aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorTom Rini <trini@konsulko.com>2021-03-07 19:43:00 -0500
committerTom Rini <trini@konsulko.com>2021-03-07 19:43:00 -0500
commit90964ab5acb29ec6617a9ff340886230b2fce8c7 (patch)
treefab87d3307e5c4902c07ff4ab630a9a709bc6daa
parente4dba4ba6f61e8128be0b4200ca2d8cebf62180b (diff)
parent7d3eff3412886f277c724a9effcbe545c4cdd5b5 (diff)
downloadu-boot-WIP/07Mar2021.zip
u-boot-WIP/07Mar2021.tar.gz
u-boot-WIP/07Mar2021.tar.bz2
Merge tag 'efi-2021-04-rc3-3' of https://source.denx.de/u-boot/custodians/u-boot-efiWIP/07Mar2021
Pull request for efi-2021-04-rc3-3 New: * Provide library functions for converting UTF-8 streams either to code page 437 or Unicode code points. Bug fixes: * Fix the capsule update unit tests. * Use the terminal size of the video console if it is the primary output.
-rw-r--r--include/charset.h34
-rw-r--r--lib/charset.c96
-rw-r--r--lib/efi_loader/efi_console.c2
-rw-r--r--lib/efi_loader/efi_unicode_collation.c21
-rw-r--r--test/py/tests/test_efi_capsule/uboot_bin_env.its4
-rw-r--r--test/unicode_ut.c114
6 files changed, 242 insertions, 29 deletions
diff --git a/include/charset.h b/include/charset.h
index cc650a2..a911160 100644
--- a/include/charset.h
+++ b/include/charset.h
@@ -14,6 +14,11 @@
#define MAX_UTF8_PER_UTF16 3
/**
+ * codepage_437 - Unicode to codepage 437 translation table
+ */
+extern const u16 codepage_437[128];
+
+/**
* console_read_unicode() - read Unicode code point from console
*
* @code: pointer to store Unicode code point
@@ -270,4 +275,33 @@ u16 *u16_strdup(const void *src);
*/
uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size);
+/**
+ * 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);
+
+/**
+ * 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 2177014..f44c58d 100644
--- a/lib/charset.c
+++ b/lib/charset.c
@@ -8,9 +8,16 @@
#include <common.h>
#include <charset.h>
#include <capitalization.h>
+#include <cp437.h>
#include <efi_loader.h>
+#include <errno.h>
#include <malloc.h>
+/**
+ * codepage_437 - Unicode to codepage 437 translation table
+ */
+const u16 codepage_437[128] = CP437;
+
static struct capitalization_table capitalization_table[] =
#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
UNICODE_CAPITALIZATION_TABLE;
@@ -25,7 +32,7 @@ static struct capitalization_table capitalization_table[] =
*
* @read_u8: - stream reader
* @src: - string buffer passed to stream reader, optional
- * Return: - Unicode code point
+ * Return: - Unicode code point, or -1
*/
static int get_code(u8 (*read_u8)(void *data), void *data)
{
@@ -71,7 +78,7 @@ static int get_code(u8 (*read_u8)(void *data), void *data)
}
return ch;
error:
- return '?';
+ return -1;
}
/**
@@ -113,14 +120,21 @@ static u8 read_console(void *data)
int console_read_unicode(s32 *code)
{
- if (!tstc()) {
- /* No input available */
- return 1;
- }
+ for (;;) {
+ s32 c;
- /* Read Unicode code */
- *code = get_code(read_console, NULL);
- return 0;
+ if (!tstc()) {
+ /* No input available */
+ return 1;
+ }
+
+ /* Read Unicode code */
+ c = get_code(read_console, NULL);
+ if (c > 0) {
+ *code = c;
+ return 0;
+ }
+ }
}
s32 utf8_get(const char **src)
@@ -466,3 +480,67 @@ uint8_t *utf16_to_utf8(uint8_t *dest, const uint16_t *src, size_t size)
return dest;
}
+
+int utf_to_cp(s32 *c, const u16 *codepage)
+{
+ if (*c >= 0x80) {
+ int j;
+
+ /* Look up codepage translation */
+ for (j = 0; j < 0x80; ++j) {
+ if (*c == codepage[j]) {
+ *c = j + 0x80;
+ return 0;
+ }
+ }
+ *c = '?';
+ return -ENOENT;
+ }
+ 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/lib/efi_loader/efi_console.c b/lib/efi_loader/efi_console.c
