Export CBS/CBB unicode functions
This way the Chromium certificate verifier can more easily use them.
Bug: chromium:1322914
Change-Id: I51dafc4e70d74da8543688b6457563d78e298150
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/62745
Auto-Submit: David Benjamin <davidben@google.com>
Reviewed-by: Bob Beck <bbe@google.com>
Commit-Queue: Bob Beck <bbe@google.com>
diff --git a/crypto/asn1/a_mbstr.c b/crypto/asn1/a_mbstr.c
index 8fc82ab..4d7ea14 100644
--- a/crypto/asn1/a_mbstr.c
+++ b/crypto/asn1/a_mbstr.c
@@ -97,22 +97,22 @@
int error;
switch (inform) {
case MBSTRING_BMP:
- decode_func = cbs_get_ucs2_be;
+ decode_func = CBS_get_ucs2_be;
error = ASN1_R_INVALID_BMPSTRING;
break;
case MBSTRING_UNIV:
- decode_func = cbs_get_utf32_be;
+ decode_func = CBS_get_utf32_be;
error = ASN1_R_INVALID_UNIVERSALSTRING;
break;
case MBSTRING_UTF8:
- decode_func = cbs_get_utf8;
+ decode_func = CBS_get_utf8;
error = ASN1_R_INVALID_UTF8STRING;
break;
case MBSTRING_ASC:
- decode_func = cbs_get_latin1;
+ decode_func = CBS_get_latin1;
error = ERR_R_INTERNAL_ERROR; // Latin-1 inputs are never invalid.
break;
@@ -162,7 +162,7 @@
}
nchar++;
- utf8_len += cbb_get_utf8_len(c);
+ utf8_len += CBB_get_utf8_len(c);
if (maxsize > 0 && nchar > (size_t)maxsize) {
OPENSSL_PUT_ERROR(ASN1, ASN1_R_STRING_TOO_LONG);
ERR_add_error_dataf("maxsize=%zu", (size_t)maxsize);
@@ -178,7 +178,7 @@
// Now work out output format and string type
int str_type;
- int (*encode_func)(CBB *, uint32_t) = cbb_add_latin1;
+ int (*encode_func)(CBB *, uint32_t) = CBB_add_latin1;
size_t size_estimate = nchar;
int outform = MBSTRING_ASC;
if (mask & B_ASN1_PRINTABLESTRING) {
@@ -190,17 +190,17 @@
} else if (mask & B_ASN1_BMPSTRING) {
str_type = V_ASN1_BMPSTRING;
outform = MBSTRING_BMP;
- encode_func = cbb_add_ucs2_be;
+ encode_func = CBB_add_ucs2_be;
size_estimate = 2 * nchar;
} else if (mask & B_ASN1_UNIVERSALSTRING) {
str_type = V_ASN1_UNIVERSALSTRING;
- encode_func = cbb_add_utf32_be;
+ encode_func = CBB_add_utf32_be;
size_estimate = 4 * nchar;
outform = MBSTRING_UNIV;
} else if (mask & B_ASN1_UTF8STRING) {
str_type = V_ASN1_UTF8STRING;
outform = MBSTRING_UTF8;
- encode_func = cbb_add_utf8;
+ encode_func = CBB_add_utf8;
size_estimate = utf8_len;
} else {
OPENSSL_PUT_ERROR(ASN1, ASN1_R_ILLEGAL_CHARACTERS);
diff --git a/crypto/asn1/a_strex.c b/crypto/asn1/a_strex.c
index dcc87f1..7e9afad 100644
--- a/crypto/asn1/a_strex.c
+++ b/crypto/asn1/a_strex.c
@@ -137,19 +137,19 @@
int get_char_error;
switch (encoding) {
case MBSTRING_UNIV:
- get_char = cbs_get_utf32_be;
+ get_char = CBS_get_utf32_be;
get_char_error = ASN1_R_INVALID_UNIVERSALSTRING;
break;
case MBSTRING_BMP:
- get_char = cbs_get_ucs2_be;
+ get_char = CBS_get_ucs2_be;
get_char_error = ASN1_R_INVALID_BMPSTRING;
break;
case MBSTRING_ASC:
- get_char = cbs_get_latin1;
+ get_char = CBS_get_latin1;
get_char_error = ERR_R_INTERNAL_ERROR; // Should not be possible.
