mirror of
https://github.com/esphome/esphome-docs.git
synced 2024-12-25 17:17:54 +01:00
Update contributing.rst (#4188)
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
This commit is contained in:
parent
a963bb3e60
commit
b6cacbdbbb
@ -5,75 +5,80 @@ Contributing
|
||||
:description: Getting started guide for contributing to the ESPHome project
|
||||
:image: github-circle.svg
|
||||
|
||||
Contributions to the ESPHome suite are very welcome! All the code for the projects
|
||||
is hosted on GitHub and you can find the sources here:
|
||||
We welcome contributions to the ESPHome suite of code and documentation! All of the project code is hosted on GitHub:
|
||||
|
||||
- `ESPHome <https://github.com/esphome/esphome>`__ (Project Source Code)
|
||||
- `ESPHome-Docs <https://github.com/esphome/esphome-docs>`__ (The documentation which you're reading here)
|
||||
|
||||
Just clone the repository locally, do the changes for your new feature/bug fix and submit
|
||||
a pull request.
|
||||
Just clone the repository locally, make changes for your new feature/bug fix and submit a pull request.
|
||||
|
||||
Contributing to ESPHome-Docs
|
||||
----------------------------
|
||||
Contributing to ``esphome-docs``
|
||||
--------------------------------
|
||||
|
||||
.. image:: /images/logo-docs.svg
|
||||
:align: center
|
||||
:width: 60.0%
|
||||
:class: dark-invert
|
||||
|
||||
One of the areas of ESPHome that can always be improved is the documentation.
|
||||
If you see an issue somewhere, a spelling mistakes or if you want to share your awesome
|
||||
setup, please feel free to submit a pull request.
|
||||
Our documentation can always be improved. We rely on contributions from our users to do so. If you notice an issue (for
|
||||
example, spelling/grammar mistakes) or if you want to share your awesome new setup, we encourage you to submit a pull
|
||||
request (PR).
|
||||
|
||||
The ESPHome documentation is built using `sphinx <http://www.sphinx-doc.org/>`__ and uses
|
||||
`reStructuredText <http://docutils.sourceforge.net/rst.html>`__ for all source files.
|
||||
`reStructuredText <http://docutils.sourceforge.net/rst.html>`__ for all source (``.rst``) files.
|
||||
|
||||
If you're not familiar with writing rST, see :ref:`rst-syntax` for a quick refresher.
|
||||
If you're not familiar with rST, see :ref:`rst-syntax` for a quick refresher.
|
||||
|
||||
Through Github
|
||||
Through GitHub
|
||||
**************
|
||||
|
||||
This guide essentially goes over the same material found in `GitHub's Editing files in another user's repository <https://docs.github.com/en/repositories/working-with-files/managing-files/editing-files#editing-files-in-another-users-repository>`__. You may also find that page helpful to read.
|
||||
This guide essentially goes over the same material found in
|
||||
`GitHub's Editing files in another user's repository <https://docs.github.com/en/repositories/working-with-files/managing-files/editing-files#editing-files-in-another-users-repository>`__.
|
||||
You may also find that page helpful to read.
|
||||
|
||||
At the bottom of each page in the docs, there is a "Edit this page on GitHub" link. Click this link and you'll see something like this:
|
||||
At the bottom of each page in the docs, there is a "Edit this page on GitHub" link. Click this link and you'll see
|
||||
something like this:
|
||||
|
||||
.. figure:: images/docs_ghedit_1.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:alt: a screenshot of an rST file opened in GitHub, with the edit button circled
|
||||
|
||||
Click the edit button to start making changes. If you're not sure about some element of syntax, see the quick-start :ref:`rst-syntax` guide.
|
||||
Click the edit button to start making changes. If you're unsure about syntax, see the quick-start :ref:`rst-syntax`
|
||||
guide.
|
||||
|
||||
Once you've made your changes, give them a useful name and press "Propose changes". At this point, you've made the changes on your own personal copy of the docs in GitHub, but you still need to submit them to us.
|
||||
Once you've made your changes, give them a useful name and press "Propose changes". At this point, you've made the
|
||||
changes on your own personal copy of the docs in GitHub, but you still need to submit them to us.
|
||||
|
||||
.. figure:: images/docs_ghedit_2.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:alt: the commit creation screen in GitHub, with the commit title and "Propose changes" button circled
|
||||
|
||||
To do that, you need to create a "Pull request":
|
||||
To do that, you need to create a "Pull request" (PR):
|
||||
|
||||
.. figure:: images/docs_ghedit_3.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:alt: the pull request prompt screen in GitHub with the "Create pull request" button circled
|
||||
|
||||
Fill out the new pull request form, replacing the ``[ ]`` with ``[x]`` to indicate that you have followed the instructions.
|
||||
Fill out the new pull request form, replacing the ``[ ]`` with ``[x]`` to indicate that you have followed the
|
||||
instructions.
|
||||
|
||||
.. figure:: images/docs_ghedit_4.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:alt: the pull request creation screen in GitHub with the "Create pull request" button circled
|
||||
|
||||
After waiting a while, you might see a green or a red mark next to your commit in your pull request:
|
||||
After a few minutes, you'll see either a green check ✅ or a red ❌ next to your commit in your pull request:
|
||||
|
||||
.. figure:: images/docs_ghedit_ci_failed.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:alt: the pull request with a commit with a red x next to it
|
||||
|
||||
This means that there is some error stopping your pull request from being fully processed. Click on the X, click on "Details" next to the lint step, and look and see what's causing your change to fail.
|
||||
If you see the red ❌, there is at least one error preventing your pull request from being fully processed. Click on the
|
||||
❌, then click on "Details" next to the lint step to determine what's wrong with your change(s).
|
||||
|
||||
.. figure:: images/docs_ghedit_ci_details.png
|
||||
:align: center
|
||||
@ -85,9 +90,15 @@ This means that there is some error stopping your pull request from being fully
|
||||
:width: 80.0%
|
||||
:alt: log messages showing reason for failed build
|
||||
|
||||
For example, in this case, you'd want to go to line 136 of ``pzemac.rst`` and adjust the number of ``===`` so that it completely underlines the section heading.
