Landscape2

Landscape2 is a tool that generates interactive landscapes websites.

You can check out how the generated landscapes look like by visiting this demo of the CNCF landscape. Additional landscapes generated by this tool can be found in the landscape2-sites repository.

Warning

Landscape2 is at an early stage of development. There may be breaking changes in the future to the command usage, flags, and configuration file formats.

How it works

Landscape2 is a CLI tool that generates static websites from the information available in the data sources provided. These data sources are passed to the tool via arguments, usually in the form of urls or local paths, and are as follows:

• Landscape data. The landscape data file is a YAML file that describes the items that will be displayed in the landscape website. For backwards compatibility reasons, this file must follow the format and conventions defined in the CNCF landscape.yml file.

• Landscape settings. The settings file is a YAML file that allows customizing some aspects of the generated landscape website, such as the logo, colors, how to group items or which ones should be featured. For more information about the settings file, please see the reference documentation.

• Landscape guide. The guide file is a YAML file that defines the content of the guide that will be displayed on the landscape website. For more information, please see the reference documentation.

• Logos location. Each landscape item must provide a valid relative reference to a logo image in SVG format in the landscape data file (item's logo field). The logos data source defines the location of those logos (base url or local path), so that the tool can get them as needed when processing the landscape items.

Data collection from external services

In addition to the information available in the landscape data file, the tool collects more data during the landscape generation from external sources (such as GitHub or Crunchbase) if the required credentials are provided. These credentials must be provided via environment variables.

• GitHub: a list of comma separated GitHub tokens with public_repo scope can be provided in the GITHUB_TOKENS environment variable. When these tokens are not provided no information from GitHub will be collected. If the expected number of items in the landscape is large it is recommended to provide more than one token to avoid hitting rate limits and speed up the collection of data (the concurrency of the process will be based on the number of tokens provided).

• Crunchbase: a Crunchbase API key can be provided in the CRUNCHBASE_API_KEY environment variable. If this token is not provided no information from Crunchbase will be collected.

Installation

The landscape2 CLI tool is distributed in a container image. This image can be used both to run the tool locally or from your CI workflows to automate the generation of landscapes. The landscape2-validate-action, which can be used to check that the landscape data file is valid, also uses this image.

Alternatively, it can also be easily built from the source.

Building from source

You can build landscape2 from the source by using Cargo, the Rust package manager. yarn is required during the installation process to build the web application, which will be embedded into the landscape2 binary as part of the build process.

$ cargo install --git https://github.com/cncf/landscape2

$ landscape2 --help

Landscape2 CLI tool

Usage: landscape2 <COMMAND>

Commands:
  build     Build landscape website
  deploy    Deploy landscape website (experimental)
  new       Create a new landscape from the built-in template
  serve     Serve landscape website
  validate  Validate landscape data sources files
  help      Print this message or the help of the given subcommand(s)

Usage

To see landscape2 in action, we will go through the process of creating, building and serving a new landscape from scratch. The following instructions will assume that the landscape2 binary is available in your PATH.

Alternatively, you can launch a container from the image provided where the landscape2 CLI tool is ready to use:

docker run -ti public.ecr.aws/g6m3a0y9/landscape2:latest sh

Creating a new landscape

The new subcommand allows us to create a new landscape from a built-in template. This template includes some sample data source files that you can use as a starting point for your landscapes.

The following command will create the directory my-landscape if it doesn't already exist and will copy into it the files in the built-in template:

$ landscape2 new --output-dir my-landscape

INFO new: landscape2::new: creating new landscape from the built-in template..
INFO new: landscape2::new: landscape created! (took: 0.003s)

✅ Landscape created successfully!

You can build it by running the following command:

👉 cd my-landscape && landscape2 build --data-file data.yml --settings-file settings.yml --guide-file guide.yml --logos-path logos --output-dir build

Building the landscape website

The build process is in charge of generating the landscape website from the information available in the datasources provided. Now we'll build the landscape we created in the previous step by using the build subcommand. Please note that the new subcommand already suggested us to do this in its output and even printed the full command to use for us.

Note

During the build process, landscape2 will try to take a screenshot of your landscape if the required settings were provided in the settings.yml file. This screenshot will be available for download from the generated web application (in PNG and PDF format), and is taken by launching Chrome/Chromium in headless mode. If Chrome/Chromium is not available, the screenshot won't be taken and a warning will be raised.

The following command will build the landscape and write the resulting files to the output-dir provided (build in this case):

$ cd my-landscape && landscape2 build --data-file data.yml --settings-file settings.yml --guide-file guide.yml --logos-path logos --output-dir build

INFO build: landscape2::build: building landscape website..
WARN build:collect_crunchbase_data: landscape2::build::crunchbase: crunchbase api key not provided: no information will be collected from crunchbase
WARN build:collect_github_data: landscape2::build::github: github tokens not provided: no information will be collected from github
INFO build: landscape2::build: landscape website built! (took: 0.555s)

✅ Landscape built successfully!

You can see it in action by running the following command:

👉 landscape2 serve --landscape-dir build

Important

Without the credentials required to collect data from external services (GitHub, Crunchbase, etc) the resulting site won't contain all the information available on the CNCF demo site. In this case, we didn't provide them intentionally, so we were warned about this in the command output (see WARN entries).

Serving a landscape

The result of the build process is a static website that you can deploy on your favorite hosting provider. To make it easier to try your landscapes, landscape2 includes a serve subcommand that will launch an HTTP server and serve the contents of your landscape. In our example, the build output displayed the command to do this, so we'll go ahead and give it a try:

$ landscape2 serve --landscape-dir build

INFO serve: landscape2::serve: http server running (press ctrl+c to stop)

🔗 Landscape available at: http://127.0.0.1:8000

If you visit http://127.0.0.1:8000 in your browser you should see the landscape you just created in action. Now you can iterate by editing the files in the my-landscape directory until your landscape is ready.

Note

The resulting website when building a landscape is a single-page application that handles routing on the client side. This means that you'll need to configure your webserver to serve the index.html file for the SPA route paths (like '/guide', '/stats', etc). One way of doing this would be to serve that file when a non existent path is requested. The serve subcommand included in landscape2 already handles this for us.

Performance considerations when building

Some operations like collecting data from external sources or processing a lot of logos images can take some time, specially in landscapes with lots of items. Landscape2 caches as much of this data as possible to make subsequent runs faster. Please keep this in mind when running the tool periodically from your workflows, and make sure the cache directory (set via --cache-dir) is saved and restored on each run. You can find some examples of how to achieve this in the workflows in the landscape2-sites repository.

Contributing

Please see CONTRIBUTING.md for more details.

Code of Conduct

This project follows the CNCF Code of Conduct.

License

Landscape2 is an Open Source project licensed under the Apache License 2.0.