break;
case MBSTRING_UTF8:
- get_char = cbs_get_utf8;
+ get_char = CBS_get_utf8;
get_char_error = ASN1_R_INVALID_UTF8STRING;
break;
default:
@@ -172,7 +172,7 @@
uint8_t utf8_buf[6];
CBB utf8_cbb;
CBB_init_fixed(&utf8_cbb, utf8_buf, sizeof(utf8_buf));
- if (!cbb_add_utf8(&utf8_cbb, c)) {
+ if (!CBB_add_utf8(&utf8_cbb, c)) {
OPENSSL_PUT_ERROR(ASN1, ERR_R_INTERNAL_ERROR);
return 1;
}
diff --git a/crypto/asn1/tasn_dec.c b/crypto/asn1/tasn_dec.c
index 24ab04f..94891c8 100644
--- a/crypto/asn1/tasn_dec.c
+++ b/crypto/asn1/tasn_dec.c
@@ -850,7 +850,7 @@
if (utype == V_ASN1_BMPSTRING) {
while (CBS_len(&cbs) != 0) {
uint32_t c;
- if (!cbs_get_ucs2_be(&cbs, &c)) {
+ if (!CBS_get_ucs2_be(&cbs, &c)) {
OPENSSL_PUT_ERROR(ASN1, ASN1_R_INVALID_BMPSTRING);
goto err;
}
@@ -859,7 +859,7 @@
if (utype == V_ASN1_UNIVERSALSTRING) {
while (CBS_len(&cbs) != 0) {
uint32_t c;
- if (!cbs_get_utf32_be(&cbs, &c)) {
+ if (!CBS_get_utf32_be(&cbs, &c)) {
OPENSSL_PUT_ERROR(ASN1, ASN1_R_INVALID_UNIVERSALSTRING);
goto err;
}
@@ -868,7 +868,7 @@
if (utype == V_ASN1_UTF8STRING) {
while (CBS_len(&cbs) != 0) {
uint32_t c;
- if (!cbs_get_utf8(&cbs, &c)) {
+ if (!CBS_get_utf8(&cbs, &c)) {
OPENSSL_PUT_ERROR(ASN1, ASN1_R_INVALID_UTF8STRING);
goto err;
}
diff --git a/crypto/bytestring/bytestring_test.cc b/crypto/bytestring/bytestring_test.cc
index 10d3469..08cfb87 100644
--- a/crypto/bytestring/bytestring_test.cc
+++ b/crypto/bytestring/bytestring_test.cc
@@ -1353,7 +1353,7 @@
std::vector<uint32_t> out;
bool ok;
} kTests[] = {
- {cbs_get_utf8, cbb_add_utf8,
+ {CBS_get_utf8, CBB_add_utf8,
// This test string captures all four cases in UTF-8.
LiteralToBytes(u8"Hello, 世界! ¡Hola, 🌎!"),
LiteralToCodePoints(U"Hello, 世界! ¡Hola, 🌎!"), true},
@@ -1362,120 +1362,120 @@
// http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt
// 2.1 First possible sequence of a certain length. (5- and 6-bit
// sequences no longer exist.)
- {cbs_get_utf8, cbb_add_utf8, {0xf8, 0x88, 0x80, 0x80, 0x80}, {}, false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8, CBB_add_utf8, {0xf8, 0x88, 0x80, 0x80, 0x80}, {}, false},
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xfc, 0x84, 0x80, 0x80, 0x80, 0x80},
{},
false},
// 3.1 Unexpected continuation bytes.
- {cbs_get_utf8, cbb_add_utf8, {0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xbf}, {}, false},
// 3.2 Lonely start characters.
- {cbs_get_utf8, cbb_add_utf8, {0xc0, ' '}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xe0, ' '}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, ' '}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xc0, ' '}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, ' '}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, ' '}, {}, false},
// 3.3 Sequences with last continuation byte missing
- {cbs_get_utf8, cbb_add_utf8, {0xc0}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xe0, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x80, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xc0}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x80, 0x80}, {}, false},
// Variation of the above with unexpected spaces.
