This adds basic support for streaming files directly from the download
server. This avoids clients timing out if they are blocked while first
waiting for a giant file to download.
The earlier path mangling code was incorrect. I've taken more time to
understand the correct use case and I've improved it. I've also split
out the parser logic and added tests, so this should either stay stable
or grow new tests and fixes if we find new issues.
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!
This adds a new caching http proxy resource that can be autogrouped into
the core http:server resource, and which caches and serves files that it
received from a different http server. It first pulls them down when
they are initially requested, which makes it possible for us to use this
for provisioning a Linux installation without having to pre-rsync the
entire package repository.
A regression in 4b0cdf9123 caused the
basic send/recv functionality to break for simple scenarios. This was
due to inadequate testing, and a partial misunderstanding of the
situation.
New testing should hopefully catch more cases, but send/recv and
compile-time checks are still not as complete as is probably possible.
This adds a new http:flag resource which can autogroup into an
http:server resource to receive actions from client HTTP requests, and
forward these values on to other resources.
When graph swapping (which is quite common) we only use the newly-made
resource if the Cmp function between the two shows a difference. If the
old resource has previously received a value via send/recv, then when it
is compared to the new value, it will almost always be different. As a
result, we need to run send/recv on the newly made graph to make sure it
has up-to-date values before we compare. This has to happen after
autogrouping since the resources can often be autogrouped and any child
grouped resource will cause a remake of all of the other children and
parents.
It turns out that the actual send/recv properties were being compared as
well, and for unknown reasons (tunnel vision perhaps) they are often not
identical. Skip comparing these for now until we find a fix or
understanding of how to make them identical.
It can be used in more places if it's not tied to the engine struct.
This also changes the signature so that more information is returned.
This can be used for logging or other useful things. Of note, this
happens to be the same struct as already exists. It's used for
convenience since it happens to match up! Of course they're related.
This gives us a simple mapping between a new resource and an old one. We
compare by kind and name because those two values are our uniqueness
constraint in the resource graphs.
This improves the autogrouping algorithm to support hierarchical
autogrouping. It's not guaranteed to work if we replace the reachability
grouper with something more efficient, but it's good enough for now.
Previously the resource could only set values in a per-hostname
namespace, but for single, user-managed values, we'd like to be able to
control things entirely. Now this resource can do that.
This fixes two small races we had while simultaneously reading from and
writing to the vertex timestamp, and simultaneously writing to the
vertex isStateOK (dirty) flag.
They were actually "safe" races, in that it doesn't matter if the
read/write race got the old or new value, or that the double write
happened. The time sequencing was correct (I believe) in both cases, but
this triggers the race detector now that we have tests for it.
We've previously not received a value from within an autogrouped
resource. It turns out this would be quite useful, and so this patch
implements the additional plumbing and testing so that this works!
Testing that an autogrouped resource can still send values has not been
done at this time.
This adds a new series of "get*" functions which can read values from
the associated "value" resources. The key name of the function must
match the name value of the resource for things to work.
Type unification isn't yet perfect in these scenarios, so you should use
casually and with caution.
This expands the Local API with the first (and in theory, only ever) API
for reading and writing simple values. This is a coordination point for
resources and functions to share things directly.
This is a new API that is similar in spirit and plumbing to the World
API, but it intended for all local machine operations and will likely
only ever have one implementation.
