Tighten ruff rules and modernize style

misc/_webservice.py

@@ -96,7 +96,9 @@ with st.expander("Optimization after OCR"):
     png_quality = st.slider(
         "PNG quality", min_value=0, max_value=100, value=75, key="png_quality"
     )
-    jbig2_threshold = st.number_input("JBIG2 threshold", value=0.85, key="jbig2_threshold")
+    jbig2_threshold = st.number_input(
+        "JBIG2 threshold", value=0.85, key="jbig2_threshold"
+    )
 
 with st.expander("Advanced options"):
     jobs = st.slider(
@@ -192,45 +194,47 @@ if uploaded:
         args.append(f"--jbig2-threshold={jbig2_threshold}")
     if jobs:
         args.append(f"--jobs={jobs}")
-    input_file = NamedTemporaryFile(delete=True, suffix=f"_{uploaded.name}")
-    input_file.write(uploaded.getvalue())
-    input_file.flush()
-    input_file.seek(0)
-    args.append(str(input_file.name))
-    output_file = NamedTemporaryFile(delete=True, suffix=".pdf")
-    args.append(str(output_file.name))
+    with NamedTemporaryFile(delete=True, suffix=f"_{uploaded.name}") as input_file:
+        input_file.write(uploaded.getvalue())
+        input_file.flush()
+        input_file.seek(0)
+        args.append(str(input_file.name))
+        with NamedTemporaryFile(delete=True, suffix=".pdf") as output_file:
+            args.append(str(output_file.name))
 
-    st.session_state['running'] = (
-        'run_button' in st.session_state and st.session_state.run_button
-    )
-    if st.button(
-        "Run OCRmyPDF",
-        disabled=st.session_state.get("running", False),
-        key='run_button',
-    ):
-        st.session_state['running'] = True
-        args = [sys.executable, '-u', '-m', "ocrmypdf"] + args
+            st.session_state['running'] = (
+                'run_button' in st.session_state and st.session_state.run_button
+            )
+            if st.button(
+                "Run OCRmyPDF",
+                disabled=st.session_state.get("running", False),
+                key='run_button',
+            ):
+                st.session_state['running'] = True
+                args = [sys.executable, '-u', '-m', "ocrmypdf"] + args
 
-        proc = subprocess.Popen(args, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
-        with st.container(border=True):
-            while proc.poll() is None:
-                line = proc.stderr.readline()
-                if line:
-                    st.html("<code>" + line.decode().strip() + "</code>")
+                proc = subprocess.Popen(
+                    args, stdout=subprocess.PIPE, stderr=subprocess.PIPE
+                )
+                with st.container(border=True):
+                    while proc.poll() is None:
+                        line = proc.stderr.readline()
+                        if line:
+                            st.html("<code>" + line.decode().strip() + "</code>")
 
-        if proc.returncode != 0:
-            st.error(f"ocrmypdf failed with exit code {proc.returncode}")
-            st.session_state['running'] = False
-            st.stop()
+                if proc.returncode != 0:
+                    st.error(f"ocrmypdf failed with exit code {proc.returncode}")
+                    st.session_state['running'] = False
+                    st.stop()
 
-        if Path(output_file.name).stat().st_size == 0:
-            st.error("No output PDF file was generated")
-            st.stop()
+                if Path(output_file.name).stat().st_size == 0:
+                    st.error("No output PDF file was generated")
+                    st.stop()
 
-        st.download_button(
-            label="Download output PDF",
-            data=output_file.read(),
-            file_name=uploaded.name,
-            mime="application/pdf",
-        )
-        st.session_state['running'] = False
+                st.download_button(
+                    label="Download output PDF",
+                    data=output_file.read(),
+                    file_name=uploaded.name,
+                    mime="application/pdf",
+                )
+                st.session_state['running'] = False

misc/batch.py

@@ -39,10 +39,7 @@ script_dir = Path(__file__).parent
 # set archive_dir to a path for backup original documents. Leave empty if not required.
 archive_dir = "/pdfbak"
 
-if len(sys.argv) > 1:
-    start_dir = Path(sys.argv[1])
-else:
-    start_dir = Path(".")
+start_dir = Path(sys.argv[1]) if len(sys.argv) > 1 else Path(".")
 
 if len(sys.argv) > 2:
     log_file = Path(sys.argv[2])

misc/bisect_pdf.py

@@ -3,6 +3,7 @@
 # SPDX-License-Identifier: MIT
 
 """Helper script for bisecting PDFs to find a page with an issue."""
+from __future__ import annotations
 
 import sys
 

misc/watcher.py

@@ -7,12 +7,12 @@
 
 from __future__ import annotations
 
+import datetime as dt
 import json
 import logging
 import shutil
 import sys
 import time
-from datetime import datetime
 from enum import Enum
 from pathlib import Path
 from typing import Annotated, Any
@@ -48,7 +48,7 @@ class LoggingLevelEnum(str, Enum):
 def get_output_path(root: Path, basename: str, output_dir_year_month: bool) -> Path:
     assert '/' not in basename, "basename must not contain '/'"
     if output_dir_year_month:
-        today = datetime.today()
+        today = dt.datetime.today()
         output_directory_year_month = root / str(today.year) / f'{today.month:02d}'
         if not output_directory_year_month.exists():
             output_directory_year_month.mkdir(parents=True, exist_ok=True)
@@ -140,7 +140,7 @@ class HandleObserverEvent(PatternMatchingEventHandler):
         ignore_patterns=None,
         ignore_directories=False,
         case_sensitive=False,
-        settings={},
+        settings=None,
     ):
         super().__init__(
             patterns=patterns,
@@ -148,7 +148,7 @@ class HandleObserverEvent(PatternMatchingEventHandler):
             ignore_directories=ignore_directories,
             case_sensitive=case_sensitive,
         )
-        self._settings = settings
+        self._settings = settings if settings else {}
 
     def on_any_event(self, event):
         if event.event_type in ['created']:
@@ -302,10 +302,7 @@ def main(
             'output_dir_year_month': output_dir_year_month,
         },
     )
-    if use_polling:
-        observer = PollingObserver()
-    else:
-        observer = Observer()
+    observer = PollingObserver() if use_polling else Observer()
     observer.schedule(handler, input_dir, recursive=True)
     observer.start()
     print(f"Watching {input_dir} for new PDFs. Press Ctrl+C to exit.")

misc/webservice.py

@@ -4,6 +4,8 @@
 
 """Run the OCRmyPDF web service."""
 
+from __future__ import annotations
+
 import os
 import sys
 
@@ -13,7 +15,7 @@ except ImportError:
     raise ImportError(
         'You need to install streamlit in the Python environment '
         'to run the web service.\n'
-    )
+    ) from None
 
 if __name__ == '__main__':
     os.execvp(