agum 2.0 → 2.1

raw patch · 3 files changed

+19/−10 lines, 3 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

Files

ChangeLog +6/−0
agum.cabal +2/−2
src/Algebra/AbelianGroup/UnificationMatching.hs +11/−8

ChangeLog view

@@ -1,3 +1,9 @@+2009-09-05  John D. Ramsdell  <ramsdell@mitre.org>++	* src/Algebra/AbelianGroup/UnificationMatching.hs:  Added+	reference to Andrew Kennedy's Ph.D. thesis as in contains a proof+	of correctness of the implemented matching algorithm.+ 2009-08-29  John D. Ramsdell  <ramsdell@mitre.org>  	* src/Algebra/AbelianGroup/UnificationMatching.hs (Substitution,

agum.cabal view

@@ -1,5 +1,5 @@ Name:			agum-Version:		2.0+Version:		2.1 Maintainer:		ramsdell@mitre.org Cabal-Version:		>= 1.2 License:		GPL@@ -14,7 +14,7 @@ 			Substitition s is more general than s\' if 			there is a substitition s\" such that s\' = 			s\" o s.-Category:		Algebra.AbelianGroup+Category:		Algebra Build-Type:		Simple Extra-Source-Files:	readme.txt ChangeLog Makefile

src/Algebra/AbelianGroup/UnificationMatching.hs view

@@ -308,16 +308,13 @@ -- --     c[0]*x[0] + c[1]*x[1] + ... + c[n-1]*x[n-1] = 0 ----- where n is the length of c.  Otherwise, (c, d) represents a--- sequence of inhomogeneous linear equations with the same--- left-hand-side:+-- where n is the length of c.  Otherwise, (c, d) represents the+-- inhomogeneous equation: -----     c[0]*x[0] + c[1]*x[1] + ... + c[n-1]*x[n-1] = d[0]---     c[0]*x[0] + c[1]*x[1] + ... + c[n-1]*x[n-1] = d[1]---     ...---     c[0]*x[0] + c[1]*x[1] + ... + c[n-1]*x[n-1] = d[m-1]+--     c[0]*x[0] + c[1]*x[1] + ... + c[n-1]*x[n-1] = g ----- where m is the length of d.+-- where g = gcd(d[0], d[1], ..., d[m-1]), and m is the length of d.+-- Thus g is the greatest common denominator of the elements of d.  type Subst = [(Int, LinEq)] @@ -431,6 +428,12 @@ divide :: Int -> [Int] -> [Int] divide small t =     map (\x -> div x small) t++-- Elementary Abelian group matching is equivalent to unification with+-- constants.  A proof of correctness of this algorithm, cast as+-- unification with constants, is in Chapter 3, Section 1 of+-- "Programming Languages and Dimensions", Andrew Kennedy's+-- Ph.D. thesis from St. Catharine's College in 1996.  -- Input and Output