Niels Lohmann 56ea5dd6bf Reserve the parser's nesting stack once ...

The stack of open arrays and objects was a std::vector<std::uint8_t> that
allocated at the first push_back and then reallocated at 1, 2, 4, 8 bytes,
so a document nested four levels deep made four allocations before it
parsed a value. A document that starts with a container now reserves 32
levels in one allocation; deeper documents grow as before. The loop that
reads the tokens is the code it was, so large documents are not affected
(a stack object with its own storage, and a reserve at every container,
both cost the token loop instructions or cycles).

Fixed cost of parsing a 124-byte object (macOS instructions): accept
-7.6 %, DOM -2.5 %; no change for a single number.

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

2026-10-22 09:27:48 +02:00

..

binary_reader.hpp

Fix compiler warnings on develop from the API performance stack (#5858)

2026-10-15 22:42:46 +02:00

input_adapters.hpp

Read FILE* input with std::fgetc under MemorySanitizer (#5874)

2026-10-21 10:22:50 +02:00

json_sax.hpp

Move object keys into the value when parsing (#5855)

2026-10-21 20:00:51 +02:00

lexer.hpp

Match literals without a stack array and keep parser states in bytes (#5854)

2026-10-21 20:00:51 +02:00

number_parse.hpp

Use standard library helpers where available, with C++11 fallbacks (#5876)

2026-10-21 20:14:59 +02:00

parser.hpp

Reserve the parser's nesting stack once

2026-10-22 09:27:48 +02:00

position_t.hpp

📄 adjust year (#5044)

2026-01-01 20:00:39 +01:00

pow5_table.hpp

Speed up the lexer: own float parser, string scan, and \u table (#5738)

2026-10-14 17:38:36 +02:00

simd_scan.hpp

Scan strings with SSE2, SSSE3, or NEON (#5853)

2026-10-21 20:00:50 +02:00

string_scan.hpp

Scan strings with SSE2, SSSE3, or NEON (#5853)

2026-10-21 20:00:50 +02:00