Use rich click

Author: ProbablyFaiz

README.md

@@ -31,51 +31,54 @@ export RRC_DATA_DIR=/path/to/output
 docker run --rm \
     -v $RRC_IMAGE_DIR:/data/images \
     -v $RRC_DATA_DIR:/data/output \
-    ghcr.io/reglab/rrc-pipeline:latest ingest
+    ghcr.io/reglab/rrc-pipeline:latest rrc ingest
 
 # 2. Run OCR
 docker run --rm --gpus all \
     -v $RRC_IMAGE_DIR:/data/images \
     -v $RRC_DATA_DIR:/data/output \
-    ghcr.io/reglab/rrc-pipeline:latest ocr
+    ghcr.io/reglab/rrc-pipeline:latest rrc ocr
 
 # 3. Detect covenants
 docker run --rm --gpus all \
     -v $RRC_IMAGE_DIR:/data/images \
     -v $RRC_DATA_DIR:/data/output \
-    ghcr.io/reglab/rrc-pipeline:latest detect
+    ghcr.io/reglab/rrc-pipeline:latest rrc detect
 
 # 4. Export results
 docker run --rm \
     -v $RRC_IMAGE_DIR:/data/images \
     -v $RRC_DATA_DIR:/data/output \
-    ghcr.io/reglab/rrc-pipeline:latest export
+    ghcr.io/reglab/rrc-pipeline:latest rrc export
 ```
 
 ## Pipeline Stages
 
-### 1. Ingest (`ingest`)
+> [!NOTE]
+> To see all available commands, run `docker run --rm ghcr.io/reglab/rrc-pipeline:latest rrc --help`.
+
+### 1. Ingest (`rrc ingest`)
 - Scans input directory for image files (jpg, jpeg, png, tiff, tif, bmp)
 - Validates images can be opened
 - Handles multi-page TIFF files
 - Creates database records for new images
 
-### 2. OCR (`ocr`)
+### 2. OCR (`rrc ocr`)
 - Transcribes images using the DocTR OCR library
 - Requires GPU acceleration
 - Processes only images without existing transcriptions
 
-### 3. Detection (`detect`)
+### 3. Detection (`rrc detect`)
 - Analyzes transcribed text using our Mistral-based covenant detection model
 - Requires GPU acceleration
 - Identifies presence of racial covenants and extracts relevant passages
 - Processes only transcribed pages without existing predictions
 
-### 4. Export (`export`)
+### 4. Export (`rrc export`)
 - Exports detection results to CSV format
 - Includes confidence scores and extracted covenant text where found
 
-### 5. Pipeline Summary (`summarize`)
+### 5. Pipeline Summary (`rrc summarize`)
 - Displays current pipeline progress and statistics
 - Shows total page counts and processing status
 - Reports covenant detection statistics

backend/pyproject.toml

@@ -26,6 +26,7 @@ dependencies = [
     "tqdm>=4.67.1",
     "click>=8.1.7",
     "rich>=13.8.1",
+    "rich-click>=1.8.9",
 ]
 
 [tool.uv]
@@ -39,11 +40,7 @@ dev-dependencies = [
 homepage = "https://github.com/reglab/rrc-pipeline"
 
 [project.scripts]
-ingest = "rrc.ingest.ingest_directory:main"
-ocr = "rrc.ocr.transcribe_pending:main"
-detect = "rrc.inference.detect_pending:main"
-export = "rrc.reporting.export_predictions:main"
-summarize = "rrc.reporting.summarize_db:main"
+rrc = "rrc.cli:cli"
 
 [build-system]
 requires = ["hatchling"]

backend/rrc/cli.py

@@ -0,0 +1,44 @@
+"""Main CLI for the RRC (Racial Restrictive Covenants) pipeline."""
+
+import rrc.utils.click as click
+from rrc.inference.detect_pending import main as detect_cmd
+from rrc.ingest.ingest_directory import main as ingest_cmd
+from rrc.ocr.transcribe_pending import main as ocr_cmd
+from rrc.reporting.export_predictions import main as export_cmd
+from rrc.reporting.summarize_db import main as summarize_cmd
+
+
+@click.group()
+@click.version_option()
+def cli():
+    """
+    RRC Pipeline - Racial Restrictive Covenants Detection
+
+    A comprehensive pipeline for identifying racial covenants in property deeds.
+    Process images through OCR, detect covenants with ML models, and export results.
+    See https://reglab.github.io/racialcovenants/ for more information.