- {cbs_get_utf8, cbb_add_utf8, {0xe0, 0x80, ' '}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x80, 0x80, ' '}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, 0x80, ' '}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x80, 0x80, ' '}, {}, false},
// 4.1 Examples of an overlong ASCII character
- {cbs_get_utf8, cbb_add_utf8, {0xc0, 0xaf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xe0, 0x80, 0xaf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x80, 0x80, 0xaf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xc0, 0xaf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, 0x80, 0xaf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x80, 0x80, 0xaf}, {}, false},
// 4.2 Maximum overlong sequences
- {cbs_get_utf8, cbb_add_utf8, {0xc1, 0xbf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xe0, 0x9f, 0xbf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x8f, 0xbf, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xc1, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, 0x9f, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x8f, 0xbf, 0xbf}, {}, false},
// 4.3 Overlong representation of the NUL character
- {cbs_get_utf8, cbb_add_utf8, {0xc0, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xe0, 0x80, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x80, 0x80, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xc0, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xe0, 0x80, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x80, 0x80, 0x80}, {}, false},
// 5.1 Single UTF-16 surrogates
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xa0, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xad, 0xbf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xae, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xb0, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xbe, 0x80}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xed, 0xbf, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xa0, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xad, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xae, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xb0, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xbe, 0x80}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xed, 0xbf, 0xbf}, {}, false},
// 5.2 Paired UTF-16 surrogates
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xa0, 0x80, 0xed, 0xb0, 0x80},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xa0, 0x80, 0xed, 0xbf, 0xbf},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xad, 0xbf, 0xed, 0xb0, 0x80},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xad, 0xbf, 0xed, 0xbf, 0xbf},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xae, 0x80, 0xed, 0xb0, 0x80},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xae, 0x80, 0xed, 0xbf, 0xbf},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xaf, 0xbf, 0xed, 0xb0, 0x80},
{},
false},
- {cbs_get_utf8,
- cbb_add_utf8,
+ {CBS_get_utf8,
+ CBB_add_utf8,
{0xed, 0xaf, 0xbf, 0xed, 0xbf, 0xbf},
{},
false},
// 5.3 Noncharacter code positions
- {cbs_get_utf8, cbb_add_utf8, {0xef, 0xbf, 0xbe}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xef, 0xbf, 0xbf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xef, 0xb7, 0x90}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xef, 0xb7, 0xaf}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x9f, 0xbf, 0xbe}, {}, false},
- {cbs_get_utf8, cbb_add_utf8, {0xf0, 0x9f, 0xbf, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xef, 0xbf, 0xbe}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xef, 0xbf, 0xbf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xef, 0xb7, 0x90}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xef, 0xb7, 0xaf}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x9f, 0xbf, 0xbe}, {}, false},
+ {CBS_get_utf8, CBB_add_utf8, {0xf0, 0x9f, 0xbf, 0xbf}, {}, false},
- {cbs_get_latin1, cbb_add_latin1, LiteralToBytes("\xa1Hola!"),
+ {CBS_get_latin1, CBB_add_latin1, LiteralToBytes("\xa1Hola!"),
LiteralToCodePoints(U"¡Hola!"), true},
// UCS-2 matches UTF-16 on the BMP.
- {cbs_get_ucs2_be, cbb_add_ucs2_be, LiteralToBytes(u"Hello, 世界!"),
+ {CBS_get_ucs2_be, CBB_add_ucs2_be, LiteralToBytes(u"Hello, 世界!"),
LiteralToCodePoints(U"Hello, 世界!"), true},
// It does not support characters beyond the BMP.
- {cbs_get_ucs2_be, cbb_add_ucs2_be,
+ {CBS_get_ucs2_be, CBB_add_ucs2_be,
LiteralToBytes(u"Hello, 世界! ¡Hola, 🌎!"),
LiteralToCodePoints(U"Hello, 世界! ¡Hola, "), false},
// Unpaired surrogates and non-characters are also rejected.
- {cbs_get_ucs2_be, cbb_add_ucs2_be, {0xd8, 0x00}, {}, false},
- {cbs_get_ucs2_be, cbb_add_ucs2_be, {0xff, 0xfe}, {}, false},
+ {CBS_get_ucs2_be, CBB_add_ucs2_be, {0xd8, 0x00}, {}, false},
+ {CBS_get_ucs2_be, CBB_add_ucs2_be, {0xff, 0xfe}, {}, false},
- {cbs_get_utf32_be, cbb_add_utf32_be,
+ {CBS_get_utf32_be, CBB_add_utf32_be,
LiteralToBytes(U"Hello, 世界! ¡Hola, 🌎!"),
LiteralToCodePoints(U"Hello, 世界! ¡Hola, 🌎!"), true},
// Unpaired surrogates and non-characters are rejected.
