This mega patch primarily introduces a new function engine. The main
reasons for this new engine are:
1) Massively improved performance with lock-contended graphs.
Certain large function graphs could have very high lock-contention which
turned out to be much slower than I would have liked. This new algorithm
happens to be basically lock-free, so that's another helpful
improvement.
2) Glitch-free function graphs.
The function graphs could "glitch" (an FRP term) which could be
undesirable in theory. In practice this was never really an issue, and
I've not explicitly guaranteed that the new graphs are provably
glitch-free, but in practice things are a lot more consistent.
3) Simpler graph shape.
The new graphs don't require the private channels. This makes
understanding the graphs a lot easier.
4) Branched graphs only run half.
Previously we would run two pure side of an if statement, and while this
was mostly meant as an early experiment, it stayed in for far too long
and now was the right time to remove this. This also means our graphs
are much smaller and more efficient too.
Note that this changed the function API slightly. Everything has been
ported. It's possible that we introduce a new API in the future, but it
is unexpected to cause removal of the two current APIs.
In addition, we finally split out the "schedule" aspect from
world.schedule(). The "pick me" aspects now happen in a separate
resource, rather than as a yucky side-effect in the function. This also
lets us more precisely choose when we're scheduled, and we can observe
without being chosen too.
As usual many thanks to Sam for helping through some of the algorithmic
graph shape issues!
The nil type is being added, but only slightly. It is not meant for real
use in the language. It will not work in all situations, and it isn't
implemented for many methods.
It is being added only for being a "dummy" placeholder value in the
engine when we have an unused value being propagated along an edge.
Most of the time, we don't need to have a dynamic call sub graph, since
the actual function call could be represented statically as it
originally was before lambda functions were implemented. Simplifying the
graph shape has important performance benefits in terms of both keep the
graph smaller (memory, etc) and in avoiding the need to run transactions
at runtime (speed) to reshape the graph.
Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
Instead of constantly making these updates, let's just remove the year
since things are stored in git anyways, and this is not an actual modern
legal risk anymore.
We forgot to omit looking deeper into private struct fields. I don't
know why we didn't catch this earlier, I can only assume some subtlety
changed, since we've previously used many of the resources this would
fail on. Maybe golang broke some API that they didn't consider stable?
This also adds a new test for this, and ensures each resource can be
inspected too!
When we look at unification variables from two different places, the
default printer will always start numbering them from ?1 and therefore
if we look at two unrelated systems, they might both print as ?1 when
they are in fact different pointers.
We don't collect them all by default since it's usually not necessary
except for debugging, but in those situations, we want a consistent
unification store which we can pass around to get sensible debug output.
This is used for representing a unification variable in our type during
type unification. For example, this allows us to have a [?1] or a
map{?1:[?2]} and so on...
Plumb through the standard context.Context so that a function can be
cancelled if someone requests this. It makes it less awkward to write
simple functions that might depend on io or network access.
If we want to use special struct types from our CLI parser, we also need
to be able to both identify, and convert them to our language type and
value representations.
For as long as we don't have fancier types in our language, these should
both be strings. Tests and extensions to these additions are welcome!
With the recent merging of embedded package imports and the entry CLI
package, it is now possible for users to build in mcl code into a single
binary. This additional permission makes it explicitly clear that this
is permitted to make it easier for those users. The condition is phrased
so that the terms can be "patched" by the original author if it's
necessary for the project. For example, if the name of the language
(mcl) changes, has a differently named new version, someone finds a
phrasing improvement or a legal loophole, or for some other
reasonable circumstance. Now go write some beautiful embedded tools!
If stringer is run from GOROOT, it can't find any packages; same error
as in https://go.dev/issue/31843. And since GOCACHE is always ignored
(see issue above) we can have a bare go generate command.
If we have rare, but special *interface{} values in resource structs, we
should be able to handle them normally. It's really not recommended that
you use these unless you know exactly why they are useful.
This lets us add a resource that has an implementation with a field
whose type is determined at compile time. This let's us write more
flexible resources.
What's missing is additional type checking so that we guarantee that a
specific resource doesn't change types during run-time.
If we use this to generate a zero value, we want to make sure it's
completely initialized in case we use it subsequently. We also improve
the docs at the same time.
This will probably change up again in the future, but this is what we
need for now to make lambdas work.
Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
We should check this for safety. An error is better than a panic. If we
try to set an unexported field, this would panic. We should prevent
being able to even type unify that though!
There were some bugs about setting resource fields that were structs
with various fields. This makes things more strict and correct. Now we
check for duplicate field names earlier (duplicates due to identical
aliases) and we also don't try and set private fields, or incorrectly
set partial structs.
Most interestingly, this also cleans up all of the resources and ensures
that each one has nicer docs and a clear struct tag for fields that we
want to use in mcl. These are mandatory now, and if you're missing the
tag, then we will ignore the field.
There were a bunch of packages that weren't well documented. With the
recent split up of the lang package, I figured it would be more helpful
for new contributors who want to learn the structure of the project.
This is mainly meant as a useful test case, but might as well have it be
fun too. As an aside, it taught me a surprising result about the %v verb
in printf, and we'll have to decide if it's an issue we care about.
https://github.com/golang/go/issues/46118
The interesting thing about this method is that it uses the simplepoly
API but has no input args-- only the output types are different. If it
had identical types in the input args, that might also have been
interesting, but it's more rare to have none. Hopefully this exercises
our type unification logic.
Map and list types are now unconditionally initialised during an Into()
call to ensure that the only data within them after the operation is
that added by the Into() function.
Prior to this change, map/list types would likely not be cleared prior
to the data being inserted into them with a few exceptions. Nil
pointers or maps/lists that were not sufficient in capacity would be
reinitialised and used to replace the existing backing data store. In
some cases this wouldn't occur meaning any residual data existing in the
container before the Into() call could persist after the data copy
completes. This behaviour is wildly inconsistent and not ideal in the
vast majority of cases. It should be assumed that the Into() call will
preserve nothing and always produce a consistent and deterministic
output.
Signed-off-by: Joe Groocock <me@frebib.net>
Some forms of reflect.Value can cause ValueOf() to panic when there is a
nil pointer somewhere within the reflect.Value, whether that be a
container type like a struct, list or map, or just a raw nil pointer.
In these cases, ValueOf() attempted to dereference the pointer without
ever checking if it was nil. mgmt lang doesn't have pointers of any
kind, so these Golang values cannot be represented in mcl types in the
current form so return a helpful error to the user.
Signed-off-by: Joe Groocock <me@frebib.net>
