amorfo77/json
1
0
Fork 0
mirror of https://github.com/nlohmann/json.git synced 2026-09-06 00:08:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Tighten the write-buffer helpers after review

Browse Source 
More of @gregmarr's review on the put_* split:

- Reattach the put_chars() doc comment, which the new helpers had been
  inserted in front of, leaving it describing put_indent().

- Compute the literal length once in put_literal() instead of spelling N - 1
  at each use.

- Add put_string(str, start, end), which keeps the pointer arithmetic and the
  bounds assertions inside the function instead of at the call site. With
  dump_float()'s to_chars() output moved onto put_buffer() as well, put_chars()
  now has no callers outside put_string()/put_buffer(): nothing passes a bare
  pointer and a count any more.

- Carry the indentation as std::size_t rather than unsigned int. It is a size,
  it is compared and combined with buffer sizes throughout, and the casts in
  put_indent() disappear. next_indent() keeps its assertion, which is far
  harder to trip on a 64-bit size_t but still reachable where that is 32 bits.

No output change: pretty and compact dumps, binary values included, are
byte-identical to develop.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
This commit is contained in:
Niels Lohmann
2026-09-05 10:24:48 +02:00
parent 8e80379b4a
commit 46d367675d
3 changed files with 92 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/nlohmann/detail/output/serializer.hpp
+45 -24
View File
@@ -108,8 +108,8 @@ class serializer
void dump(const BasicJsonType& val, void dump(const BasicJsonType& val,
const bool pretty_print, const bool pretty_print,
const bool ensure_ascii, const bool ensure_ascii,
const unsigned int indent_step, const std::size_t indent_step,
const unsigned int current_indent = 0) const std::size_t current_indent = 0)
{ {
dump_internal(val, pretty_print, ensure_ascii, indent_step, current_indent); dump_internal(val, pretty_print, ensure_ascii, indent_step, current_indent);
flush(); flush();
@@ -127,8 +127,8 @@ class serializer
void dump_internal(const BasicJsonType& val, void dump_internal(const BasicJsonType& val,
const bool pretty_print, const bool pretty_print,
const bool ensure_ascii, const bool ensure_ascii,
const unsigned int indent_step, const std::size_t indent_step,
const unsigned int current_indent = 0) const std::size_t current_indent = 0)
{ {
switch (val.m_data.m_type) switch (val.m_data.m_type)
{ {
@@ -385,11 +385,13 @@ class serializer
@brief the indentation level to use for the children of the current value @brief the indentation level to use for the children of the current value
A very large @a indent_step can wrap the unsigned accumulation on deep A very large @a indent_step can wrap the unsigned accumulation on deep
nesting, which would silently truncate the indentation. nesting, which would silently truncate the indentation. Far harder to reach
now that the accumulator is a std::size_t, but still reachable where that is
32 bits wide.
*/ */
static unsigned int next_indent(const unsigned int current_indent, const unsigned int indent_step) static std::size_t next_indent(const std::size_t current_indent, const std::size_t indent_step)
{ {
const unsigned int new_indent = current_indent + indent_step; const std::size_t new_indent = current_indent + indent_step;
JSON_ASSERT(new_indent >= current_indent); JSON_ASSERT(new_indent >= current_indent);
return new_indent; return new_indent;
} }
@@ -449,7 +451,7 @@ class serializer
put_buffer(string_buffer, bytes); put_buffer(string_buffer, bytes);
bytes = 0; bytes = 0;
} }
put_chars(s.data() + i, run); put_string(s, i, i + run);
bytes_after_last_accept = 0; bytes_after_last_accept = 0;
undumped_chars = 0; undumped_chars = 0;
i += run; i += run;
@@ -706,13 +708,6 @@ class serializer
write_buffer[write_buffer_pos++] = c; write_buffer[write_buffer_pos++] = c;
} }
/*!
@brief append @a length characters to the write buffer
Runs that do not fit the buffer are written straight through the output
adapter (after flushing what is pending), so large string/number payloads
are not copied an extra time.
*/
/*! /*!
