Hi John, I missed this in between. I had come across this on github, and stole the benchmarking module you were using at some point. I hadn't noticed it was on hackage. I will study it! What kept defeating me with these optimization schemes is that these libraries inevitably involve the user's writing recursive definitions themselves. I don't know how the fancy conduit system does if you put a user defined function in the pipeline.
On Wednesday, August 26, 2015 at 1:20:27 PM UTC-4, John Wiegley wrote: > > >>>>> Gabriel Gonzalez <gabri...@gmail.com <javascript:>> writes: > > > I'm really happy about this! I've been wanting a package like this for > some > > time, for two main reasons: > > This looks like a great library, Michael. I had started down a similar > road > with http://hackage.haskell.org/package/fusion, and working on a > companion > library pipes-fusion (not yet released), based on work done by Snoyman for > conduit. But I like the generality of Stream, and its resulting ability to > replace FreeT, while fusion's streams are more fixed to the Stream (Of a) > construction. Plus, I had not yet demonstrated that stream fusion still > occurs > for really large pipelines, where the inliner may become overwhelmed. > > Looking forward to further developments! > > John > -- You received this message because you are subscribed to the Google Groups "Haskell Pipes" group. To unsubscribe from this group and stop receiving emails from it, send an email to haskell-pipes+unsubscr...@googlegroups.com. To post to this group, send email to haskell-pipes@googlegroups.com.