|
||||
In this example, you need to go to line 136 of ``pzemac.rst`` and adjust the number of ``===`` so that it completely
|
||||
underlines the section heading.
|
||||
|
||||
Once you make that change, the pull request will be built again, and hopefully this time where will be no other errors.
|
||||
Once you make that change, the pull request will be tested & built again; ideally, this time where will be no remaining
|
||||
errors. If, however, more errors are discovered, simply repeat the process above to correct them.
|
||||
|
||||
.. note::
|
||||
|
||||
All tests must be passing before we will review (and merge) a pull request.
|
||||
|
||||
Build
|
||||
*****
|
||||
@ -100,7 +111,8 @@ Build
|
||||
|
||||
docker run --rm -v "${PWD}/":/workspaces/esphome-docs -p 8000:8000 -it ghcr.io/esphome/esphome-docs
|
||||
|
||||
With ``PWD`` referring to the root of the ``esphome-docs`` git repository. Then go to ``<CONTAINER_IP>:8000`` in your browser.
|
||||
...with ``PWD`` referring to the root of the ``esphome-docs`` git repository. Then, to see the preview, go to
|
||||
``<HOST_IP>:8000`` in your browser.
|
||||
|
||||
This way, you don't have to install the dependencies to build the documentation.
|
||||
|
||||
@ -121,16 +133,21 @@ Then, use the provided Makefile to build the changes and start a live-updating w
|
||||
Notes
|
||||
*****
|
||||
|
||||
Some notes about the docs:
|
||||
|
||||
- Use the English language (duh...)
|
||||
- An image tells a thousand words, please use them wherever possible. But also don't forget to shrink them, for example
|
||||
I often use https://tinypng.com/
|
||||
- Try to use examples as often as possible (also while it's great to use highly accurate,
|
||||
and domain-specific lingo, it should not interfere with new users understanding the content)
|
||||
- Fixes/improvements for the docs themselves should go to the ``current`` branch of the
|
||||
esphomedocs repository. New features should be added against the ``next`` branch.
|
||||
- Always create new branches in your fork for each pull request.
|
||||
- An image is worth a thousand words, please use them wherever possible. Regardless, don't forget to optimize them so
|
||||
pages load quickly; you might try using https://tinypng.com/
|
||||
- Try to use examples as often as possible.
|
||||
- When using highly accurate, domain-specific terminology, be sure that it does not interfere with new users
|
||||
understanding the content.
|
||||
- Be sure to target the correct **base branch** of the ``esphome-docs`` repository:
|
||||
|
||||
- **Fixes/improvements** for documentation must target the ``current`` branch.
|
||||
- **New features** must target the ``next`` branch.
|
||||
|
||||
- **Create new branches in your fork** for each pull request; to avoid confusion (and other potential issues), do not
|
||||
make changes directly in the ``next`` and ``current`` branches in your fork.
|
||||
- Wrap lines in all files at no more than 120 characters. This makes reviewing PRs faster and easier. Exceptions should
|
||||
be made only for lines which contain long links or other specific content/metadata that cannot be wrapped.
|
||||
|
||||
.. _rst-syntax:
|
||||
|
||||
@ -139,10 +156,12 @@ Syntax
|
||||
|
||||
Here's a quick RST primer:
|
||||
|
||||
Title hierarchy is based on order of occurrence, not on type of character used to underline it. This
|
||||
documents establish the following character order for better consistency.
|
||||
Title hierarchy is based on order of occurrence, not on type of character used to underline it. For consistency, we
|
||||
adhere to the following order:
|
||||
|
||||
- **Headers**: You can write titles like this:
|
||||
- **Headers**:
|
||||
|
||||
You can write titles like this:
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
@ -153,20 +172,27 @@ documents establish the following character order for better consistency.
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
My Sub Section
|
||||
--------------
|
||||
My Section
|
||||
----------
|
||||
|
||||
and sub-section headers like this:
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
My Sub-sub Section
|
||||
******************
|
||||
My Sub-section
|
||||
**************
|
||||
|
||||
and sub-sub-section headers like this:
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
My Sub-sub-section
|
||||
^^^^^^^^^^^^^^^^^^
|
||||
|
||||
.. note::
|
||||
|
||||
The length of the bar below the text **must** match the title Text length.
|
||||
Also, titles should be in Title Case
|
||||
- The length of the bar below the text **must** match the title text length.
|
||||
- Section titles should use Title Case.
|
||||
|
||||
- **Links**: To create a link to an external resource (for example https://www.google.com), use
|
||||
``\`Link text <link_url>\`__``. For example:
|
||||
@ -177,8 +203,8 @@ documents establish the following character order for better consistency.
|
||||
|
||||
`Google.com <https://www.google.com>`__
|
||||
|
||||
- **References**: To reference another document, use the ``:doc:`` and ``:ref:`` roles (references
|
||||
are set up globally and can be used between documents):
|
||||
- **References**: To reference another document, use the ``:doc:`` and ``:ref:`` roles (references are set up globally
|
||||
and can be used between documents):
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
@ -223,9 +249,9 @@ documents establish the following character order for better consistency.
|
||||
|
||||
.. note::
|
||||
|
||||
Please note the empty line after the ``code-block`` directive. That is necessary.
|
||||
Note that a blank line is *required* after every ``code-block`` directive.
|
||||
|
||||
- **Images**: To show images, use the ``figure`` directive:
|
||||
- **Images**: Use the ``figure`` directive to display an image:
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
@ -243,13 +269,11 @@ documents establish the following character order for better consistency.
|
||||
|
||||
.. note::
|
||||
|
||||
All images in the documentation need to be as small as possible to ensure
|
||||
fast page load times. For normal figures the maximum size should be at most
|
||||
about 1000x800 px or so. Additionally, please use online tools like
|
||||
All images in the documentation need to be as small as possible to minimize page load times. Typically, the
|
||||
maximum size should be roughly 1000x800 px or so. Additionally, please use online tools like
|
||||
https://tinypng.com/ or https://tinyjpg.com/ to further compress images.
|
||||
|
||||
- **Notes and warnings**: You can create simple notes and warnings using the ``note`` and ``warning``
|
||||
directives:
|
||||
- **Notes and warnings**: You can create simple notes and warnings using the ``note`` and ``warning`` directives:
|
||||
|
||||
.. code-block:: rst
|
||||
|
||||
@ -305,28 +329,30 @@ documents establish the following character order for better consistency.
