fix: apply Phase 1-2 correctness and constant extraction from PR #273

- Replace datetime.utcnow() with datetime.now(timezone.utc) in 4 files
- Extract duplicate literals to constants in 5 files
  - models.py: "files.id" → _FILES_ID_FK
  - upload_to_email.py: "logo.png" → _LOGO_FILENAME
  - general.py: "%B %d, %Y" → _DATE_DISPLAY_FORMAT
  - files.py: "File not found" → _FILE_NOT_FOUND
  - upload_to_google_drive.py: Google token URL → _GOOGLE_TOKEN_URL

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>

app/api/diagnostic.py

@@ -64,7 +64,7 @@ async def test_notification(request: Request):
     # Add request_time to request.state
     import datetime
 
-    request.state.request_time = datetime.datetime.utcnow().isoformat()
+    request.state.request_time = datetime.datetime.now(datetime.timezone.utc).isoformat()
     """
     Send a test notification through all configured notification channels
     """

app/models.py

@@ -4,6 +4,9 @@ from sqlalchemy import Boolean, Column, DateTime, ForeignKey, Integer, String, T
 
 from app.database import Base
 
+# Foreign key constants
+_FILES_ID_FK = "files.id"
+
 
 class DocumentMetadata(Base):
     __tablename__ = "documents"
@@ -50,7 +53,7 @@ class FileRecord(Base):
     is_duplicate = Column(Boolean, default=False, nullable=False, index=True)
 
     # If this is a duplicate, record the ID of the original file for reference
-    duplicate_of_id = Column(Integer, ForeignKey("files.id"), nullable=True)
+    duplicate_of_id = Column(Integer, ForeignKey(_FILES_ID_FK), nullable=True)
 
     # Timestamp when we inserted this record
     created_at = Column(DateTime(timezone=True), server_default=func.now())
@@ -65,7 +68,7 @@ class FileProcessingStep(Base):
     __tablename__ = "file_processing_steps"
 
     id = Column(Integer, primary_key=True, index=True)
-    file_id = Column(Integer, ForeignKey("files.id"), nullable=False, index=True)
+    file_id = Column(Integer, ForeignKey(_FILES_ID_FK), nullable=False, index=True)
     step_name = Column(String, nullable=False, index=True)  # e.g., "hash_file", "upload_to_dropbox"
     status = Column(String, nullable=False)  # "pending", "in_progress", "success", "failure", "skipped"
     started_at = Column(DateTime(timezone=True), nullable=True)  # When step started
@@ -80,7 +83,7 @@ class FileProcessingStep(Base):
 class ProcessingLog(Base):
     __tablename__ = "processing_logs"
     id = Column(Integer, primary_key=True, index=True)
-    file_id = Column(Integer, ForeignKey("files.id"), nullable=True)  # Optional file association
+    file_id = Column(Integer, ForeignKey(_FILES_ID_FK), nullable=True)  # Optional file association
     task_id = Column(String, index=True)  # Celery task ID
     step_name = Column(String)  # e.g., "OCR", "convert_to_pdf", "upload_s3"
     status = Column(String)  # "pending", "in_progress", "success", "failure"

app/tasks/monitor_stalled_steps.py

@@ -6,7 +6,7 @@ that have been stuck in "in_progress" state for too long and mark them as failed
 """
 
 import logging
-from datetime import datetime
+from datetime import datetime, timezone
 
 from app.celery_app import celery
 from app.database import SessionLocal
@@ -37,12 +37,12 @@ def monitor_stalled_steps():
 
             if stalled_count > 0:
                 logger.warning(
-                    f"[{datetime.utcnow().isoformat()}] "
+                    f"[{datetime.now(timezone.utc).isoformat()}] "
                     f"Recovered {stalled_count} stalled step(s). "
                     f"Marked as failed due to timeout."
                 )
             else:
-                logger.debug(f"[{datetime.utcnow().isoformat()}] No stalled steps found.")
+                logger.debug(f"[{datetime.now(timezone.utc).isoformat()}] No stalled steps found.")
 
             return {"recovered": stalled_count}
 

app/tasks/upload_to_email.py

@@ -20,6 +20,9 @@ from app.utils import log_task_progress
 
 logger = logging.getLogger(__name__)
 
+# Constants
+_LOGO_FILENAME = "logo.png"
+
 
 def get_email_template(template_name="default.html"):
     """
@@ -87,16 +90,16 @@ def attach_logo(msg):
     """Attach the DocuElevate logo to the email with proper Content-ID."""
     try:
         # Try to find logo in workdir first (for customization)
-        custom_logo_path = os.path.join(settings.workdir, "templates", "email", "logo.png")
+        custom_logo_path = os.path.join(settings.workdir, "templates", "email", _LOGO_FILENAME)
         if os.path.exists(custom_logo_path):
             logo_path = custom_logo_path
         else:
             # Use built-in logo
             app_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
-            logo_path = os.path.join(app_dir, "static", "logo.png")
+            logo_path = os.path.join(app_dir, "static", _LOGO_FILENAME)
             # Fallback to logo in frontend/static if app/static doesn't exist
             if not os.path.exists(logo_path):
-                logo_path = os.path.join(app_dir, "..", "frontend", "static", "logo.png")
+                logo_path = os.path.join(app_dir, "..", "frontend", "static", _LOGO_FILENAME)
 
         if os.path.exists(logo_path):
             with open(logo_path, "rb") as img:
@@ -106,7 +109,7 @@ def attach_logo(msg):
             mimetype = "image/svg+xml" if logo_path.endswith(".svg") else "image/png"
             logo_attach = MIMEImage(logo_data, mimetype)
             logo_attach.add_header("Content-ID", "<logo>")
-            logo_attach.add_header("Content-Disposition", "inline", filename="logo.png")
+            logo_attach.add_header("Content-Disposition", "inline", filename=_LOGO_FILENAME)
             msg.attach(logo_attach)
             logger.info(f"Logo attached from {logo_path}")
             return True

app/tasks/upload_to_google_drive.py

@@ -20,6 +20,9 @@ from app.utils import log_task_progress
 
 logger = logging.getLogger(__name__)
 
