Skip to content
This repository has been archived by the owner on Nov 15, 2023. It is now read-only.

add new documentation guidelines #14115

Merged
merged 21 commits into from
May 26, 2023
Merged
Show file tree
Hide file tree
Changes from 17 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 16 additions & 10 deletions docs/CODEOWNERS
Validating CODEOWNERS rules …
Original file line number Diff line number Diff line change
Expand Up @@ -35,11 +35,17 @@
/client/sysinfo/ @koute
/client/tracing/ @koute

# Documentation audit
/primitives/runtime @paritytech/docs-audit
/primitives/arithmetic @paritytech/docs-audit
# /primitives/core (to be added later)
# /primitives/io (to be added later)

# FRAME
/frame/ @paritytech/frame-coders
/frame/nfts/ @jsidorenko
/frame/ @paritytech/frame-coders @paritytech/docs-audit
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
/frame/nfts/ @jsidorenko @paritytech/docs-audit
/frame/state-trie-migration/ @paritytech/frame-coders @cheme
/frame/uniques/ @jsidorenko
/frame/uniques/ @jsidorenko @paritytech/docs-audit

# GRANDPA, BABE, consensus stuff
/client/consensus/babe/ @andresilva
Expand All @@ -57,12 +63,12 @@
/primitives/merkle-mountain-range/ @acatangiu

# Contracts
/frame/contracts/ @athei
/frame/contracts/ @athei @paritytech/docs-audit

# NPoS and election
/frame/election-provider-multi-phase/ @paritytech/staking-core
/frame/election-provider-support/ @paritytech/staking-core
/frame/elections-phragmen/ @paritytech/staking-core
/frame/nomination-pools/ @paritytech/staking-core
/frame/staking/ @paritytech/staking-core
/primitives/npos-elections/ @paritytech/staking-core
/frame/election-provider-multi-phase/ @paritytech/staking-core @paritytech/docs-audit
/frame/election-provider-support/ @paritytech/staking-core @paritytech/docs-audit
/frame/elections-phragmen/ @paritytech/staking-core @paritytech/docs-audit
/frame/nomination-pools/ @paritytech/staking-core @paritytech/docs-audit
/frame/staking/ @paritytech/staking-core @paritytech/docs-audit
/primitives/npos-elections/ @paritytech/staking-core @paritytech/docs-audit
1 change: 1 addition & 0 deletions docs/CONTRIBUTING.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ There are a few basic ground-rules for contributors (including the maintainer(s)
. **All modifications** must be made in a **pull-request** to solicit feedback from other contributors.
. A pull-request *must not be merged until CI* has finished successfully.
. Contributors should adhere to the link:STYLE_GUIDE.md[house coding style].
. Contributors should adhere to the link:DOCUMENTATION_GUIDELINES.md[house documenting style], when applicable.


== Merge Process
Expand Down
254 changes: 254 additions & 0 deletions docs/DOCUMENTATION_GUIDELINES.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,254 @@
wra# Substrate Documentation Guidelines
kianenigma marked this conversation as resolved.
Show resolved Hide resolved

This document is only focused on documenting parts of substrate that relates to its external API. The list of such crates can be found in [CODEOWNERS](./CODEOWNERS). Search for the crates that are auto-assigned to a team called `docs-audit`.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We could have some docs labels similar to needs-audit like needs-docs and the CI puts them automatically.
Not sure how much that gets us, but would make it easier for people to see that they need to update it.

Copy link
Contributor

@sacha-l sacha-l May 11, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I support this idea fwiw, yes

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would defer this decision to be made as a part of paritytech/labels#29 by @the-right-joyce and @juangirini. As far as I know, the agreed that using labels for things like audit was not the right choice as it is not a permanent label (eg 'needs audit' -> 'audited').

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It depends on how you want to see it. needs-docs could very well be a permanent label, if it means that the PR is worth documenting instead of whether it was already documented or not then it is a permanent label. Similarly we could think of needs-audit.
In that context needs-docs might need a better naming, but I see it useful too.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I find the label be a redundant side effect of the auto-review rule, isn't that the case? we already will assign this team to review automatically. I think this is semantically the same as the label.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

True, good point. Though the auto-review rule will only ask the @paritytech/docs-audit team for a review. Should we add a rule to require at least one approval from them on top of the approvals we already need?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Eventually, yes, but let's first see if the docs-audit team has enough bandwidth for this.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we got rid off the audit labels, for the security audits we will use Github projects in future. The process still needs to be documented, but as a preparation I already have here the project set up: https://github.com/orgs/paritytech/projects/103
I'm syncing with Regina from SRLabs regarding that.

