Martin Jesper Low Madsen

I'm experiencing the same thing. For me it seems like an issue with bundling (by debugging stack traces) where the `useTranslate` context loses its reference. In this scenario, the provider...

Figured out why. It's because react-i18next is listed as a regular dependency and not a peer dependency. Using pnpm/yarn 2/3 introduces this issue. I can see from previous issues that...

Already have, however, the module resolution in a symlink setup does not function well in this case. ``` diff --git a/package.json b/package.json index 02388f7e443548dc6e428429719346293099412c..cc1169a110b2b1cb53e81d78837cfec7a69570a3 100644 --- a/package.json +++ b/package.json @@...

I decided to drop next-i18next altogether due to the lack of language detection support (pass on lng config on the client) and go with the custom solution I was using...

Hmm, I actually ended up forcing Yarn 2 as the packager - it kind of depends on how long Yarn 1 remains the default packager for git sources - I...

Just throwing my response on Slack here for reference: > For mocks through interfaces, we cannot distinguish them at runtime they'll just be represented as anonymous objects, but for overloads...

I'm not sure I understand what you want to reuse the marker for. Do you mean storing the identity as the marker instead of the boolean value as it is...

Hehe, I'm not gonna hate You :p A document sounds like a good idea. To fully support literals, we would have to adjust the emitted descriptor to proxy those as...

I've given this some more thought and there are (as I see it) two directions we can go. 1. The opposite direction of this PR (that I originally intended), preserving...

> Hi, yes, we should create a document where we explain the concern about this. > > This is the examples I'm not sure how you will identify correctly. >...