Replace typer with cyclopts CLI library in misc scripts

Migrate watcher.py and pdf_text_diff.py from typer to cyclopts for CLI argument parsing. Update pyproject.toml to reflect the dependency change in the watcher optional feature.
2026-01-13 00:43:14 -08:00
parent bf76c8270c
commit 4c7086c609
3 changed files with 65 additions and 69 deletions
@@ -5,33 +5,45 @@

 from __future__ import annotations

+from pathlib import Path
 from subprocess import run
 from tempfile import NamedTemporaryFile
 from typing import Annotated

-import typer
+import cyclopts
+
+app = cyclopts.App()


+@app.default
 def main(
-    pdf1: Annotated[typer.FileBinaryRead, typer.Argument()],
-    pdf2: Annotated[typer.FileBinaryRead, typer.Argument()],
-    engine: Annotated[str, typer.Option()] = 'pdftotext',
+    pdf1: Annotated[Path, cyclopts.Parameter()],
+    pdf2: Annotated[Path, cyclopts.Parameter()],
+    *,
+    engine: Annotated[str, cyclopts.Parameter()] = 'pdftotext',
 ):
    """Compare text in PDFs."""
-    text1 = run(
-        ['pdftotext', '-layout', '-', '-'], stdin=pdf1, capture_output=True, check=True
-    )
-    text2 = run(
-        ['pdftotext', '-layout', '-', '-'], stdin=pdf2, capture_output=True, check=True
-    )
+    with open(pdf1, 'rb') as f1, open(pdf2, 'rb') as f2:
+        text1 = run(
+            ['pdftotext', '-layout', '-', '-'],
+            stdin=f1,
+            capture_output=True,
+            check=True,
+        )
+        text2 = run(
+            ['pdftotext', '-layout', '-', '-'],
+            stdin=f2,
+            capture_output=True,
+            check=True,
+        )

-    with NamedTemporaryFile() as f1, NamedTemporaryFile() as f2:
-        f1.write(text1.stdout)
-        f1.flush()
-        f2.write(text2.stdout)
-        f2.flush()
+    with NamedTemporaryFile() as t1, NamedTemporaryFile() as t2:
+        t1.write(text1.stdout)
+        t1.flush()
+        t2.write(text2.stdout)
+        t2.flush()
        diff = run(
-            ['diff', '--color=always', '--side-by-side', f1.name, f2.name],
+            ['diff', '--color=always', '--side-by-side', t1.name, t2.name],
            capture_output=True,
        )
        run(['less', '-R'], input=diff.stdout, check=True)
@@ -42,4 +54,4 @@ def main(


 if __name__ == '__main__':
-    typer.run(main)
+    app()
@@ -17,8 +17,8 @@ from enum import Enum
 from pathlib import Path
 from typing import Annotated, Any

+import cyclopts
 import pikepdf
-import typer
 from dotenv import load_dotenv
 from watchdog.events import PatternMatchingEventHandler
 from watchdog.observers import Observer
@@ -30,7 +30,7 @@ load_dotenv()


 # pylint: disable=logging-format-interpolation
-app = typer.Typer(name="ocrmypdf-watcher")
+app = cyclopts.App(name="ocrmypdf-watcher")

 log = logging.getLogger('ocrmypdf-watcher')

@@ -153,110 +153,94 @@ class HandleObserverEvent(PatternMatchingEventHandler):
            execute_ocrmypdf(file_path=Path(event.src_path), **self._settings)