When it comes to the docs update, I'd also say this should be part of the review process. The same way we're changing the release process and want reviewers to add/adapt the changelog if needed, I think it should also be in their responsibility to think if the PR is important for docs as well and respectively add @paritytech/docs-audit as reviewers and the docs team are the ones who decide if the docs should be updated or not - does that make sense?


These are crates that are often used by external developers and need more thorough documentation. These are the crates most concerned with FRAME development.

- [General/Non-Pallet Crates](#generalnon-pallet-crates)
- [What to Document?](#what-to-document)
- [Rust Docs vs. Code Comments](#rust-docs-vs-code-comments)
- [How to Document?](#how-to-document)
- [TLDR](#tldr)
- [Proc-Macros](#proc-macros)
- [Other Guidelines](#other-guidelines)
- [Document Through Code](#document-through-code)
- [Formatting Matters](#formatting-matters)
- [Pallet Crates](#pallet-crates)
- [Top Level Pallet Docs (`lib.rs`)](#top-level-pallet-docs-librs)
- [Polkadot and Substrate](#polkadot-and-substrate)
- [Dispatchable](#dispatchable)
- [Storage Items](#storage-items)


## General/Non-Pallet Crates

First, consider the case for all such crates, except for those that are pallets.

### What to Document?

The first question is, what should you document? Use the following filter:

1. In the crates assigned to `docs-audit` in [CODEOWNERS](./CODEOWNERS),
2. All `pub` item need to be documented. If it is not `pub`, it does not appear in the rust-docs, and is not public facing.
* Within `pub` items, sometimes they are only `pub` in order to be used by another internal crate, and you can foresee that this will not be used by anyone else other than you. These need **not** be documented thoroughly, and are left to your discretion to identify.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
* Within `pub` items, sometimes they are only `pub` in order to be used by another internal crate, and you can foresee that this will not be used by anyone else other than you. These need **not** be documented thoroughly, and are left to your discretion to identify.
* There are some `pub` items that are only `pub` in order to be accessible for another internal crate. These should still be documented although you do not need to be as thorough as you are with the `pub` items that are for external developer use.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we ask to add a comment like "//no-doc" to establish that the author intentionally didn't add the docs?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@hummusonrails I am a bit hesitant about your suggestions that seem paraphrasing to me. Do you have strong opinions that the original text is wrong or poorly worded, or are you adding optional suggestions? I accepted most, but might close some. Please let me know if you prefer otherwise.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@kianenigma In this one, I attempted to make it more neutral and formal, whereas your version was more informal and personal in tone. I'm of the opinion that documentation should lean more into the former. That's why I made that suggestion. Hope that helps clarify!

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

One common convention is to expose such items via a #[doc(hidden)] public module called __private from the crate root. In fact this is used in at least 5 popular crates I can think of, including the entire syn/quote ecosystem. Maybe worth suggesting a convention here?

* Reminder: `trait` items are public by definition, if the trait is public.
3. All public modules (`mod`) should have reasonable module-level documentation (`//!`).


#### Rust Docs vs. Code Comments

Note that anything starting with `///` is an external rust-doc, and everything starting with `//` does not appear in the rust-docs. It's important to not confuse the two in your documentation.

```rust
/// Computes the square root of the input, returning `Ok(_)` if successful.
///
/// # Errors
/// ...
///
// Details about the complexity, how you implemented this, and some quirks that
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
// are NOT relevant to the external interface, so it starts with '//'.
// This can also be moved inside the function.
pub fn sqrt(x: u32) -> Result<u32, ()> {
todo!();
}
```

### How to Document?

There are a few very good sources that you can look into:

- https://doc.rust-lang.org/rustdoc/how-to-write-documentation.html
- https://web.mit.edu/rust-lang_v1.25/arch/amd64_ubuntu1404/share/doc/rust/html/book/first-edition/documentation.html
- https://blog.guillaume-gomez.fr/articles/2020-03-12+Guide+on+how+to+write+documentation+for+a+Rust+crate

As mentioned [here](https://web.mit.edu/rust-lang_v1.25/arch/amd64_ubuntu1404/share/doc/rust/html/book/first-edition/documentation.html#writing-documentation-comments) and [here](https://blog.guillaume-gomez.fr/articles/2020-03-12+Guide+on+how+to+write+documentation+for+a+Rust+crate), always start with a **single sentence** demonstrating what is being documented. All additional documentation should be added *after a newline*. Strive to make the first sentence succinct and short. The reason for this is the first paragraph of docs about an item (everything before the first newline) is used as the excerpt that rust doc displays about this item when it appears in tables, such as the table listing all functions in a module. If this excerpt is too long, the module docs will be very difficult to read.

