Improve documentation

https://supabase.com/docs/guides/api#realtime-api-1

Describe the problem

Although it is complicated, realtime does work on tables in other schemas.

![image](https://user-images.githubusercontent.com/54564956/165872771-bec8da69-a107-4556-95e9-7a9c

Issue split from hasura/graphql-engine#6951

our pytest output has a pretty low signal:noise ratio, and it's often difficult to understand the cause of a test failure, if there is one. [Example: a 8600-line log for what turned out to be a flaky test that passed on a second run](https://app.circleci.com/pipelines/github/hasura/graphql-engine-mono/10008/workflows/5e17c7

Bug description

On windows, I run prisma format and note the unusual file ending. The lines are all LF, but the very last line is CRLF.

This causes issue on my Linux CI where it formats it ending in LF's only, causing a diff to occur and the build to fail.

How to reproduce

1. On windows do prisma format
2. Open in HxD or similar
3. See attached:

![image](https://user-images.g

With recent changes RecordMetadata#getTimestampIndex() is allowed to return non--1 value in the ORDER BY designated_ts DESC case (also see RecordCursorFactory#hasDescendingOrder()). This can be easily seen on filtered scan queries, but not only. Previously, the contract was different: a non--1 value was returned only for empty ORDER BY and ORDER BY designated_ts ASC cases.

This change b

The macros can emit code directly referencing serde_json::Value which may trigger compilation errors if the user doesn't have it in their own dependencies: https://discord.com/channels/665528275556106240/665528275556106243/969377568685981716

I'm surprised we haven't encountered this before now, probably because if someone wants to use serde_json::Value in their code deliberately then they p

find and findAndModify commands support maxTimeMS argument that specifies the command's time limit in milliseconds. We should support it.

The implementation should be relatively easy: get that optional parameter, and (if it is not zero) derive the Go context with a timeout and use it. In common handler code, check if the error is (possibly wrapped) context timeout, and return the right er

These files when added magically but not part of git tracked files causes update failure.

reference: frappe/bench#1225

related: frappe/frappe#16616 (you'll find which function create this file in this PR)

The problem faced currently?

While adding an ingress route, or a target in ingress route.
If by mistake white space gets added to the route or host. We get the error specified route not found or unable to reach the specified target

How can we solve it?

Mission control should remove the white space if present before saving the config

If you want this feature to be implemented, gi