Commands

CLI subcommand implementations for the run.py unified entry point. Each module exposes a single cmd_* function that argparse dispatches to.

Available Commands

Command	Module	Description
guide	guide.py	Prints the complete setup and usage guide (prerequisites, environment variables, config files, workflow).
setup-connection	setup_connection.py	Create/verify the external connection, register the schema, configure search settings, and wait for the connection to reach ready state. Does not ingest any content. Useful for initial setup or re-registering an updated schema.
full-deployment	deploy.py	End-to-end deployment: create connection → register schema → configure search → identity crawl (if USE_GROUP_ACL) → ingest items with ACLs.
ingest	ingest.py	Re-ingest items into an existing connection (assumes prior full-deployment).
ingest-item	ingest_item.py	Ingest a single Salesforce record by its ID (--id).
ingest-object	ingest_object.py	Ingest all records of one Salesforce object type (--type).
identity-dry-run	identity_dry_run.py	Preview identity crawl changes without calling Graph APIs. Use --save to write to SQLite.

Shared Utilities (__init__.py)

Function	Description
setup_logging(prefix, verbose)	Configures root logger (file: INFO+, console: WARNING+ or INFO+ with --verbose), a "progress" logger that always prints to console, and a summary log file. Returns (log_file, summary_file).
reset_logging()	Removes all handlers from root and progress loggers. Called before each iteration in --continuous mode so every run gets a fresh log file.
write_summary(...)	Writes a run summary (connection status, ingestion stats, timing) to both a log file and console.
build_parser()	Constructs the argparse.ArgumentParser with all subcommands and the global --verbose flag.

Options

Global

Flag	Description
--verbose	Print all log levels (INFO+) to console. Without this flag only WARNING+ are shown; the log file always captures everything.

Continuous Mode (full-deployment and ingest only)

Flag	Default	Description
--continuous	—	Keep running with scheduled full and incremental crawls.
--full-crawl-hours <int>	24	Full crawl interval in hours (min 12, max 168).
--incremental-hours <int>	4	Incremental crawl interval in hours (min 1, max 168).

Identity crawl only runs on full sync cycles (not incremental).

Usage

python run.py guide                                                         # Show setup guide
python run.py setup-connection                                              # Create connection + schema (no ingestion)
python run.py setup-connection --verbose                                    # Setup with detailed console output
python run.py full-deployment                                               # Deploy (full sync)
python run.py full-deployment --verbose                                     # Deploy (detailed console)
python run.py full-deployment --continuous                                   # Deploy + continuous (24h full, 4h incremental)
python run.py full-deployment --continuous --full-crawl-hours 48 --incremental-hours 2
python run.py ingest                                                        # Re-ingest only (full sync)
python run.py ingest --continuous                                           # Continuous (24h full, 4h incremental)
python run.py identity-dry-run --verbose                                    # Preview identity changes
python run.py identity-dry-run --save --verbose                             # Preview + save to SQLite
python run.py ingest-item --id 500f6000008iCNYAA2                           # Ingest one record
python run.py ingest-object --type Case                                     # Ingest all Cases