[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Chicken-users] Problems with srfi-4-comprehensions

From: Alex Shinn
Subject: Re: [Chicken-users] Problems with srfi-4-comprehensions
Date: Sat, 24 Mar 2007 15:50:04 +0900


On 3/24/07, Ivan Raikov <address@hidden> wrote:

  Thanks for adding these extensions to loopy-loop. I still haven't looked
at it, as I am still fighting with SRFI-42, but you have certainly made
the case for at least considering loopy-loop.

Technically, I cheated, I didn't add what you wanted in that
specific example (generating a uniform vector), just the ability
to iterate over uniform vectors.

I've so far mostly ignored accumulators other than the basic
list.  If you want to build a vector by accumulating a list and
then calling list->vector, that's fine, (and that's what SRFI-42
does) but if you want something efficient you need a means to
presize the vector, mutate the vector in place, and the user
needs to be aware that mutation ruins re-entrancy and multiple

Taylor Campbell made a variation of the macro at

which adds efficient vector accumulators and other features.  For
vectors you need to compute and pass the length ahead of time.

I've been thinking of easier to use vector and matrix operations,
but haven't come up with anything I like.  Everything I've seen
is either easy or efficient, but not both.  In particular, I'd
like to be able to write a natural looking matrix multiply
algorithm that is fast.

Are your vectors actually rows/columns of a matrix?  If you could
give a more concrete example of how you want to use this I may be
able to come up with something.

Oh, I finally added documentation for writing your own extensions


reply via email to

[Prev in Thread] Current Thread [Next in Thread]