Got it - makes sense!

@jcmcdonald temper your expectations - it doesn't do the thing yet :)

@wdede feel free to apply these changes yourself and update this differential - or we can do it in a session!

Hi @emlarson - did you run every command found at C++ and C Development Environment? That error indicates that Linux was asked to locate a binary called cmake at line 85 of the Makefile and it was not found on your path. You could install it by itself with sudo apt install cmake, but if it's not on your system, you may be missing other environment components that would still cause your build to fail at a later step. The line in question from that page that installs this build tool is sudo apt install build-essential cmake valgrind kcachegrind vim emacs cccc. Hope this helps!

It was probably a mistake not to start off like this, cascading changes.

• Begin ParseResult<T>

Hi @emlarson - you want the private key for the ssh-add command, just ~/.ssh/id_rsa or similar. The public key is what would be uploaded to Phabricator/GitHub/etc.

I thought that would "just work" but it didn't.

• Pure UnitParser::chop() but it doesn't build.

Yup, just a big ol' clang-format. Thanks!

In D288#9829, @jcmcdonald wrote:

@bdlovy Simultaneous diffs can be tricky. I'd go ahead and move the changes from one into the other. The idea is to have diffs that land faster, and making them small helps with that, but sometimes you just can't keep it small.

Alternatively, you could land one, and then rebase the other off the latest master.

• Found the problem, notes for the fix

Ah, meant to create a new diff, forgot I branched off the branch. We just talked about small diffs - @jcmcdonald would you prefer I abandon this and start over as two separate, or just rename this and do it all at once?

• Fix some but not all of the tests

ah thanks - I was building the stable branch, just needed to checkout master.

@jcmcdonald I'm unable to build eventpp locally as found in libdeps when -Werror is used - found that odd, it's not even our code. By removing that flag flag from PawLIB and SIMPLExpress I could get a working build. Did the clang upgrade change how we need to be including that code?

Thanks! Tracking for rR Ratscript and rS SIMPLEXpress in T1323.

D281 covers code and tests, and I already did the Wiki update. Sphinx docs are also non-existant for this project - I will be producing those all at once, and they will naturally reflect the new syntax. This is done.

@jcmcdonald Thanks for wrangling this monster into submission. Much appreciated. I'm goping to shoot for much (much) more focused DRs - but in this case, S-sB00 was never expected to pass to satisfy the original goal, just S-sB01.

Looks good to me!

• Address lints

• use new enum classes in Unit::specifiers()

@jcmcdonald against all odds, I think I've actually addressed everything necessary for review! Still leaving:

• Incorporate comments - version numbers, authors, swap c for ch

• ReservedCharacter enum class
• Pure ParseResult::parse()
• Notes for splitting up large function

Just need to purify the parser functions and split up my monster function.

• Remove case brackets, swap tuple for class

• Whitespace tweaks
• Use LetterCase enum class

• SpecifierType enum

• Additional refactors

• Convert spaces to tabs
• Format notes

Address low hanging fruit.

This is awesome - thanks so much!

• Remove '^', use '/' for open and close