implement r6rs hungry escaped EOL

* libguile/private-options.h (SCM_HUNGRY_EOL_ESCAPES_P): New private
  option.
* libguile/read.c: Define SCM_HUNGRY_EOL_ESCAPES_P, defaulting to #f.
  (skip_intraline_whitespace): New helper.
  (scm_read_string): If SCM_HUNGRY_EOL_ESCAPES_P,
  skip_intraline_whitespace after an escaped EOL.

* test-suite/tests/reader.test ("read-options"): Add test.

libguile/private-options.h

@@ -4,7 +4,7 @@
  * We put this in a private header, since layout of data structures
  * is an implementation detail that we want to hide.
  * 
- * Copyright (C) 2007, 2009, 2010 Free Software Foundation, Inc.
+ * Copyright (C) 2007, 2009, 2010, 2011 Free Software Foundation, Inc.
  * 
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public License
@@ -63,6 +63,7 @@ SCM_INTERNAL scm_t_option scm_read_opts[];
 #define SCM_KEYWORD_STYLE      scm_read_opts[3].val
 #define SCM_R6RS_ESCAPES_P     scm_read_opts[4].val
 #define SCM_SQUARE_BRACKETS_P  scm_read_opts[5].val
+#define SCM_HUNGRY_EOL_ESCAPES_P scm_read_opts[6].val
 
 #define SCM_N_READ_OPTIONS 6
 

libguile/read.c

@@ -29,6 +29,7 @@
 #include <string.h>
 #include <unistd.h>
 #include <unicase.h>
+#include <unictype.h>
 
 #include "libguile/_scm.h"
 #include "libguile/bytevectors.h"
@@ -75,6 +76,8 @@ scm_t_option scm_read_opts[] = {
     "Use R6RS variable-length character and string hex escapes."},
   { SCM_OPTION_BOOLEAN, "square-brackets", 1,
     "Treat `[' and `]' as parentheses, for R6RS compatibility."},
+  { SCM_OPTION_BOOLEAN, "hungry-eol-escapes", 0,
+    "In strings, consume leading whitespace after an escaped end-of-line."},
   { 0, },
 };
 
@@ -486,6 +489,22 @@ scm_read_sexp (scm_t_wchar chr, SCM port)
         }                                                          \
     } while (0)
 
+static void
+skip_intraline_whitespace (SCM port)
+{
+  scm_t_wchar c;
+  
+  do
+    {
+      c = scm_getc (port);
+      if (c == EOF)
+        return;
+    }
+  while (c == '\t' || uc_is_general_category (c, UC_SPACE_SEPARATOR));
+
+  scm_ungetc (c, port);
+}                                         
+
 static SCM
 scm_read_string (int chr, SCM port)
 #define FUNC_NAME "scm_lreadr"
@@ -524,6 +543,8 @@ scm_read_string (int chr, SCM port)
             case '\\':
               break;
             case '\n':
+              if (SCM_HUNGRY_EOL_ESCAPES_P)
+                skip_intraline_whitespace (port);
               continue;
             case '0':
               c = '\0';

test-suite/tests/reader.test

@@ -1,6 +1,6 @@
 ;;;; reader.test --- Reader test.    -*- coding: iso-8859-1; mode: scheme -*-
 ;;;;
-;;;; Copyright (C) 1999, 2001, 2002, 2003, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
+;;;; Copyright (C) 1999, 2001, 2002, 2003, 2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc.
 ;;;; Jim Blandy <jimb@red-bean.com>
 ;;;;
 ;;;; This library is free software; you can redistribute it and/or
@@ -364,8 +364,24 @@
            (with-output-to-string
              (lambda ()
                (write (integer->char #x80))))))
-       "#\\x80"))))
+       "#\\x80")))
 
+  (with-test-prefix "hungry escapes"
+    (pass-if "default not hungry"
+      ;; Assume default setting of not hungry.
+      (equal? (with-input-from-string "\"foo\\\n  bar\""
+                read)
+              "foo  bar"))
+    (pass-if "hungry"
+      (dynamic-wind
+        (lambda ()
+          (read-enable 'hungry-eol-escapes))
+        (lambda ()
+          (equal? (with-input-from-string "\"foo\\\n  bar\""
+                    read)
+                  "foobar"))
+        (lambda ()
+          (read-disable 'hungry-eol-escapes))))))
 
 
 (with-test-prefix "#;"