|
||||
2. Ordered item #2
|
||||
|
||||
- **imgtable**: ESPHome uses a custom RST directive to show the table on the front page (see
|
||||
`index.rst <https://github.com/esphome/esphome-docs/blob/current/index.rst>`__).
|
||||
New pages need to be added to the ``imgtable`` list. The syntax is CSV with <PAGE NAME>, <FILE NAME> (without RST),
|
||||
<IMAGE> (in top-level images/ directory), <COMMENT> (optional - short text to describe the component). The aspect ratio of these images should be 8:10 (or 10:8) but exceptions are possible.
|
||||
`index.rst <https://github.com/esphome/esphome-docs/blob/current/index.rst>`__). New pages need to be added to the
|
||||
``imgtable`` list. The syntax is CSV with <PAGE NAME>, <FILE NAME> (without RST), <IMAGE> (in the top-level
|
||||
``images/`` directory), <COMMENT> (optional; short text to describe the component). The aspect ratio of these images
|
||||
should be 8:10 (or 10:8) but exceptions are possible.
|
||||
|
||||
Because these images are served on the main page, they need to be compressed heavily. SVGs are preferred over JPGs
|
||||
and JPGs should be max. 300x300px.
|
||||
and JPGs should be no more than 300x300px.
|
||||
|
||||
If you have imagemagick installed, you can use this command to convert the thumbnail:
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
convert -sampling-factor 4:2:0 -strip -interlace Plane -quality 80% -resize 300x300 in.jpg out.jpg
|
||||
|
||||
reStructured text can do a lot more than this, so if you're looking for a more complete guide
|
||||
please have a look at the `Sphinx reStructuredText Primer <http://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html>`__.
|
||||
reStructured text can do a lot more than this; if you're looking for a more complete guide, please have a look at the
|
||||
`Sphinx reStructuredText Primer <http://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html>`__.
|
||||
|
||||
.. _setup_dev_env:
|
||||
|
||||
Setting Up Development Environment
|
||||
----------------------------------
|
||||
Setting Up a Development Environment
|
||||
------------------------------------
|
||||
|
||||
For developing new features to ESPHome, you will first need to set up a development environment.
|
||||
This is only possible for ``pip`` installs.
|
||||
If you want to develop (a) new feature(s) for ESPHome, you need to set up a development environment.
|
||||
Note that ``pip`` must be installed before running the ``setup`` script.
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
@ -341,31 +367,29 @@ This is only possible for ``pip`` installs.
|
||||
git checkout -b my-new-feature
|
||||
cd ..
|
||||
|
||||
The environment is now ready for use, but you need to activate the Python virtual environment
|
||||
every time you are using it.
|
||||
The environment is now ready for use, but you'll need to activate the Python virtual environment (venv) each time you
|
||||
wish to use it.
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
# Activate venv
|
||||
source venv/bin/activate
|
||||
|
||||
Now you can open ESPHome in your IDE of choice (mine is CLion) with the PlatformIO
|
||||
addons (see PlatformIO docs for more info). Then develop the new feature with the
|
||||
guidelines below.
|
||||
Now you can open ESPHome in your IDE of choice (many of us are using `VSCode <https://code.visualstudio.com/download>`__)
|
||||
with the PlatformIO addons (see PlatformIO docs for more info) and develop the new feature with the guidelines below.
|
||||
|
||||
All PRs are automatically checked for some basic formatting/code mistakes with Github Actions.
|
||||
These checks *must* pass for your PR to be mergeable.
|
||||
All PRs are automatically checked and tested for some common formatting/code errors with Github Actions. *These checks*
|
||||
**must all pass** *before we will review (and eventually merge) your PR.*
|
||||
|
||||
Setting Up Git Environment
|
||||
--------------------------
|
||||
|
||||
ESPHome's code-base is hosted on GitHub, and contributing is done exclusively through
|
||||
"Pull Requests" (PRs) in the GitHub interface. So you need to set up your git environment
|
||||
first.
|
||||
ESPHome's codebase is hosted on GitHub; contributing is done exclusively through "Pull Requests" (PRs) in the
|
||||
GitHub interface. To use this, you need to set up your ``git`` environment first.
|
||||
|
||||
When you want to create a patch for ESPHome, first go to the repository you want to contribute to
|
||||
(esphome, etc) and click fork in the top right corner. This will create
|
||||
a fork of the repository that you can modify and create git branches on.
|
||||
If you want to contribute changes/fixes you've made back to ESPHome, first, go to the repository you want to contribute
|
||||
to (``esphome``, for example) and click "fork" in the top right corner. This will create a fork of the repository that
|
||||
you can modify and create branches on.
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
@ -396,27 +420,155 @@ a fork of the repository that you can modify and create git branches on.
|
||||
git push -u origin <BRANCH_NAME>
|
||||
# For example: git push -u origin gpio-switch-fix
|
||||
|
||||
Then go to your repository fork in GitHub and wait for a create pull request message to show
|
||||
up in the top (alternatively go to branches and create it from there). Fill out the
|
||||
Pull Request template outlining your changes; if your PR is not ready to merge yet please
|
||||
mark it as a draft PR in the dropdown of the green "create PR" button.
|
||||
Submitting a Pull Request
|
||||
*************************
|
||||
|
||||
**Review Process:** ESPHome's code base tries to have a high code standard. At the bottom
|
||||
of the Pull Request you will be able to see the "Github Actions" continuous integration check which
|
||||
will automatically go through your patch and try to spot errors. If the CI check fails,
|
||||
please see the Github Actions log and fix all errors that appear there. Only PRs that pass the automated
|
||||
checks can be merged!
|
||||
After you have pushed your changes to GitHub, go to your repository fork and look for a "create pull request" button
|
||||
near the top of the page (or, alternatively, go to branches and create it from there). As you create the PR:
|
||||
|
||||
**Catching up with reality**: Sometimes other commits have been made to the same files
|
||||
you edited. Then your changes need to be re-applied on top of the latest changes with
|
||||
a "rebase". More info `here <https://developers.home-assistant.io/docs/en/development_catching_up.html>`__.