-@app.command()
+@app.default
 def main(
    input_dir: Annotated[
        Path,
-        typer.Argument(
-            envvar='OCR_INPUT_DIRECTORY',
-            exists=True,
-            file_okay=False,
-            dir_okay=True,
-            readable=True,
-            resolve_path=True,
+        cyclopts.Parameter(
+            env_var='OCR_INPUT_DIRECTORY',
        ),
-    ] = '/input',
+    ] = Path('/input'),
    output_dir: Annotated[
        Path,
-        typer.Argument(
-            envvar='OCR_OUTPUT_DIRECTORY',
-            exists=True,
-            file_okay=False,
-            dir_okay=True,
-            writable=True,
-            resolve_path=True,
+        cyclopts.Parameter(
+            env_var='OCR_OUTPUT_DIRECTORY',
        ),
-    ] = '/output',
+    ] = Path('/output'),
    archive_dir: Annotated[
        Path,
-        typer.Argument(
-            envvar='OCR_ARCHIVE_DIRECTORY',
-            exists=True,
-            file_okay=False,
-            dir_okay=True,
-            writable=True,
-            resolve_path=True,
+        cyclopts.Parameter(
+            env_var='OCR_ARCHIVE_DIRECTORY',
        ),
-    ] = '/processed',
+    ] = Path('/processed'),
+    *,
    output_dir_year_month: Annotated[
        bool,
-        typer.Option(
-            envvar='OCR_OUTPUT_DIRECTORY_YEAR_MONTH',
+        cyclopts.Parameter(
+            env_var='OCR_OUTPUT_DIRECTORY_YEAR_MONTH',
            help='Create a subdirectory in the output directory for each year/month',
        ),
    ] = False,
    on_success_delete: Annotated[
        bool,
-        typer.Option(
-            envvar='OCR_ON_SUCCESS_DELETE',
+        cyclopts.Parameter(
+            env_var='OCR_ON_SUCCESS_DELETE',
            help='Delete the input file after successful OCR',
        ),
    ] = False,
    on_success_archive: Annotated[
        bool,
-        typer.Option(
-            envvar='OCR_ON_SUCCESS_ARCHIVE',
+        cyclopts.Parameter(
+            env_var='OCR_ON_SUCCESS_ARCHIVE',
            help='Archive the input file after successful OCR',
        ),
    ] = False,
    deskew: Annotated[
        bool,
-        typer.Option(
-            envvar='OCR_DESKEW',
+        cyclopts.Parameter(
+            env_var='OCR_DESKEW',
            help='Deskew the input file before OCR',
        ),
    ] = False,
    ocr_json_settings: Annotated[
-        str,
-        typer.Option(
-            envvar='OCR_JSON_SETTINGS',
+        str | None,
+        cyclopts.Parameter(
+            env_var='OCR_JSON_SETTINGS',
            help='JSON settings to pass to OCRmyPDF (JSON string or file path)',
        ),
    ] = None,
    poll_new_file_seconds: Annotated[
        int,
-        typer.Option(
-            envvar='OCR_POLL_NEW_FILE_SECONDS',
+        cyclopts.Parameter(
+            env_var='OCR_POLL_NEW_FILE_SECONDS',
            help='Seconds to wait before polling a new file',
-            min=0,
        ),
    ] = 1,
    use_polling: Annotated[
        bool,
-        typer.Option(
-            envvar='OCR_USE_POLLING',
+        cyclopts.Parameter(
+            env_var='OCR_USE_POLLING',
            help='Use polling instead of filesystem events',
        ),
    ] = False,
    retries_loading_file: Annotated[
        int,
-        typer.Option(
-            envvar='OCR_RETRIES_LOADING_FILE',
+        cyclopts.Parameter(
+            env_var='OCR_RETRIES_LOADING_FILE',
            help='Number of times to retry loading a file before giving up',
-            min=0,
        ),
    ] = 5,
    loglevel: Annotated[
        LoggingLevelEnum,
-        typer.Option(
-            envvar='OCR_LOGLEVEL',
+        cyclopts.Parameter(
+            env_var='OCR_LOGLEVEL',
            help='Logging level',
        ),
    ] = LoggingLevelEnum.INFO,
    patterns: Annotated[
        str,
-        typer.Option(
-            envvar='OCR_PATTERNS',
+        cyclopts.Parameter(
+            env_var='OCR_PATTERNS',
            help='File patterns to watch',
        ),
    ] = '*.pdf,*.PDF',
@@ -322,7 +306,7 @@ def main(
        observer = Observer()
    observer.schedule(handler, input_dir, recursive=True)
    observer.start()
-    typer.echo(f"Watching {input_dir} for new PDFs. Press Ctrl+C to exit.")
+    print(f"Watching {input_dir} for new PDFs. Press Ctrl+C to exit.")
    try:
        while True:
            time.sleep(30)
@@ -53,7 +53,7 @@ Changelog = "https://github.com/ocrmypdf/OCRmyPDF/docs/release_notes.md"

 [project.optional-dependencies]
 # User-installable features - use `uv sync --extra <name>` or `pip install ocrmypdf[name]`
-watcher = ["watchdog>=1.0.2", "typer-slim[standard]", "python-dotenv"]
+watcher = ["watchdog>=1.0.2", "cyclopts>=3", "python-dotenv"]
 webservice = ["streamlit>=1.41.0"]

 [project.scripts]