[PATCH 4/6] lib/charset: UTF-8 stream conversion

[Heinrich Schuchardt]

Provide functions to convert an UTF-8 stream to code page 437 or UTF-32.

Add unit tests.

Signed-off-by: Heinrich Schuchardt <xypron.glpk at gmx.de>
---
 include/charset.h | 18 +++++++++++
 lib/charset.c     | 55 +++++++++++++++++++++++++++------
 test/unicode_ut.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 142 insertions(+), 9 deletions(-)

diff --git a/include/charset.h b/include/charset.h
index 52e7d1474e..a911160f19 100644
--- a/include/charset.h
+++ b/include/charset.h
@@ -286,4 +286,22 @@ uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size);
  */
 int utf_to_cp(s32 *c, const u16 *codepage);

+/**
+ * utf8_to_cp437_stream() - convert UTF-8 stream to codepage 437
+ *
+ * @c:		next UTF-8 character to convert
+ * @buffer:	buffer, at least 5 characters
+ * Return:	next codepage 437 character or 0
+ */
+int utf8_to_cp437_stream(u8 c, char *buffer);
+
+/**
+ * utf8_to_utf32_stream() - convert UTF-8 stream to UTF-32
+ *
+ * @c:		next UTF-8 character to convert
+ * @buffer:	buffer, at least 5 characters
+ * Return:	next codepage 437 character or 0
+ */
+int utf8_to_utf32_stream(u8 c, char *buffer);
+
 #endif /* __CHARSET_H_ */
diff --git a/lib/charset.c b/lib/charset.c
index 946d5ee23e..f44c58d9d8 100644
--- a/lib/charset.c
+++ b/lib/charset.c
@@ -481,15 +481,6 @@ uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size)
 	return dest;
 }

-/**
- * utf_to_cp() - translate Unicode code point to 8bit codepage
- *
- * Codepoints that do not exist in the codepage are rendered as question mark.
- *
- * @c:		pointer to Unicode code point to be translated
- * @codepage:	Unicode to codepage translation table
- * Return:	0 on success, -ENOENT if codepoint cannot be translated
- */
 int utf_to_cp(s32 *c, const u16 *codepage)
 {
 	if (*c >= 0x80) {
@@ -507,3 +498,49 @@ int utf_to_cp(s32 *c, const u16 *codepage)
 	}
 	return 0;
 }
+
+int utf8_to_cp437_stream(u8 c, char *buffer)
+{
+	char *end;
+	const char *pos;
+	s32 s;
+	int ret;
+
+	for (;;) {
+		pos = buffer;
+		end = buffer + strlen(buffer);
+		*end++ = c;
+		*end = 0;
+		s = utf8_get(&pos);
+		if (s > 0) {
+			*buffer = 0;
+			ret = utf_to_cp(&s, codepage_437);
+			return s;
+			}
+		if (pos == end)
+			return 0;
+		*buffer = 0;
+	}
+}
+
+int utf8_to_utf32_stream(u8 c, char *buffer)
+{
+	char *end;
+	const char *pos;
+	s32 s;
+
+	for (;;) {
+		pos = buffer;
+		end = buffer + strlen(buffer);
+		*end++ = c;
+		*end = 0;
+		s = utf8_get(&pos);
+		if (s > 0) {
+			*buffer = 0;
+			return s;
+		}
+		if (pos == end)
+			return 0;
+		*buffer = 0;
+	}
+}
diff --git a/test/unicode_ut.c b/test/unicode_ut.c
index 154361aea7..6f6aea5f60 100644
--- a/test/unicode_ut.c
+++ b/test/unicode_ut.c
@@ -47,6 +47,9 @@ static const char d3[] = {0xe6, 0xbd, 0x9c, 0xe6, 0xb0, 0xb4, 0xe8, 0x89,
 /* Three letters translating to two utf-16 word each */
 static const char d4[] = {0xf0, 0x90, 0x92, 0x8d, 0xf0, 0x90, 0x92, 0x96,
 			  0xf0, 0x90, 0x92, 0x87, 0x00};
+/* Letter not in code page 437 */
+static const char d5[] = {0xCE, 0x92, 0x20, 0x69, 0x73, 0x20, 0x6E, 0x6F,
+			  0x74, 0x20, 0x42, 0x00};

 /* Illegal utf-8 strings */
 static const char j1[] = {0x6a, 0x31, 0xa1, 0x6c, 0x00};
@@ -631,6 +634,81 @@ static int unicode_test_utf_to_cp(struct unit_test_state *uts)
 }
 UNICODE_TEST(unicode_test_utf_to_cp);

+static void utf8_to_cp437_stream_helper(const char *in, char *out)
+{
+	char buffer[5];
+	int ret;
+
+	*buffer = 0;
+	for (; *in; ++in) {
+		ret = utf8_to_cp437_stream(*in, buffer);
+		if (ret)
+			*out++ = ret;
+	}
+	*out = 0;
+}
+
+static int unicode_test_utf8_to_cp437_stream(struct unit_test_state *uts)
+{
+	char buf[16];
+
+	utf8_to_cp437_stream_helper(d1, buf);
+	ut_asserteq_str("U-Boot", buf);
+	utf8_to_cp437_stream_helper(d2, buf);
+	ut_asserteq_str("kafb\xa0tur", buf);
+	utf8_to_cp437_stream_helper(d5, buf);
+	ut_asserteq_str("? is not B", buf);
+	utf8_to_cp437_stream_helper(j2, buf);
+	ut_asserteq_str("j2l", buf);
+
+	return 0;
+}
+UNICODE_TEST(unicode_test_utf8_to_cp437_stream);
+
+static void utf8_to_utf32_stream_helper(const char *in, s32 *out)
+{
+	char buffer[5];
+	int ret;
+
+	*buffer = 0;
+	for (; *in; ++in) {
+		ret = utf8_to_utf32_stream(*in, buffer);
+		if (ret)
+			*out++ = ret;
+	}
+	*out = 0;
+}
+
+static int unicode_test_utf8_to_utf32_stream(struct unit_test_state *uts)
+{
+	s32 buf[16];
+
+	const u32 u1[] = {0x55, 0x2D, 0x42, 0x6F, 0x6F, 0x74, 0x0000};
+	const u32 u2[] = {0x6B, 0x61, 0x66, 0x62, 0xE1, 0x74, 0x75, 0x72, 0x00};
+	const u32 u3[] = {0x0392, 0x20, 0x69, 0x73, 0x20, 0x6E, 0x6F, 0x74,
+			  0x20, 0x42, 0x00};
+	const u32 u4[] = {0x6A, 0x32, 0x6C, 0x00};
+
+	memset(buf, 0, sizeof(buf));
+	utf8_to_utf32_stream_helper(d1, buf);
+	ut_asserteq_mem(u1, buf, sizeof(u1));
+
+	memset(buf, 0, sizeof(buf));
+	utf8_to_utf32_stream_helper(d2, buf);
+	ut_asserteq_mem(u2, buf, sizeof(u2));
+
+	memset(buf, 0, sizeof(buf));
+	utf8_to_utf32_stream_helper(d5, buf);
+	ut_asserteq_mem(u3, buf, sizeof(u3));
+
+	memset(buf, 0, sizeof(buf));
+	utf8_to_utf32_stream_helper(j2, buf);
+	ut_asserteq_mem(u4, buf, sizeof(u4));
+
+	return 0;
+}
+UNICODE_TEST(unicode_test_utf8_to_utf32_stream);
+
 #ifdef CONFIG_EFI_LOADER
 static int unicode_test_efi_create_indexed_name(struct unit_test_state *uts)
 {
--
2.30.0