|
||||
- Complete the Pull Request template:
|
||||
|
||||
- Include a brief (but complete) summary of your changes.
|
||||
- PRs without a descrption/summary of the changes will not be reviewed or merged, although exceptions may
|
||||
occasionally be made for small PRs and/or PRs made by frequent contributors/maintainers.
|
||||
- **Do not delete the template.**
|
||||
|
||||
- **Mark your PR as a draft** if it is not ready to be reviewed or merged yet. Your PR should be considered a draft if:
|
||||
|
||||
- You still plan to make more changes to the code/documentation.
|
||||
- Changes have been requested to the PR but you have not completed making (or discussing) the requested changes yet.
|
||||
- You are waiting on feedback from the community and/or maintainers to complete your changes.
|
||||
|
||||
This lets reviewers know that the PR may continue to change so they will not spend valuable time reviewing it until
|
||||
it is ready. We do this because, if a PR is reviewed and then it changes, it must be re-reviewed. Reviewing a single
|
||||
PR multiple times is not a productive use of time and we try as much as possible to avoid doing so.
|
||||
|
||||
Review Process
|
||||
**************
|
||||
|
||||
ESPHome's maintainers work hard to maintain a high standard for its code, so reviews can take some time. At the bottom
|
||||
of each pull request you will see the "Github Actions" continuous integration (CI) checks which will automatically
|
||||
analyize all code changed in your branch. These checks try to spot (and suggest corrections for) errors. If any CI
|
||||
check fails, please look at the Github Actions log and fix all errors that appear there.
|
||||
|
||||
**All automated checks must be passing** before a given PR will be reviewed and (eventually) merged!
|
||||
|
||||
**When will my PR be reviewed/merged?**
|
||||
|
||||
ESPHome is a big project; we encourage everybody to test, review and comment on PRs. Despite this, reviews can (and
|
||||
often do) take some time.
|
||||
|
||||
**But howwww looonnnggg???**
|
||||
|
||||
Small PRs are easier to review and are often reviewed first. If you want your PR to be reviewed (and merged) quickly,
|
||||
here are some tips:
|
||||
|
||||
- We would rather review ten ten-line PRs than one 100-line PR.
|
||||
- Be sure to follow all :ref:`codebase_standards` as you make changes -- when reviewers have to spend time
|
||||
commenting on/correcting your PR because you didn't name variables correctly or didn't prefix member variable
|
||||
accesses with ``this->``, it wastes time we could be using to review other PRs which *do* follow the standards.
|
||||
- If you wish to take on a big project, such as refactoring a substantial section of the codebase or integrating
|
||||
another open source project with ESPHome, please discuss this with us on `Discord <https://discord.gg/KhAMKrd>`__ or
|
||||
`create a discussion on GitHub <https://github.com/esphome/esphome/discussions>`__ **before** you do all the work and
|
||||
attempt to submit a massive PR.
|
||||
- While we realize it's not *always* possible, avoid submitting PRs which are thousands of lines in size. Such PRs are
|
||||
simply too complex and take excessive amounts of time to review. Break your work into multiple, smaller PRs to make
|
||||
the changes more tenable for reviewers.
|
||||
- If you are not sure about how you should proceed with some changes, **please**
|
||||
`discuss it with us on Discord <https://discord.gg/KhAMKrd>`__ before you go do a bunch of work that we can't (for
|
||||
whatever reason) accept...and then you have to go back and re-do it all to get your PR merged. It's easier to make
|
||||
corrections early-on -- and we want to help you!
|
||||
|
||||
.. _prs-are-being-drafted-when-changes-are-needed:
|
||||
|
||||
Why Was My PR was Marked as a Draft?
|
||||
************************************
|
||||
|
||||
If your PR was reviewed and changes were requested, our bot will automatically mark your PR as a draft. This means that
|
||||
the PR is not ready to be merged or further reviewed for the moment.
|
||||
|
||||
When a PR is marked as a draft, it tells other reviewers that this particular PR is a work-in-progress and it doesn't
|
||||
require their attention yet.
|
||||
|
||||
Once you have made the requested changes, you can mark the PR as ready for review again by clicking the "Ready for
|
||||
review" button:
|
||||
|
||||
.. figure:: images/pr-draft-ready.png
|
||||
:align: center
|
||||
:width: 100.0%
|
||||
:alt: The ready for review button in the bottom of a PR in draft mode
|
||||
|
||||
Before you click the "Ready for review" button, ensure that:
|
||||
|
||||
- You have addressed all requested changes
|
||||
- There are no merge conflicts
|
||||
- All CI jobs and checks are passing successfully
|
||||
|
||||
Once you've clicked the "Ready for review" button, the PR will return to a normal state again and our bot will
|
||||
automatically notify the reviewers who requested the changes that the PR is ready to go!
|
||||
|
||||
Updating Your Branches
|
||||
**********************
|
||||
|
||||
Sometimes you'll want (or need) to bring changes that were made in ESPHome's ``dev`` branch back into your (local copy
|
||||
of a) branch.
|
||||
|
||||
The examples that follow in this section assume that you have:
|
||||
|
||||
- already used ``git remote`` to add ``upstream`` as shown earlier, and
|
||||
- your feature branch (the branch from which you created your PR) currently checked out
|
||||
|
||||
.. _feature_branches:
|
||||
|
||||
Feature Branches
|
||||
^^^^^^^^^^^^^^^^
|
||||
|
||||
There are a couple of ways you can update your (local) feature branch. The easiest is by clicking the "Update branch"
|
||||
button in GitHub:
|
||||
|
||||
.. image:: images/update_branch.png
|
||||
:align: center
|
||||
:width: 80.0%
|
||||
:class: light-invert
|
||||
|
||||
...then run ``git pull`` to pull these changes back down from GitHub.
|
||||
|
||||
If you prefer to do it the command-line/terminal way, you can do this, instead:
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
# Fetch the latest upstream changes and apply them
|
||||
# Fetch the latest upstream changes
|
||||
git fetch upstream dev
|
||||
# Merge in the changes we fetched above
|
||||
git merge upstream/dev
|
||||
|
||||
Your Local Copy of ``dev``
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
As you create new branches for your work, you'll want to be sure they include all of the latest changes from ESPHome's
|
||||
``dev`` branch -- it's not a good practice to create a new feature branch from an outdated ``dev`` branch.
