Docker images for YAML Runtimes

• Quickstart
• Dependencies
• Usage
• Pull images
• Play
• Daemon
• Build
• Test
• Example
• Architecture
• List of Libraries

This project provides several Dockerfiles for runtime environments for a list of YAML processors.

The goal of this project is to be able to play with most available YAML libraries out there and compare them.

Libraries can be tested for bugs, and it makes it easier to try out bugfixes for a language for which you don't have any development libraries installed.

It was part of the YAML Editor and then outsourced in its own project.

The YAML editor allows to receive the output of a list of YAML processors in vim, with a very helpful automatic tiled layout and nice shortcuts.

It currently can work with only one docker container at a time, that's one reason why we build the alpine-runtime-all image.

Each library has one or more programs to process YAML input and output parsing events or JSON. In the docker images they can be found under the /yaml/bin directory.

Quickstart

source /path/to/yaml-runtimes/.rc

To play with PyYAML:

yamlrun-docker-pull python
docker run -i --rm yamlio/alpine-runtime-python py-pyyaml-event <t/data/input.yaml
docker run -i --rm yamlio/alpine-runtime-python py-pyyaml-json <t/data/input.yaml
docker run -i --rm yamlio/alpine-runtime-python py-pyyaml-yaml <t/data/input.yaml
docker run -i --rm yamlio/alpine-runtime-python py-pyyaml-py <t/data/input.yaml

To play with libyaml:

yamlrun-docker-pull static
docker run -i --rm yamlio/alpine-runtime-static c-libyaml-event <t/data/input.yaml
docker run -i --rm yamlio/alpine-runtime-static c-libyaml-yaml <t/data/input.yaml

Dependencies

For some operations, you need to install the following dependencies:

• perl
• YAML::PP perl module
• jq

To install YAML::PP, first install the cpanm client (debian example):

apt-get install cpanminus

Then install the module with cpanm:

# Install YAML::PP into the local/ directory
cpanm -l local --notest YAML::PP

The scripts will automatically use the local/ directory to search for modules. You could also do this manually by setting PERL5LIB:

export PERL5LIB=$PWD/local/lib/perl5

Usage

To list all libraries:

make list

Pull images

# pull alpine-runtime-static
make docker-pull-perl
# pull alpine-runtime-node
make docker-pull-node
# pull alpine-runtime-all
make docker-pull-all

Play

To play around with the several processors, call them like this:

docker run -i --rm yamlio/alpine-runtime-static c-libfyaml-event <t/data/input.yaml

To get a list of all available views:

make list-views-static
make list-views-all

Daemon

By default, for every test a docker run will be executed. To make testing a bit faster, you can run the containers in background:

# Start all containers
yamlrun-docker-start
# Only start alpine-runtime-perl container
yamlrun-docker-start perl

# Test
make testv

# Stop all containers
yamlrun-docker-stop
# Only stop alpine-runtime-perl container
yamlrun-docker-stop perl

# List containers
yamlrun-docker-status
# Restart all
yamlrun-docker-restart

Then you can run this instead:

docker exec -i alpine-runtime-static c-libfyaml-event <t/data/input.yaml

Also the tests (see below) will run docker exec instead automatically.

Build

To build all images, do

make build

Note that this can take a while.

You can also just build a single environment or library:

# build all javascript libraries
make node
# build C libyaml
make c-libyaml
# build perl YAML::PP
make perl-pp

To list all images, do

make list-images

Test

To see if the build was successful and all programs work, run

make test
# or verbose:
make testv

This will test each program with a simple YAML file.

To test only one runtime or library:

make test LIBRARY=c-libyaml
make testv RUNTIME=perl
make testv RUNTIME=all LIBRARY=hs-hsyaml
# Test all libraries in alpine-runtime-all image
make testv RUNTIME=all

Example

If you want to test a certain library, for example c-libfyaml, the steps would be:

make c-libfyaml
make list-images
make testv LIBRARY=c-libfyaml
docker run -i --rm yamlio/alpine-runtime-static c-libfyaml-event <t/data/input.yaml
docker run -i --rm yamlio/alpine-runtime-static c-libfyaml-json <t/data/input.yaml

Architecture

So far all docker images are based on Alpine Linux.

For each environment there is a builder image and a runtime image. The libraries are built in the builder containers, and all necessary files are then copied into the runtime images.

List of Libraries

Currently only official releases of the libraries are build. Allowing to build from sources like git might be added at some point.

The list of libraries and their configuration can be found in list.yaml.

Type make list to see the following list:

ID	Language	Name	Version	Runtime
c-libfyaml	C	libfyaml	0.7.12	static
c-libyaml	C	libyaml	0.2.5	static
cpp-rapidyaml	C++	rapidyaml	0.4.0	static
cpp-yamlcpp	C++	yaml-cpp	0.8.0	static
dotnet-yamldotnet	C#	YamlDotNet	11.2.1	dotnet
go-yaml	Go	go-yaml	v2	static
hs-hsyaml	Haskell	HsYAML	0.2.1.0	haskell
hs-reference	Haskell	YAMLReference	master	haskell
js-jsyaml	Javascript	js-yaml	4.1.0	node
js-yaml	Javascript	yaml	2.4.5	node
lua-lyaml	Lua	lyaml	6.2.8	lua
nim-nimyaml	Nim	NimYAML	0.16.0	static
perl-pp	Perl	YAML::PP	0.38.0	perl
perl-pplibyaml	Perl	YAML::PP::LibYAML	0.005	perl
perl-refparser	Perl	Generated RefParser	0.0.5	perl
perl-syck	Perl	YAML::Syck	1.34	perl
perl-tiny	Perl	YAML::Tiny	1.74	perl
perl-xs	Perl	YAML::XS (libyaml)	0.89	perl
perl-yaml	Perl	YAML.pm	1.31	perl
py-pyyaml	Python	PyYAML	5.3.1	python
py-ruamel	Python	ruamel.yaml	0.18.6	python
ruby-psych	Ruby	psych	4.0.3	ruby
rust-yamlrust	Rust	yaml	0.4.4	static

Contributing

See the Contributing Guidelines