feat(p26): PPT 視覺審核 daily 22:00 cron — minicpm-v 自動掃當天新生 .pptx
All checks were successful
CD Pipeline / deploy (push) Successful in 2m54s

Operation Ollama-First v5.0 / Phase 26 — PPT 自我審視整合

services/ppt_vision_service.py 擴充:
- check_ppt_file(pptx_path, max_slides=5) — 整檔視覺檢查
  • LibreOffice headless 轉每張 slide 為 png
  • 對前 N 張跑 check_image
  • 彙總 issues + 平均 confidence
  • fail-safe:LibreOffice 不在 / 轉檔失敗 → 回 skip 不阻擋
- audit_recent_ppts(reports_dir, hours=24, max_files=10)
  • 掃 reports/ 過去 24h 新生 .pptx(getmtime filter)
  • 對每個檔跑 check_ppt_file
  • 彙總總 issues
- push_ppt_audit_to_telegram(summary)
  • 有 issues 才推 Telegram(避免「無問題」洗版)
  • 每檔最多 3 張 slide / 每張 2 個 issue 列出

run_scheduler.py — 每日 22:00 cron
- run_ppt_vision_audit task wrapper
- PPT_VISION_ENABLED=false 時 service 內部 skip(不打 LLM)

設計哲學:
不動既有 5 個 prs.save() 呼叫點(risk 高)→ 改寫獨立 daily cron 集中處理
零侵入 PPT 生成主流程 / 零 risk regression / feature flag OFF 預設

部署需求:
LibreOffice headless(apt install libreoffice)— 不在則 cron task 自動 skip + log

tests/test_ppt_vision_audit.py (9 tests 全綠)
- flag OFF skip / 目錄不存在 / 無 .pptx
- 舊檔(>hours)filter / LibreOffice 不在 fail-safe
- check_ppt_file flag/missing 容錯
- Telegram 推播:無 issues 不推 / 有 issues 推

regression: ppt_vision_service 既有 6 tests 全綠

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
OoO
2026-05-04 11:16:11 +08:00
parent bd32e04dad
commit 72a7c385d5
3 changed files with 353 additions and 0 deletions

View File

@@ -0,0 +1,149 @@
"""
tests/test_ppt_vision_audit.py
─────────────────────────────────────────────────────────────────
Operation Ollama-First v5.0 / Phase 26 — PPT 視覺審核整合驗證
"""
import os
import tempfile
from unittest.mock import patch, MagicMock
import pytest
@pytest.fixture(autouse=True)
def _reset_env(monkeypatch):
monkeypatch.delenv('PPT_VISION_ENABLED', raising=False)
yield
def test_audit_disabled_by_default():
"""flag OFF 時直接 skiperrors 含 disabled marker"""
from services.ppt_vision_service import audit_recent_ppts
summary = audit_recent_ppts(reports_dir='/tmp')
assert summary['total_issues'] == 0
assert any('PPT_VISION_ENABLED=false' in e for e in summary['errors'])
def test_audit_missing_dir(monkeypatch):
monkeypatch.setenv('PPT_VISION_ENABLED', 'true')
from services.ppt_vision_service import audit_recent_ppts
summary = audit_recent_ppts(reports_dir='/tmp/nonexistent_xyz_ppts')
assert summary['total_issues'] == 0
assert any('not found' in e for e in summary['errors'])
def test_audit_no_recent_ppts(monkeypatch):
"""目錄存在但無 .pptx → 0 audit"""
monkeypatch.setenv('PPT_VISION_ENABLED', 'true')
from services.ppt_vision_service import audit_recent_ppts
with tempfile.TemporaryDirectory() as tmpdir:
# 放一個 .txt 干擾
with open(os.path.join(tmpdir, 'not_ppt.txt'), 'w') as f:
f.write('hello')
summary = audit_recent_ppts(reports_dir=tmpdir)
assert summary['audited_files'] == []
assert summary['total_issues'] == 0
def test_audit_filter_old_files(monkeypatch):
"""超過 hours 視窗的舊檔不應 audit"""
monkeypatch.setenv('PPT_VISION_ENABLED', 'true')
from services.ppt_vision_service import audit_recent_ppts
with tempfile.TemporaryDirectory() as tmpdir:
old_path = os.path.join(tmpdir, 'old.pptx')
with open(old_path, 'w') as f:
f.write('fake')
# 改 mtime 到 2 天前
old_time = __import__('time').time() - 2 * 86400
os.utime(old_path, (old_time, old_time))
summary = audit_recent_ppts(reports_dir=tmpdir, hours=24)
assert summary['audited_files'] == []
def test_check_ppt_file_libreoffice_not_installed(monkeypatch):
"""LibreOffice 不在 → fail-safe skip 不 raise"""
monkeypatch.setenv('PPT_VISION_ENABLED', 'true')
from services.ppt_vision_service import PPTVisionService
with tempfile.NamedTemporaryFile(suffix='.pptx', delete=False) as f:
f.write(b'fake pptx')
ppt_path = f.name
try:
with patch('services.ppt_vision_service.subprocess.run',
side_effect=FileNotFoundError('libreoffice')):
svc = PPTVisionService()
result = svc.check_ppt_file(ppt_path)
assert result['success'] is False
assert 'libreoffice not installed' in (result['error'] or '')
finally:
os.unlink(ppt_path)
def test_check_ppt_file_disabled():
"""flag OFF check_ppt_file 也 skip"""
from services.ppt_vision_service import PPTVisionService
with tempfile.NamedTemporaryFile(suffix='.pptx', delete=False) as f:
f.write(b'fake pptx')
ppt_path = f.name
try:
svc = PPTVisionService()
result = svc.check_ppt_file(ppt_path)
assert result['success'] is False
assert 'PPT_VISION_ENABLED=false' in (result['error'] or '')
finally:
os.unlink(ppt_path)
def test_check_ppt_file_missing(monkeypatch):
monkeypatch.setenv('PPT_VISION_ENABLED', 'true')
from services.ppt_vision_service import PPTVisionService
svc = PPTVisionService()
result = svc.check_ppt_file('/tmp/this_pptx_does_not_exist_xyz.pptx')
assert result['success'] is False
assert 'not found' in (result['error'] or '')
def test_push_telegram_skips_when_no_issues():
"""無 issues 不推(避免靜默無問題洗版)"""
from services.ppt_vision_service import push_ppt_audit_to_telegram
summary = {'total_issues': 0, 'audited_files': [{'path': 'a.pptx', 'issues': 0}]}
assert push_ppt_audit_to_telegram(summary) is False
def test_push_telegram_with_issues():
"""有 issues 才推 Telegram"""
from services.ppt_vision_service import push_ppt_audit_to_telegram
summary = {
'total_issues': 3,
'audited_files': [
{
'path': '/tmp/test.pptx',
'slides_checked': 2,
'issues': 3,
'issues_by_slide': [(1, ['⚠️ 圖表被切', '⚠️ 文字溢出']), (2, ['⚠️ 配色衝突'])],
},
],
}
with patch('services.telegram_templates._send_telegram_raw') as mock_send:
result = push_ppt_audit_to_telegram(summary)
assert result is True
mock_send.assert_called_once()
msg = mock_send.call_args[0][0]
assert 'PPT 視覺審核' in msg
assert '3 issues' in msg or '3' in msg