Tree transformer&TreeTransform #43
Conversation
|
What about |
|
|
||
| def transformAnnotated(tree: Annotated, annot: Tree, arg: Tree)(implicit ctx: Context): Annotated = cpy.Annotated(tree, annot, arg) | ||
|
|
||
| def transformSharedTree(tree: SharedTree, shared: Tree)(implicit ctx: Context): SharedTree = cpy.SharedTree(tree, shared) |
There was a problem hiding this comment.
Potentially yes. But now it's still here.
|
@odersky I had a close look at optimizations in your implementation and incorporated similar ideas:
|
| tree = tree match { | ||
| case tree: Ident => | ||
| var t: Tree = tree | ||
| val mutatedInfo = mutateTransformers(info, _.prepareForIdent(tree), info.nx.nxPrepIdent, cur) |
There was a problem hiding this comment.
idea: preemptively create mutators
|
Result of discussion: split big method, have transformers responsible of calling remaining transformations on created nodes. |
|
LGTM, although I suggest squashing the 4 commits because their history is erased in the last one anyway. |
|
Agree on sqashing and another small thing: I would suggest to reorder prepare/transform methods so that each is a block. It's easier to see what kinds of methods are supported that way, and it's also simpler to just copy a block of transform methods in order to modify them in a subclass. Otherwise LGTM. |
1) using fast tracks in case node type isn't altered by Transformation; 2) using pre-computed hints(nxTransformXXX arrays) to quickly jump to next transformation interested in transforming particular Tree type; 3) using pre-computed hints(nxPrepareXXX arrays) to know which transformations are going to 'prepare' for transforming particular Tree type; 4) recomputing those hints in case some transformation changed implementation class during 'prepare'; 5) TreeTransform is now responsible of calling transformFollowing on nodes created by it.
Tree transformer&TreeTransform
Remove flow typing from main branch
Backport "Apply implicit conversion from derived Conversion instance defined as implicit rather than given" to 3.3 LTS
To speedup discussion I propose my implementation