From 96245a828aa9c717b847812812546667e090ee3f Mon Sep 17 00:00:00 2001
From: Ben Sima <ben@bsima.me>
Date: Tue, 19 May 2020 17:25:49 -0700
Subject: Add more re functions and docs

---
 bs/re.scm | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 51 insertions(+), 7 deletions(-)

(limited to 'bs')

diff --git a/bs/re.scm b/bs/re.scm
index 6ecc074..e2bc276 100644
--- a/bs/re.scm
+++ b/bs/re.scm
@@ -1,11 +1,55 @@
 #!r6rs
 (library (bs re (0))
-  (export match group)
+
+  ;; Similar to Python's `re` module. One big difference is that the
+  ;; string object to be searched is always the first argument to a
+  ;; function. This allows us to pipe a string through multiple
+  ;; transformations. For example:
+  ;;
+  ;; (-> " string "
+  ;;     string.strip
+  ;;     (re.search "in")
+  ;;     (re.group 1))
+  ;;
+  ;; TODO: consider putting the string as the last argument, to fit with
+  ;; other container types like `list`, then use ->>
+  ;; TODO: port srfi-115 to guile
+  ;; TODO: make `match` and `search` do different things
+
+  (export match group sub search
+          I IGNORECASE M MULTILINE)
   (import (rnrs base (6))
           (ice-9 regex))
-  (define (match s pat)
-    (string-match pat s))
-  (define (group m n)
-    (if m
-        (match:substring m n)
-        #f)))
+
+  (define I regexp/icase)
+  (define IGNORECASE regexp/icase)
+  (define M regexp/newline)
+  (define MULTILINE regexp/newline)
+
+  ;; Compile `pattern` into a regular expression object.
+  (define (compile pattern . flags)
+    (apply make-regexp (cons pattern flags)))
+
+  ;; If zero or more characters at the beginning of `string` match the
+  ;; regular expression `pattern`, return a corresponding match object.
+  (define (match string pattern)
+    (regexp-exec pattern string))
+
+  ;;
+  (define (group match-obj n)
+    (if match-obj
+        (match:substring match-obj n)
+        #f))
+
+  ;; Return the string obtained by replacing the leftmost
+  ;; non-overlapping occurrences of `pattern` in `string` by the
+  ;; replacement `repl`. If the pattern isn’t found, string is returned
+  ;; unchanged.
+  (define (sub string pattern repl)
+    (regexp-replace pattern string repl))
+
+  ;; Scan through `string` looking for the first location where the
+  ;; regular expression `pattern` produces a match, and return a
+  ;; corresponding match object. Returns `#f` if no match is found.
+  (define (search string pattern)
+    (regexp-exec pattern string)))
-- 
cgit v1.2.3