Merge bitcoin/bitcoin#23595: util: Add ParseHex<std::byte>() helper

facd1fb911abfc595a3484ee53397eff515d4c40 refactor: Use Span of std::byte in CExtKey::SetSeed (MarcoFalke)
fae1006019188700e0c497a63fc1550fe00ca8bb util: Add ParseHex<std::byte>() helper (MarcoFalke)
fabdf81983e2542d60542b80fb94ccb1acdd204a test: Add test for embedded null in hex string (MarcoFalke)

Pull request description:

  This adds the hex->`std::byte` helper after the `std::byte`->hex helper was added in commit 9394964f6b9d1cf1220a4eca17ba18dc49ae876d

ACKs for top commit:
  pk-b2:
    ACK https://github.com/bitcoin/bitcoin/pull/23595/commits/facd1fb911abfc595a3484ee53397eff515d4c40
  laanwj:
    Code review ACK facd1fb911abfc595a3484ee53397eff515d4c40

Tree-SHA512: e2329fbdea2e580bd1618caab31f5d0e59c245a028e1236662858e621929818870b76ab6834f7ac6a46d7874dfec63f498380ad99da6efe4218f720a60e859be

8 files changed, 28 insertions, 13 deletions

diff --git a/src/key.cpp b/src/key.cpp
index d1d521f97d..9b0971a2dd 100644
--- a/src/key.cpp
+++ b/src/key.cpp
@@ -340,11 +340,11 @@ bool CExtKey::Derive(CExtKey &out, unsigned int _nChild) const {
     return key.Derive(out.key, out.chaincode, _nChild, chaincode);
 }
 
-void CExtKey::SetSeed(Span<const uint8_t> seed)
+void CExtKey::SetSeed(Span<const std::byte> seed)
 {
     static const unsigned char hashkey[] = {'B','i','t','c','o','i','n',' ','s','e','e','d'};
     std::vector<unsigned char, secure_allocator<unsigned char>> vout(64);
-    CHMAC_SHA512{hashkey, sizeof(hashkey)}.Write(seed.data(), seed.size()).Finalize(vout.data());
+    CHMAC_SHA512{hashkey, sizeof(hashkey)}.Write(UCharCast(seed.data()), seed.size()).Finalize(vout.data());
     key.Set(vout.data(), vout.data() + 32, true);
     memcpy(chaincode.begin(), vout.data() + 32, 32);
     nDepth = 0;
diff --git a/src/key.h b/src/key.h
index b21e658107..12d03778a0 100644
--- a/src/key.h
+++ b/src/key.h
@@ -85,7 +85,7 @@ public:
 
     //! Simple read-only vector-like interface.
     unsigned int size() const { return (fValid ? keydata.size() : 0); }
-    const unsigned char* data() const { return keydata.data(); }
+    const std::byte* data() const { return reinterpret_cast<const std::byte*>(keydata.data()); }
     const unsigned char* begin() const { return keydata.data(); }
     const unsigned char* end() const { return keydata.data() + size(); }
 
@@ -178,7 +178,7 @@ struct CExtKey {
     void Decode(const unsigned char code[BIP32_EXTKEY_SIZE]);
     bool Derive(CExtKey& out, unsigned int nChild) const;
     CExtPubKey Neuter() const;
-    void SetSeed(Span<const uint8_t> seed);
+    void SetSeed(Span<const std::byte> seed);
 };
 
 /** Initialize the elliptic curve support. May not be called twice without calling ECC_Stop first. */
diff --git a/src/test/bip32_tests.cpp b/src/test/bip32_tests.cpp
index 0fa6b7784f..64cc924239 100644
--- a/src/test/bip32_tests.cpp
+++ b/src/test/bip32_tests.cpp
@@ -120,8 +120,9 @@ const std::vector<std::string> TEST5 = {
     "xprv9s21ZrQH143K3QTDL4LXw2F7HEK3wJUD2nW2nRk4stbPy6cq3jPPqjiChkVvvNKmPGJxWUtg6LnF5kejMRNNU3TGtRBeJgk33yuGBxrMPHL"
 };
 