- {cbs_get_utf32_be, cbb_add_utf32_be, {0x00, 0x00, 0xd8, 0x00}, {}, false},
- {cbs_get_utf32_be, cbb_add_utf32_be, {0x00, 0x00, 0xff, 0xfe}, {}, false},
+ {CBS_get_utf32_be, CBB_add_utf32_be, {0x00, 0x00, 0xd8, 0x00}, {}, false},
+ {CBS_get_utf32_be, CBB_add_utf32_be, {0x00, 0x00, 0xff, 0xfe}, {}, false},
// Test that the NUL character can be encoded.
- {cbs_get_latin1, cbb_add_latin1, {0}, {0}, true},
- {cbs_get_utf8, cbb_add_utf8, {0}, {0}, true},
- {cbs_get_ucs2_be, cbb_add_ucs2_be, {0, 0}, {0}, true},
- {cbs_get_utf32_be, cbb_add_utf32_be, {0, 0, 0, 0}, {0}, true},
+ {CBS_get_latin1, CBB_add_latin1, {0}, {0}, true},
+ {CBS_get_utf8, CBB_add_utf8, {0}, {0}, true},
+ {CBS_get_ucs2_be, CBB_add_ucs2_be, {0, 0}, {0}, true},
+ {CBS_get_utf32_be, CBB_add_utf32_be, {0, 0, 0, 0}, {0}, true},
};
for (const auto &t : kTests) {
SCOPED_TRACE(Bytes(t.in));
@@ -1524,24 +1524,24 @@
ASSERT_TRUE(CBB_init(cbb.get(), 0));
for (uint32_t v : kBadCodePoints) {
SCOPED_TRACE(v);
- EXPECT_FALSE(cbb_add_utf8(cbb.get(), v));
- EXPECT_FALSE(cbb_add_latin1(cbb.get(), v));
- EXPECT_FALSE(cbb_add_ucs2_be(cbb.get(), v));
- EXPECT_FALSE(cbb_add_utf32_be(cbb.get(), v));
+ EXPECT_FALSE(CBB_add_utf8(cbb.get(), v));
+ EXPECT_FALSE(CBB_add_latin1(cbb.get(), v));
+ EXPECT_FALSE(CBB_add_ucs2_be(cbb.get(), v));
+ EXPECT_FALSE(CBB_add_utf32_be(cbb.get(), v));
}
// Additional values that are out of range.
- EXPECT_FALSE(cbb_add_latin1(cbb.get(), 0x100));
- EXPECT_FALSE(cbb_add_ucs2_be(cbb.get(), 0x10000));
+ EXPECT_FALSE(CBB_add_latin1(cbb.get(), 0x100));
+ EXPECT_FALSE(CBB_add_ucs2_be(cbb.get(), 0x10000));
- EXPECT_EQ(1u, cbb_get_utf8_len(0));
- EXPECT_EQ(1u, cbb_get_utf8_len(0x7f));
- EXPECT_EQ(2u, cbb_get_utf8_len(0x80));
- EXPECT_EQ(2u, cbb_get_utf8_len(0x7ff));
- EXPECT_EQ(3u, cbb_get_utf8_len(0x800));
- EXPECT_EQ(3u, cbb_get_utf8_len(0xffff));
- EXPECT_EQ(4u, cbb_get_utf8_len(0x10000));
- EXPECT_EQ(4u, cbb_get_utf8_len(0x10ffff));
+ EXPECT_EQ(1u, CBB_get_utf8_len(0));
+ EXPECT_EQ(1u, CBB_get_utf8_len(0x7f));
+ EXPECT_EQ(2u, CBB_get_utf8_len(0x80));
+ EXPECT_EQ(2u, CBB_get_utf8_len(0x7ff));
+ EXPECT_EQ(3u, CBB_get_utf8_len(0x800));
+ EXPECT_EQ(3u, CBB_get_utf8_len(0xffff));
+ EXPECT_EQ(4u, CBB_get_utf8_len(0x10000));
+ EXPECT_EQ(4u, CBB_get_utf8_len(0x10ffff));
}
TEST(CBSTest, BogusTime) {
diff --git a/crypto/bytestring/internal.h b/crypto/bytestring/internal.h
index ba23244..ff7a4a5 100644
--- a/crypto/bytestring/internal.h
+++ b/crypto/bytestring/internal.h
@@ -67,28 +67,6 @@
int CBB_finish_i2d(CBB *cbb, uint8_t **outp);
-// Unicode utilities.
