Updated solidity to 0.6.2

[g-r-a-n-t]

This is part of some other work I'm doing here.

[spalladino]

Hey there! I'm interested in picking this up for a side project, and when trying to run the tests on this branch with cargo test I'm getting the following error after the 2nd call to compile:

{"errors":[{"component":"general","formattedMessage":"Internal exception in StandardCompiler::compile: /solidity/libsolidity/interface/CompilerStack.cpp(90): Throw in function solidity::frontend::CompilerStack::CompilerStack(const Callback&)\nDynamic exception type: boost::exception_detail::clone_impl<solidity::langutil::InternalCompilerError>\nstd::exception::what: You shall not have another CompilerStack aside me.\n[solidity::util::tag_comment*] = You shall not have another CompilerStack aside me.\n","message":"Internal exception in StandardCompiler::compile: /solidity/libsolidity/interface/CompilerStack.cpp(90): Throw in function solidity::frontend::CompilerStack::CompilerStack(const Callback&)\nDynamic exception type: boost::exception_detail::clone_impl<solidity::langutil::InternalCompilerError>\nstd::exception::what: You shall not have another CompilerStack aside me.\n[solidity::util::tag_comment*] = You shall not have another CompilerStack aside me.\n","severity":"error","type":"InternalCompilerError"}]}

Where the You shall not have another CompilerStack aside me seems to come from here.

@g-r-a-n-t did you face this issue as well? @axic what would be needed to fully clear out the outstanding compiler stack, so as to call compile more than once and run multiple tests? grant seems to be properly calling free on the compile result and a full reset as well (see here). What else may be missing?

[g-r-a-n-t]

Hi @spalladino, thanks for reaching out!

More details on why this is happening can be found here. Basically, we can't call solc from multiple threads anymore. This is problematic with Rust tests because they run in parallel by default.

I think it would be best to just lock calls to solidity_compile within solc-rust, effectively shielding users from the fact that solc is not re-entrant. It should be pretty straight forward. Once this is fixed, I think we're good to merge this.

Let me know if you would like more info or suggestions :)

[spalladino]

Cool, thanks a lot Grant! Just gave a shot at it here. Let me know if you'd prefer if I sent the PR to your fork instead.

[g-r-a-n-t]

see #2