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

*To*: srfi-32@xxxxxxxxxxxxxxxxx*Subject*: a typo; min/max; partial orders*From*: Doug Orleans <dougo@xxxxxxxxxxx>*Date*: Thu, 12 Dec 2002 17:55:54 -0500 (EST)*Delivered-to*: srfi-32@xxxxxxxxxxxxxxxxx

Hi, I don't know if discussion is still allowed for SRFI 32, but it's still listed as being in draft mode on the SRFI web site so I figured I'd post. First of all, there's a typo in the draft posted at http://srfi.schemers.org/srfi-32/srfi-32.txt : This is due to the definition of equality, given a < comparator: (and (not (< x y)) (not (< y x))) The definition is rather different, given a <= comparator: (and (<= x y) (<= x y)) The last line should read (<= y x)) Now, my requests: it would be nice to have generalized versions of the R5RS `min' and `max' functions, that take a list or vector and a comparison operator. Even more generally, they could take a count N, and return the N least or greatest elements of the list or vector. Would it be reasonable to put these in SRFI 32, or do they belong somewhere else? What I *actually* need right now is a function to compute the list of minima of a list given a partial order. Topological sort and order-preserving insertion would be nice too. Olin Shivers wrote, in this forum, "worrying about partial orders is a snare & a delusion". Does this mean that partial-order algorithms belongs in a separate SRFI? --dougo@xxxxxxxxxxx

- Prev by Date:
**3-way compares and short-circuit sorting** - Next by Date:
**Brief (and late) Comments** - Previous by thread:
**3-way compares and short-circuit sorting** - Next by thread:
**Brief (and late) Comments** - Index(es):