@brief append @a indent indentation characters to the write buffer @brief append @a indent indentation characters to the write buffer
@@ -725,7 +720,7 @@ class serializer
flushing does not disturb what the buffer holds, so re-filling it between flushing does not disturb what the buffer holds, so re-filling it between
flushes would be redundant work. flushes would be redundant work.
*/ */
void put_indent(unsigned int indent) void put_indent(std::size_t indent)
{ {
// closing braces at the outermost level ask for no indentation at all // closing braces at the outermost level ask for no indentation at all
if (indent == 0) if (indent == 0)
@@ -738,10 +733,10 @@ class serializer
// fill whatever room is left in the buffer; this is the whole job // fill whatever room is left in the buffer; this is the whole job
// whenever the indentation is narrower than the buffer, which is the // whenever the indentation is narrower than the buffer, which is the
// case for every sane indent_step // case for every sane indent_step
const std::size_t head = (std::min)(static_cast<std::size_t>(indent), capacity - write_buffer_pos); const std::size_t head = (std::min)(indent, capacity - write_buffer_pos);
std::memset(write_buffer.data() + write_buffer_pos, indent_char, head); std::memset(write_buffer.data() + write_buffer_pos, indent_char, head);
write_buffer_pos += head; write_buffer_pos += head;
indent -= static_cast<unsigned int>(head); indent -= head;
if (JSON_HEDLEY_LIKELY(indent == 0)) if (JSON_HEDLEY_LIKELY(indent == 0))
{ {
@@ -758,7 +753,7 @@ class serializer
{ {
write_buffer_pos = capacity; write_buffer_pos = capacity;
flush(); flush();
indent -= static_cast<unsigned int>(capacity); indent -= capacity;
} }
// the buffer still holds indentation characters throughout, so the tail // the buffer still holds indentation characters throughout, so the tail
@@ -778,14 +773,30 @@ class serializer
void put_literal(const char (&s)[N]) void put_literal(const char (&s)[N])
{ {
static_assert(N >= 2, "put_literal expects a non-empty string literal"); static_assert(N >= 2, "put_literal expects a non-empty string literal");
static_assert(N - 1 < write_buffer_size, "string literal must fit into the write buffer"); // the array bound counts the terminating NUL, which is not written
constexpr std::size_t length = N - 1;
static_assert(length < write_buffer_size, "string literal must fit into the write buffer");
if (JSON_HEDLEY_UNLIKELY(write_buffer_pos + (N - 1) > write_buffer.size())) if (JSON_HEDLEY_UNLIKELY(write_buffer_pos + length > write_buffer.size()))
{ {
flush(); flush();
} }
std::memcpy(write_buffer.data() + write_buffer_pos, s, N - 1); std::memcpy(write_buffer.data() + write_buffer_pos, s, length);
write_buffer_pos += N - 1; write_buffer_pos += length;
}
/*!
@brief append the characters of @a str in [@a start, @a end)
Keeps the pointer arithmetic and the bounds checking inside the function
rather than at the call site, which is all @ref put_chars could offer.
*/
template<typename StringType>
void put_string(const StringType& str, std::size_t start, std::size_t end)
{
JSON_ASSERT(start <= end);
JSON_ASSERT(end <= str.size());
put_chars(str.data() + start, end - start);
} }
/*! /*!
@@ -801,6 +812,16 @@ class serializer
put_chars(buffer.data(), length); put_chars(buffer.data(), length);
} }
/*!
@brief append @a length characters to the write buffer
Runs that do not fit the buffer are written straight through the output
adapter (after flushing what is pending), so large string/number payloads
are not copied an extra time.
The callers all reach this through @ref put_literal, @ref put_buffer or
@ref put_string, which each derive the length from something that knows it.
