Pick any line to see why it is the way it is: the commit, the pull request and issue it came from, and what the agent was thinking.
| The runner base keeps its old comment, so the pushed base image still matches its folder | 1 | #!/usr/bin/env bash |
| 2 | # Sets up what Deployments needs on a Cloudflare account, once. Safe to run | |
| 3 | # again: each step skips what exists. Then deploy with scripts/deploy.sh. | |
| 4 | # | |
| 5 | # Needs the Workers for Platforms add-on on the account, and a zone for | |
| 6 | # apps (g1t uses g1t.page) named in services/pages/wrangler.jsonc. | |
| 7 | # | |
| 8 | # Two steps need more than `wrangler login` can do: a wildcard DNS record | |
| 9 | # on the apps' zone, and an API token for the deployments service. Set | |
| 10 | # CLOUDFLARE_EMAIL and CLOUDFLARE_API_KEY (a Global API Key) and this | |
| 11 | # script makes both; otherwise it says what to make by hand. | |
| 12 | set -euo pipefail | |
| 13 | ||
| 14 | ROOT="$(cd "$(dirname "$0")/.." && pwd)" | |
| 15 | export CLOUDFLARE_API_TOKEN="${CLOUDFLARE_DEPLOY_TOKEN:-}" | |
| 16 | ZONE="${G1T_APPS_ZONE:-g1t.page}" | |
| 17 | NAMESPACE="g1t-deployments" | |
| 18 | TOKEN_FILE="$ROOT/.credentials/deployments-cloudflare-token.txt" | |
| 19 | w() { npx wrangler "$@"; } | |
| 20 | ||
| 21 | cd "$ROOT/services/deployments" | |
| 22 | ||
| 23 | echo "== D1 database" | |
| 24 | if grep -q TO_BE_CREATED wrangler.jsonc; then | |
| 25 | id=$(w d1 create g1t-deployments </dev/null 2>&1 | grep -oE '[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}' | head -1 || true) | |
| 26 | [ -n "$id" ] || { echo "Could not create the database; is it there already? (npx wrangler d1 list)"; exit 1; } | |
| 27 | sed -i "s/TO_BE_CREATED/$id/" wrangler.jsonc | |
| 28 | echo "Created $id and wrote it into services/deployments/wrangler.jsonc. Commit that." | |
| 29 | else | |
| 30 | echo "Already set." | |
| 31 | fi | |
| 32 | ||
| 33 | echo "== Dispatch namespace and event queue" | |
| 34 | w dispatch-namespace list 2>/dev/null | grep -q "$NAMESPACE" || w dispatch-namespace create "$NAMESPACE" | |
| 35 | w queues list 2>/dev/null | grep -q g1t-events-deployments || w queues create g1t-events-deployments | |
| The docs folder is gone, and what it held lives where people read it: how a self-hosted g1t runs and how to deploy g1t to Cloudflare are pages on docs.g1t.sh under Run g1t yourself, and speed, rate limits and operating g1t.sh are sections of CONTRIBUTING.md; code that cited a file in docs/ now points to the page or section that covers it, or says what it means itself, and applied migrations and the runner images are left as they were. | 36 | # Where every consumer sends what it gave up on |
| 37 | # (docs.g1t.sh/guides/deploy-to-cloudflare/#the-dead-letter-queue). | |
| Merge branch 'worktree-agent-ad8a36dfcd4176015' into spend-guardrails | 38 | w queues list 2>/dev/null | grep -q g1t-events-dlq || w queues create g1t-events-dlq |
| The runner base keeps its old comment, so the pushed base image still matches its folder | 39 | |
| 40 | echo "== DNS record and the service's token" | |
| 41 | if [ -n "${CLOUDFLARE_API_KEY:-}" ] && [ -n "${CLOUDFLARE_EMAIL:-}" ]; then | |
| 42 | G1T_APPS_ZONE="$ZONE" TOKEN_FILE="$TOKEN_FILE" python "$ROOT/scripts/cloudflare-setup.py" | |
| 43 | else | |
| 44 | cat <<EOF | |
| 45 | Set CLOUDFLARE_EMAIL and CLOUDFLARE_API_KEY to do this for you, or by hand: | |
| 46 | 1. On $ZONE, add a proxied DNS record: type AAAA, name *, content 100:: | |
| 47 | 2. Create an API token with Workers Scripts: Edit and Account Analytics: | |
| 48 | Read on the account, and save it as one line in | |
| 49 | $TOKEN_FILE | |
| 50 | EOF | |
| 51 | fi | |
| 52 | ||
| 53 | if [ -f "$TOKEN_FILE" ]; then | |
| 54 | echo "== Storing the token as the deployments service's secret" | |
| 55 | w deploy | |
| 56 | tr -d '\r\n' <"$TOKEN_FILE" | w secret put CLOUDFLARE_API_TOKEN | |
| 57 | fi | |
| 58 | echo "== Done. Now: scripts/deploy.sh" |
This file's history is long; its oldest lines are credited to the oldest commit read.