Brian R. Murphy
Brian R. Murphy
I'm still not absolutely certain that paranoid mode is enabled for the aptos-transactional-test-harness, though I've had debugging code test the flag value at initialization and later. third_party transactional tests are...
> "some editors" Which editors? Supporting editors that don't work with utf8 feels like we're still writing webapps for IE6 Emacs + rust-analyzer has problems with UTF8. There are bugs...
> "Rust input is interpreted as a sequence of Unicode code points encoded in UTF-8." `https://doc.rust-lang.org/reference/input-format.html` We have no need to support editors that can't deal with valid Rust source....
To regenerate these files, we can run `cargo test --features "evm-backend" --test build_testsuite_evm -p move-cli -p move-unit-test` but this currently doesn't even compile.
After https://github.com/aptos-labs/aptos-core/pull/13125 lands these will at least compile. More work is required to fix the test outputs. Notably, the attribute maps don't seem to be readily available in `test_runner.rs`. Notably,...
We should not remove tests until we have equivalent tests in the Aptos world if we have similar functionality. Disassembler, etc. Evm can die, sure. On Mon, Apr 29, 2024,...
The move-cli tests were the only way I was able to reproduce and debug the V2 disassembler problem. We need a replacement.
@rahxephon89 @wrwg
I note that the usual expected output extension `"exp"` is defined in third_party/move/move-command-line-common/src/testing.rs: ``` /// Extension for expected output files pub const EXP_EXT: &str = "exp"; /// Extension for expected...