clippy: Box result in RpcCustomError::SendTransactionPreflightFailure

[kskalski]

Problem

Clippy in Rust toolchain 1.94 reports clippy:result-large-err warning for large error in rpc code.

error: the `Err`-variant returned from this closure is very large
    --> rpc/src/rpc.rs:1384:45
     |
1384 |   ...                   .spawn_blocking(move || {
     |  _______________________________________^
1385 | | ...                       confirmed_block
1386 | | ...                           .encode_with_options(encoding, encoding_options)
1387 | | ...                           .map_err(RpcCustomError::from)
1388 | | ...                   })
     | |_______________________^ the `Err`-variant is at least 376 bytes
     |
     = help: try reducing the size of `solana_rpc_client_api::custom_error::RpcCustomError`, for example by boxing large elements or replacing it with `Box<solana_rpc_client_api::custom_error::RpcCustomError>`

Summary of Changes

Box result field in SendTransactionPreflightFailure variant

[mergify]

If this PR represents a change to the public RPC API:

1. Make sure it includes a complementary update to rpc-client/ (example)
2. Open a follow-up PR to update the JavaScript client @solana/kit (example)

Thank you for keeping the RPC clients in sync with the server API @kskalski.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.2%. Comparing base (9554223) to head (2db6977).
⚠️ Report is 3 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #11191   +/-   ##
=======================================
  Coverage    83.2%    83.2%           
=======================================
  Files         841      841           
  Lines      318322   318323    +1     
=======================================
+ Hits       265020   265027    +7     
+ Misses      53302    53296    -6     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[brooksprumo]

Which errors? In the past I've boxed large errors that were within our own error enums to handle similar lints.

[kskalski]

Looks like bigtable is no longer causing issues, but rpc does solana_rpc_client_api::custom_error::RpcCustomError

[brooksprumo]

How about boxing the result field inside the SendTransactionPreflightFailure error variant? That type is RpcSimulateTransactionResult, which is 352 bytes.

[brooksprumo]

Ah right... this is a public-facing API.

https://docs.rs/solana-rpc-client-api/latest/solana_rpc_client_api/custom_error/enum.RpcCustomError.html#variant.SendTransactionPreflightFailure

But we haven't upgraded the crate since v3.1?

I think this might be a semver-breaking change then. Ugh.

Here's the first PR where I did something similar: #6290

And the PR where I did the boxing (which mentions the api-breaking-ness): #6293

Maybe the clippy-allow is the only way forward until the next major version.

[kskalski]

But we haven't upgraded the crate since v3.1?

I think until 4.0 ships we use -beta labels, there are versions with that on https://docs.rs/solana-rpc-client-api/4.0.0-beta.1/solana_rpc_client_api/custom_error/enum.RpcCustomError.html

I guess the optimal course is to do this change and BP to 4.0. Otherwise, maybe need to go back to the previous way of just masking the lint in the call sites, which were not too many.

[brooksprumo]

May be worth asking in Discord on the correct way to handle this.

[kskalski]

It was suggested on discord that we just allow the lint, since rpc is going to be removed anyway. I have it on another PR.