/// Target: http://172.24.12.39:9393/
/// </summary>
[Collection("Live API")]
public class PdfAttachmentLiveTests : LiveApiTestBase
{
    // ──────────────────────────────────────────────────────────────
    // CheckAsync(Stream) → multipart → /api/pdf/attachments/check
    // ──────────────────────────────────────────────────────────────

    [Fact]
    public async Task Check_Stream_PdfWithoutAttachments_ReturnsHasAttachmentsFalse()
    {
        using var stream = LoadPdfStream("valid.pdf");

        var result = await AttachmentClient.CheckAsync(stream);

        result.Should().NotBeNull();
        result.HasAttachments.Should().BeFalse();
        result.AttachmentCount.Should().Be(0);
        result.Attachments.Should().BeEmpty();
    }

    [Fact]
    public async Task Check_Stream_PdfWith6Attachments_ReturnsAllAttachmentsWithMetadata()
    {
        using var stream = LoadPdfStream("pdfWithMoreThanOneAttachment.pdf");

        var result = await AttachmentClient.CheckAsync(stream);

        result.HasAttachments.Should().BeTrue();
        result.AttachmentCount.Should().Be(6, "PDF has exactly 6 attachments");
        result.Attachments.Should().HaveCount(6);

        foreach (var attachment in result.Attachments)
        {
            attachment.FileName.Should().NotBeNullOrEmpty();
            attachment.MimeType.Should().NotBeNullOrEmpty();
            attachment.Size.Should().BeGreaterThan(0);
        }
    }

    [Fact]
    public async Task Check_Stream_PdfWithSwissQrCode_HasNoAttachments()
    {
        // QR code PDF has no embedded file attachments
        using var stream = LoadPdfStream("pdfWithSwissQRCode.pdf");

        var result = await AttachmentClient.CheckAsync(stream);

        result.HasAttachments.Should().BeFalse();
        result.AttachmentCount.Should().Be(0);
    }

    [Fact]
    public async Task Check_Stream_EmptyStream_Throws()
    {
        using var stream = new MemoryStream(Array.Empty<byte>());

        var act = async () => await AttachmentClient.CheckAsync(stream);

        await act.Should().ThrowAsync<HttpRequestException>();
    }

    [Fact]
    public async Task Check_Stream_CorruptedContent_Throws()
    {
        byte[] garbage = "This is not a valid PDF content"u8.ToArray();
        using var stream = new MemoryStream(garbage);

        var act = async () => await AttachmentClient.CheckAsync(stream);

        await act.Should().ThrowAsync<HttpRequestException>(
            "corrupted PDF should return 500 → client throws");
    }

    // ──────────────────────────────────────────────────────────────
    // CheckAsync(byte[]) → Base64 JSON → /api/pdf/attachments/check
    // ──────────────────────────────────────────────────────────────

    [Fact]
    public async Task Check_Bytes_PdfWithoutAttachments_ReturnsHasAttachmentsFalse()
    {
        byte[] bytes = LoadPdfBytes("valid.pdf");

        var result = await AttachmentClient.CheckAsync(bytes);

        result.HasAttachments.Should().BeFalse();
        result.AttachmentCount.Should().Be(0);
    }

    [Fact]
    public async Task Check_Bytes_PdfWith6Attachments_ReturnsAllAttachmentsWithMetadata()
    {
        byte[] bytes = LoadPdfBytes("pdfWithMoreThanOneAttachment.pdf");

        var result = await AttachmentClient.CheckAsync(bytes);

        result.HasAttachments.Should().BeTrue();
        result.AttachmentCount.Should().Be(6);
        result.Attachments.Should().HaveCount(6);

        foreach (var attachment in result.Attachments)
        {
            attachment.FileName.Should().NotBeNullOrEmpty();
            attachment.MimeType.Should().NotBeNullOrEmpty();
            attachment.Size.Should().BeGreaterThan(0);
        }
    }

    [Fact]
    public async Task Check_Bytes_EmptyArray_Throws()
    {
        var act = async () => await AttachmentClient.CheckAsync(Array.Empty<byte>());

        await act.Should().ThrowAsync<HttpRequestException>();
    }

    // ──────────────────────────────────────────────────────────────
    // ExtractAsync(Stream) → multipart → /api/pdf/attachments/extract
    // ──────────────────────────────────────────────────────────────

    [Fact]
    public async Task Extract_Stream_PdfWith6Attachments_Returns6NamedStreams()
    {
        using var stream = LoadPdfStream("pdfWithMoreThanOneAttachment.pdf");

        var attachments = await AttachmentClient.ExtractAsync(stream);

        attachments.Should().HaveCount(6, "PDF contains 6 attachments");

        foreach (var (name, content) in attachments)
        {
            name.Should().NotBeNullOrEmpty();
            content.Length.Should().BeGreaterThan(0);
            content.Dispose();
        }
    }

    [Fact]
    public async Task Extract_Stream_AttachmentNamesAreUnique()
    {
        using var stream = LoadPdfStream("pdfWithMoreThanOneAttachment.pdf");

        var attachments = await AttachmentClient.ExtractAsync(stream);

        try
        {
            attachments.Keys.Should().OnlyHaveUniqueItems(
                "ZIP entry names must be unique within the archive");
        }
        finally
        {
            foreach (var s in attachments.Values) s.Dispose();
        }
    }

    [Fact]
    public async Task Extract_Stream_EmptyStream_Throws()
    {
        using var stream = new MemoryStream(Array.Empty<byte>());

        var act = async () => await AttachmentClient.ExtractAsync(stream);

        await act.Should().ThrowAsync<HttpRequestException>();
    }

    [Fact]
    public async Task Extract_Stream_CorruptedContent_Throws()
    {
        byte[] garbage = "This is not a valid PDF content"u8.ToArray();
        using var stream = new MemoryStream(garbage);

        var act = async () => await AttachmentClient.ExtractAsync(stream);

        await act.Should().ThrowAsync<HttpRequestException>();
    }

    // ──────────────────────────────────────────────────────────────
    // ExtractAsync(byte[]) → Base64 JSON → /api/pdf/attachments/extract
    // ──────────────────────────────────────────────────────────────

    [Fact]
    public async Task Extract_Bytes_PdfWith6Attachments_Returns6NamedStreams()
    {
        byte[] bytes = LoadPdfBytes("pdfWithMoreThanOneAttachment.pdf");

        var attachments = await AttachmentClient.ExtractAsync(bytes);

        attachments.Should().HaveCount(6);

        foreach (var (name, content) in attachments)
        {
            name.Should().NotBeNullOrEmpty();
            content.Length.Should().BeGreaterThan(0);
            content.Dispose();
        }
    }

    [Fact]
    public async Task Extract_Bytes_AttachmentContentsAreNonEmpty()
    {
        byte[] bytes = LoadPdfBytes("pdfWithMoreThanOneAttachment.pdf");

        var attachments = await AttachmentClient.ExtractAsync(bytes);

        try
        {
            foreach (var (_, content) in attachments)
            {
                ((MemoryStream)content).ToArray().Should().NotBeEmpty(
                    "every extracted attachment must have content");
            }
        }
        finally
        {
            foreach (var s in attachments.Values) s.Dispose();
        }
    }

    [Fact]
    public async Task Extract_Bytes_EmptyArray_Throws()
    {
        var act = async () => await AttachmentClient.ExtractAsync(Array.Empty<byte>());

        await act.Should().ThrowAsync<HttpRequestException>();
    }
}