Implement initial Lumi-as-a-library
This is the initial step towards redefining Lumi as a library that runs
atop vanilla Node.js/V8, rather than as its own runtime.
This change is woefully incomplete but this includes some of the more
stable pieces of my current work-in-progress.
The new structure is that within the sdk/ directory we will have a client
library per language. This client library contains the object model for
Lumi (resources, properties, assets, config, etc), in addition to the
"language runtime host" components required to interoperate with the
Lumi resource monitor. This resource monitor is effectively what we call
"Lumi" today, in that it's the thing orchestrating plans and deployments.
Inside the sdk/ directory, you will find nodejs/, the Node.js client
library, alongside proto/, the definitions for RPC interop between the
different pieces of the system. This includes existing RPC definitions
for resource providers, etc., in addition to the new ones for hosting
different language runtimes from within Lumi.
These new interfaces are surprisingly simple. There is effectively a
bidirectional RPC channel between the Lumi resource monitor, represented
by the lumirpc.ResourceMonitor interface, and each language runtime,
represented by the lumirpc.LanguageRuntime interface.
The overall orchestration goes as follows:
1) Lumi decides it needs to run a program written in language X, so
it dynamically loads the language runtime plugin for language X.
2) Lumi passes that runtime a loopback address to its ResourceMonitor
service, while language X will publish a connection back to its
LanguageRuntime service, which Lumi will talk to.
3) Lumi then invokes LanguageRuntime.Run, passing information like
the desired working directory, program name, arguments, and optional
configuration variables to make available to the program.
4) The language X runtime receives this, unpacks it and sets up the
necessary context, and then invokes the program. The program then
calls into Lumi object model abstractions that internally communicate
back to Lumi using the ResourceMonitor interface.
5) The key here is ResourceMonitor.NewResource, which Lumi uses to
serialize state about newly allocated resources. Lumi receives these
and registers them as part of the plan, doing the usual diffing, etc.,
to decide how to proceed. This interface is perhaps one of the
most subtle parts of the new design, as it necessitates the use of
promises internally to allow parallel evaluation of the resource plan,
letting dataflow determine the available concurrency.
6) The program exits, and Lumi continues on its merry way. If the program
fails, the RunResponse will include information about the failure.
Due to (5), all properties on resources are now instances of a new
Property<T> type. A Property<T> is just a thin wrapper over a T, but it
encodes the special properties of Lumi resource properties. Namely, it
is possible to create one out of a T, other Property<T>, Promise<T>, or
to freshly allocate one. In all cases, the Property<T> does not "settle"
until its final state is known. This cannot occur before the deployment
actually completes, and so in general it's not safe to depend on concrete
resolutions of values (unlike ordinary Promise<T>s which are usually
expected to resolve). As a result, all derived computations are meant to
use the `then` function (as in `someValue.then(v => v+x)`).
Although this change includes tests that may be run in isolation to test
the various RPC interactions, we are nowhere near finished. The remaining
work primarily boils down to three things:
1) Wiring all of this up to the Lumi code.
2) Fixing the handful of known loose ends required to make this work,
primarily around the serialization of properties (waiting on
unresolved ones, serializing assets properly, etc).
3) Implementing lambda closure serialization as a native extension.
This ongoing work is part of pulumi/pulumi-fabric#311.
2017-08-26 21:07:54 +02:00
|
|
|
#!/bin/bash
|
2017-02-10 18:08:06 +01:00
|
|
|
#
|
2018-07-12 22:29:35 +02:00
|
|
|
# This script regenerates all Protobuf/gRPC client files.
|
2017-02-10 18:08:06 +01:00
|
|
|
#
|
2018-07-12 22:29:35 +02:00
|
|
|
# For now, it must be run manually, and the results are checked into source control. Eventually we
|
|
|
|
|
# might choose to automate this process as part of the overall build so that it's less manual and
|
|
|
|
|
# hence error prone.
|
2017-02-10 18:08:06 +01:00
|
|
|
#
|
2018-07-12 22:29:35 +02:00
|
|
|
# This script relies only on Docker. The container holds the installation of gRPC, tools, etc., for
|
2019-06-18 23:39:42 +02:00
|
|
|
# different languages, so nothing is else required to be installed on your machine.
|
2017-02-10 18:08:06 +01:00
|
|
|
set -e
|
|
|
|
|
|
2018-08-04 19:27:36 +02:00
|
|
|
# First build our Protobuf/gRPC compiler Docker image, so dev machines don't need it.
|
|
|
|
|
echo "* Building Protobuf/gRPC compilers:"
|
|
|
|
|
docker build -t pulumi/protobuf-builder .