About [special sections](https://web.mit.edu/rust-lang_v1.25/arch/amd64_ubuntu1404/share/doc/rust/html/book/first-edition/documentation.html#special-sections), we will most likely not need to think about panic and safety in any runtime related code. Our code is never `unsafe`, and will (almost) never panic.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
About [special sections](https://web.mit.edu/rust-lang_v1.25/arch/amd64_ubuntu1404/share/doc/rust/html/book/first-edition/documentation.html#special-sections), we will most likely not need to think about panic and safety in any runtime related code. Our code is never `unsafe`, and will (almost) never panic.
About [special sections](https://web.mit.edu/rust-lang_v1.25/arch/amd64_ubuntu1404/share/doc/rust/html/book/first-edition/documentation.html#special-sections), we will most likely not need to think about panic and safety in any runtime related code. Our code is never `unsafe`, and will (almost) never panic. If you see a panic in the code that is a reason to raise a question to the developer who introduced it rather than work to document it.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you elaborate more what you are suggesting? in general, you might see things like expect() in the code but that is acceptable. Did you take that into account?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was trying to make explicit that a code documenter should flag a panic and seek clarification before documenting as this in the "How to document" section. Very open to alternative ways to do so!


Use `# Examples as much as possible. These are great ways to further demonstrate what your APIs are doing, and add free test coverage. As an additional benefit, any code in rust-docs is treated as an "integration tests", not unit tests, which tests your crate in a different way than unit tests. So, it is both a win for "more documentation" and a win for "more test coverage".

You can also consider having an `# Error` section optionally. Of course, this only applies if there is a `Result` being returned, and if the `Error` variants are overly complicated.

Strive to include correct links to other items in your written docs as much as possible. In other words, avoid \`some_func\` and instead use \[\`some_fund\`\]. Read more about how to correctly use links in your rust-docs [here](https://doc.rust-lang.org/rustdoc/write-documentation/linking-to-items-by-name.html#valid-links) and [here](https://rust-lang.github.io/rfcs/1946-intra-rustdoc-links.html#additions-to-the-documentation-syntax).
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
Strive to include correct links to other items in your written docs as much as possible. In other words, avoid `` `some_func` `` and instead use ``[`some_func`]``.


> While you are linking, you might become conscious of the fact that you are in need of linking to (too many) foreign items in order to explain your API. This is leaning more towards API-Design rather than documentation, but it is a warning that the subject API might be slightly wrong. For example, most "glue" traits[^1] in `frame/support` should be designed and documented without making hard assumptions about particular pallets that implement them.

#### TLDR

0. Have the goal of enforcing `#![deny(missing_docs)]` mentally, even if it is not enforced by the compiler 🙈.
1. Start with a single, clear and concise sentence. Follow up with more context, after a newline, if needed.
2. Use examples as much as reasonably possible.
3. Use links as much as possible.
4. Think about context. If you are explaining a lot of foreign topics while documenting a trait that should not explicitly depend on them, you have likely not designed it properly.

#### Proc-Macros

Note that there are special considerations when documenting proc macros. Doc links will appear to function _within_ your proc macro crate, but often will no longer function when these proc macros are re-exported elsewhere in your project. The exception is doc links to _other proc macros_ which will function just fine if they are also being re-exported. It is also often necessary to disambiguate between a proc macro and a function of the same name, which can be done using the `macro@my_macro_name` syntax in your link. Read more about how to correctly use links in your rust-docs [here](https://doc.rust-lang.org/rustdoc/write-documentation/linking-to-items-by-name.html#valid-links) and [here](https://rust-lang.github.io/rfcs/1946-intra-rustdoc-links.html#additions-to-the-documentation-syntax).


### Other Guidelines

The above five guidelines must always be reasonably respected in the documentation.

The following are a set of notes that may not necessarily hold in all circumstances:


#### Document Through Code

You should make sure that your code is properly-named and well-organized so that your code functions as a form of documentation. However, within the complexity of our projects in Polkadot/Substrate that is not enough. Particularly, things like examples, errors and panics cannot be documented only through properly-named and well-organized code.

