refactor: Refactor evmbin CLI

[ltfschoen]

• teststate.json uses the Test struct and each name contains the State struct (see ethjson package)
• create2callPrecompiles.json https://github.com/ethereum/tests/blob/725dbc73a54649e22a00330bd0f4d6699a5060e5/GeneralStateTests/stCreate2/create2callPrecompiles.json#L2 state test JSON file from Ethereum Foundation that's now in the /res directory, as doesn't return a state root mistmatch error like the teststate.json file does.
• Run parity-evm binary using state test for example:

cargo build -p evmbin --release
 ./target/release/parity-evm state-test ./evmbin/res/teststate.json --only add12 --chain eip150
 ./target/release/parity-evm state-test ./evmbin/res/create2callPrecompiles.json --only create2callPrecompiles --chain Constantinople

• Run stats:

./target/release/parity-evm stats --to "0000000000000000000000000000000000000004" --from "0000000000000000000000000000000000000003" --code "05" --input "06" --gas "21" --gas-price "2" --chain "./evmbin/res/testchain.json" --json

Output:
{"depth":1,"gas":"0x21","gasCost":"0x0","memory":"0x","op":5,"opName":"SDIV","pc":0,"stack":[],"storage":{}}
{"error":"EVM: Stack underflow SDIV 2/0","gasUsed":"0x21","time":87}

Note: operation SDIV from code: 05 does division and expects two items on the stack, try changing to code: 01 for ADD (addition). For more info do a search for "SDIV" across the codebase and read about all the different EVM bytecode instructions.

• Build docs: cargo doc -p evmbin --release --open --no-deps --document-private-items

[parity-cla-bot]

It looks like @ltfschoen signed our Contributor License Agreement. 👍

Many thanks,

Parity Technologies CLA Bot

[dvdplm]

good stuff here.

[dvdplm]

Should these be doc-comments? I'm unsure of how the docs are rendered for private fields of a public struct.

[ltfschoen]

Thanks. I'm looking into it. If I update main.rs so that info.rs is public: pub mod info; then it shows that module in the rustdocs (it currently isn't at the moment), but if I then make this a doc comment (i.e. /// Chain ...) and then build the docs with cargo doc -p evmbin --release --open --no-deps i get the following error.

error: documentation comments cannot be applied to method arguments
   --> evmbin/./src/info.rs:105:2
    |
105 |     /// Chain specification name associated with the transaction
    |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ doc comments are not allowed here

[ltfschoen]

@dvdplm I tried to make the input parameters into this run_transaction into a struct called InputData, since we can use doc comments on structs. The approach that I tried has been pushed in this commit 8a6e644 (a separate branch to try it out), however I'm not sure how to overcome the ^^ undeclared lifetime error. In the InputData struct i'm not sure how to declare the lifetime parameter in addition to the generic type T (i.e. change pub struct InputData<T> { to pub struct InputData<'a, T> { or something?), and the existing code mut informant: T, is giving the following error and i'm not sure how to overcome that either:

122 |     mut informant: T,
    |     ^^^ expected identifier, found keyword
help: you can escape reserved keywords to use them as identifiers
    |
122 |     r#mut informant: T,

[dvdplm]

See 8a6e644#r33959733 – but I admit I'm not sure of the context here and why we need lifetimes on these values (prolly me being ignorant).

[ltfschoen]

Thanks for the tips! I somehow managed to get it to compile and I've created a separate PR #10769. However there's quite a bit of code duplication and I haven't figured out how to approach resolving yet.

[ordian]

@ltfschoen a merge with master would fix cargo audit failure (and resolve the merge conflict).

[debris]

I merged this pr with master. It looks good so far, but the tests are not finished. @ltfschoen can you please fix the tests so we can merge it in?

concerns addressed

[debris]

@dvdplm not entirely, tests in evmbin/src/main.rs are still not finished

[dvdplm]

@dvdplm not entirely, tests in evmbin/src/main.rs are still not finished

Ouch, sorry about that. @ordian can you take a look at that please? :/