MessagePack reader: read runs of scalar array elements in one loop

As for CBOR: when an array opens and the input is contiguous, read the
integers, floats, booleans, and nils that follow in a tight loop instead
of going through parse_msgpack_internal and parse_msgpack_value for every
element. The loop stops at any other kind of value and when fewer than 9
bytes are left; those values are read by the general path as before.
Non-contiguous inputs are not affected.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

include/nlohmann/detail/input/binary_reader.hpp

@@ -2130,7 +2130,7 @@ class binary_reader
             case 0x9D:
             case 0x9E:
             case 0x9F:
-                return enter_array(conditional_static_cast<std::size_t>(static_cast<unsigned int>(current) & 0x0Fu));
+                return enter_msgpack_array(conditional_static_cast<std::size_t>(static_cast<unsigned int>(current) & 0x0Fu));
 
             // fixstr
             case 0xA0:
@@ -2261,13 +2261,13 @@ class binary_reader
             case 0xDC: // array 16
             {
                 std::uint16_t len{};
-                return get_number(len) && enter_array(static_cast<std::size_t>(len));
+                return get_number(len) && enter_msgpack_array(static_cast<std::size_t>(len));
             }
 
             case 0xDD: // array 32
             {
                 std::uint32_t len{};
-                return get_number(len) && enter_array(conditional_static_cast<std::size_t>(len));
+                return get_number(len) && enter_msgpack_array(conditional_static_cast<std::size_t>(len));
             }
 
             case 0xDE: // map 16
@@ -2609,6 +2609,138 @@ class binary_reader
         }
     }
 
+    /*!
+    @brief read the integers, floats, booleans, and nils that follow in a
+           MessagePack array, without the per-element bookkeeping of
+           @ref parse_msgpack_internal
+
+    Stops at the end of the array, at any other kind of value, and when fewer
+    than 9 bytes (the longest of these values) are left in the input, leaving
+    those to @ref parse_msgpack_value, which reads them the same way.
+
+    @param[in,out] remaining  the number of elements of the array not yet read
+    @return whether the SAX parser accepted the values
+    */
+    bool parse_msgpack_scalar_run(std::size_t& remaining)
+    {
+        while (remaining != 0 && ia.bulk_remaining() >= 9)
+        {
+            const auto* const data = reinterpret_cast<const unsigned char*>(ia.bulk_data());
+            const unsigned char head = data[0];
+            std::size_t size = 0;
+            bool accepted = true;
+            switch (head)
+            {
+                case 0xC0: // nil
+                    size = 1;
+                    accepted = sax->null();
+                    break;
+                case 0xC2: // false
+                    size = 1;
+                    accepted = sax->boolean(false);
+                    break;
+                case 0xC3: // true
+                    size = 1;
+                    accepted = sax->boolean(true);
+                    break;
+                case 0xCA: // float 32
+                    size = 5;
+                    accepted = emit_float(load_big_endian<float>(data + 1));
+                    break;
+                case 0xCB: // float 64
+                    size = 9;
+                    accepted = emit_float(load_big_endian<double>(data + 1));
+                    break;
+                case 0xCC: // uint 8
+                    size = 2;
+                    accepted = emit_unsigned(data[1]);
+                    break;
+                case 0xCD: // uint 16
+                    size = 3;
+                    accepted = emit_unsigned(load_big_endian<std::uint16_t>(data + 1));
+                    break;
+                case 0xCE: // uint 32
+                    size = 5;
+                    accepted = emit_unsigned(load_big_endian<std::uint32_t>(data + 1));
+                    break;
+                case 0xCF: // uint 64
+                    size = 9;
+                    accepted = emit_unsigned(load_big_endian<std::uint64_t>(data + 1));
+                    break;
+                case 0xD0: // int 8
+                    size = 2;
+                    accepted = emit_signed(static_cast<std::int8_t>(data[1]));
+                    break;
+                case 0xD1: // int 16
+                    size = 3;
+                    accepted = emit_signed(load_big_endian<std::int16_t>(data + 1));
+                    break;
+                case 0xD2: // int 32
+                    size = 5;
+                    accepted = emit_signed(load_big_endian<std::int32_t>(data + 1));
+                    break;
+                case 0xD3: // int 64
+                    size = 9;
+                    accepted = emit_signed(load_big_endian<std::int64_t>(data + 1));
+                    break;
+                default:
+                    if (head <= 0x7F)
+                    {
+                        // positive fixint
+                        size = 1;
+                        accepted = sax->number_unsigned(static_cast<number_unsigned_t>(head));
+                    }
+                    else if (head >= 0xE0)
+                    {
+                        // negative fixint
+                        size = 1;
+                        accepted = sax->number_integer(static_cast<std::int8_t>(head));
+                    }
+                    else
+                    {
+                        // any other value is read by parse_msgpack_value
+                        return true;
+                    }
+                    break;
+            }
+            // as if read by get() and get_number(): current is the last byte
+            ia.bulk_skip(size);
+            chars_read += size;
+            current = data[size - 1];
+            --remaining;
+            if (JSON_HEDLEY_UNLIKELY(!accepted))
+            {
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /*!
+    @brief open a MessagePack array and read the scalars it begins with (see
+           @ref parse_msgpack_scalar_run)
+    @param[in] len  the number of elements
+    @return whether the SAX parser accepted the array and the values read
+    */
+    bool enter_msgpack_array(const std::size_t len)
+    {
+        return enter_array(len) && (len == 0 || parse_msgpack_array_start(std::integral_constant<bool, bulk_scan> {}));
+    }
+
+    /// read the scalars the array just opened begins with; out of line, so
+    /// that @ref parse_msgpack_value stays small enough to be inlined
+    JSON_HEDLEY_NEVER_INLINE
+    bool parse_msgpack_array_start(std::true_type /*bulk*/)
+    {
+        return parse_msgpack_scalar_run(container_stack.back().remaining);
+    }
+
+    /// input that is not contiguous: every value is read by parse_msgpack_value
+    bool parse_msgpack_array_start(std::false_type /*bulk*/) const noexcept
+    {
+        return true;
+    }
+
     /*!
     @brief read a MessagePack value and everything nested inside it
 