> Our north star is self-documenting code that also happens to be well-documented and littered with examples.


* Your written documents should *complement* the code, not *repeat* it. As an example, a documentation on top of a code example should never look like the following:

```rust
/// Sends request and handles the response.
trait SendRequestAndHandleResponse {

}
```

In the above example, the documentation has added no useful information not already contained within the properly-named trait and is redundant.


#### Formatting Matters

The way you format your documents (newlines, heading and so on) makes a difference. Consider the below examples:

```rust
/// This function works with input u32 x and multiplies it by two. If
/// we optimize the other variant of it, we would be able to achieve more
/// efficiency but I have to think about it. Probably can panic if the input
/// overflows u32.
fn multiply_by_2(x: u32) -> u32 { .. }
```

```rust
/// Multiplies an input of type [`u32`] by two.
///
/// # Panics
///
/// Panics if the input overflows.
///
/// # Complexity
///
/// Is implemented using some algorithm that yields complexity of O(1).
// More efficiency can be achieved if we improve this via such and such.
fn multiply_by_2(x: u32) -> u32 { .. }
```

They are both roughly conveying the same set of facts, but one is easier to follow because it was formatted cleanly. Especially for traits and types that you can foresee will be seen and used a lot, try and write a well formatted version.

Similarly, make sure your comments are wrapped at 100 characters line-width (as defined by our [`rustfmt.toml`](../rustfmt.toml)), no **more and no less**! The more is fixed by `rustfmt` and our CI, but if you (for some unknown reason) wrap your lines at 59 characters, it will pass the CI, and it will not look good 🫣. Consider using a plugin like [rewrap](https://marketplace.visualstudio.com/items?itemName=stkb.rewrap) (for Visual Studio Code) to properly do this.

[^1]: Those that help two pallets talk to each other.

## Pallet Crates

The guidelines so far have been general in nature, and are applicable to crates that are pallets and crates that're not pallets.

The following is relevant to how to document parts of a crate that is a pallet. See [`pallet-fast-unstake`](../frame/fast-unstake/src/lib.rs) as one examples of adhering these guidelines.

### Top Level Pallet Docs (`lib.rs`)

For the top-level pallet docs, consider the following template:

```
//! # <Pallet Name>
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
//!
//! <single-liner about the pallet>.
//!
//! ## Overview
//!
//! <should be high-level details that are relevant to the most broad audience>
//!
//! <The audience here is potentially non-coders who just want to know what this pallet does, not how it does it>
//!
//! <potentially a few paragraphs, focus on what external folks should know about the pallet>
//!
//! ### Example
//!
//! <Your pallet must have a few tests that cover important user journeys. Use https://crates.io/crates/docify to reuse these as examples>.
//!
//! ## Pallet API
//!
//! <Reminder: inside the [`pallet`] module, a template that leads the reader to the relevant items is auto-generated. There is no need to repeat things like "See Config trait for ...", which are generated inside [`pallet`] here anyways. You can use the below line as-is:>
//!
//! See the [`pallet`] module for more information about the interfaces this pallet exposes, including its configuration trait, dispatchables, storage items, events and errors.
//!
//! <The audience of this is those who want to know how this pallet works, to the extent of being able to build something on top of it, like a DApp or another pallet>
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
//! <The audience of this is those who want to know how this pallet works, to the extent of being able to build something on top of it, like a DApp or another pallet>
//! <Readers of this section are developers who want to know how this pallet works, to the extent of being able to build another pallet that uses this pallet, or to build a DApp on top of it>

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you expect authors to write 1-2 sentences addressing this? I saw in the fast-unstake pallet you just left it at:

See pallet module for more information.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

good point, let me re-asses.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Updated. I think in 99% of the time this section can be copy-pasted. We could even consider auto-generating it, but given that we want it to come after ## Overview, it might be hard to do.

//!
//! This section can most often be left as-is.
//!
//! ## Low Level / Implementation Details
//!
//! <The format of this section is up to you, but we suggest the Design-oriented approach that follows>
//!
//! <The audience of this would be your future self, or anyone who wants to gain a deep understanding of how the pallet works so that they can eventually propose optimizations to it>
//!
//! ### Design Goals (optional)
//!
//! <Describe your goals with the pallet design.>
//!
//! ### Design (optional)
//!
//! <Describe how you've reached those goals. This should describe the storage layout of your pallet and what was your approach in designing it that way.>
//!
//! ### Terminology (optional)
sacha-l marked this conversation as resolved.
Show resolved Hide resolved
//!
//! <Optionally, explain any non-obvious terminology here. You can link to it if you want to use the terminology further up>
```

This template's details (heading 3s and beyond) are left flexible, and at the discretion of the developer to make the best final choice about. For example, you might want to include `### Terminology` or not. Moreover, you might find it more useful to include it in `## Overview`.

