Google Cloud functions for BigTable
This is a reference implementaion for getting data out of BigTable.
Contents
This directory holds GCP Cloud Functions, one per file, along with shared utilities in shared.go
. The file names correspond to the hosted endpoints. ie endpoint .../*-notionaltvl
is the file notional-tvl.go
Debugging with VSCode
prereqs
- Golang >= 1.16 installed and available on your path.
- The Go VSCode extension, and gopls installed.
IDE setup
- open a new VSCode window
- File menu --> "Open Workspace from File..."
- Select
event_database/cloud_functions/workspace.code-workspace
Opening the workspace file as described above will open both cloud_functions
and functions_server
, so that you get all the VSCode goodness of intellesense, ability to run the code with the Go debugger, set breakpoints, etc.
Add your environment variables to functions_server/.vscode/launch.json
Start the debug server by pressing F5
. You can check your server is up by requesting http://localhost:8080/readyz.
deploying
First deploy (creation) must include all the flags to configure the environment:
gcloud functions --project your-project deploy testnet --region europe-west3 --entry-point Entry --runtime go116 --trigger-http --allow-unauthenticated --service-account=your-readonly@your-project.iam.gserviceaccount.com --update-env-vars GCP_PROJECT=your-project,BIGTABLE_INSTANCE=wormhole-testnet
gcloud functions --project your-project deploy processvaa-testnet --region europe-west3 --entry-point ProcessVAA --runtime go116 --trigger-topic new-vaa-testnet --service-account=your-readonly@your-project.iam.gserviceaccount.com --update-env-vars GCP_PROJECT=your-project,BIGTABLE_INSTANCE=wormhole-testnet
Subsequent deploys (updates) only need include flags to indentify the resource for updating: project, region, name.
gcloud functions --project your-project deploy testnet --region europe-west3 --entry-point Entry
gcloud functions --project your-project deploy processvaa-testnet --region europe-west3 --entry-point ProcessVAA
invocation
All routes accept their input(s) as query parameters, or request body. Just two different ways of querying:
GET
curl "https://region-project-id.cloudfunctions.net/testnet/readrow?emitterChain=2&emitterAddress=000000000000000000000000e982e462b094850f12af94d21d470e21be9d0e9c&sequence=0000000000000006"
POST
curl -X POST https://region-project-id.cloudfunctions.net/testnet/readrow \
-H "Content-Type:application/json" \
-d \
'{"emitterChain":"2", "emitterAddress":"000000000000000000000000e982e462b094850f12af94d21d470e21be9d0e9c", "sequence":"0000000000000006"}'
See ./bigtable-endpoints.md for API patterns