+    """
+    pass
+
+
+# Add the original commands to the CLI group with new names and emojis in help
+cli.add_command(ingest_cmd, name="ingest")
+cli.add_command(ocr_cmd, name="ocr")
+cli.add_command(detect_cmd, name="detect")
+cli.add_command(export_cmd, name="export")
+cli.add_command(summarize_cmd, name="summarize")
+
+# Update the help text for each command to add emojis
+ingest_cmd.help = ingest_cmd.help or "Ingest images from a directory into the database"
+ocr_cmd.help = ocr_cmd.help or "Transcribe pending pages using OCR"
+detect_cmd.help = (
+    detect_cmd.help or "Detect covenants in transcribed pages using ML models"
+)
+export_cmd.help = export_cmd.help or "Export covenant predictions to CSV files"
+summarize_cmd.help = (
+    summarize_cmd.help or "Display a summary of the current database state"
+)
+
+
+if __name__ == "__main__":
+    cli()

backend/rrc/inference/detect_pending.py

@@ -1,11 +1,11 @@
 from pathlib import Path
 
-import click
 import tqdm
 from rich.console import Console
 from sqlalchemy import func, select
 from sqlalchemy.orm import Session, joinedload
 
+import rrc.utils.click as click
 import rrc.utils.io
 from rrc.db.models import CovenantPrediction, Page, Provenance
 from rrc.db.session import get_session

backend/rrc/inference/service.py

@@ -93,6 +93,8 @@ def __init__(self, options: dict[str, Any] = None):
         self.model_download_dir = self.options.get("model_download_dir")
 
     def __enter__(self):
+        if DEFAULT_DEVICE is None:
+            raise RuntimeError("No CUDA or MPS device available")
         self.vllm_model = vllm.LLM(
             model=self.model_name_or_path,
             device=DEFAULT_DEVICE,

backend/rrc/ingest/ingest_directory.py

@@ -3,12 +3,12 @@
 from pathlib import Path
 from typing import TypeVar
 
-import click
 import PIL.Image
 import tqdm
 from rich.console import Console
 from sqlalchemy import select
 
+import rrc.utils.click as click
 from rrc.db.models import Page
 from rrc.db.session import get_session
 from rrc.utils.io import get_image_path

backend/rrc/ocr/service.py

@@ -45,6 +45,8 @@ def __init__(self, options: dict[str, Any] | None = None):
         super().__init__(options)
 
     def __enter__(self):
+        if DEFAULT_DEVICE is None:
+            raise RuntimeError("No CUDA or MPS device available")
         self.model = ocr_predictor(
             det_arch=_DOCTR_DET_ARCH,
             reco_arch=_DOCTR_RECO_ARCH,

backend/rrc/ocr/transcribe_pending.py

@@ -1,9 +1,9 @@
-import click
 import tqdm
 from rich.console import Console
 from sqlalchemy import select
 from sqlalchemy.orm import Session
 
+import rrc.utils.click as click
 from rrc.db.models import Page, Provenance, Transcription
 from rrc.db.session import get_session
 from rrc.ocr.service import DoctrOCRService

backend/rrc/reporting/export_predictions.py

@@ -2,12 +2,12 @@
 from datetime import datetime
 from pathlib import Path
 
-import click
 import tqdm
 from rich.console import Console
 from sqlalchemy import select
 from sqlalchemy.orm import joinedload
 
+import rrc.utils.click as click
 import rrc.utils.io
 from rrc.db.models import CovenantPrediction
 from rrc.db.session import get_session

backend/rrc/reporting/summarize_db.py

@@ -1,9 +1,9 @@
-import click
 from rich.console import Console
 from rich.progress_bar import ProgressBar
 from rich.table import Table
 from sqlalchemy import func, select
 
+import rrc.utils.click as click
 from rrc.db.models import CovenantPrediction, Page, Transcription
 from rrc.db.session import get_session