|
| 1 | +# Mypy Roadmap |
| 2 | + |
| 3 | +The goal of the roadmap is to document areas the mypy core team is |
| 4 | +planning to work on in the future or is currently working on. PRs |
| 5 | +targeting these areas are very welcome, but please check first with a |
| 6 | +core team member that nobody else is working on the same thing. |
| 7 | + |
| 8 | +**Note:** This doesn’t include everything that the core team will work |
| 9 | +on, and everything is subject to change. Near-term plans are likely |
| 10 | +more accurate. |
| 11 | + |
| 12 | +## April-June 2017 |
| 13 | + |
| 14 | +- Add more comprehensive testing for `--incremental` and `--quick` |
| 15 | + modes to improve reliability. At least write more unit tests with |
| 16 | + focus on areas that have previously had bugs. |
| 17 | + ([issue](https://github.com/python/mypy/issues/3455)) |
| 18 | + |
| 19 | +- Speed up `--quick` mode to better support million+ line codebases |
| 20 | + through some of these: |
| 21 | + |
| 22 | + - Make it possible to use remote caching for incremental cache |
| 23 | + files. This would speed up a cold run with no local cache data. |
| 24 | + We need to update incremental cache to use hashes to determine |
| 25 | + whether files have changes to allow |
| 26 | + [sharing cache data](https://github.com/python/mypy/issues/3403). |
| 27 | + |
| 28 | + - See if we can speed up deserialization of incremental cache |
| 29 | + files. Initial experiments aren’t very promising though so there |
| 30 | + might not be any easy wins left. |
| 31 | + ([issue](https://github.com/python/mypy/issues/3456)) |
| 32 | + |
| 33 | +- Improve support for complex signatures such as `open(fn, 'rb')` and |
| 34 | + specific complex decorators such as `contextlib.contextmanager` |
| 35 | + through type checker plugins/hooks. |
| 36 | + ([issue](https://github.com/python/mypy/issues/1240)) |
| 37 | + |
| 38 | +- Document basic properties of all type operations used within mypy, |
| 39 | + including compatibility, proper subtyping, joins and meets. |
| 40 | + ([issue](https://github.com/python/mypy/issues/3454)) |
| 41 | + |
| 42 | +- Make TypedDict an officially supported mypy feature. This makes it |
| 43 | + possible to give precise types for dictionaries that represent JSON |
| 44 | + objects, such as `{"path": "/dir/fnam.ext", "size": 1234}`. |
| 45 | + ([issue](https://github.com/python/mypy/issues/3453)) |
| 46 | + |
| 47 | +- Make error messages more useful and informative. |
| 48 | + ([issue](https://github.com/python/mypy/labels/topic-usability)) |
| 49 | + |
| 50 | +- Resolve [#2008](https://github.com/python/mypy/issues/2008) (we are |
| 51 | + converging on approach 4). |
| 52 | + |
| 53 | +## July-December 2017 |
| 54 | + |
| 55 | +- Invest some effort into systematically filling in missing |
| 56 | + annotations and stubs in typeshed, with focus on features heavily |
| 57 | + used at Dropbox. Better support for ORMs will be a separate |
| 58 | + project. |
| 59 | + |
| 60 | +- Improve opt-in warnings about `Any` types to make it easier to keep |
| 61 | + code free from unwanted `Any` types. For example, warn about using |
| 62 | + `list` (instead of `List[x]`) and calling `open` if we can’t infer a |
| 63 | + precise return type, or using types imported from ignored modules |
| 64 | + (they are implicitly `Any`). |
| 65 | + |
| 66 | +- Add support for protocols and structural subtyping (PEP 544). |
| 67 | + |
| 68 | +- Switch completely to pytest and remove the custom testing framework. |
| 69 | + ([issue](https://github.com/python/mypy/issues/1673)) |
| 70 | + |
| 71 | +- Make it possible to run mypy as a daemon to avoid reprocessing the |
| 72 | + entire program on each run. This will improve performance |
| 73 | + significantly. Even when using the incremental mode, processing a |
| 74 | + large number of files is not cheap. |
| 75 | + |
| 76 | +- Refactor and simplify specific tricky parts of mypy internals, such |
| 77 | + as the [conditional type binder](https://github.com/python/mypy/issues/3457), |
| 78 | + [symbol tables](https://github.com/python/mypy/issues/3458) or |
| 79 | + the various [semantic analysis passes](https://github.com/python/mypy/issues/3459). |
| 80 | + |
| 81 | +- Implement a general type system plugin architecture. It should be |
| 82 | + able to support some typical ORM features at least, such as |
| 83 | + metaclasses that add methods with automatically inferred signatures |
| 84 | + and complex descriptors such as those used by Django models. |
| 85 | + ([issue](https://github.com/python/mypy/issues/1240)) |
| 86 | + |
| 87 | +- Add support for statically typed |
| 88 | + [protobufs](https://developers.google.com/protocol-buffers/). |
| 89 | + |
| 90 | +- Provide much faster, reliable interactive feedback through |
| 91 | + fine-grained incremental type checking, built on top the daemon |
| 92 | + mode. |
| 93 | + |
| 94 | +- Start work on editor plugins and support for selected IDE features. |
0 commit comments