programme-pulse-chat/tests/test_claude_client.py
DJP b70d148b94 Productionise Programme Pulse
Backend
- Routes moved under /api/, JWT bearer auth via @before_request
- DEV_AUTH_BYPASS escape hatch for local dev
- In-memory chat history and report state replaced with Postgres tables
  (preferences, chat_messages, reports, feedback_events) keyed on user
- SQLAlchemy 2.x + Alembic migrations run on container start
- Graceful Airtable failure handling — bad creds no longer 500 the API
- Per-user data isolation via g.user_email from validated token

Frontend
- React + Vite + TypeScript SPA at /programme-pulse/
- MSAL.js (PKCE, sessionStorage, ID token to backend)
- VITE_DEV_AUTH_BYPASS mirrors backend bypass for local dev
- Streaming chat via fetch ReadableStream + SSE parsing
- Charts via chart.js, markdown via react-markdown + remark-gfm
- Full UI parity with the original templates/index.html

Deploy (optical-dev split-build pattern)
- Dockerfile + docker-compose.yml (name: programme-pulse pinned;
  app + Postgres; 127.0.0.1 binding only)
- deploy/apache-programme-pulse.conf.tmpl with flushpackets=on for SSE
- deploy/deploy.sh mirrors OSOP — port auto-pick (5051..5099),
  apache conf render, frontend build in throwaway node container,
  rsync to /var/www/html/programme-pulse, /api/health poll

Tests
- 49 passing; new tests for DB-backed preferences and JWT auth helpers
- SQLite-backed test fixture in tests/conftest.py

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-07 11:08:28 -04:00

130 lines
4.2 KiB
Python

import pytest
from unittest.mock import MagicMock
from src.claude_client import ClaudeClient, MODEL
@pytest.fixture
def mock_anthropic(mocker):
return mocker.patch("src.claude_client.Anthropic")
def _make_response(mock_anthropic, text: str):
text_block = MagicMock()
text_block.type = "text"
text_block.text = text
mock_response = MagicMock()
mock_response.content = [text_block]
mock_anthropic.return_value.messages.create.return_value = mock_response
return mock_response
def test_generate_report_returns_text_content(mock_anthropic):
_make_response(mock_anthropic, "## Executive Summary\n\nTest content.")
client = ClaudeClient("fake_key")
result = client.generate_report("system", "user prompt")
assert result == "## Executive Summary\n\nTest content."
def test_generate_report_skips_thinking_blocks(mock_anthropic):
thinking_block = MagicMock()
thinking_block.type = "thinking"
thinking_block.text = "internal reasoning not for output"
text_block = MagicMock()
text_block.type = "text"
text_block.text = "Report output only"
mock_response = MagicMock()
mock_response.content = [thinking_block, text_block]
mock_anthropic.return_value.messages.create.return_value = mock_response
client = ClaudeClient("fake_key")
result = client.generate_report("system", "user prompt")
assert "internal reasoning" not in result
assert result == "Report output only"
def test_generate_report_uses_correct_model(mock_anthropic):
_make_response(mock_anthropic, "report")
client = ClaudeClient("fake_key")
client.generate_report("sys", "user")
call_kwargs = mock_anthropic.return_value.messages.create.call_args.kwargs
assert call_kwargs["model"] == MODEL
def test_generate_report_uses_max_tokens(mock_anthropic):
_make_response(mock_anthropic, "report")
client = ClaudeClient("fake_key")
client.generate_report("sys", "user")
call_kwargs = mock_anthropic.return_value.messages.create.call_args.kwargs
assert call_kwargs["max_tokens"] == 4096
def test_generate_report_passes_system_prompt(mock_anthropic):
_make_response(mock_anthropic, "report")
client = ClaudeClient("fake_key")
client.generate_report("my system prompt", "user message")
call_kwargs = mock_anthropic.return_value.messages.create.call_args.kwargs
assert call_kwargs["system"] == "my system prompt"
def test_generate_report_concatenates_multiple_text_blocks(mock_anthropic):
block_a = MagicMock()
block_a.type = "text"
block_a.text = "Part one. "
block_b = MagicMock()
block_b.type = "text"
block_b.text = "Part two."
mock_response = MagicMock()
mock_response.content = [block_a, block_b]
mock_anthropic.return_value.messages.create.return_value = mock_response
client = ClaudeClient("fake_key")
result = client.generate_report("sys", "user")
assert result == "Part one. Part two."
def test_chat_returns_text(mocker):
mock_anthropic = mocker.patch("src.claude_client.Anthropic")
mock_client = MagicMock()
mock_anthropic.return_value = mock_client
mock_block = MagicMock()
mock_block.type = "text"
mock_block.text = "Here is the answer."
mock_client.messages.create.return_value.content = [mock_block]
from src.claude_client import ClaudeClient
client = ClaudeClient("fake_key")
result = client.chat(
messages=[{"role": "user", "content": "What is blocked?"}],
system_prompt="You are an assistant."
)
assert result == "Here is the answer."
def test_chat_passes_messages_and_system(mocker):
mock_anthropic = mocker.patch("src.claude_client.Anthropic")
mock_client = MagicMock()
mock_anthropic.return_value = mock_client
mock_block = MagicMock()
mock_block.type = "text"
mock_block.text = "response"
mock_client.messages.create.return_value.content = [mock_block]
from src.claude_client import ClaudeClient
client = ClaudeClient("fake_key")
messages = [{"role": "user", "content": "hello"}]
client.chat(messages=messages, system_prompt="sys")
call_kwargs = mock_client.messages.create.call_args.kwargs
assert call_kwargs["system"] == "sys"
assert call_kwargs["messages"] == messages
assert call_kwargs["max_tokens"] == 8192