From bc6b0bee8fe4120642029daaa8ce6c069ef667b8 Mon Sep 17 00:00:00 2001
From: Benjamin Barenblat <bbaren@mit.edu>
Date: Wed, 26 Aug 2015 17:29:32 -0400
Subject: Rework replacement API to rely on transformation

Redesign library API around highly general regex-based transformations.
Instead of specifying a string to substitute for each match, you
now execute an entire function over the match (and over nonmatching
regions as well).  The resulting C++ code is much simpler, with more
functionality pushed into Ur, and the engine now supports certain
types of regex transformations needed to mimic Perl.
---
 src/regex.urs | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 55 insertions(+), 8 deletions(-)

(limited to 'src/regex.urs')

diff --git a/src/regex.urs b/src/regex.urs
index bd7696b..6e7bdcc 100644
--- a/src/regex.urs
+++ b/src/regex.urs
@@ -16,18 +16,65 @@ specific language governing permissions and limitations under the License. *)
 
 This library implements ECMAScript regular expressions. *)
 
+type substring = {Start : int, Len : int}
+type match = {Whole : substring, Groups : list substring}
+
 (* Searching *)
 
-(* Matches a regular expression against any part of a string.  Returns 'Some
-strs', where 'strs' is a list of subexpression matches, if a match succeeds, and
-'None' otherwise. *)
+(* Matches a regular expression against any part of a string.  Returns
+'Some match' if a match succeeds and 'None' otherwise. *)
 val match : string (* needle *)
-         -> string (* haystack *)
-         -> option (list string)
+            -> string (* haystack *)
+            -> option match
+
+(* Finds _all_ matches for a regular expression in a string. *)
+val all_matches : string (* needle *)
+                  -> string (* haystack *)
+                  -> list match
+
+(* Replacement *)
 
 (* Replaces all substrings in 'haystack' that match 'needle' with the string
 'replacement.' *)
 val replace : string (* needle *)
-           -> string (* replacement *)
-           -> string (* haystack *)
-           -> string
+              -> string (* replacement *)
+              -> string (* haystack *)
+              -> string
+
+(* Transforms a string by applying a function to replace every match in the
+string. *)
+val transform_matches : string (* needle *)
+                        -> (match -> string) (* transformation *)
+                        -> string (* haystack *)
+                        -> string
+
+(* Executes a general regex-guided transformation over a string.  Matches
+'needle' against any part of 'haystack', splitting 'haystack' into matching and
+nonmatching regions.  Then, runs the provided transformation functions over the
+regions and concatenates the results.
+
+The number of nonmatching regions is always exactly one more than the number of
+matching regions.  If two matching regions abut or a matching region adjoins the
+edge of a string, this function will insert an empty nonmatching region as
+appropriate.
+
+An example may make this a bit clearer:
+
+    let
+      val haystack "axbxax"
+    in
+      transform "x"
+                (fn nm => "_" ^ String.substring haystack nm ^ "_")
+                (fn  m => "*" ^ String.substring haystack  m ^ "_")
+                haystack
+    end
+
+evaluates to
+
+    "_a_*x*_b_*x*__"
+*)
+val transform : string (* needle *)
+                -> (substring -> string) (* non-matching transformation *)
+                -> (match -> string) (* matching transformation *)
+                -> string (* haystack *)
+                -> string
-- 
cgit v1.2.3