|
|
|
|
|
|
2018-07-12 22:29:35 +02:00
|
|
|
DOCKER_RUN="docker run -it --rm -v $(pwd)/../python:/python -v $(pwd)/../nodejs:/nodejs -v $(pwd):/local pulumi/protobuf-builder"
|
|
|
|
|
PROTOC="$DOCKER_RUN protoc"
|
|
|
|
|
|
|
|
|
|
# `status.proto` is in our source tree so that we can implement initialization failure in the
|
|
|
|
|
# dynamic client (written in JS) using the protobuf notion of "details" -- arbitrary protobuf
|
|
|
|
|
# messages packaged up inside of an error. Hence, `JS_PROTO_FILES` includes it and `PROTO_FILES`
|
|
|
|
|
# does not.
|
2018-06-30 01:14:49 +02:00
|
|
|
PROTO_FILES=$(find . -name "*.proto" -not -name "status.proto")
|
|
|
|
|
JS_PROTO_FILES=$(find . -name "*.proto")
|
2017-02-10 18:28:46 +01:00
|
|
|
|
2018-08-04 19:27:36 +02:00
|
|
|
echo "* Generating Protobuf/gRPC SDK files:"
|
2018-07-12 22:29:35 +02:00
|
|
|
echo -e "\tVERSION: $($PROTOC --version)"
|
|
|
|
|
echo -e "Generated by version $($PROTOC --version) of protoc" > ./grpc_version.txt
|
2017-02-10 18:28:46 +01:00
|
|
|
|
2017-12-21 18:24:48 +01:00
|
|
|
GO_PULUMIRPC=./go
|
Implement initial Lumi-as-a-library
This is the initial step towards redefining Lumi as a library that runs
atop vanilla Node.js/V8, rather than as its own runtime.
This change is woefully incomplete but this includes some of the more
stable pieces of my current work-in-progress.
The new structure is that within the sdk/ directory we will have a client
library per language. This client library contains the object model for
Lumi (resources, properties, assets, config, etc), in addition to the
"language runtime host" components required to interoperate with the
Lumi resource monitor. This resource monitor is effectively what we call
"Lumi" today, in that it's the thing orchestrating plans and deployments.
Inside the sdk/ directory, you will find nodejs/, the Node.js client
library, alongside proto/, the definitions for RPC interop between the
different pieces of the system. This includes existing RPC definitions
for resource providers, etc., in addition to the new ones for hosting
different language runtimes from within Lumi.
These new interfaces are surprisingly simple. There is effectively a
bidirectional RPC channel between the Lumi resource monitor, represented
by the lumirpc.ResourceMonitor interface, and each language runtime,
represented by the lumirpc.LanguageRuntime interface.
The overall orchestration goes as follows:
1) Lumi decides it needs to run a program written in language X, so
it dynamically loads the language runtime plugin for language X.
2) Lumi passes that runtime a loopback address to its ResourceMonitor
service, while language X will publish a connection back to its
LanguageRuntime service, which Lumi will talk to.
3) Lumi then invokes LanguageRuntime.Run, passing information like
the desired working directory, program name, arguments, and optional
configuration variables to make available to the program.
4) The language X runtime receives this, unpacks it and sets up the
necessary context, and then invokes the program. The program then
calls into Lumi object model abstractions that internally communicate
back to Lumi using the ResourceMonitor interface.
5) The key here is ResourceMonitor.NewResource, which Lumi uses to
serialize state about newly allocated resources. Lumi receives these
and registers them as part of the plan, doing the usual diffing, etc.,
to decide how to proceed. This interface is perhaps one of the
most subtle parts of the new design, as it necessitates the use of
promises internally to allow parallel evaluation of the resource plan,
letting dataflow determine the available concurrency.
6) The program exits, and Lumi continues on its merry way. If the program
fails, the RunResponse will include information about the failure.
Due to (5), all properties on resources are now instances of a new
Property<T> type. A Property<T> is just a thin wrapper over a T, but it
encodes the special properties of Lumi resource properties. Namely, it
is possible to create one out of a T, other Property<T>, Promise<T>, or
to freshly allocate one. In all cases, the Property<T> does not "settle"
until its final state is known. This cannot occur before the deployment
actually completes, and so in general it's not safe to depend on concrete
resolutions of values (unlike ordinary Promise<T>s which are usually
expected to resolve). As a result, all derived computations are meant to
use the `then` function (as in `someValue.then(v => v+x)`).
