Telescope upgrade

[pyramation]

No description provided.

[pyramation]

Fixed all the comments, casing, and issues from before, should be in better shape!

• do we need to include the ICA types from Add ICA types #36

[webmaster128]

do we need to include the ICA types from #36

Yeah, let's include them. You find them in wasmd-0.28/third_party/proto/ibc/applications/interchain_accounts now

[webmaster128]

Amazing! Let's get the ICA types in here (rebase to latest main and do a git submodule init && git submodule update then you have them). Otherwise 🏅

[webmaster128]

🥰

[pyramation]

the check is giving:

src/cosmos/tx/v1beta1/service.ts:924:14 - error TS7056: The inferred type of this node exceeds the maximum length the compiler will serialize. An explicit type annotation is needed.

I think I know how to solve this, but will make a large diff.

UPDATE: I'm able to fix this, it does make a large diff. See comments below.

[pyramation]

the check is giving:

src/cosmos/tx/v1beta1/service.ts:924:14 - error TS7056: The inferred type of this node exceeds the maximum length the compiler will serialize. An explicit type annotation is needed.

I think I know how to solve this, but will make a large diff.

https://app.circleci.com/pipelines/github/confio/cosmjs-types/123/workflows/569fd5f5-81d4-49c0-b0ec-27d44cbead66/jobs/115

btw — we can try doing this for only the google module if you want to keep the Exact<DeepPartial business. The plugin system should support that.

I found this error before and it's the meta typing (that I found quite unnecessary) that causes the memory issues.

FYI seems that you were around when they did this stuff and may have more context:

• Proposal: add an option to use shallow partial stephenh/ts-proto#378
• your comment here (maybe not completely related but I recall reading this thread which helped me solve my issue a while ago)

other related issues/prs

• feat: Use exact types for fromPartial stephenh/ts-proto#412
• Problems with Exact types stephenh/ts-proto#454

[pyramation]

OK, I think that may be it!

Only difference should be Exact vs DeepPartial, etc., but this method works and is much faster for the TS compiler to deal with.

We can roll back the Exact/DeepPartial stuff, but then the CI can't process things because of TS compiler memory.

[webmaster128]

Alright, let's give this a shot. I created a branch 0.5 for patching the old setup as long as we need it. This goes to main and 0.6.0-alpha release such that we can start testing it.

Amazing stuff 👏 Thanks a lot, Dan!

[pyramation]

wow this marks the end of an era! and a new beginning. Excited to start improving things like DecCoin and Long 🙌🏻