index edcfce7..c400355 100644
--- a/lib/efi_loader/efi_console.c
+++ b/lib/efi_loader/efi_console.c
@@ -311,7 +311,7 @@ static void query_console_size(void)
const char *stdout_name = env_get("stdout");
int rows = 25, cols = 80;
- if (stdout_name && !strcmp(stdout_name, "vidconsole") &&
+ if (stdout_name && !strncmp(stdout_name, "vidconsole", 10) &&
IS_ENABLED(CONFIG_DM_VIDEO)) {
struct stdio_dev *stdout_dev =
stdio_get_by_name("vidconsole");
diff --git a/lib/efi_loader/efi_unicode_collation.c b/lib/efi_loader/efi_unicode_collation.c
index f6c875b..36be798 100644
--- a/lib/efi_loader/efi_unicode_collation.c
+++ b/lib/efi_loader/efi_unicode_collation.c
@@ -23,7 +23,7 @@ static const char illegal[] = "+,<=>:;\"/\\|?*[]\x7f";
static const u16 codepage[] = CP1250;
#else
/* Unicode code points for code page 437 characters 0x80 - 0xff */
-static const u16 codepage[] = CP437;
+static const u16 *codepage = codepage_437;
#endif
/* GUID of the EFI_UNICODE_COLLATION_PROTOCOL2 */
@@ -300,23 +300,10 @@ static bool EFIAPI efi_str_to_fat(struct efi_unicode_collation_protocol *this,
break;
}
c = utf_to_upper(c);
- if (c >= 0x80) {
- int j;
-
- /* Look for codepage translation */
- for (j = 0; j < 0x80; ++j) {
- if (c == codepage[j]) {
- c = j + 0x80;
- break;
- }
- }
- if (j >= 0x80) {
- c = '_';
- ret = true;
- }
- } else if (c && (c < 0x20 || strchr(illegal, c))) {
- c = '_';
+ if (utf_to_cp(&c, codepage) ||
+ (c && (c < 0x20 || strchr(illegal, c)))) {
ret = true;
+ c = '_';
}
fat[i] = c;
diff --git a/test/py/tests/test_efi_capsule/uboot_bin_env.its b/test/py/tests/test_efi_capsule/uboot_bin_env.its
index 31e2f80..fc65907 100644
--- a/test/py/tests/test_efi_capsule/uboot_bin_env.its
+++ b/test/py/tests/test_efi_capsule/uboot_bin_env.its
@@ -10,7 +10,7 @@
#address-cells = <2>;
images {
- u-boot-bin@100000 {
+ u-boot-bin {
description = "U-Boot binary on SPI Flash";
data = /incbin/("BINFILE1");
compression = "none";
@@ -21,7 +21,7 @@
algo = "sha1";
};
};
- u-boot-env@150000 {
+ u-boot-env {
description = "U-Boot environment on SPI Flash";
data = /incbin/("BINFILE2");
compression = "none";
diff --git a/test/unicode_ut.c b/test/unicode_ut.c
index 6130ef0..6f6aea5 100644
--- a/test/unicode_ut.c
+++ b/test/unicode_ut.c
@@ -47,11 +47,15 @@ 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};
static const char j2[] = {0x6a, 0x32, 0xc3, 0xc3, 0x6c, 0x00};
static const char j3[] = {0x6a, 0x33, 0xf0, 0x90, 0xf0, 0x00};
+static const char j4[] = {0xa1, 0x00};
static int unicode_test_u16_strlen(struct unit_test_state *uts)
{
@@ -165,6 +169,12 @@ static int unicode_test_utf8_get(struct unit_test_state *uts)
ut_asserteq(0x0001048d, code);
ut_asserteq_ptr(s, d4 + 4);
+ /* Check illegal character */
+ s = j4;
+ code = utf8_get((const char **)&s);
+ ut_asserteq(-1, code);
+ ut_asserteq_ptr(j4 + 1, s);
+
return 0;
}
UNICODE_TEST(unicode_test_utf8_get);
@@ -595,6 +605,110 @@ static int unicode_test_u16_strsize(struct unit_test_state *uts)
}
UNICODE_TEST(unicode_test_u16_strsize);
+static int unicode_test_utf_to_cp(struct unit_test_state *uts)
+{
+ int ret;
+ s32 c;
+
+ c = '\n';
+ ret = utf_to_cp(&c, codepage_437);
+ ut_asserteq(0, ret);
+ ut_asserteq('\n', c);
+
+ c = 'a';
+ ret = utf_to_cp(&c, codepage_437);
+ ut_asserteq(0, ret);
+ ut_asserteq('a', c);
+
+ c = 0x03c4; /* Greek small letter tau */
+ ret = utf_to_cp(&c, codepage_437);
+ ut_asserteq(0, ret);
+ ut_asserteq(0xe7, c);
+
+ c = 0x03a4; /* Greek capital letter tau */
+ ret = utf_to_cp(&c, codepage_437);
+ ut_asserteq(-ENOENT, ret);
+ ut_asserteq('?', c);
+
+ return 0;
+}
+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)
{