|
||||
|
||||
For this reason, you'll periodically want to update your local ``dev`` branch. A more detailed explanation can be found
|
||||
`here <https://developers.home-assistant.io/docs/en/development_catching_up.html>`__, but here's the TL;DR:
|
||||
|
||||
.. code-block:: bash
|
||||
|
||||
# Fetch the latest upstream changes
|
||||
git fetch upstream dev
|
||||
git rebase upstream/dev
|
||||
|
||||
Note that you can use this procedure for other branches, too, such as ``next`` or ``current`` from ``esphome-docs``.
|
||||
|
||||
.. warning::
|
||||
|
||||
Using ``git rebase`` will result in your changes having to be *force-pushed* back up to GitHub.
|
||||
|
||||
**Do not force-push** your branch once your PR is being reviewed; GitHub allows reviewers to mark files a "viewed"
|
||||
and, when you force-push, this history **is lost**, forcing your reviewer to re-review files they may have already
|
||||
reviewed!
|
||||
|
||||
If you must update your branch, use a method described in :ref:`feature_branches`, instead.
|
||||
|
||||
.. _contributing_to_esphome:
|
||||
|
||||
Contributing to ESPHome
|
||||
@ -427,21 +579,19 @@ Contributing to ESPHome
|
||||
:width: 60.0%
|
||||
:class: dark-invert
|
||||
|
||||
This is a guide to contributing to the ESPHome codebase. ESPHome uses two languages for its project:
|
||||
Python and C++.
|
||||
This is a guide to contributing to the ESPHome codebase. ESPHome uses two languages for its project: Python and C++.
|
||||
|
||||
The user configuration is read, validated and transformed into a custom firmware
|
||||
with the Python side of the firmware.
|
||||
The Python side of ESPHome reads a YAML configuration file, validates it and transforms it into a custom firmware which
|
||||
includes only the code needed to perform as defined in the configuration file.
|
||||
|
||||
The C++ codebase is what's actually running on the ESP and called the "runtime". This part of
|
||||
the codebase should first set up the communication interface to a sensor/component/etc. and
|
||||
communicate with the ESPHome core via the defined interfaces (like Sensor, BinarySensor, Switch).
|
||||
The C++ part of the codebase is what's actually running on the microcontroller; this is called the "runtime". This part
|
||||
of the codebase should first set up the communication interface to a sensor/component/etc. and then communicate with
|
||||
the ESPHome core via the defined interfaces (like ``Sensor``, ``BinarySensor`` and ``Switch``, among others).
|
||||
|
||||
1. Directory Structure
|
||||
**********************
|
||||
Directory Structure
|
||||
*******************
|
||||
|
||||
After you've :ref:`set up development environment <setup_dev_env>`, you will have a folder structure
|
||||
like this:
|
||||
After you've :ref:`set up a development environment <setup_dev_env>`, you will have a folder structure like this:
|
||||
|
||||
.. code-block:: text
|
||||
|
||||
@ -464,10 +614,10 @@ like this:
|
||||
│ │ ├── switch.py
|
||||
│ ...
|
||||
|
||||
As you can see, all components are in the "components" folder. Each component is in its own
|
||||
subfolder which contains the Python code (.py) and the C++ code (.h and .cpp).
|
||||
All components are in the "components" folder. Each component is in its own subfolder which contains the Python code
|
||||
(``.py``) and the C++ code (``.h`` and ``.cpp``).
|
||||
|
||||
Suppose the user types in the following:
|
||||
Consider a YAML configuration file containing the following:
|
||||
|
||||
.. code-block:: yaml
|
||||
|
||||
@ -476,34 +626,35 @@ Suppose the user types in the following:
|
||||
sensor:
|
||||
- platform: hello2
|
||||
|
||||
In both cases, ESPHome will automatically look for corresponding entries in the "components"
|
||||
folder and find the directory with the given name. For example the first entry will make ESPHome
|
||||
look at the ``esphome/components/hello1/__init__.py`` file and the second entry will result in
|
||||
``esphome/components/hello2/sensor.py``.
|
||||
In both cases, ESPHome will automatically look for corresponding entries in the "components" folder and find the
|
||||
directory with the given name. In this example, the first entry causes ESPHome to look for the
|
||||
``esphome/components/hello1/__init__.py`` file and the second entry tells ESPHome to look for
|
||||
``esphome/components/hello2/sensor.py`` or ``esphome/components/hello2/sensor/__init__.py``.
|
||||
|
||||
Let's leave what's written in those files for (2.), but for now you should also know that
|
||||
whenever a component is loaded, all the C++ source files in the folder of the component
|
||||
are automatically copied into the generated PlatformIO project. So you just need to add the C++
|
||||
source files in the folder and the ESPHome core will copy them with no additional code required
|
||||
by the component developer.
|
||||
Let's leave what's written in those files for :ref:`the next section <config_validation>`, but for now you should also
|
||||
know that, whenever a component is loaded, all the C++ source files in the folder of the component are automatically
|
||||
copied into the generated PlatformIO project. All you need to do is add the C++ source files in the component's folder
|
||||
and the ESPHome core will copy them with no additional code required by the component developer.
|
||||
|
||||
.. note::
|
||||
|
||||
For testing you can use :doc:`/components/external_components`.
|
||||
For testing, you can use :doc:`/components/external_components`.
|
||||
|
||||
ESPHome also has a ``custom_components`` mechanism like `Home Assistant does
|
||||
<https://developers.home-assistant.io/docs/creating_component_index>`__. However this is
|
||||
discouraged in favor of :doc:`/components/external_components`.
|
||||
<https://developers.home-assistant.io/docs/creating_component_index>`__. Note, however, that
|
||||
**custom componenets are deprecated** in favor of :doc:`/components/external_components`.
|
||||
|
||||
2. Config Validation
|
||||
********************
|
||||
.. _config_validation:
|
||||
|
||||
The first thing ESPHome does is read and validate the user config. For this ESPHome has a powerful
|
||||
"config validation" mechanism. Each component defines a config schema that is validated against
|
||||
the user config.