Although this change includes tests that may be run in isolation to test
the various RPC interactions, we are nowhere near finished. The remaining
work primarily boils down to three things:
1) Wiring all of this up to the Lumi code.
2) Fixing the handful of known loose ends required to make this work,
primarily around the serialization of properties (waiting on
unresolved ones, serializing assets properly, etc).
3) Implementing lambda closure serialization as a native extension.
This ongoing work is part of pulumi/pulumi-fabric#311.
2017-08-26 21:07:54 +02:00
|
|
|
GO_PROTOFLAGS="plugins=grpc"
|
2017-12-21 18:24:48 +01:00
|
|
|
echo -e "\tGo: $GO_PULUMIRPC [$GO_PROTOFLAGS]"
|
|
|
|
|
mkdir -p $GO_PULUMIRPC
|
2018-07-12 22:29:35 +02:00
|
|
|
$PROTOC --go_out=$GO_PROTOFLAGS:$GO_PULUMIRPC $PROTO_FILES
|
Implement initial Lumi-as-a-library
This is the initial step towards redefining Lumi as a library that runs
atop vanilla Node.js/V8, rather than as its own runtime.
This change is woefully incomplete but this includes some of the more
stable pieces of my current work-in-progress.
The new structure is that within the sdk/ directory we will have a client
library per language. This client library contains the object model for
Lumi (resources, properties, assets, config, etc), in addition to the
"language runtime host" components required to interoperate with the
Lumi resource monitor. This resource monitor is effectively what we call
"Lumi" today, in that it's the thing orchestrating plans and deployments.
Inside the sdk/ directory, you will find nodejs/, the Node.js client
library, alongside proto/, the definitions for RPC interop between the
different pieces of the system. This includes existing RPC definitions
for resource providers, etc., in addition to the new ones for hosting
different language runtimes from within Lumi.
These new interfaces are surprisingly simple. There is effectively a
bidirectional RPC channel between the Lumi resource monitor, represented
by the lumirpc.ResourceMonitor interface, and each language runtime,
represented by the lumirpc.LanguageRuntime interface.
The overall orchestration goes as follows:
1) Lumi decides it needs to run a program written in language X, so
it dynamically loads the language runtime plugin for language X.
2) Lumi passes that runtime a loopback address to its ResourceMonitor
service, while language X will publish a connection back to its
LanguageRuntime service, which Lumi will talk to.
3) Lumi then invokes LanguageRuntime.Run, passing information like
the desired working directory, program name, arguments, and optional
configuration variables to make available to the program.
4) The language X runtime receives this, unpacks it and sets up the
necessary context, and then invokes the program. The program then
calls into Lumi object model abstractions that internally communicate
back to Lumi using the ResourceMonitor interface.
5) The key here is ResourceMonitor.NewResource, which Lumi uses to
serialize state about newly allocated resources. Lumi receives these
and registers them as part of the plan, doing the usual diffing, etc.,
to decide how to proceed. This interface is perhaps one of the
most subtle parts of the new design, as it necessitates the use of
promises internally to allow parallel evaluation of the resource plan,
letting dataflow determine the available concurrency.
6) The program exits, and Lumi continues on its merry way. If the program
fails, the RunResponse will include information about the failure.
Due to (5), all properties on resources are now instances of a new
Property<T> type. A Property<T> is just a thin wrapper over a T, but it
encodes the special properties of Lumi resource properties. Namely, it
is possible to create one out of a T, other Property<T>, Promise<T>, or
to freshly allocate one. In all cases, the Property<T> does not "settle"
until its final state is known. This cannot occur before the deployment
actually completes, and so in general it's not safe to depend on concrete
resolutions of values (unlike ordinary Promise<T>s which are usually
expected to resolve). As a result, all derived computations are meant to
use the `then` function (as in `someValue.then(v => v+x)`).
Although this change includes tests that may be run in isolation to test
the various RPC interactions, we are nowhere near finished. The remaining
work primarily boils down to three things:
1) Wiring all of this up to the Lumi code.
2) Fixing the handful of known loose ends required to make this work,
primarily around the serialization of properties (waiting on
unresolved ones, serializing assets properly, etc).
3) Implementing lambda closure serialization as a native extension.
This ongoing work is part of pulumi/pulumi-fabric#311.
