CSV Drop

A lightweight macOS app for uploading CSV files to BigQuery.

Behind the scenes, the app uses bqcsv to upload a CSV file.

No API keys or OAuth setup required — the app reuses your existing gcloud / bq credentials.

Installation

Run the setup script to install CSV Drop, bqcsv, and verify the Google Cloud SDK:

curl -fsSL https://raw.githubusercontent.com/sqlhabit/csv-drop/refs/heads/main/bin/install | bash

The script downloads the latest GitHub release to /Applications/CSVDrop.app, installs bqcsv (via Homebrew Python when available), and checks for bq / gcloud.

If you already installed CSV Drop from the Mac App Store, install only the CLI dependencies:

curl -fsSL https://raw.githubusercontent.com/sqlhabit/csv-drop/refs/heads/main/bin/install | bash -s -- --skip-app

Prerequisites

1. Google Cloud SDK with bq and gcloud installed.
2. bqcsv CLI installed (the setup script above handles this).
3. Authenticated CLI session:

gcloud auth login
gcloud auth application-default login   # optional, if bq asks for ADC

Development

Debug builds can run against a local bqcsv checkout instead of an installed CLI:

cp Development.xcconfig.example Development.xcconfig
# edit Development.xcconfig and set BQCSV_DEV_REPO to your checkout

The repo root must contain src/cli.py. The app runs python3 -m src.cli with PYTHONPATH set to that directory. Python is resolved from the installed bqcsv shebang (pyenv, asdf, Homebrew), a repo .venv, or BQCSV_PYTHON / PYTHON_PATH. Development.xcconfig is gitignored; the shared Xcode scheme passes BQCSV_DEV_REPO into the app at launch.

If BQCSV_DEV_REPO is unset, Debug falls back to the same lookup as Release (BQCSV_PATH, Homebrew, pyenv, PATH).