+# Google OAuth constants
+_GOOGLE_TOKEN_URL = "https://oauth2.googleapis.com/token"
+
 
 def get_drive_service_oauth():
     """
@@ -40,7 +43,7 @@ def get_drive_service_oauth():
         credentials = OAuthCredentials(
             None,  # No access token initially, will be refreshed
             refresh_token=settings.google_drive_refresh_token,
-            token_uri="https://oauth2.googleapis.com/token",
+            token_uri=_GOOGLE_TOKEN_URL,
             client_id=settings.google_drive_client_id,
             client_secret=settings.google_drive_client_secret,
             # Use only drive.file scope

app/utils/logging.py

@@ -1,7 +1,7 @@
 import logging
 import threading
 from collections import defaultdict
-from datetime import datetime
+from datetime import datetime, timezone
 
 from app.database import SessionLocal
 from app.models import FileProcessingStep, ProcessingLog
@@ -106,7 +106,7 @@ def log_task_progress(task_id, step_name, status, message=None, file_id=None, de
                 .first()
             )
 
-            now = datetime.utcnow()
+            now = datetime.now(timezone.utc)
 
             if not step_record:
                 # Create new step record

app/utils/step_timeout.py

@@ -6,7 +6,7 @@ This prevents files from getting stuck in "pending" state when processing crashe
 """
 
 import logging
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 from typing import Optional
 
 from sqlalchemy.orm import Session
@@ -54,7 +54,7 @@ def mark_stalled_steps_as_failed(
     if timeout_seconds is None:
         timeout_seconds = get_step_timeout()
 
-    now = datetime.utcnow()
+    now = datetime.now(timezone.utc)
     cutoff_time = now - timedelta(seconds=timeout_seconds)
 
     # Query for stalled steps

app/views/files.py

@@ -13,6 +13,9 @@ from app.views.base import APIRouter, get_db, logger, require_login, templates
 
 router = APIRouter()
 
+# Error message constants
+_FILE_NOT_FOUND = "File not found"
+
 
 @router.get("/files")
 @require_login
@@ -433,7 +436,7 @@ def preview_original_file(request: Request, file_id: int, db: Session = Depends(
 
     file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
     if not file_record:
-        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="File not found")
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail=_FILE_NOT_FOUND)
 
     if not file_record.original_file_path or not os.path.exists(file_record.original_file_path):
         raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Original file not found on disk")
@@ -460,7 +463,7 @@ def preview_processed_file(request: Request, file_id: int, db: Session = Depends
 
     file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
     if not file_record:
-        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="File not found")
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail=_FILE_NOT_FOUND)
 
     if not file_record.processed_file_path or not os.path.exists(file_record.processed_file_path):
         raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Processed file not found on disk")
@@ -487,7 +490,7 @@ def get_original_text(request: Request, file_id: int, db: Session = Depends(get_
 
     file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
     if not file_record:
-        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="File not found")
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail=_FILE_NOT_FOUND)
 
     if not file_record.original_file_path or not os.path.exists(file_record.original_file_path):
         raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Original file not found on disk")
@@ -527,7 +530,7 @@ def get_processed_text(request: Request, file_id: int, db: Session = Depends(get
 
     file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
     if not file_record:
-        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="File not found")
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail=_FILE_NOT_FOUND)
 
     if not file_record.processed_file_path or not os.path.exists(file_record.processed_file_path):
         raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Processed file not found on disk")

app/views/general.py

@@ -14,6 +14,9 @@ from app.views.base import APIRouter, get_db, logger, require_login, templates
 
 router = APIRouter()
 
+# Date format constant
+_DATE_DISPLAY_FORMAT = "%B %d, %Y"
+
 
 @router.get("/", include_in_schema=False)
 async def serve_index(request: Request, db: Session = Depends(get_db)):
@@ -82,7 +85,7 @@ async def serve_about(request: Request):
 async def serve_privacy(request: Request):
     """Serve the privacy policy page."""
     # Pass the current date for the "Last Updated" field
-    current_date = date.today().strftime("%B %d, %Y")
+    current_date = date.today().strftime(_DATE_DISPLAY_FORMAT)
     return templates.TemplateResponse("privacy.html", {"request": request, "current_date": current_date})
 
 
@@ -148,12 +151,12 @@ Please visit http://www.apache.org/licenses/LICENSE-2.0 for the complete license
 @router.get("/cookies", include_in_schema=False)
 async def serve_cookies(request: Request):
     """Serve the cookie policy page."""
-    current_date = date.today().strftime("%B %d, %Y")
+    current_date = date.today().strftime(_DATE_DISPLAY_FORMAT)
     return templates.TemplateResponse("cookies.html", {"request": request, "current_date": current_date})
 
 
 @router.get("/terms", include_in_schema=False)
 async def serve_terms(request: Request):
     """Serve the terms of service page."""
-    current_date = date.today().strftime("%B %d, %Y")
+    current_date = date.today().strftime(_DATE_DISPLAY_FORMAT)
     return templates.TemplateResponse("terms.html", {"request": request, "current_date": current_date})