|
||||
Config Validation
|
||||
*****************
|
||||
|
||||
To do this, all ESPHome Python modules that can be configured by the user have a special field
|
||||
called ``CONFIG_SCHEMA``. An example of such a schema is shown below:
|
||||
The first task ESPHome performs is to read and validate the provided YAML configuration file. ESPHome has a powerful
|
||||
"config validation" mechanism for this purpose. Each component defines a config schema which is used to validate the
|
||||
provided configuration file.
|
||||
|
||||
To do this, all ESPHome Python modules that can be configured by the user define a special variable named
|
||||
``CONFIG_SCHEMA``. An example of such a schema is shown below:
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
@ -517,70 +668,71 @@ called ``CONFIG_SCHEMA``. An example of such a schema is shown below:
|
||||
cv.Optional(CONF_MY_OPTIONAL_KEY, default=10): cv.int_,
|
||||
}).extend(cv.COMPONENT_SCHEMA)
|
||||
|
||||
This variable is automatically loaded by the ESPHome core and validated against.
|
||||
This variable is automatically loaded by the ESPHome core and is used to validate the provided configuration.
|
||||
The underlying system ESPHome uses for this is `voluptuous <https://github.com/alecthomas/voluptuous>`__.
|
||||
Going into how to validate is out of scope for this guide, but the best way to learn is to look
|
||||
at examples of how similar components validate user input.
|
||||
How validation works is out of scope for this guide; the easiest way to learn is to look at how similar components
|
||||
validate user input.
|
||||
|
||||
A few point on validation:
|
||||
A few notes on validation:
|
||||
|
||||
- ESPHome puts a lot of effort into **strict validation** - If possible, all validation methods should be as strict
|
||||
as possible and detect wrong user input at the validation stage (and not later).
|
||||
- All default values should be defined in the schema (and not in C++ codebase or other code parts).
|
||||
- Config keys should be descriptive - If the meaning of a key is not immediately obvious you should
|
||||
always prefer long_but_descriptive_keys.
|
||||
- ESPHome puts a lot of effort into **strict validation**. All validation methods should be as strict as possible and
|
||||
detect incorrect user input at the validation stage, mitigating compiler warnings and/or errors.
|
||||
- All default values should be defined in the schema -- not in C++ codebase.
|
||||
- Prefer naming configuration keys in a way which is descriptive instead of short. Put another way, if the meaning of a
|
||||
key is not immediately obvious, don't be afraid to use ``long_but_descriptive_keys``. There is no reason to use
|
||||
obscure shorthand. As an example, ``scrn_btn_inpt`` is indeed shorter but more difficult to understand, particularly
|
||||
for new users; avoid naming keys and variables in this way.
|
||||
|
||||
3. Code Generation
|
||||
******************
|
||||
Code Generation
|
||||
***************
|
||||
|
||||
After the user input has been successfully validated, the last step of the Python codebase
|
||||
is called: Code generation.
|
||||
The last step the Python codebase performs is called *code generation*. This runs only after the user input has been
|
||||
successfully validated.
|
||||
|
||||
As you may know, ESPHome converts the user's configuration into C++ code (you can see the generated
|
||||
code under ``<NODE_NAME>/src/main.cpp``). Each component must define its own ``to_code`` method
|
||||
that converts the user input to C++ code.
|
||||
As you may know, ESPHome "converts" the user's YAML configuration into C++ code (you can see the generated code under
|
||||
``<NODE_NAME>/src/main.cpp``). Each component must define its own ``to_code`` method that "converts" the user input to
|
||||
C++ code.
|
||||
|
||||
This method is also automatically loaded and invoked by the ESPHome core. An example of
|
||||
such a method can be seen below:
|
||||
This method is also automatically loaded and invoked by the ESPHome core. Here's an example of such a method:
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
import esphome.codegen as cg
|
||||
|
||||
def to_code(config):
|
||||
async def to_code(config):
|
||||
var = cg.new_Pvariable(config[CONF_ID])
|
||||
yield cg.register_component(var)
|
||||
await cg.register_component(var, config)
|
||||
|
||||
cg.add(var.set_my_required_key(config[CONF_MY_REQUIRED_KEY]))
|
||||
|
||||
Again, going into all the details of ESPHome code generation would be out-of-scope. However,
|
||||
ESPHome's code generation is 99% syntactic sugar - and again it's probably best to study other
|
||||
components and just copy what they do.
|
||||
The details of ESPHome code generation is out-of-scope for this document. However, ESPHome's code generation is 99%
|
||||
syntactic sugar - and (again) it's probably best to study similar components and just copy what they do.
|
||||
|
||||
There's one important concept for the ``to_code`` method: coroutines with ``yield``.
|
||||
First the problem that leads to coroutines: In ESPHome, components can declare (via ``cg.Pvariable``) and access variables
|
||||
(``cg.get_variable()``) - but sometimes when one part of the code base requests a variable
|
||||
it has not been declared yet because the code for the component creating the variable has not run yet.
|
||||
There's one important concept for the ``to_code`` method: coroutines with ``await``.
|
||||
|
||||
To allow for ID references, ESPHome uses so-called ``coroutines``. When you see a ``yield`` statement
|
||||
in a ``to_code`` method, ESPHome will call the provided method - and if that method needs to wait
|
||||
for a variable to be declared first, ``yield`` will wait until that variable has been declared.
|
||||
After that, ``yield`` returns and the method will execute on the next line.
|
||||
The problem that necessitates coroutines is this: in ESPHome, components can declare (via ``cg.Pvariable``) and access
|
||||
variables (``cg.get_variable()``) -- but sometimes, when one part of the codebase requests a variable, it has not been
|
||||
declared yet because the code for the component creating the variable has not yet run.
|
||||
|
||||
Next, there's a special method - ``cg.add`` - that you will often use. ``cg.add()`` does a very simple
|
||||
thing: Any C++ declared in the parentheses of ``cg.add()`` will be added to the generated code.
|
||||
If you do not call "add" a piece of code explicitly, it will not be added to the main.cpp file!
