This README explains the user cases, source file organization, and procedures for building the Blockstack documentation. You can find the documentation at https://docs.blockstack.com

Blockstack is a ecosystem build around a platform. There are several types of users to support with the documentation. Types are exist when they can operate within a vertical of the ecosystem. These are the users that can appear within this ecosystem and that the docs must support.

Finally, a key user set but seldom mentioned for any company docs is the company employees. These users are expected to make use of the documentation when onboarding or to support other users.

Our documentation is written in Markdown (.md), built using Jekyll, and deployed to a Netlify server. Serving the content from Netlify allows us to use functionality (plugins/javascript) not supported with standard GitHub pages.

Blockstack versions it source files in a public GitHub repo (duh :smile). You can submit changes by cloning, forking, and submitting a pull request. You can also make use of the Edit this page on GitHub link from any https://docs.blockstack.org page.

Some content is single sourced. Modifying single source content requires an understanding of that concept, how it works with Liquid tags, and the organization of this repo's source files.

UIKit provides the documentation theme. There is explicit use of HTML plus UIKit components directly in files where needed for special layouts. And there is use of CSV and JSON source files transformed with Liquid template language to produce reference content.

Directories that contain information used to build content.

These are the other directories in the site structure:

If you are making significant changes to the documentation, you should build and view the entire set locally before submitting a PR.

To run locally:

1. Install Jekyll into your workstation environment

2. Build and serve locally.

   bundle exec jekyll serve --config _config.yml,staticman.yml
   

   Use this format to turn on production features:

   JEKYLL_ENV=production bundle exec jekyll serve --config _config.yml
   

You can also do a test deploy using a tool like Surge.

cd _site
surge

Make sure you delete the deployed Surge domain when you are done. Using the teardown command.

  surge – single command web publishing. (v0.21.3)

  Usage:
    surge <project> <domain>

  Options:
    -a, --add           adds user to list of collaborators (email address)
    -r, --remove        removes user from list of collaborators (email address)
    -V, --version       show the version number
    -h, --help          show this help message

  Additional commands:
    surge whoami        show who you are logged in as
    surge logout        expire local token
    surge login         only performs authentication step
    surge list          list all domains you have access to
    surge teardown      tear down a published project
    surge plan          set account plan

  Guides:
    Getting started     surge.sh/help/getting-started-with-surge
    Custom domains      surge.sh/help/adding-a-custom-domain
    Additional help     surge.sh/help

  When in doubt, run surge from within your project directory.

The documentation is deployed to Netlify using the following command:

```
JEKYLL_ENV=production bundle exec jekyll build --config _config.yml

The _data/cliRef.json file is generated from the blockstack-cli subcommand docs. This data file is consumed by the _includes/commandline.md file which is used to serve up the reference.

1. Install the latest version of the cli according to the instructions at: https://github.com/blockstack/cli-blockstack

2. Generate the json for the cli in the docs.blockstack repo.

   $ blockstack-cli docs | python -m json.tool > _data/cliRef.json 
   

3. Make sure the generated docs are clean by building the documentation.

   If you run into any problem in the generation usually it results from a problem in the repo. You can make a pull request back to the repo to fix anything.

As of 8/12/19 Clarity is in the develop branch of core. You can build the Clarity command line from the Docker image. core/src/vm/docs/mod.rs

1. Pull the latest developer preview from the Docker Hub.

   $ docker pull blockstack/blockstack-core:clarity-developer-preview
   

2. Build the lastest JSON.

   docker run -it blockstack/blockstack-core:clarity-developer-preview blockstack-core docgen | jsonpp > ~/repos/docs.blockstack/_data/clarityRef.json
   

3. Build the documentation and verify the Clarity reference is building correctly.

4. Make changes in core

5. Build the docker image

6. Run doc gen with the new image

   $ docker run --name docsbuild -it blockstack-test blockstack-core docgen | jsonpp > ~/repos/docs.blockstack/_data/clarityRef.json
   

   This generates the JSON source files which are consumed through Liquid templates into markdown.

7. Rebuild the documentation site with Jekyll.

8. Review the changes in the Clarity documentation to ensure that no breaking changes were introduced through code changes.

9. If you find the documentation is no longer formatting correctly or there are errors in the reference, create a PR against the blockstack-core repo.

You can view the source code.

1. Pull the Blockstack core clarity-developer-preview image from Docker Hub.

    $ docker pull blockstack/blockstack-core:clarity-developer-preview

2. Start the Blockstack Core test environment with a Bash shell.

   $ docker run -it -v $HOME/blockstack-dev-data:/data/ blockstack/blockstack-core:clarity-developer-preview bash

   The command launches a container with the Clarity test environment and opens a bash shell into the container.

3. Run the clarity-cli in the shell.

   root@5b9798633251:/src/blockstack-core# clarity-cli
   Usage: clarity-cli [command]
   where command is one of:
   
   initialize         to initialize a local VM state database.
   mine_block         to simulated mining a new block.
   get_block_height   to print the simulated block height.
   check              to typecheck a potential contract definition.
   launch             to launch a initialize a new contract in the local state database.
   eval               to evaluate (in read-only mode) a program in a given contract context.
   eval_raw           to typecheck and evaluate an expression without a contract or database context.
   repl               to typecheck and evaluate expressions in a stdin/stdout loop.
   execute            to execute a public function of a defined contract.
   generate_address   to generate a random Stacks public address for testing purposes.

The FAQ system servers single-sourced content that support the FAQs that appear in blockstack.org, app.co, and stacks.com site. We have FAQs that fall into these categories:

• general
• appusers
• dappdevs
• appminers
• coredevs
• opensource
• miscquest
• wallet (Wallet users)
• tokens (Stacks owners)

FAQs are usually written internally by a team that are unfamiliar with markdown or HTML used in websites. So, FAQs are produced using this process:

1. Draft new or revised FAQs in a Google or Paper document.
2. Review the drafts and approve them.
3. Convert the FAQ document to HTML.
4. Strip out the unnecessary codes such as id or class designations. This leaves behind plain html.
5. Add the new FAQs to the _data/theFAQS.json file. Currently this is manually done through cut and paste.
6. Copy the JSON for appminers categories to the _data/appFAQ.json file.
7. Run the Jekyll build and verify the content builds correctly by viewing this LOCAL_HOST/faqs/allfaqs
8. Push your changes to the site and redeploy.

Single sourcing the content ensures that FAQs are the same regardless of where and on which property they appear in. These source files play a role in the FAQ single-sourcing layout:

• UIKit
• Liquid templates