setup_github.sh

#!/usr/bin/env bash
#
# setup_github.sh — initialize this folder as a git repo and push it to GitHub.
#
# WHAT THIS DOES:
#   1. Safety-checks for files that should never be public (.db, .env, real data)
#   2. git init + first commit
#   3. wires up the GitHub remote and pushes
#
# WHAT IT DOES NOT DO:
#   It does not store or handle your GitHub password/token. When git asks you to
#   authenticate during the push, that happens directly between your machine and
#   GitHub — this script never sees it.
#
# BEFORE YOU RUN:
#   1. Create an EMPTY repo on github.com named "careask"
#      (do NOT add a README, .gitignore, or license there — you already have them)
#   2. Run this script from inside the careask folder:
#        bash setup_github.sh
#
set -e

REPO_NAME="careask"
GITHUB_USER="bbiddick7"
REMOTE_URL="https://github.com/${GITHUB_USER}/${REPO_NAME}.git"

echo "==> CareAsk GitHub setup"
echo

# --- 1. Safety check: refuse to proceed if sensitive files are present ---
echo "==> Checking for files that should never be committed..."
PROBLEM=0
if ls db/*.db >/dev/null 2>&1; then
  echo "  ! Found a .db file in db/ — this is gitignored, but confirming it won't be tracked."
fi
if [ -f .env ]; then
  echo "  ! Found a .env file. It is gitignored, but double-check it holds no real secrets you don't want backed up."
fi
# Look for anything that smells like real data outside the synthetic generator.
if grep -rliE "ssn|social security|medicaid id|real_patient|phi" --include=*.csv --include=*.json . 2>/dev/null | grep -v node_modules; then
  echo "  !! Potential real-data file detected above. STOP and review before pushing."
  PROBLEM=1
fi
if [ "$PROBLEM" -eq 1 ]; then
  echo
  echo "Aborting. Remove or review the flagged files, then re-run."
  exit 1
fi
echo "  ok — nothing obviously sensitive detected."
echo

# --- 2. Confirm git identity ---
if ! git config user.name >/dev/null 2>&1; then
  echo "==> Git needs your identity. Setting it now."
  read -rp "  Your name: " GIT_NAME
  read -rp "  Your GitHub email: " GIT_EMAIL
  git config --global user.name "$GIT_NAME"
  git config --global user.email "$GIT_EMAIL"
fi

# --- 3. Init + commit ---
if [ -d .git ]; then
  echo "==> This folder is already a git repo; skipping init."
else
  echo "==> Initializing git repo..."
  git init
fi

echo "==> Staging files..."
git add .

echo
echo "==> These files will be committed:"
git status --short
echo
read -rp "Look right? Press ENTER to commit, or Ctrl-C to stop and review: " _

git commit -m "Initial commit: CareAsk PHI-safe NL-to-SQL reference architecture" || {
  echo "Nothing to commit (already committed?). Continuing."
}

# --- 4. Wire remote + push ---
git branch -M main

if git remote get-url origin >/dev/null 2>&1; then
  echo "==> Remote 'origin' already set to: $(git remote get-url origin)"
else
  echo "==> Adding remote: ${REMOTE_URL}"
  git remote add origin "${REMOTE_URL}"
fi

echo
echo "==> Pushing to GitHub. You'll be asked to authenticate."
echo "    Username: ${GITHUB_USER}"
echo "    Password: paste your Personal Access Token (NOT your account password)."
echo "    (If you don't have a token yet, press Ctrl-C and create one first:"
echo "     GitHub → Settings → Developer settings → Personal access tokens.)"
echo
git push -u origin main

echo
echo "==> Done. Your repo should be live at:"
echo "    https://github.com/${GITHUB_USER}/${REPO_NAME}"
echo
echo "Next: add topics (healthcare, hipaa, llm, python, nlp-to-sql) and"
echo "confirm the About description matches your LinkedIn project entry."