|
||||
To allow for ID references, ESPHome uses so-called ``coroutines``. When you see an ``await`` statement in a ``to_code``
|
||||
method, ESPHome will call the provided method and, if that method needs to wait for a variable to be declared first,
|
||||
``await`` will wait until that variable has been declared. After that, ``await`` returns and the method will execute on
|
||||
the next line.
|
||||
|
||||
4. Runtime
|
||||
**********
|
||||
Next, there's a special method - ``cg.add`` - that you will often use. ``cg.add()`` performs a very simple task: Any
|
||||
C++ declared in the parentheses of ``cg.add()`` will be added to the generated code. Note that, if you do not call
|
||||
"add" to insert a piece of code explicitly, it will not be added to the ``main.cpp`` file!
|
||||
|
||||
Okay, the Python part of the codebase is now complete - now let's talk about the C++ part of
|
||||
creating a new component.
|
||||
Runtime
|
||||
*******
|
||||
|
||||
The two major parts of any component roughly are:
|
||||
At this point, the Python part of the codebase has completed its work. Let's move on and discuss the C++ part of
|
||||
components.
|
||||
|
||||
- Setup Phase
|
||||
- Run Phase
|
||||
Most components consist of two primary parts/steps:
|
||||
|
||||
- Setup Phase
|
||||
- Run Phase
|
||||
|
||||
When you create a new component, your new component will inherit from :apiclass:`Component`.
|
||||
That class has a special ``setup()`` method that will be called once to set up the component -
|
||||
@ -592,83 +744,96 @@ if communication works (if not, it should call ``mark_failed()``).
|
||||
|
||||
Again, look at examples of other components to learn more.
|
||||
|
||||
The next thing that will be called with your component is ``loop()`` (or ``update()`` for a
|
||||
:apiclass:`PollingComponent`). In these methods you should retrieve the latest data from the
|
||||
component and publish them with the provided methods. One thing to note in these methods
|
||||
is that anything in ``loop()`` or ``setup()`` **should not block**. Specifically methods like
|
||||
``delay(10)`` should be avoided and delays above ~10ms are not permitted. The reason for this
|
||||
is that ESPHome uses a central single-threaded loop for all components - if your component
|
||||
blocks the whole loop will be slowed down.
|
||||
The next method that will be called with your component is ``loop()`` (or ``update()`` for a
|
||||
:apiclass:`PollingComponent`). These methods should retrieve the latest data from your component and publish them with
|
||||
the provided methods.
|
||||
|
||||
Finally, your component should have a ``dump_config`` method that prints the user configuration.
|
||||
Finally, your component must have a ``dump_config`` method that prints the complete user configuration.
|
||||
|
||||
5. Extras
|
||||
*********
|
||||
A Note About Delays in Code
|
||||
***************************
|
||||
|
||||
**Code in** ``loop()``, ``update()`` **and** ``setup()`` **must not block**.
|
||||
|
||||
Methods like ``delay()`` should be avoided and **delays longer than 10 ms are not permitted**. Because ESPHome uses a
|
||||
single-threaded loop for all components, if your component blocks, it will delay the whole loop, negatively impacting
|
||||
other components. This can result in a variety of problems such as network connections being lost.
|
||||
|
||||
If your code **must** wait for something to happen (for example, your sensor requires hundreds of milliseconds to
|
||||
initialize and/or take a reading), then you'll need to implement a state machine to facilitate this. For example, your
|
||||
code can send the "take reading" command, return, and, when the next iteration of ``loop()`` or ``update()`` is called,
|
||||
it then attempts to read back the measurement from the sensor.
|
||||
|
||||
``loop()`` is called every 16 ms (assuming no other components delay this, which may happen from time to time) and
|
||||
``update()`` is called at an interval defined in the user configuration for the component, but only for
|
||||
:apiclass:`PollingComponent`.
|
||||
|
||||
For any :apiclass:`Component` (which is nearly everything), the well-known ``set_timeout`` method is also available;
|
||||
this can be a handy alternative to implemeting a state machine.
|
||||
|
||||
Extras
|
||||
******
|
||||
|
||||
.. note::
|
||||
|
||||
This serves as documentation for some of ESPHome's internals and is not necessarily part of the
|
||||
development guide.
|
||||
This serves as documentation for some of ESPHome's internals and is not necessarily part of the development guide.
|
||||
|
||||
All Python modules have some magic symbols that will automatically be loaded by the ESPHome
|
||||
loader. These are:
|
||||
All Python modules have some magic symbols that will automatically be loaded by the ESPHome loader. These are:
|
||||
|
||||
- ``CONFIG_SCHEMA``: The configuration schema to validate the user config against.
|
||||
- ``to_code``: The function that will be called with the validated configuration and should
|
||||
create the necessary C++ source code.
|
||||
- ``DEPENDENCIES``: Mark the component to depend on other components. If the user hasn't explicitly
|
||||
added these components in their configuration, a validation error will be generated.
|
||||
- ``to_code``: The function that will be called with the validated configuration and should create the necessary C++
|
||||
source code.
|
||||
- ``DEPENDENCIES``: Mark the component to depend on other components. If the user hasn't explicitly added these
|
||||
components in their configuration, a validation error will be generated.
|
||||
- ``AUTO_LOAD``: Automatically load a component if the user hasn't added it manually.
|
||||
- ``MULTI_CONF``: Mark this component to accept an array of configurations. If this is an
|
||||
integer instead of a boolean, validation will only permit the given number of entries.
|
||||
- ``CONFLICTS_WITH``: Mark a list of components as conflicting with this component. If the user
|
||||
has one of them in the config, a validation error will be generated.
|
||||
- ``MULTI_CONF``: Mark this component to accept an array of configurations. If this is an integer instead of a boolean,
|
||||
validation will only permit the given number of entries.
|
||||
- ``CONFLICTS_WITH``: Mark a list of components as conflicting with this component. If the user has one of them in
|
||||
their config, a validation error will be generated.
|
||||
- ``CODEOWNERS``: GitHub usernames or team names of people that are responsible for this component. You should add at
|
||||
least your GitHub username here, as well as anyone who helped you to write code that is being included.
|
||||
|
||||
- ``ESP_PLATFORMS``: Provide a list of allowed ESP types this component works with.
|
||||
- ``CODEOWNERS``: GitHub usernames or team names of people that are responsible for this component.