-
-// The following functions read one Unicode code point from |cbs| with the
-// corresponding encoding and store it in |*out|. They return one on success and
-// zero on error.
-OPENSSL_EXPORT int cbs_get_utf8(CBS *cbs, uint32_t *out);
-OPENSSL_EXPORT int cbs_get_latin1(CBS *cbs, uint32_t *out);
-OPENSSL_EXPORT int cbs_get_ucs2_be(CBS *cbs, uint32_t *out);
-OPENSSL_EXPORT int cbs_get_utf32_be(CBS *cbs, uint32_t *out);
-
-// cbb_get_utf8_len returns the number of bytes needed to represent |u| in
-// UTF-8.
-OPENSSL_EXPORT size_t cbb_get_utf8_len(uint32_t u);
-
-// The following functions encode |u| to |cbb| with the corresponding
-// encoding. They return one on success and zero on error.
-OPENSSL_EXPORT int cbb_add_utf8(CBB *cbb, uint32_t u);
-OPENSSL_EXPORT int cbb_add_latin1(CBB *cbb, uint32_t u);
-OPENSSL_EXPORT int cbb_add_ucs2_be(CBB *cbb, uint32_t u);
-OPENSSL_EXPORT int cbb_add_utf32_be(CBB *cbb, uint32_t u);
-
-
#if defined(__cplusplus)
} // extern C
#endif
diff --git a/crypto/bytestring/unicode.c b/crypto/bytestring/unicode.c
index 6f9467f..10fba07 100644
--- a/crypto/bytestring/unicode.c
+++ b/crypto/bytestring/unicode.c
@@ -38,7 +38,7 @@
// TOP_BITS returns a byte with the top |n| bits set.
#define TOP_BITS(n) ((uint8_t)~BOTTOM_BITS(8 - (n)))
-int cbs_get_utf8(CBS *cbs, uint32_t *out) {
+int CBS_get_utf8(CBS *cbs, uint32_t *out) {
uint8_t c;
if (!CBS_get_u8(cbs, &c)) {
return 0;
@@ -80,7 +80,7 @@
return 1;
}
-int cbs_get_latin1(CBS *cbs, uint32_t *out) {
+int CBS_get_latin1(CBS *cbs, uint32_t *out) {
uint8_t c;
if (!CBS_get_u8(cbs, &c)) {
return 0;
@@ -89,7 +89,7 @@
return 1;
}
-int cbs_get_ucs2_be(CBS *cbs, uint32_t *out) {
+int CBS_get_ucs2_be(CBS *cbs, uint32_t *out) {
// Note UCS-2 (used by BMPString) does not support surrogates.
uint16_t c;
if (!CBS_get_u16(cbs, &c) ||
@@ -100,11 +100,11 @@
return 1;
}
-int cbs_get_utf32_be(CBS *cbs, uint32_t *out) {
+int CBS_get_utf32_be(CBS *cbs, uint32_t *out) {
return CBS_get_u32(cbs, out) && is_valid_code_point(*out);
}
-size_t cbb_get_utf8_len(uint32_t u) {
+size_t CBB_get_utf8_len(uint32_t u) {
if (u <= 0x7f) {
return 1;
}
@@ -117,7 +117,7 @@
return 4;
}
-int cbb_add_utf8(CBB *cbb, uint32_t u) {
+int CBB_add_utf8(CBB *cbb, uint32_t u) {
if (!is_valid_code_point(u)) {
return 0;
}
@@ -142,14 +142,14 @@
return 0;
}
-int cbb_add_latin1(CBB *cbb, uint32_t u) {
+int CBB_add_latin1(CBB *cbb, uint32_t u) {
return u <= 0xff && CBB_add_u8(cbb, (uint8_t)u);
}
-int cbb_add_ucs2_be(CBB *cbb, uint32_t u) {
+int CBB_add_ucs2_be(CBB *cbb, uint32_t u) {
return u <= 0xffff && is_valid_code_point(u) && CBB_add_u16(cbb, (uint16_t)u);
}
-int cbb_add_utf32_be(CBB *cbb, uint32_t u) {
+int CBB_add_utf32_be(CBB *cbb, uint32_t u) {
return is_valid_code_point(u) && CBB_add_u32(cbb, u);
}
diff --git a/crypto/pkcs8/pkcs8.c b/crypto/pkcs8/pkcs8.c
index 4bc337b..07d5de8 100644
--- a/crypto/pkcs8/pkcs8.c
+++ b/crypto/pkcs8/pkcs8.c
@@ -85,15 +85,15 @@
CBS_init(&cbs, (const uint8_t *)in, in_len);
while (CBS_len(&cbs) != 0) {
uint32_t c;
- if (!cbs_get_utf8(&cbs, &c) ||
- !cbb_add_ucs2_be(&cbb, c)) {
+ if (!CBS_get_utf8(&cbs, &c) ||
+ !CBB_add_ucs2_be(&cbb, c)) {
OPENSSL_PUT_ERROR(PKCS8, PKCS8_R_INVALID_CHARACTERS);
goto err;
}
}
// Terminate the result with a UCS-2 NUL.
