infix operator precedence
authorhaftmann
Tue Sep 20 08:20:22 2005 +0200 (2005-09-20)
changeset 17495ddb14cbec6a2
parent 17494 e70600834f44
child 17496 26535df536ae
infix operator precedence
NEWS
     1.1 --- a/NEWS	Tue Sep 20 00:16:29 2005 +0200
     1.2 +++ b/NEWS	Tue Sep 20 08:20:22 2005 +0200
     1.3 @@ -710,6 +710,16 @@
     1.4    x |> f                f #> g
     1.5    (x, y) |-> f          f #-> g
     1.6  
     1.7 +* Pure/library.ML: introduced/changed precedence of infix operators:
     1.8 +
     1.9 +  infix 1 |> |-> ||> ||>> |>> |>>> #> #->;
    1.10 +  infix 2 ?;
    1.11 +  infix 3 o oo ooo oooo;
    1.12 +  infix 4 ~~ upto downto;
    1.13 +
    1.14 +Maybe INCOMPATIBILITY when any of those is used in conjunction with other
    1.15 +infix operators.
    1.16 +
    1.17  * Pure/library.ML: natural list combinators fold, fold_rev, and
    1.18  fold_map support linear functional transformations and nesting.  For
    1.19  example:
    1.20 @@ -736,17 +746,14 @@
    1.21  
    1.22  * Pure/General: structure AList (cf. Pure/General/alist.ML) provides
    1.23  basic operations for association lists, following natural argument
    1.24 -order.  The old functions may be expressed as follows:
    1.25 +order; ; moreover the explicit equality predicate passed here avoids
    1.26 +potentially expensive polymorphic runtime equality checks.
    1.27 +The old functions may be expressed as follows:
    1.28  
    1.29    assoc = uncurry (AList.lookup (op =))
    1.30    assocs = these oo AList.lookup (op =)
    1.31    overwrite = uncurry (AList.update (op =)) o swap
    1.32  
    1.33 -* Pure/General: structure AList (cf. Pure/General/alist.ML) provides
    1.34 -basic operations for association lists, following natural argument
    1.35 -order; moreover the explicit equality predicate passed here avoids
    1.36 -potentially expensive polymorphic runtime equality checks.
    1.37 -
    1.38  * Pure/General: structure OrdList (cf. Pure/General/ord_list.ML)
    1.39  provides a reasonably efficient light-weight implementation of sets as
    1.40  lists.