-void RunTest(const TestVector &test) {
-    std::vector<unsigned char> seed = ParseHex(test.strHexMaster);
+void RunTest(const TestVector& test)
+{
+    std::vector<std::byte> seed{ParseHex<std::byte>(test.strHexMaster)};
     CExtKey key;
     CExtPubKey pubkey;
     key.SetSeed(seed);
diff --git a/src/test/fuzz/hex.cpp b/src/test/fuzz/hex.cpp
index cc1bc1c8cf..e637975b48 100644
--- a/src/test/fuzz/hex.cpp
+++ b/src/test/fuzz/hex.cpp
@@ -25,6 +25,8 @@ FUZZ_TARGET_INIT(hex, initialize_hex)
 {
     const std::string random_hex_string(buffer.begin(), buffer.end());
     const std::vector<unsigned char> data = ParseHex(random_hex_string);
+    const std::vector<std::byte> bytes{ParseHex<std::byte>(random_hex_string)};
+    assert(AsBytes(Span{data}) == Span{bytes});
     const std::string hex_data = HexStr(data);
     if (IsHex(random_hex_string)) {
         assert(ToLower(random_hex_string) == hex_data);
diff --git a/src/test/key_io_tests.cpp b/src/test/key_io_tests.cpp
index b06157e99f..e70b8b3dfd 100644
--- a/src/test/key_io_tests.cpp
+++ b/src/test/key_io_tests.cpp
@@ -35,7 +35,7 @@ BOOST_AUTO_TEST_CASE(key_io_valid_parse)
             continue;
         }
         std::string exp_base58string = test[0].get_str();
-        std::vector<unsigned char> exp_payload = ParseHex(test[1].get_str());
+        const std::vector<std::byte> exp_payload{ParseHex<std::byte>(test[1].get_str())};
         const UniValue &metadata = test[2].get_obj();
         bool isPrivkey = find_value(metadata, "isPrivkey").get_bool();
         SelectParams(find_value(metadata, "chain").get_str());
diff --git a/src/test/util_tests.cpp b/src/test/util_tests.cpp
index 56a1155469..fda56ccff7 100644
--- a/src/test/util_tests.cpp
+++ b/src/test/util_tests.cpp
@@ -153,7 +153,7 @@ static const unsigned char ParseHex_expected[65] = {
     0xde, 0x5c, 0x38, 0x4d, 0xf7, 0xba, 0x0b, 0x8d, 0x57, 0x8a, 0x4c, 0x70, 0x2b, 0x6b, 0xf1, 0x1d,
     0x5f
 };
-BOOST_AUTO_TEST_CASE(util_ParseHex)
+BOOST_AUTO_TEST_CASE(parse_hex)
 {
     std::vector<unsigned char> result;
     std::vector<unsigned char> expected(ParseHex_expected, ParseHex_expected + sizeof(ParseHex_expected));
@@ -169,6 +169,14 @@ BOOST_AUTO_TEST_CASE(util_ParseHex)
     result = ParseHex(" 89 34 56 78");
     BOOST_CHECK(result.size() == 4 && result[0] == 0x89 && result[1] == 0x34 && result[2] == 0x56 && result[3] == 0x78);
 
+    // Embedded null is treated as end
+    const std::string with_embedded_null{" 11 "s
+                                         " \0 "
+                                         " 22 "s};
+    BOOST_CHECK_EQUAL(with_embedded_null.size(), 11);
+    result = ParseHex(with_embedded_null);
+    BOOST_CHECK(result.size() == 1 && result[0] == 0x11);
+
     // Stop parsing at invalid value
     result = ParseHex("1234 invalid 1234");
     BOOST_CHECK(result.size() == 2 && result[0] == 0x12 && result[1] == 0x34);
diff --git a/src/util/strencodings.cpp b/src/util/strencodings.cpp
index bcedd4f517..675fe7d2d7 100644
--- a/src/util/strencodings.cpp
+++ b/src/util/strencodings.cpp
@@ -77,10 +77,10 @@ bool IsHexNumber(std::string_view str)
     return str.size() > 0;
 }
 
-std::vector<unsigned char> ParseHex(std::string_view str)
+template <typename Byte>
+std::vector<Byte> ParseHex(std::string_view str)
 {
-    // convert hex dump to vector
-    std::vector<unsigned char> vch;
+    std::vector<Byte> vch;
     auto it = str.begin();
     while (it != str.end() && it + 1 != str.end()) {
         if (IsSpace(*it)) {
@@ -90,10 +90,12 @@ std::vector<unsigned char> ParseHex(std::string_view str)
         auto c1 = HexDigit(*(it++));
         auto c2 = HexDigit(*(it++));
         if (c1 < 0 || c2 < 0) break;
-        vch.push_back(uint8_t(c1 << 4) | c2);
+        vch.push_back(Byte(c1 << 4) | Byte(c2));
     }
     return vch;
 }
+template std::vector<std::byte> ParseHex(std::string_view);
+template std::vector<uint8_t> ParseHex(std::string_view);
 
 void SplitHostPort(std::string_view in, uint16_t& portOut, std::string& hostOut)
 {
diff --git a/src/util/strencodings.h b/src/util/strencodings.h
index ebb6d88952..ee58383528 100644
--- a/src/util/strencodings.h
+++ b/src/util/strencodings.h
@@ -55,7 +55,9 @@ enum class ByteUnit : uint64_t {
 * @return           A new string without unsafe chars
 */
 std::string SanitizeString(std::string_view str, int rule = SAFE_CHARS_DEFAULT);
-std::vector<unsigned char> ParseHex(std::string_view str);
+/** Parse the hex string into bytes (uint8_t or std::byte). Ignores whitespace. */
+template <typename Byte = uint8_t>
+std::vector<Byte> ParseHex(std::string_view str);
 signed char HexDigit(char c);
 /* Returns true if each character in str is a hex character, and has an even
  * number of hex digits.*/