- if (!cbb_add_ucs2_be(&cbb, 0) ||
+ if (!CBB_add_ucs2_be(&cbb, 0) ||
!CBB_finish(&cbb, out, out_len)) {
goto err;
}
diff --git a/crypto/pkcs8/pkcs8_x509.c b/crypto/pkcs8/pkcs8_x509.c
index 87c0961..92bdb9d 100644
--- a/crypto/pkcs8/pkcs8_x509.c
+++ b/crypto/pkcs8/pkcs8_x509.c
@@ -339,8 +339,8 @@
}
while (CBS_len(&value) != 0) {
uint32_t c;
- if (!cbs_get_ucs2_be(&value, &c) ||
- !cbb_add_utf8(&cbb, c)) {
+ if (!CBS_get_ucs2_be(&value, &c) ||
+ !CBB_add_utf8(&cbb, c)) {
OPENSSL_PUT_ERROR(PKCS8, PKCS8_R_INVALID_CHARACTERS);
CBB_cleanup(&cbb);
goto err;
@@ -972,8 +972,8 @@
CBS_init(&name_cbs, (const uint8_t *)name, name_len);
while (CBS_len(&name_cbs) != 0) {
uint32_t c;
- if (!cbs_get_utf8(&name_cbs, &c) ||
- !cbb_add_ucs2_be(&value, c)) {
+ if (!CBS_get_utf8(&name_cbs, &c) ||
+ !CBB_add_ucs2_be(&value, c)) {
OPENSSL_PUT_ERROR(PKCS8, PKCS8_R_INVALID_CHARACTERS);
return 0;
}
diff --git a/include/openssl/bytestring.h b/include/openssl/bytestring.h
index 33e13ef..7dce9c4 100644
--- a/include/openssl/bytestring.h
+++ b/include/openssl/bytestring.h
@@ -635,6 +635,28 @@
OPENSSL_EXPORT int CBB_flush_asn1_set_of(CBB *cbb);
+// Unicode utilities.
+
+// The following functions read one Unicode code point from |cbs| with the
+// corresponding encoding and store it in |*out|. They return one on success and
+// zero on error.
+OPENSSL_EXPORT int CBS_get_utf8(CBS *cbs, uint32_t *out);
+OPENSSL_EXPORT int CBS_get_latin1(CBS *cbs, uint32_t *out);
+OPENSSL_EXPORT int CBS_get_ucs2_be(CBS *cbs, uint32_t *out);
+OPENSSL_EXPORT int CBS_get_utf32_be(CBS *cbs, uint32_t *out);
+
+// CBB_get_utf8_len returns the number of bytes needed to represent |u| in
+// UTF-8.
+OPENSSL_EXPORT size_t CBB_get_utf8_len(uint32_t u);
+
+// The following functions encode |u| to |cbb| with the corresponding
+// encoding. They return one on success and zero on error.
+OPENSSL_EXPORT int CBB_add_utf8(CBB *cbb, uint32_t u);
+OPENSSL_EXPORT int CBB_add_latin1(CBB *cbb, uint32_t u);
+OPENSSL_EXPORT int CBB_add_ucs2_be(CBB *cbb, uint32_t u);
+OPENSSL_EXPORT int CBB_add_utf32_be(CBB *cbb, uint32_t u);
+
+
#if defined(__cplusplus)
} // extern C