Nonetheless, the high level flow of going from the most high level explanation to the most low level explanation is important to follow.

As a rule of thumb, the Heading 2s (`##`) in this template can be considered a strict rule, while the Heading 3s (`###`) and beyond are flexible.

#### Polkadot and Substrate

Optionally, in order to demonstrate the relation between the two, you can start the pallet documentation with:

```
//! > Made with *Substrate*, for *Polkadot*.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sounds a bit exclusive, when in reality used not only by Polkadot.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

well to be fair "made for" doesn't mean it couldn't be used for other things, but this is a good point

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a trial just for marketing to see. We are still yet to decide if we want to add this in all pallets and crates.

//!
//! [![github]](https://github.com/paritytech/substrate/frame/fast-unstake) -
//! [![polkadot]](https://polkadot.network)
//!
//! [polkadot]: https://img.shields.io/badge/polkadot-E6007A?style=for-the-badge&logo=polkadot&logoColor=white
//! [github]: https://img.shields.io/badge/github-8da0cb?style=for-the-badge&labelColor=555555&logo=github
```

### Dispatchable

For each dispatchable (`fn` item inside `pallet::call`), consider the following template:
kianenigma marked this conversation as resolved.
Show resolved Hide resolved

```
/// <One-liner explaining what the dispatchable does>
///
/// ## Dispatch Origin
///
/// The dispatch origin of this call must be <details (e.g. Root, Signed, Unsigned)>
///
/// ## Details
///
/// <All other details, namely any errors that could occur within this dispatch and the events this dispatch could emit>
///
/// ## Errors (optional)
///
/// <If an extensive list of errors can be returned, list them individually instead of mentioning them in the section above>
///
/// ## Events (optional)
///
/// <Events are akin to the "return type" of dispatchables, optionally mention them>
pub fn name_of_dispatchable(origin: OriginFor<T>, ...) -> DispatchResult {}
```

### Storage Items

1. If a map-like type is being used, always note the choice of your hashers as private code docs (`// Hasher X chosen because ...`). Recall that this is not relevant information to external people, so it must be documented as `//`.
2. Consider explaining why a storage type is always bounded.
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
3. Consider explaining the crypto-economics of how a deposit is being taken in return of the storage being used.
5 changes: 2 additions & 3 deletions docs/PULL_REQUEST_TEMPLATE.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,12 +18,11 @@ Before you submit, please check that:
- [ ] Github project assignment
kianenigma marked this conversation as resolved.
Show resolved Hide resolved
- [ ] **Related Issues:** You mentioned a related issue if this PR is related to it, e.g. `Fixes #228` or `Related #1337`.
- [ ] **2 Reviewers:** You asked at least two reviewers to review. If you aren't sure, start with GH suggestions.
- [ ] **Style Guide:** Your PR adheres to [the style guide](https://github.com/paritytech/substrate/blob/master/docs/STYLE_GUIDE.md)
- [ ] **Style Guide:** Your PR adheres to [the style guide](https://github.com/paritytech/substrate/blob/master/docs/STYLE_GUIDE.md).
- In particular, mind the maximal line length of 100 (120 in exceptional circumstances).
- There is no commented code checked in unless necessary.
- Any panickers in the runtime have a proof or were removed.
- [ ] **Runtime Version:** You bumped the runtime version if there are breaking changes in the **runtime**.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why is this no longer needed?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is not relevant anymore, contributors don't bump this. only the release team manages it. @coderobe can you confirm?

- [ ] **Docs:** You updated any rustdocs which may need to change.
- [ ] **Docs:** You updated any rustdocs which may need to change, as described [here](https://github.com/paritytech/substrate/blob/master/docs/DOCUMENTATION_GUIDELINES.md).
- [ ] **Polkadot Companion:** Has the PR altered the external API or interfaces used by Polkadot?
- [ ] If so, do you have the corresponding Polkadot PR ready?
- [ ] Optionally: Do you have a corresponding Cumulus PR?
Expand Down