[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Guile-commits] 01/01: Allow decoding of UTF-8 containing U+FFFD, the re
From: |
Mark H. Weaver |
Subject: |
[Guile-commits] 01/01: Allow decoding of UTF-8 containing U+FFFD, the replacement character. |
Date: |
Thu, 13 Aug 2015 05:44:15 +0000 |
mhw pushed a commit to branch master
in repository guile.
commit 00884bb79fff41fdf5f22f24a74e366a94a14c9b
Author: Mark H Weaver <address@hidden>
Date: Thu Aug 13 01:31:36 2015 -0400
Allow decoding of UTF-8 containing U+FFFD, the replacement character.
* libguile/strings.c (scm_from_utf8_stringn): Use 'u8_mbtoucr' and check
for a decoding error by its 'nbytes' return value. Previously we used
'u8_mbtouc' and improperly assumed that a U+FFFD character indicated a
decoding error.
* libguile/symbols.c (utf8_string_equals_wide_string): Likewise.
* test-suite/tests/bytevectors.test (exception:decoding-error): New
variable.
("2.9 Operations on Strings"): Add tests.
---
libguile/strings.c | 7 ++++---
libguile/symbols.c | 8 ++++----
test-suite/tests/bytevectors.test | 14 +++++++++++++-
3 files changed, 21 insertions(+), 8 deletions(-)
diff --git a/libguile/strings.c b/libguile/strings.c
index 90dc83a..ee43e81 100644
--- a/libguile/strings.c
+++ b/libguile/strings.c
@@ -1,4 +1,5 @@
-/* Copyright (C) 1995,1996,1998,2000,2001, 2004, 2006, 2008, 2009, 2010, 2011,
2012, 2013, 2014 Free Software Foundation, Inc.
+/* Copyright (C) 1995, 1996, 1998, 2000, 2001, 2004, 2006,
+ * 2008-2015 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
@@ -1673,9 +1674,9 @@ scm_from_utf8_stringn (const char *str, size_t len)
ascii = 0;
- nbytes = u8_mbtouc (&c, ustr + i, len - i);
+ nbytes = u8_mbtoucr (&c, ustr + i, len - i);
- if (c == 0xfffd)
+ if (nbytes < 0)
/* Bad UTF-8. */
decoding_error (__func__, errno, str, len);
diff --git a/libguile/symbols.c b/libguile/symbols.c
index f93833b..71d9827 100644
--- a/libguile/symbols.c
+++ b/libguile/symbols.c
@@ -1,5 +1,5 @@
-/* Copyright (C) 1995, 1996, 1997, 1998, 2000, 2001, 2003, 2004,
- * 2006, 2009, 2011, 2013 Free Software Foundation, Inc.
+/* Copyright (C) 1995-1998, 2000, 2001, 2003, 2004, 2006, 2009, 2011,
+ * 2013, 2015 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
@@ -164,10 +164,10 @@ utf8_string_equals_wide_string (const scm_t_uint8
*narrow, size_t nlen,
ucs4_t c;
int nbytes;
- nbytes = u8_mbtouc (&c, narrow + byte_idx, nlen - byte_idx);
+ nbytes = u8_mbtoucr (&c, narrow + byte_idx, nlen - byte_idx);
if (nbytes == 0)
break;
- else if (c == 0xfffd)
+ else if (nbytes < 0)
/* Bad UTF-8. */
return 0;
else if (c != wide[char_idx])
diff --git a/test-suite/tests/bytevectors.test
b/test-suite/tests/bytevectors.test
index 4cc5b67..f8f020a 100644
--- a/test-suite/tests/bytevectors.test
+++ b/test-suite/tests/bytevectors.test
@@ -1,6 +1,6 @@
;;;; bytevectors.test --- R6RS bytevectors. -*- mode: scheme; coding: utf-8;
-*-
;;;;
-;;;; Copyright (C) 2009-2014 Free Software Foundation, Inc.
+;;;; Copyright (C) 2009-2015 Free Software Foundation, Inc.
;;;;
;;;; Ludovic Courtès
;;;;
@@ -24,6 +24,9 @@
:use-module (rnrs bytevectors)
:use-module (srfi srfi-4))
+(define exception:decoding-error
+ (cons 'decoding-error "input (locale conversion|decoding) error"))
+
;;; Some of the tests in here are examples taken from the R6RS Standard
;;; Libraries document.
@@ -501,6 +504,15 @@
(= (string-length str)
(- (bytevector-length utf8) 2)))))
+ (pass-if-equal "utf8->string [replacement character]"
+ '(104 105 65533)
+ (map char->integer
+ (string->list (utf8->string #vu8(104 105 239 191 189)))))
+
+ (pass-if-exception "utf8->string [invalid encoding]"
+ exception:decoding-error
+ (utf8->string #vu8(104 105 239 191 50)))
+
(pass-if "utf16->string"
(let* ((utf16 (uint-list->bytevector (map char->integer
(string->list "hello, world"))