# Contributing Hasura GraphQL engine docs are built using [Sphinx](http://www.sphinx-doc.org/en/master/). Sphinx files are written using the [RST markup language](http://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html). ## Docs issues in the repo Issues in the repo for documentation are labelled as `c/docs` (see [list](https://github.com/hasura/graphql-engine/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen++label%3Ac%2Fdocs)). Issues also labelled as `good first issue` are aimed at those making their first contribution to the repo (see [list](https://github.com/hasura/graphql-engine/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen++label%3Ac%2Fdocs+label%3A%22good+first+issue%22)). Issues also marked as `help wanted` (see [list](https://github.com/hasura/graphql-engine/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen++label%3Ac%2Fdocs+label%3A%22help+wanted%22)) are those that require community contributions. Please note that some of these issues, labelled with both `c/docs` and `c/console`/`c/server`, are part of a change/task that requires modifications in some component of GraphQL engine and the docs. Feel free to open pull requests to address these issues or to add/fix docs features/content, even if a corresponding issue doesn't exist. If you are unsure about whether to go ahead and work on something like the latter, please get in touch with the maintainers in the `GraphQL engine`->`contrib` channel in the community [Discord](https://discord.gg/vBPpJkS). ## Setup requirements - [Python 3](https://www.python.org/downloads/) - [Pip for Python 3](https://pip.pypa.io/en/stable/installing/) ## Steps for contributing - Fork the repo and clone it: ```bash git clone https://github.com//graphql-engine ``` - Move to the `docs` folder via the command line and checkout to a new branch: ```bash cd docs git checkout -b ``` - Install dependencies: ``` pip3 install -r requirements.txt ``` - For development, live reload and auto build while you're editing and saving files: ```bash make livehtml ``` - Make the required changes. - (Optional) Build docs to produce HTML files and verify: ```bash ENV= make html-images ``` - The generated docs are in `_build/html`. - View the built files by running a webserver: ```bash cd _build/html && http-server ``` or ```bash cd _build/html && python3 -m http.server 8080 ``` - Commit the changes. Follow the common guidelines for commit messages at the [main contributing guide](../CONTRIBUTING.md#common-guidelines). - Push the changes to your fork and submit a pull request. ## Some guidelines while adding docs ### Header section - Page titles should be self sufficient. Users might not have the context of the hierarchy of the page in the docs tree. A user can land on a page via search as well. e.g. Say you are adding a new deployment guide for AWS under `Guides -> Deployment -> AWS`. The title of this page should not be just `AWS` but instead `AWS deployment guide for Hasura GraphQL engine`. It's ok to alias it to just `AWS` in the sidebar as there the user has the context of the page hierarchy. - Ensure every new added page has a ``Table of contents`` section with the appropriate depth. ### Content - Add appropriate cross-links in content to assist users. i.e. if you refer to some functionality that is documented in some other docs page, add a link to that page. e.g. if you have a statement like "create a relationship between tables X and Y ...", make "create a relationship" a link to the `Create relationships` page. - Try to make each section within a page self-sufficient. i.e. avoid structuring all pages as step-by-step guides unless it really is the intent. This ensures that we can refer to sections externally (from other docs pages, console, etc.) and expect that the user will be able to follow the section without being lost on context that was set in earlier sections of the page. Adding statements such as "As we have described in the above section ..." might help to set up the needed context. ### GraphQL request examples While adding GraphQL request examples: - Use a tab width of 2 for nesting the requests and responses for optimal use of the space and maintaining consistency. - Nest query arguments for logical readability. Unfortunately GraphiQL prettify does not do a good job of doing this by default. - Ensure that the order of fields in the responses is the same as in the requests for better readability. ### Images - Before adding images to docs, first compress them via some tool to ensure users don't have to unnecessarily download more data than needed. You can use www.tinypng.com for this. Sometimes you can compress images by over 75% without losing any visible quality. - Use the `thumbnail` directive for images to allow click-to-zoom. ### Syntax - Ensure heading underlines are the same length as the headings. Short underlines will throw warnings during builds. - Use bold in headings in place of string literals for aesthetics (i.e. ** in place of ``) - While adding code blocks ensure the right language type is set. Sometimes adding placeholders breaks the language's syntax in which case you'll have to set the language type to `none` to avoid warnings during builds. - Use `:ref:` instead of `:doc:` to link to pages to avoid having to set the relative path and chances of broken links while moving pages ### Reference links - For external links, add a double `_` in the end to avoid `Duplicate explicit target name` warnings , e.g. \``Google __`\` - If you link to an external resource, make sure to link to the most current version of the same, e.g. `https://www.postgresql.org/docs/current/intro-whatis.html` rather than `https://www.postgresql.org/docs/9.6/intro-whatis.html` ### Pre-commit checks - Just before committing your changes, delete your local `_build` folder completely and then build docs again. Scan the output and look for any syntax warnings (e.g. title underline too short, could not lex literal block, etc.). Ideally there should not be any syntax warnings that are being thrown. ## Notes - Docs are currently deployed manually. Changes will not reflect immediately after a PR gets merged. - The search is powered by [Algolia](https://www.algolia.com/) and is updated everyday. Your local changes will not be reflected in search results.