ZTransducer should provide a way to split on Chunk[A] #3913 #3914
Conversation
|
|
9b177a2 to
8db603a
Compare
7c836aa to
be1cb10
Compare
98323c9 to
e5d80d6
Compare
There was a problem hiding this comment.
Hi @TobiasPfeifer, sorry for the delay here. The changes look good and the coverage is fantastic. Left one question regarding the signature.
| /** | ||
| * Splits elements on a delimiter and transforms the splits into desired output. | ||
| */ | ||
| def splitOnChunk[A, O: ClassTag](delimiter: Chunk[A], fn: Chunk[A] => O): ZTransducer[Any, Nothing, A, O] = |
There was a problem hiding this comment.
Not sure we need fn here? That's easy enough for the user to apply with ZTransducer#map, no?
There was a problem hiding this comment.
Type signature would then be ZTransducer[Any, Nothing, A, Chunk[A]] that results in a ZStream[R, E, Chunk[Chunk[A]]]
Flattening the Chunk[Chunk[A]] would just reverse the split. So mapping the nested Chunk[A] is the only useful continuation I can think of.
Is there another useful thing to do with Chunk[Chunk[A]] than mapping it to some Chunk[O]?
ZTransducer.splitOn does string concatenation (=mapping) as well to not produce Chunk[Chunk[String]] that has to be mapped to Chunk[String]
There was a problem hiding this comment.
The resulting stream would be ZStream[R, E, Chunk[A]], not Chunk[Chunk[A]] - applying flattenChunks would not reverse the split (because the delimiter is discarded) but rather provide a stream of all the individual elements.
Does that make sense?
There was a problem hiding this comment.
you are right. I was confused when working on a ZStream[R, E, Chunk[A]], but need to call flattenChunks in this case before splitting anyways.
changes applied already and added another test for consecutive delimiters
|
Thank you @TobiasPfeifer! |
#3913