single_include/nlohmann/json.hpp

@@ -19907,7 +19907,7 @@ class binary_reader
             case 0x9D:
             case 0x9E:
             case 0x9F:
-                return enter_array(conditional_static_cast<std::size_t>(static_cast<unsigned int>(current) & 0x0Fu));
+                return enter_msgpack_array(conditional_static_cast<std::size_t>(static_cast<unsigned int>(current) & 0x0Fu));
 
             // fixstr
             case 0xA0:
@@ -20038,13 +20038,13 @@ class binary_reader
             case 0xDC: // array 16
             {
                 std::uint16_t len{};
-                return get_number(len) && enter_array(static_cast<std::size_t>(len));
+                return get_number(len) && enter_msgpack_array(static_cast<std::size_t>(len));
             }
 
             case 0xDD: // array 32
             {
                 std::uint32_t len{};
-                return get_number(len) && enter_array(conditional_static_cast<std::size_t>(len));
+                return get_number(len) && enter_msgpack_array(conditional_static_cast<std::size_t>(len));
             }
 
             case 0xDE: // map 16
@@ -20386,6 +20386,138 @@ class binary_reader
         }
     }
 
+    /*!
+    @brief read the integers, floats, booleans, and nils that follow in a
+           MessagePack array, without the per-element bookkeeping of
+           @ref parse_msgpack_internal
+
+    Stops at the end of the array, at any other kind of value, and when fewer
+    than 9 bytes (the longest of these values) are left in the input, leaving
+    those to @ref parse_msgpack_value, which reads them the same way.
+
+    @param[in,out] remaining  the number of elements of the array not yet read
+    @return whether the SAX parser accepted the values
+    */
+    bool parse_msgpack_scalar_run(std::size_t& remaining)
+    {
+        while (remaining != 0 && ia.bulk_remaining() >= 9)
+        {
+            const auto* const data = reinterpret_cast<const unsigned char*>(ia.bulk_data());
+            const unsigned char head = data[0];
+            std::size_t size = 0;
+            bool accepted = true;
+            switch (head)
+            {
+                case 0xC0: // nil
+                    size = 1;
+                    accepted = sax->null();
+                    break;
+                case 0xC2: // false
+                    size = 1;
+                    accepted = sax->boolean(false);
+                    break;
+                case 0xC3: // true
+                    size = 1;
+                    accepted = sax->boolean(true);
+                    break;
+                case 0xCA: // float 32
+                    size = 5;
+                    accepted = emit_float(load_big_endian<float>(data + 1));
+                    break;
+                case 0xCB: // float 64
+                    size = 9;
+                    accepted = emit_float(load_big_endian<double>(data + 1));
+                    break;
+                case 0xCC: // uint 8
+                    size = 2;
+                    accepted = emit_unsigned(data[1]);
+                    break;
+                case 0xCD: // uint 16
+                    size = 3;
+                    accepted = emit_unsigned(load_big_endian<std::uint16_t>(data + 1));
+                    break;
+                case 0xCE: // uint 32
+                    size = 5;
+                    accepted = emit_unsigned(load_big_endian<std::uint32_t>(data + 1));
+                    break;