2017-08-26 21:07:54 +02:00
|
|
|
|
2019-01-30 02:07:47 +01:00
|
|
|
# Protoc for JavaScript has a bug where it emits Google Closure Compiler directives in the module prologue that mutate
|
|
|
|
|
# the global object, which causes side-by-side bugs in pulumi/pulumi (pulumi/pulumi#2401). The protoc compiler
|
|
|
|
|
# absolutely should not be emitting commonjs modules that mutate global, but alas, it does, and we have to sed the
|
|
|
|
|
# output to not do that.
|
|
|
|
|
#
|
|
|
|
|
# We're replacing the literal code string
|
|
|
|
|
# var global = Function('return this')();
|
|
|
|
|
# with
|
|
|
|
|
# var proto = { pulumirpc: {} }, global = proto;
|
|
|
|
|
#
|
|
|
|
|
# This sets up the remainder of the protobuf file so that it works fine, but doesn't mess with global.
|
|
|
|
|
$DOCKER_RUN /bin/bash -c 'set -x && JS_PULUMIRPC=/nodejs/proto && \
|
|
|
|
|
JS_PROTOFLAGS="import_style=commonjs,binary" && \
|
|
|
|
|
JS_HACK_PROTOS=$(find . -name "*.proto" -not -name "status.proto") && \
|
|
|
|
|
echo -e "\tJS: $JS_PULUMIRPC [$JS_PROTOFLAGS]" && \
|
|
|
|
|
TEMP_DIR=/tmp/nodejs-build && \
|
|
|
|
|
echo -e "\tJS temp dir: $TEMP_DIR" && \
|
|
|
|
|
mkdir -p "$TEMP_DIR" && \
|
|
|
|
|
protoc --js_out=$JS_PROTOFLAGS:$JS_PULUMIRPC --grpc_out=minimum_node_version=6:$JS_PULUMIRPC --plugin=protoc-gen-grpc=/usr/local/bin/grpc_tools_node_protoc_plugin status.proto && \
|
|
|
|
|
protoc --js_out=$JS_PROTOFLAGS:$TEMP_DIR --grpc_out=minimum_node_version=6:$TEMP_DIR --plugin=protoc-gen-grpc=/usr/local/bin/grpc_tools_node_protoc_plugin $JS_HACK_PROTOS && \
|
|
|
|
|
sed -i "s/^var global = .*;/var proto = { pulumirpc: {} }, global = proto;/" "$TEMP_DIR"/*.js && \
|
|
|
|
|
cp "$TEMP_DIR"/*.js "$JS_PULUMIRPC"'
|
2018-06-26 20:14:03 +02:00
|
|
|
|
|
|
|
|
function on_exit() {
|
|
|
|
|
rm -rf "$TEMP_DIR"
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
# Protoc for Python has a bug where, if your proto files are all in the same directory relative
|
|
|
|
|
# to one another, imports of said proto files will produce imports that don't work using Python 3.
|
|
|
|
|
#
|
|
|
|
|
# Since our proto files are all in the same directory, this little bit of sed rewrites the broken
|
|
|
|
|
# imports that protoc produces, of the form
|
|
|
|
|
# import foo_pb2 as bar
|
|
|
|
|
# to the form
|
|
|
|
|
# from . import foo_pb2 as bar
|
|
|
|
|
# This form is semantically equivalent and is accepted by both Python 2 and Python 3.
|
|
|
|
|
trap on_exit EXIT
|
|
|
|
|
|
|
|
|
|
echo -e "\tPython temp dir: $TEMP_DIR"
|
2018-07-12 22:29:35 +02:00
|
|
|
$DOCKER_RUN /bin/bash -c 'PY_PULUMIRPC=/python/lib/pulumi/runtime/proto/ && \
|
|
|
|
|
echo -e "\tPython: $PY_PULUMIRPC" && \
|
|
|
|
|
TEMP_DIR="/tmp/python-build" && \
|
|
|
|
|
echo -e "\tPython temp dir: $TEMP_DIR" && \
|
|
|
|
|
mkdir -p "$TEMP_DIR" && \
|
|
|
|
|
python -m grpc_tools.protoc -I./ --python_out="$TEMP_DIR" --grpc_python_out="$TEMP_DIR" *.proto && \
|
|
|
|
|
sed -i "s/^import \([^ ]*\)_pb2 as \([^ ]*\)$/from . import \1_pb2 as \2/" "$TEMP_DIR"/*.py && \
|
|
|
|
|
cp "$TEMP_DIR"/*.py "$PY_PULUMIRPC"'
|
2017-02-10 18:08:06 +01:00
|
|
|
|
2018-08-04 19:27:36 +02:00
|
|
|
echo "* Done."
|
