Aleksandr Razumov
Aleksandr Razumov
Currently in most cases FLOOD_WAIT handling is user responsibility. However, it is not convenient. We can implement some sort of invoker wrapper: 1. Separate flood-waits per request/scope? 2. Automatic retries...
Add tracing of request via context. Ref: https://github.com/open-telemetry
Like [pebble.EventListener](https://pkg.go.dev/github.com/cockroachdb/pebble#EventListener)
Current go Telegram server lacks lots of functions and is not suitable for sophisticated testing. I'm looking into features like: - [ ] Session state/key storage - [ ] Correct...
https://tools.ietf.org/html/rfc6062
Please use [semantic versioning](https://semver.org/) with tags, like this: ```bash $ git tag v0.0.1 $ git push origin v0.0.1 # ... * [new tag] v0.0.1 -> v0.0.1 ``` So we can...
Currently it is disabled because I've protected whole gortc.io with cloudflare.
So, instead of writing ``` query := fmt.Sprintf("SELECT * FROM table WHERE foo = '%s' AND bar != %d", "foo", 1234) ``` We can do something like ``` query :=...
To allow code coverage calculation [codecov should be set up](https://docs.codecov.com/docs) for this repo. Github action can be tested [manually](https://github.com/ClickHouse/ch-go/actions/workflows/cover.yml) via workflow dispatch.