dovecot-2.2: lib: Added uni_utf8_partial_strlen_n()

[dovecot at dovecot.org]

details:   http://hg.dovecot.org/dovecot-2.2/rev/f191dbcaec5f
changeset: 18145:f191dbcaec5f
user:      Timo Sirainen <tss at iki.fi>
date:      Sat Jan 10 04:30:40 2015 +0200
description:
lib: Added uni_utf8_partial_strlen_n()

diffstat:

 src/lib/test-unichar.c |  16 ++++++++++++++++
 src/lib/unichar.c      |  18 ++++++++++++++----
 src/lib/unichar.h      |   6 ++++++
 3 files changed, 36 insertions(+), 4 deletions(-)

diffs (83 lines):

diff -r 7459c0891a85 -r f191dbcaec5f src/lib/test-unichar.c
--- a/src/lib/test-unichar.c	Sat Jan 10 04:25:21 2015 +0200
+++ b/src/lib/test-unichar.c	Sat Jan 10 04:30:40 2015 +0200
@@ -5,6 +5,20 @@
 #include "buffer.h"
 #include "unichar.h"
 
+static void test_unichar_uni_utf8_partial_strlen_n(void)
+{
+	static const char input[] = "\xC3\xA4\xC3\xA4";
+	size_t pos;
+
+	test_begin("uni_utf8_partial_strlen_n()");
+	test_assert(uni_utf8_partial_strlen_n(input, 1, &pos) == 0 && pos == 0);
+	test_assert(uni_utf8_partial_strlen_n(input, 2, &pos) == 1 && pos == 2);
+	test_assert(uni_utf8_partial_strlen_n(input, 3, &pos) == 1 && pos == 2);
+	test_assert(uni_utf8_partial_strlen_n(input, 4, &pos) == 2 && pos == 4);
+	test_assert(uni_utf8_partial_strlen_n(input, (size_t)-1, &pos) == 2 && pos == 4);
+	test_end();
+}
+
 void test_unichar(void)
 {
 	static const char overlong_utf8[] = "\xf8\x80\x95\x81\xa1";
@@ -32,4 +46,6 @@
 	test_assert(!uni_utf8_str_is_valid(overlong_utf8));
 	test_assert(uni_utf8_get_char(overlong_utf8, &chr2) < 0);
 	test_end();
+
+	test_unichar_uni_utf8_partial_strlen_n();
 }
diff -r 7459c0891a85 -r f191dbcaec5f src/lib/unichar.c
--- a/src/lib/unichar.c	Sat Jan 10 04:25:21 2015 +0200
+++ b/src/lib/unichar.c	Sat Jan 10 04:30:40 2015 +0200
@@ -192,18 +192,28 @@
 	return uni_utf8_strlen_n(input, (size_t)-1);
 }
 
-unsigned int uni_utf8_strlen_n(const void *_input, size_t size)
+unsigned int uni_utf8_strlen_n(const void *input, size_t size)
+{
+	size_t partial_pos;
+
+	return uni_utf8_partial_strlen_n(input, size, &partial_pos);
+}
+
+unsigned int uni_utf8_partial_strlen_n(const void *_input, size_t size,
+				       size_t *partial_pos_r)
 {
 	const unsigned char *input = _input;
-	unsigned int len = 0;
+	unsigned int count, len = 0;
 	size_t i;
 
 	for (i = 0; i < size && input[i] != '\0'; ) {
-		i += uni_utf8_char_bytes(input[i]);
-		if (i > size)
+		count = uni_utf8_char_bytes(input[i]);
+		if (i + count > size)
 			break;
+		i += count;
 		len++;
 	}
+	*partial_pos_r = i;
 	return len;
 }
 
diff -r 7459c0891a85 -r f191dbcaec5f src/lib/unichar.h
--- a/src/lib/unichar.h	Sat Jan 10 04:25:21 2015 +0200
+++ b/src/lib/unichar.h	Sat Jan 10 04:30:40 2015 +0200
@@ -55,6 +55,12 @@
 unsigned int uni_utf8_strlen(const char *input) ATTR_PURE;
 /* Returns UTF-8 string length with maximum input size. */
 unsigned int uni_utf8_strlen_n(const void *input, size_t size) ATTR_PURE;
+/* Same as uni_utf8_strlen_n(), but if input ends with a partial UTF-8
+   character, don't include it in the return value and set partial_pos_r to
+   where the character begins. Otherwise partial_pos_r is set to the end
+   of the input. */
+unsigned int uni_utf8_partial_strlen_n(const void *input, size_t size,
+				       size_t *partial_pos_r);
 
 /* Returns the number of bytes belonging to this UTF-8 character. The given
    parameter is the first byte of the UTF-8 sequence. Invalid input is