|
1 | 1 | # Walkthrough: a typical contribution
|
| 2 | + |
| 3 | +There are _a lot_ of ways to contribute to the rust compiler, including fixing |
| 4 | +bugs, improving performance, helping design features, providing feedback on |
| 5 | +existing features, etc. This chapter does not claim to scratch the surface. |
| 6 | +Instead, it walks through the design and implementation of a new feature. Not |
| 7 | +all of the steps and processes described here are needed for every |
| 8 | +contribution, and I will try to point those out as they arise. |
| 9 | + |
| 10 | +In general, if you are interested in making a contribution and aren't sure |
| 11 | +where to start, please feel free to ask! |
| 12 | + |
| 13 | +## Overview |
| 14 | + |
| 15 | +The feature I will discuss in this chapter is the `?` Kleene operator for |
| 16 | +macros. Basically, we want to be able to write something like this: |
| 17 | + |
| 18 | +```rust |
| 19 | +macro_rules! foo { |
| 20 | + ($arg:ident $(, $optional_arg:ident)?) => { |
| 21 | + println!("{}", $arg); |
| 22 | + |
| 23 | + $( |
| 24 | + println!("{}", $optional_arg); |
| 25 | + )? |
| 26 | + } |
| 27 | +} |
| 28 | + |
| 29 | +fn main() { |
| 30 | + let x = 0; |
| 31 | + foo!(x); // ok! prints "0" |
| 32 | + foo!(x, x); // ok! prints "0 0" |
| 33 | +} |
| 34 | +``` |
| 35 | + |
| 36 | +So basically, the `$(pat)?` matcher in the macro means "this pattern can occur |
| 37 | +0 or 1 times", similar to other regex syntaxes. |
| 38 | + |
| 39 | +There were a number of steps to go from an idea to stable rust feature. Here is |
| 40 | +a quick list. We will go through each of these in order below. As I mentioned |
| 41 | +before, not all of these are needed for every type of contribution. |
| 42 | + |
| 43 | +- **Idea discussion/Pre-RFC** A Pre-RFC is an early draft or design discussion |
| 44 | + of a feature. This stage is intended to flesh out the design space a bit and |
| 45 | + get a grasp on the different merits and problems with an idea. It's a great |
| 46 | + way to get early feedback on your idea before presenting it the wider |
| 47 | + audience. You can find the original discussion [here][prerfc]. |
| 48 | +- **RFC** This is when you formally present your idea to the community for |
| 49 | + consideration. You can find the RFC [here][rfc]. |
| 50 | +- **Implementation** Implement your idea unstabley in the compiler. You can |
| 51 | + find the original implementation [here][impl1]. |
| 52 | +- **Possibly iterate/refine** As the community gets experience with your |
| 53 | + feature on the nightly compiler and in `libstd`, there may be additional |
| 54 | + feedback about design choice that might be adjusted. This particular feature |
| 55 | + went [through][impl2] a [number][impl3] of [iterations][impl4]. |
| 56 | +- **Stabilization** When your feature has baked enough, a rust team member may |
| 57 | + [propose to stabilize it][merge]. If there is consensus, this is done. |
| 58 | +- **Relax** Your feature is now a stable rust feature! |
| 59 | + |
| 60 | +[prerfc]: https://internals.rust-lang.org/t/pre-rfc-at-most-one-repetition-macro-patterns/6557 |
| 61 | +[rfc]: https://github.com/rust-lang/rfcs/pull/2298 |
| 62 | +[impl1]: https://github.com/rust-lang/rust/pull/47752 |
| 63 | +[impl2]: https://github.com/rust-lang/rust/pull/49719 |
| 64 | +[impl3]: https://github.com/rust-lang/rust/pull/51336 |
| 65 | +[impl4]: https://github.com/rust-lang/rust/pull/51587 |
| 66 | +[merge]: https://github.com/rust-lang/rust/issues/48075#issuecomment-433177613 |
| 67 | + |
| 68 | +## Pre-RFC and RFC |
| 69 | + |
| 70 | +> NOTE: In general, if you are not proposing a _new_ feature or substantial |
| 71 | +> change to rust or the ecosystem, you don't need to follow the RFC process. |
| 72 | +> Instead, you can just jump to [implementation](#impl). |
| 73 | +> |
| 74 | +> You can find the official guidelines for when to open an RFC [here][rfcwhen]. |
| 75 | +
|
| 76 | +[rfcwhen]: https://github.com/rust-lang/rfcs#when-you-need-to-follow-this-process |
| 77 | + |
| 78 | +An RFC is a document that describes the feature or change you are proposing in |
| 79 | +detail. Anyone can write an RFC; the process is the same for everyone, |
| 80 | +including rust team members. |
| 81 | + |