*/
JSON_HEDLEY_NON_NULL(2) JSON_HEDLEY_NON_NULL(2)
void put_chars(const char* s, std::size_t length) void put_chars(const char* s, std::size_t length)
{ {
@@ -1050,7 +1071,7 @@ class serializer
auto* begin = number_buffer.data(); auto* begin = number_buffer.data();
auto* end = ::nlohmann::detail::to_chars(begin, begin + number_buffer.size(), x); auto* end = ::nlohmann::detail::to_chars(begin, begin + number_buffer.size(), x);
put_chars(begin, static_cast<size_t>(end - begin)); put_buffer(number_buffer, static_cast<std::size_t>(end - begin));
} }
JSON_HEDLEY_NON_NULL(1) JSON_HEDLEY_NON_NULL(1)
include/nlohmann/json.hpp
+1 -1
View File
@@ -1345,7 +1345,7 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
if (indent >= 0) if (indent >= 0)
{ {
s.dump(*this, true, ensure_ascii, static_cast<unsigned int>(indent)); s.dump(*this, true, ensure_ascii, static_cast<std::size_t>(indent));
} }
else else
{ {
single_include/nlohmann/json.hpp
+46 -25
View File
@@ -21233,8 +21233,8 @@ class serializer
void dump(const BasicJsonType& val, void dump(const BasicJsonType& val,
const bool pretty_print, const bool pretty_print,
const bool ensure_ascii, const bool ensure_ascii,
const unsigned int indent_step, const std::size_t indent_step,
const unsigned int current_indent = 0) const std::size_t current_indent = 0)
{ {
dump_internal(val, pretty_print, ensure_ascii, indent_step, current_indent); dump_internal(val, pretty_print, ensure_ascii, indent_step, current_indent);
flush(); flush();
@@ -21252,8 +21252,8 @@ class serializer
void dump_internal(const BasicJsonType& val, void dump_internal(const BasicJsonType& val,
const bool pretty_print, const bool pretty_print,
const bool ensure_ascii, const bool ensure_ascii,
const unsigned int indent_step, const std::size_t indent_step,
const unsigned int current_indent = 0) const std::size_t current_indent = 0)
{ {
switch (val.m_data.m_type) switch (val.m_data.m_type)
{ {
@@ -21510,11 +21510,13 @@ class serializer
@brief the indentation level to use for the children of the current value @brief the indentation level to use for the children of the current value
A very large @a indent_step can wrap the unsigned accumulation on deep A very large @a indent_step can wrap the unsigned accumulation on deep
nesting, which would silently truncate the indentation. nesting, which would silently truncate the indentation. Far harder to reach
now that the accumulator is a std::size_t, but still reachable where that is
32 bits wide.
*/ */
static unsigned int next_indent(const unsigned int current_indent, const unsigned int indent_step) static std::size_t next_indent(const std::size_t current_indent, const std::size_t indent_step)
{ {
const unsigned int new_indent = current_indent + indent_step; const std::size_t new_indent = current_indent + indent_step;
JSON_ASSERT(new_indent >= current_indent); JSON_ASSERT(new_indent >= current_indent);
return new_indent; return new_indent;
} }
@@ -21574,7 +21576,7 @@ class serializer
put_buffer(string_buffer, bytes); put_buffer(string_buffer, bytes);
bytes = 0; bytes = 0;
} }
put_chars(s.data() + i, run); put_string(s, i, i + run);
bytes_after_last_accept = 0; bytes_after_last_accept = 0;
undumped_chars = 0; undumped_chars = 0;
i += run; i += run;
@@ -21831,13 +21833,6 @@ class serializer
write_buffer[write_buffer_pos++] = c; write_buffer[write_buffer_pos++] = c;
} }
/*!
@brief append @a length characters to the write buffer
Runs that do not fit the buffer are written straight through the output
adapter (after flushing what is pending), so large string/number payloads
are not copied an extra time.
*/
/*! /*!
@brief append @a indent indentation characters to the write buffer @brief append @a indent indentation characters to the write buffer
@@ -21850,7 +21845,7 @@ class serializer
flushing does not disturb what the buffer holds, so re-filling it between flushing does not disturb what the buffer holds, so re-filling it between
flushes would be redundant work. flushes would be redundant work.
