Allow macros to generate method symbols, add missing method type constructors #8090
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dottybot
reviewed
Jan 25, 2020
|
sorry, git accident |
nicolasstucki
suggested changes
Jan 27, 2020
| * | ||
| * @note As a macro can only splice code into the point at which it is expanded, all generated symbols must be | ||
| * direct or indirect children of the reflection context's owner. */ | ||
| def newMethod(parent: Symbol, name: String, tpe: Type, flags: Flags = Flags.EmptyFlags, privateWithin: Option[Symbol] = None)(given ctx: Context): Symbol = |
There was a problem hiding this comment.
We are trying to avoid default parameters for compatibility reasons. Remove them and maybe add an overloaded version without the last two.
/** ... */
def newMethod(parent: Symbol, name: String, tpe: Type)(given ctx: Context): Symbol =
newMethod(parent, name, tpe, Flags.EmptyFlags, None)
/** ... */
def newMethod(parent: Symbol, name: String, tpe: Type, flags: Flags, privateWithin: Option[Symbol])(given ctx: Context): Symbol =…tructors Since there already exists the machinery to generate a new tree for an existing method symbol, this change simply adds the ability to generate a fresh method symbol with the given name, type and flags. Then, `DefDef.apply` can be used to generate the full tree. Missing type constructors for `PolyType` and `ByNameType` are added, as well as corresponding `.param(Int)` accessors in order to make references to a `MethodType` or a `PolyType`'s parameter types. Testing is achieved by synthesizing a series of method definitions, along with references to these methods and some basic correctness assertions. The purpose of each test is described in a comment above the corresponding block of code. Note 1: it is possible to additionaly specify flags and private in the `Symbol.newMethod` function. These seem unnecessary for just defining local functions, but may be useful in synthesizing more complex declarations that may be supported later, such as local classes. Note 2: implicit/given method types are not exposed by this change. Since these types are structurally identical to normal method types, perhaps some optional flags could be added to the MethodType constructor for this purpose?
fhackett
force-pushed
the
fhackett-synth-defdef
branch
from
97945bc to
5ff0325
Compare
nicolasstucki
approved these changes
Jan 28, 2020
|
Thank you @fhackett |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Partially addressing #6280, and the issue underlying #7626. Review by @nicolasstucki.
Since there already exists the machinery to generate a new tree for an existing method symbol, this change simply adds the ability to generate a fresh method symbol with the given name, type and flags. Then,
DefDef.applycan be used to generate the full tree.Missing type constructors for
PolyTypeandByNameTypeare added, as well as corresponding.param(Int)accessors in order to make references to aMethodTypeor aPolyType's parameter types.Testing is achieved by synthesizing a series of method definitions, along with references to these methods and some basic correctness assertions. The purpose of each test is described in a comment above the corresponding block of code.
Note 1: there are optional parameters for flags and private in the
Symbol.newMethodfunction. These seem unnecessary for just defining local functions, but may be useful in synthesizing more complex declarations that may be supported later, such as local classes.Note 2: implicit/given method types are not exposed by this change. Since these types are structurally identical to normal method types, perhaps some optional flags could be added to the MethodType constructor for this purpose?