| 82 | +To open an RFC, open a PR on the |
| 83 | +[rust-lang/rfcs](https://github.com/rust-lang/rfcs) repo on GitHub. You can |
| 84 | +find detailed instructions in the |
| 85 | +[README](https://github.com/rust-lang/rfcs#what-the-process-is). |
| 86 | + |
| 87 | +Before opening an RFC, you should do the research to "flesh out" your idea. |
| 88 | +Hastily-proposed RFCs tend not to be accepted. You should generally have a good |
| 89 | +description of the motivation, impact, disadvantages, and potential |
| 90 | +interactions with other features. |
| 91 | + |
| 92 | +If that sounds like a lot of work, it's because it is. But no fear! Even if |
| 93 | +you're not a compiler hacker, you can get great feedback by doing a _pre-RFC_. |
| 94 | +This is an _informal_ discussion of the idea. The best place to do this is |
| 95 | +internals.rust-lang.org. Your post doesn't have to follow any particular |
| 96 | +structure. It doesn't even need to be a cohesive idea. Generally, you will get |
| 97 | +tons of feedback that you can integrate back to produce a good RFC. |
| 98 | + |
| 99 | +(Another pro-tip: try searching the RFCs repo and internals for prior related |
| 100 | +ideas. A lot of times an idea has already been considered and was either |
| 101 | +rejected or postponed to be tried again later. This can save you and everybody |
| 102 | +else some time) |
| 103 | + |
| 104 | +In the case of our example, a participant in the pre-RFC thread pointed out a |
| 105 | +syntax ambiguity and a potential resolution. Also, the overall feedback seemed |
| 106 | +positive. In this case, the discussion converged pretty quickly, but for some |
| 107 | +ideas, a lot more discussion can happen (e.g. see [this RFC][nonascii] which |
| 108 | +received a whopping 684 comments!). If that happens, don't be discouraged; it |
| 109 | +means the community is interested in your idea, but it perhaps needs some |
| 110 | +adjustments. |
| 111 | + |
| 112 | +[nonascii]: https://github.com/rust-lang/rfcs/pull/2457 |
| 113 | + |
| 114 | +The RFC for our `?` macro feature did receive some discussion on the RFC thread |
| 115 | +too. As with most RFCs, there were a few questions that we couldn't answer by |
| 116 | +discussion: we needed experience using the feature to decide. Such questions |
| 117 | +are listed in the "Unresolved Questions" section of the RFC. Also, over the |
| 118 | +course of the RFC discussion, you will probably want to update the RFC document |
| 119 | +itself to reflect the course of the discussion (e.g. new alternatives or prior |
| 120 | +work may be added or you may decide to change parts of the proposal itself). |
| 121 | + |
| 122 | +In the end, when the discussion seems to reach a consensus and die down a bit, |
| 123 | +a rust team member may propose [to merge the RFC][rfcmerge]. This means that |
| 124 | +they want the other members of the appropriate teams to review and comment on |
| 125 | +the RFC. More changes may be proposed. At some point, when everyone is |
| 126 | +satisfied, the RFC enters the "final comment period" (FCP), which is the last |
| 127 | +chance for people to bring up objections. When the FCP is over, the RFC is |
| 128 | +"merged" (or accepted). |
| 129 | + |
| 130 | +[rfcmerge]: https://github.com/rust-lang/rfcs/pull/2298#issuecomment-360582667 |
| 131 | + |
| 132 | +Some other possible outcomes might be for a team member to propose to |
| 133 | + |
| 134 | +- _Close_: this feature in its current form is not a good fit for rust. Don't |
| 135 | + be discouraged if this happens to your RFC, and don't take it personally. |
| 136 | + This is not a reflection on you, but rather a community decision that rust |
| 137 | + will go a different direction. |
| 138 | +- _Postpone_: there is interest in going this direction but not at the moment. |
| 139 | + This happens most often because the appropriate rust team doesn't have the |
| 140 | + bandwidth to shepherd the feature through the process to stabilization. Often |
| 141 | + this is the case when the feature doesn't fit into the team's roadmap. |
| 142 | + Postponed ideas may be revisited later. |
| 143 | + |
| 144 | +<a name="impl"></a> |
| 145 | + |
| 146 | +## Implementation |
| 147 | + |
| 148 | +TODO |
0 commit comments