*/ */
void put_indent(unsigned int indent) void put_indent(std::size_t indent)
{ {
// closing braces at the outermost level ask for no indentation at all // closing braces at the outermost level ask for no indentation at all
if (indent == 0) if (indent == 0)
@@ -21863,10 +21858,10 @@ class serializer
// fill whatever room is left in the buffer; this is the whole job // fill whatever room is left in the buffer; this is the whole job
// whenever the indentation is narrower than the buffer, which is the // whenever the indentation is narrower than the buffer, which is the
// case for every sane indent_step // case for every sane indent_step
const std::size_t head = (std::min)(static_cast<std::size_t>(indent), capacity - write_buffer_pos); const std::size_t head = (std::min)(indent, capacity - write_buffer_pos);
std::memset(write_buffer.data() + write_buffer_pos, indent_char, head); std::memset(write_buffer.data() + write_buffer_pos, indent_char, head);
write_buffer_pos += head; write_buffer_pos += head;
indent -= static_cast<unsigned int>(head); indent -= head;
if (JSON_HEDLEY_LIKELY(indent == 0)) if (JSON_HEDLEY_LIKELY(indent == 0))
{ {
@@ -21883,7 +21878,7 @@ class serializer
{ {
write_buffer_pos = capacity; write_buffer_pos = capacity;
flush(); flush();
indent -= static_cast<unsigned int>(capacity); indent -= capacity;
} }
// the buffer still holds indentation characters throughout, so the tail // the buffer still holds indentation characters throughout, so the tail
@@ -21903,14 +21898,30 @@ class serializer
void put_literal(const char (&s)[N]) void put_literal(const char (&s)[N])
{ {
static_assert(N >= 2, "put_literal expects a non-empty string literal"); static_assert(N >= 2, "put_literal expects a non-empty string literal");
static_assert(N - 1 < write_buffer_size, "string literal must fit into the write buffer"); // the array bound counts the terminating NUL, which is not written
constexpr std::size_t length = N - 1;
static_assert(length < write_buffer_size, "string literal must fit into the write buffer");
if (JSON_HEDLEY_UNLIKELY(write_buffer_pos + (N - 1) > write_buffer.size())) if (JSON_HEDLEY_UNLIKELY(write_buffer_pos + length > write_buffer.size()))
{ {
flush(); flush();
} }
std::memcpy(write_buffer.data() + write_buffer_pos, s, N - 1); std::memcpy(write_buffer.data() + write_buffer_pos, s, length);
write_buffer_pos += N - 1; write_buffer_pos += length;
}
/*!
@brief append the characters of @a str in [@a start, @a end)
Keeps the pointer arithmetic and the bounds checking inside the function
rather than at the call site, which is all @ref put_chars could offer.
*/
template<typename StringType>
void put_string(const StringType& str, std::size_t start, std::size_t end)
{
JSON_ASSERT(start <= end);
JSON_ASSERT(end <= str.size());
put_chars(str.data() + start, end - start);
} }
/*! /*!
@@ -21926,6 +21937,16 @@ class serializer
put_chars(buffer.data(), length); put_chars(buffer.data(), length);
} }
/*!
@brief append @a length characters to the write buffer
Runs that do not fit the buffer are written straight through the output
adapter (after flushing what is pending), so large string/number payloads
are not copied an extra time.
The callers all reach this through @ref put_literal, @ref put_buffer or
@ref put_string, which each derive the length from something that knows it.
*/
JSON_HEDLEY_NON_NULL(2) JSON_HEDLEY_NON_NULL(2)
void put_chars(const char* s, std::size_t length) void put_chars(const char* s, std::size_t length)
{ {
@@ -22175,7 +22196,7 @@ class serializer
auto* begin = number_buffer.data(); auto* begin = number_buffer.data();
auto* end = ::nlohmann::detail::to_chars(begin, begin + number_buffer.size(), x); auto* end = ::nlohmann::detail::to_chars(begin, begin + number_buffer.size(), x);
put_chars(begin, static_cast<size_t>(end - begin)); put_buffer(number_buffer, static_cast<std::size_t>(end - begin));
} }
JSON_HEDLEY_NON_NULL(1) JSON_HEDLEY_NON_NULL(1)
@@ -24032,7 +24053,7 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
if (indent >= 0) if (indent >= 0)
{ {
s.dump(*this, true, ensure_ascii, static_cast<unsigned int>(indent)); s.dump(*this, true, ensure_ascii, static_cast<std::size_t>(indent));
} }
else else
{ {