|
||||
You should add at least your GitHub username here, as well as anyone who helped you to write code
|
||||
that is being included.
|
||||
.. _codebase_standards:
|
||||
|
||||
Codebase Standards
|
||||
------------------
|
||||
|
||||
Standard for the esphome-core codebase:
|
||||
ESPHome's maintainers work hard to maintain a high standard for its code. We try our best to adhere to these standards:
|
||||
|
||||
- The C++ code style is based on the
|
||||
`Google C++ Style Guide <https://google.github.io/styleguide/cppguide.html>`__ with a few modifications.
|
||||
|
||||
- function, method and variable names are ``lower_snake_case``
|
||||
- class/struct/enum names should be ``UpperCamelCase``
|
||||
- constants should be ``UPPER_SNAKE_CASE``
|
||||
- fields should be ``protected`` and ``lower_snake_case_with_trailing_underscore_`` (DO NOT use private)
|
||||
- Function, method and variable names are ``lower_snake_case``
|
||||
- Class/struct/enum names should be ``UpperCamelCase``
|
||||
- Constants should be ``UPPER_SNAKE_CASE``
|
||||
- Fields should be ``protected`` and ``lower_snake_case_with_trailing_underscore_`` (DO NOT use ``private``)
|
||||
- It's preferred to use long variable/function names over short and non-descriptive ones.
|
||||
- All uses of class members and member functions should be prefixed with
|
||||
``this->`` to distinguish them from global functions in code review.
|
||||
- All uses of class members and member functions should be prefixed with ``this->`` to distinguish them from global
|
||||
functions/variables.
|
||||
- Use two spaces, not tabs.
|
||||
- Using ``#define`` s is discouraged and should be replaced with constants.
|
||||
- Using ``#define`` is discouraged and should be replaced with constants or enums (if appropriate).
|
||||
- Use ``using type_t = int;`` instead of ``typedef int type_t;``
|
||||
- Wrap lines in all files at no more than 120 characters. This makes reviewing PRs faster and easier. Exceptions
|
||||
should be made only for lines where wrapping them would result in a syntax issue.
|
||||
|
||||
- New components should dump their configuration using ``ESP_LOGCONFIG``
|
||||
at startup in ``dump_config()``
|
||||
- Components should dump their configuration using ``ESP_LOGCONFIG`` at startup in ``dump_config()``.
|
||||
- ESPHome uses a unified formatting tool for all source files (but this tool can be difficult to install).
|
||||
When creating a new PR in GitHub, see the Github Actions output to see what formatting needs to be changed
|
||||
and what potential problems are detected.
|
||||
- Use of external libraries should be kept to a minimum:
|
||||
|
||||
- The number of external libraries should be kept to a minimum. If the component you're developing has a simple
|
||||
communication interface, please consider implementing the library natively in ESPHome.
|
||||
- If the component you're developing has a simple communication interface, please consider implementing the library
|
||||
natively in ESPHome.
|
||||
- Libraries which directly manipulate pins or don't do any I/O generally do not cause problems.
|
||||
- Libraries which use hardware interfaces (I²C, for example), should be configured/wrapped to use ESPHome's own
|
||||
communication abstractions.
|
||||
- If the library accesses a global variable/state (``Wire`` is a good example) then there's likely a problem because
|
||||
the component may not be modular. Put another way, this approach may mean that it's not possible to create multiple
|
||||
instances of the component for use wihtin ESPHome.
|
||||
|
||||
- This depends on the communication interface of course - if the library is directly working
|
||||
with pins or doesn't do any I/O itself, it's ok. However if it's something like I²C, then ESPHome's
|
||||
own communication abstractions should be used. Especially if the library accesses a global variable/state
|
||||
like ``Wire`` there's a problem because then the component may not modular (i.e. not possible
|
||||
to create two instances of a component on one ESP)
|
||||
|
||||
- Components **must** use the provided abstractions like ``sensor``, ``switch`` etc.
|
||||
Components should specifically **not** directly access other components like for example
|
||||
publish to MQTT topics.
|
||||
|
||||
- Implementations for new devices should contain reference links for the datasheet and other sample
|
||||
implementations.
|
||||
- Components **must** use the provided abstractions like ``sensor``, ``switch``, etc. Components specifically should
|
||||
**not** directly access other components -- for example, to publish to MQTT topics.
|
||||
- Implementations for new devices should contain reference links for the datasheet and other sample implementations.
|
||||
- Please test your changes :)
|
||||
|
||||
.. note::
|
||||
@ -694,30 +859,6 @@ Standard for the esphome-core codebase:
|
||||
# Run lint only over changed files from powershell
|
||||
docker run --rm -v "$($current_dir):/esphome" -it ghcr.io/esphome/esphome-lint script/quicklint
|
||||
|
||||
|
||||
PRs are being drafted when changes are needed
|
||||
---------------------------------------------
|
||||
|
||||
If there have been changes requested to your PR, our bot will automatically mark your PR as a draft.
|
||||
This means that the PR is not ready to be merged or further reviewed for the moment.
|
||||
|
||||
Draft PRs tell other reviewers that look at the list of all PRs that this PR is currently in progress and doesn't require their attention yet.
|
||||
|
||||
Once you have made the requested changes, you can mark the PR as ready for review again by clicking the "Ready for review button":
|
||||
|
||||
.. figure:: images/pr-draft-ready.png
|
||||
:align: center
|
||||
:width: 100.0%
|
||||
:alt: The ready for review button in the bottom of a PR in draft mode
|
||||
|
||||
Before you click the "Ready for review" button, ensure you have addressed all requested changes,
|
||||
there are no merge conflicts, and that all our CI jobs and checks are passing successfully.
|
||||
|
||||
Once you've clicked the "Ready for review" button, the PR will return to a normal state again,
|
||||
and our bot will automatically notify the reviewers who requested the changes that the PR is ready to go!
|
||||
|
||||
|
||||
|
||||
See Also
|
||||
--------
|
||||
|
||||
|
BIN
guides/images/update_branch.png
Normal file
BIN
guides/images/update_branch.png
Normal file
Binary file not shown.
After Width: | Height: | Size: 56 KiB |
Loading…
Reference in New Issue
Block a user