+                case 0xCF: // uint 64
+                    size = 9;
+                    accepted = emit_unsigned(load_big_endian<std::uint64_t>(data + 1));
+                    break;
+                case 0xD0: // int 8
+                    size = 2;
+                    accepted = emit_signed(static_cast<std::int8_t>(data[1]));
+                    break;
+                case 0xD1: // int 16
+                    size = 3;
+                    accepted = emit_signed(load_big_endian<std::int16_t>(data + 1));
+                    break;
+                case 0xD2: // int 32
+                    size = 5;
+                    accepted = emit_signed(load_big_endian<std::int32_t>(data + 1));
+                    break;
+                case 0xD3: // int 64
+                    size = 9;
+                    accepted = emit_signed(load_big_endian<std::int64_t>(data + 1));
+                    break;
+                default:
+                    if (head <= 0x7F)
+                    {
+                        // positive fixint
+                        size = 1;
+                        accepted = sax->number_unsigned(static_cast<number_unsigned_t>(head));
+                    }
+                    else if (head >= 0xE0)
+                    {
+                        // negative fixint
+                        size = 1;
+                        accepted = sax->number_integer(static_cast<std::int8_t>(head));
+                    }
+                    else
+                    {
+                        // any other value is read by parse_msgpack_value
+                        return true;
+                    }
+                    break;
+            }
+            // as if read by get() and get_number(): current is the last byte
+            ia.bulk_skip(size);
+            chars_read += size;
+            current = data[size - 1];
+            --remaining;
+            if (JSON_HEDLEY_UNLIKELY(!accepted))
+            {
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /*!
+    @brief open a MessagePack array and read the scalars it begins with (see
+           @ref parse_msgpack_scalar_run)
+    @param[in] len  the number of elements
+    @return whether the SAX parser accepted the array and the values read
+    */
+    bool enter_msgpack_array(const std::size_t len)
+    {
+        return enter_array(len) && (len == 0 || parse_msgpack_array_start(std::integral_constant<bool, bulk_scan> {}));
+    }
+
+    /// read the scalars the array just opened begins with; out of line, so
+    /// that @ref parse_msgpack_value stays small enough to be inlined
+    JSON_HEDLEY_NEVER_INLINE
+    bool parse_msgpack_array_start(std::true_type /*bulk*/)
+    {
+        return parse_msgpack_scalar_run(container_stack.back().remaining);
+    }
+
+    /// input that is not contiguous: every value is read by parse_msgpack_value
+    bool parse_msgpack_array_start(std::false_type /*bulk*/) const noexcept
+    {
+        return true;
+    }
+
     /*!
     @brief read a MessagePack value and everything nested inside it
 

tests/src/unit-msgpack.cpp

@@ -27,12 +27,15 @@ using nlohmann::json;
 #include <sstream>
 #include <iomanip>
 #include <limits>
+#include <list>
 #include <set>
 #include "make_test_data_available.hpp"
 #include "round_trip_corpus.hpp"
 #include "test_utils.hpp"
 #include "sax_countdown.hpp"
+#include "sax_event_loggers.hpp"
 using utils::SaxCountdown;
+using utils::SaxEventLogger;
 
