[nasretdinov]

So it's really the issue due a partial variable shadowing during assignment, not errgroup-specific. Got bitten by it multiple times unfortunately, but I don't think there's an easy lint or vet check that could prevent such errors from happening. Having said that, overwriting a variable during := assignment is typically only useful for errors, so potentially you might want to have a lint check that complains when you've overwritten something else. There's already a check that the value is unused so when re-assigning the error value you're less likely to miss anything

[eddythompson80]

No? Its about errgroup ctx being automatically cancelled once the group is done (in success or err). It kinda makes sense, and its documented. I understand how it can be surprising, but if you think about it for a second, it makes sense. The context created is meant to manage the lifecycle of the errgroup. Once the group is done, the ctx is done.

[nasretdinov]

It's only about context _returned by errgroup_ being cancelled, you shouldn't be overwriting it in the way in which it was used by an article, which is really hard to spot due to variable shadowing.

[masklinn]

> variable shadowing

There is no shadowing.

It's also harder to spot that this is a bad pattern because it's exactly what the package examples currently do.

[yubblegum]

> So it's really the issue due a partial variable shadowing during assignment, not errgroup-specific.

Did we read the same article? The "alternative fix" is to paper over documented bahavior of a context. Of course it makes sense that if wait returns context should be canceled. The one character fix is throwing away information ..

[masklinn]

> The one character fix is throwing away information ..

It's duplicate information, since Wait essentially provides the same thing (and in fact the context returned is mostly to solely for the callbacks to use, if the callbacks don't need it, it should be thrown away).

[dondraper36]

That's why I often use eg, egCtx :)