This adds the ability to name a vat, this can be used for debugging
purposes and possibly in the...

The unmarshallers didn't match what syrup-decode expected.

- Fixed for what syrup-decode expect...

No need for the cond arrow plus identity hack

- Catch exceptions in the general case of each vat loop
- But also more specifically catch exc...

Left a todo to check the order

This lead the followup-question-finder to accidentally be listified

Otherwise the ordering is wrong!

Otherwise else, zero valued continuation error, so annoying

This exports the spawn-mycapn function from the captp module so that you
can use it to build net...

This removes some of the TODOs which were being used in place of crypto
and adds generating the ...

We'll want a safety check at some point but I guess we have two
conflicting structures between t...

- Somehow the list reading stuff didn't end up in the guile version,
fixed
- Switch to usin...

There was a bug where encoding pairs, syrup records and sets could
result in a syrup encoding er...

Just avoids having to deal with yet another path thing...

The other version might accidentally read a unicode character instead
of just one byte

This brief idea isn't even supported or used anymore. It was used in
the brief period where aut...

This ports the all-of joiners from the racket version. These joiners are
used to create one prom...

This is to support cross compilation (assuming I got it right)

Eventually we'll bundle a captp explainer here or somewhere else.
This is the simplest fit for now.

This removes the old macro for defining both the record
and (un)marshalling functions and replac...

- Remove no-longer needed import of (fibers internal) from vat.scm
- Switch default-vat-schedu...

- methods* and extend-methods now take the fallback as their first
argument
- however, meth...

We're maintaining the autotools files on our own now!

Neither name is great, but this is better

They don't put a comment before the return and so I guess we won't
either.

Mostly ported from the Racket docs, a bit from the Spritely core
whitepaper.

GitLab doesn't seem to recognize our symlink from COPYING to
LICENSE.txt :P

`(usleep 500)` isn't the same as half a second, `(usleep 500000)` is.
Regardless, I guess we'll ...

It seems that this test is failing not because of promise pipelining
on errors specifically, but...

This version was in the Racket version of Goblins, and appears to test
a different case!

Previous version of test was using `#:promise? #t` alongside a `#:catch`,
but `#:catch` will ins...

This mostly changes uses of vat to the new syntax that assumes the
`'run` method. This also conv...