 
 TEST_CASE("MessagePack")
@@ -2212,3 +2215,97 @@ TEST_CASE("MessagePack large strings and binaries (chunked reader)")
         }
     }
 }
+
+TEST_CASE("MessagePack arrays of scalars from contiguous and non-contiguous input")
+{
+    // contiguous input reads the integers, floats, booleans, and nils an
+    // array begins with in one loop, other input value by value; both must
+    // give the same events, value, or error for the complete document and for
+    // every truncation of it, and stop at every event a SAX handler rejects
+    const std::vector<std::uint8_t> msgpack =
+    {
+        0x83,                                                       // map(3)
+        0xA1, 'a',                                                  // "a"
+        0xDC, 0x00, 0x16,                                           // array 16 (22)
+        0x00, 0x7F, 0xCC, 0xFF, 0xCD, 0x01, 0x00,                   // 0, 127, 255, 256
+        0xCE, 0x00, 0x01, 0x00, 0x00,                               // 65536
+        0xCF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,       // 18446744073709551615
+        0xFF, 0xE0, 0xD0, 0x80, 0xD1, 0xFF, 0x00,                   // -1, -32, -128, -256
+        0xD2, 0xFF, 0xFE, 0xFF, 0xFF,                               // -65537
+        0xD3, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,       // -9223372036854775808
+        0xC2, 0xC3, 0xC0,                                           // false, true, nil
+        0xCA, 0x3F, 0xC0, 0x00, 0x00,                               // 1.5 (float 32)
+        0xCB, 0xBF, 0xB9, 0x99, 0x99, 0x99, 0x99, 0x99, 0x9A,       // -0.1
+        0xA1, 'x',                                                  // "x" (ends the loop)
+        0x01,                                                       // 1
+        0xCB, 0x40, 0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,       // 2.5
+        0xD9, 0x01, 'y',                                            // "y" (str 8)
+        0xC0,                                                       // nil
+        0xA1, 'b',                                                  // "b"
+        0x93,                                                       // fixarray(3)
+        0x92, 0xCB, 0x3F, 0xF4, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // [1.25,
+        0xCB, 0xC0, 0x0E, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,       //  -3.75]
+        0x92, 0xCB, 0x3F, 0xF4, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // [1.25,
+        0xCB, 0xC0, 0x0E, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,       //  -3.75]
+        0x90,                                                       // []
+        0xA1, 'c',                                                  // "c"
+        0x94, 0x80, 0x01, 0xC4, 0x01, 0x05, 0x91, 0x03              // [{}, 1, bin 8, [3]]
+    };
+
+    const json j = json::from_msgpack(msgpack);
+    CHECK(j["a"].size() == 22);
+    CHECK(j["a"][5] == 18446744073709551615u);
+    CHECK(j["a"][7] == -32);
+    CHECK(j["a"][11] == (std::numeric_limits<std::int64_t>::min)());
+    CHECK(j["a"][15] == 1.5);
+    CHECK(j["a"][16] == -0.1);
+    CHECK(j["a"][17] == "x");
+    CHECK(j["b"] == json::parse("[[1.25, -3.75], [1.25, -3.75], []]"));
+    CHECK(j["c"][2].is_binary());
+
+    for (std::size_t length = 0; length <= msgpack.size(); ++length)
+    {
+        CAPTURE(length)
+        const std::vector<std::uint8_t> contiguous(msgpack.begin(), msgpack.begin() + static_cast<std::ptrdiff_t>(length));
+        const std::list<std::uint8_t> bytewise(contiguous.begin(), contiguous.end());
+
+        CHECK(json::from_msgpack(contiguous, true, false).dump() == json::from_msgpack(bytewise.begin(), bytewise.end(), true, false).dump());
+#if !defined(JSON_NOEXCEPTION)
+        std::string contiguous_error;
+        std::string bytewise_error;
+        try
+        {
+            static_cast<void>(json::from_msgpack(contiguous));
+        }
+        catch (const json::parse_error& e)
+        {
+            contiguous_error = e.what();
+        }
+        try
+        {
+            static_cast<void>(json::from_msgpack(bytewise.begin(), bytewise.end()));
+        }
+        catch (const json::parse_error& e)
+        {
+            bytewise_error = e.what();
+        }
+        CHECK(contiguous_error == bytewise_error);
+#endif
+
+        SaxEventLogger from_contiguous;
+        SaxEventLogger from_bytewise;
+        CHECK(json::sax_parse(contiguous, &from_contiguous, json::input_format_t::msgpack)
+              == json::sax_parse(bytewise.begin(), bytewise.end(), &from_bytewise, json::input_format_t::msgpack));
+        CHECK(from_contiguous.events == from_bytewise.events);
+    }
+
+    for (int events = 0; events <= 60; ++events)
+    {
+        CAPTURE(events)
+        const std::list<std::uint8_t> bytewise(msgpack.begin(), msgpack.end());
+        SaxCountdown from_contiguous(events);
+        SaxCountdown from_bytewise(events);
+        CHECK(json::sax_parse(msgpack, &from_contiguous, json::input_format_t::msgpack)
+              == json::sax_parse(bytewise.begin(), bytewise.end(), &from_bytewise, json::input_format_t::msgpack));
+    }
+}