test(ci): add service-specific tests to PR workflow and update test data

.github/workflows/pr-test.yaml

@@ -9,8 +9,21 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  e2e-test:
+  test:
     runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        test-group:
+          - name: "Agent Service"
+            command: "python services/AgentService/test_api.py"
+          - name: "PDF Service"
+            command: "python services/PDFService/test_api.py"
+          - name: "TTS Service"
+            command: "python services/TTSService/test_api.py"
+          - name: "Monologue E2E"
+            command: "python tests/test.py --monologue --target bofa-context.pdf --context citi-context.pdf"
+          - name: "Podcast E2E"
+            command: "python tests/test.py --target bofa-context.pdf --context citi-context.pdf"
 
     steps:
       - uses: actions/checkout@v4
@@ -32,10 +45,12 @@ jobs:
       - name: Install test dependencies
         run: |
           python -m pip install --upgrade pip
-          pip install -r tests/requirements-test.txt  # Make sure this file exists with your test dependencies
+          pip install -r tests/requirements-test.txt
 
-      - name: Run monologue E2E tests
-        run: python tests/test.py --monologue --target bofa-context.pdf --context citi-context.pdf
+      - name: Pip install shared package
+        run: |
+          cd shared
+          pip install .
 
-      - name: Run podcast E2E tests
-        run: python tests/test.py --target bofa-context.pdf --context citi-context.pdf
+      - name: Run ${{ matrix.test-group.name }} tests
+        run: ${{ matrix.test-group.command }}

services/AgentService/test_api.py

@@ -4,6 +4,7 @@
 import time
 from shared.api_types import TranscriptionRequest
 from shared.pdf_types import PDFMetadata
+from datetime import datetime
 
 
 def test_transcribe_api():
@@ -13,35 +14,42 @@ def test_transcribe_api():
 
     # Create a proper TranscriptionRequest
     pdf_metadata_1 = PDFMetadata(
-        filename="sample.pdf", markdown="Sample markdown content", summary=""
+        filename="sample.pdf",
+        type="target",
+        markdown="Sample markdown content",
+        summary="",
+        status="success",
+        created_at=datetime.utcnow(),
     )
 
     pdf_metadata_2 = PDFMetadata(
-        filename="sample2.pdf", markdown="Sample markdown content 2", summary=""
-    )
-
-    pdf_metadata_3 = PDFMetadata(
-        filename="sample3.pdf", markdown="Sample markdown content 3", summary=""
+        filename="sample2.pdf",
+        type="context",
+        markdown="Sample markdown content 2",
+        summary="",
+        status="success",
+        created_at=datetime.utcnow(),
     )
 
     request = TranscriptionRequest(
         # TranscriptionParams fields
+        userId="test-agent-service",
         name="Test Podcast",
         duration=2,  # Duration in minutes
+        monologue=True,
         speaker_1_name="Host",
-        speaker_2_name="Guest",
         voice_mapping={
-            "speaker-1": "iP95p4xoKVk53GoZ742B",  # Example voice ID
-            "speaker-2": "9BWtsMINqrJLrRacOk9x",  # Example voice ID
+            "speaker-1": "iP95p4xoKVk53GoZ742B",
         },
-        guide="Sample focus instructions",  # Optional
+        guide="Sample focus instructions",
         # TranscriptionRequest specific fields
-        pdf_metadata=[pdf_metadata_1, pdf_metadata_2, pdf_metadata_3],
+        pdf_metadata=[pdf_metadata_1, pdf_metadata_2],
         job_id="test-job-123",
+        vdb_task=False,
     )
 
     # Send POST request
-    response = requests.post(TRANSCRIBE_URL, json=request.model_dump())
+    response = requests.post(TRANSCRIBE_URL, json=request.model_dump(mode="json"))
 
     # Check if the request was successful
     assert (

services/PDFService/test_api.py

@@ -190,9 +190,11 @@ def test_health_endpoint():
 
 def main():
     """Main entry point for the test script"""
-    if len(sys.argv) < 2:
-        print("Usage: python test_api.py <path_to_pdf_file1> [path_to_pdf_file2 ...]")
-        sys.exit(1)
+    # Hardcode two sample files
+    sample_files = [
+        "samples/bofa-context.pdf",
+        "samples/citi-context.pdf"
+    ]
 
     print("Running PDF Service API tests...")
 
@@ -201,7 +203,7 @@ def main():
         sys.exit(1)
 
     print("\nTest 2: Valid PDF conversion")
-    if not test_convert_pdf_endpoint(sys.argv[1:]):
+    if not test_convert_pdf_endpoint(sample_files):
         sys.exit(1)
 
     print("\nTest 3: Invalid file handling")

services/TTSService/test_api.py

@@ -51,7 +51,7 @@ def test_tts_api():
 
     try:
         # Load sample JSON data
-        with open("sample.json", "r") as f:
+        with open("services/TTSService/sample.json", "r") as f:
             data = json.load(f)
 
         # Add job_id if not present

tests/requirements-test.txt

@@ -1,2 +1,4 @@
 requests
-websockets
+websockets
+ujson
+fastapi