Skip to content

Optimizations for CPU and MCU environments #989

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
MarcAntoineCRUE wants to merge 2 commits into
base: master
Choose a base branch
from
+182 −65
Open

Optimizations for CPU and MCU environments #989

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b9d2b6d
refactor cJSON_Delete to use iterative approach for deep nesting and …
MarcAntoineCRUE Feb 22, 2026
096d532
optimize number parsing in parse_number function with a fast path for…
MarcAntoineCRUE Feb 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
247 changes: 182 additions & 65 deletions cJSON.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,7 +249,7 @@ static cJSON *cJSON_New_Item(const internal_hooks * const hooks)
return node;
}

/* Delete a cJSON structure. */
/* Delete a cJSON structure. Iterative to avoid stack overflow on deep nesting. */
CJSON_PUBLIC(void) cJSON_Delete(cJSON *item)
{
cJSON *next = NULL;
Expand All @@ -258,17 +258,29 @@ CJSON_PUBLIC(void) cJSON_Delete(cJSON *item)
next = item->next;
if (!(item->type & cJSON_IsReference) && (item->child != NULL))
{
cJSON_Delete(item->child);
/* Instead of recursing, descend into child iteratively:
* Push current sibling chain onto the child's tail so we
* process it after the subtree is freed. */
cJSON *child = item->child;
cJSON *last_child = child;
item->child = NULL;

/* Find the last sibling in the child chain */
while (last_child->next != NULL)
{
last_child = last_child->next;
}
/* Append remaining siblings after child chain */
last_child->next = next;
next = child;
}
if (!(item->type & cJSON_IsReference) && (item->valuestring != NULL))
{
global_hooks.deallocate(item->valuestring);
item->valuestring = NULL;
}
if (!(item->type & cJSON_StringIsConst) && (item->string != NULL))
{
global_hooks.deallocate(item->string);
item->string = NULL;
}
global_hooks.deallocate(item);
item = next;
Expand Down Expand Up @@ -308,17 +320,80 @@ static cJSON_bool parse_number(cJSON * const item, parse_buffer * const input_bu
{
double number = 0;
unsigned char *after_end = NULL;
unsigned char *number_c_string;
unsigned char decimal_point = get_decimal_point();
size_t i = 0;
size_t number_string_length = 0;
cJSON_bool has_decimal_point = false;
cJSON_bool has_exponent = false;
/* Use a stack buffer to avoid heap allocation for number parsing.
* 64 bytes is more than enough for any valid JSON number
* (max ~25 chars for doubles like -1.7976931348623157e+308). */
unsigned char number_c_string_stack[64];
unsigned char *number_c_string = NULL;
cJSON_bool number_c_string_allocated = false;

if ((input_buffer == NULL) || (input_buffer->content == NULL))
{
return false;
}

/* ===== Fast path: simple integers (no decimal, no exponent) =====
* Avoids strtod() entirely for the most common JSON number type.
* strtod() is expensive: locale handling, NaN/Inf detection, etc.
* Handles values up to 9 digits (<=999,999,999) to stay C89-safe with unsigned long. */
{
const unsigned char *p = buffer_at_offset(input_buffer);
unsigned long fast_int = 0;
cJSON_bool fast_negative = false;
size_t fast_len = 0;
size_t digit_count = 0;

if (*p == '-')
{
fast_negative = true;
p++;
fast_len++;
}

/* Must have at least one digit */
if (*p >= '0' && *p <= '9')
{
/* Accumulate up to 9 digits - guaranteed no overflow in unsigned long */
while (*p >= '0' && *p <= '9' && digit_count < 9)
{
fast_int = fast_int * 10u + (unsigned long)(*p - '0');
p++;
fast_len++;
digit_count++;
}

/* Fast path valid if: no more digits, no decimal, no exponent */
if ((*p < '0' || *p > '9') && *p != '.' && *p != 'e' && *p != 'E')
{
double fast_value = fast_negative ? -(double)fast_int : (double)fast_int;

item->valuedouble = fast_value;
if (fast_value >= INT_MAX)
{
item->valueint = INT_MAX;
}
else if (fast_value <= (double)INT_MIN)
{
item->valueint = INT_MIN;
}
else
{
item->valueint = (int)fast_value;
}
item->type = cJSON_Number;
input_buffer->offset += fast_len;
return true;
}
/* else: fall through to full strtod path below */
}
}
/* ===== End fast path ===== */

/* copy the number into a temporary buffer and replace '.' with the decimal point
* of the current locale (for strtod)
* This also takes care of '\0' not necessarily being available for marking the end of the input */
Expand All @@ -338,9 +413,13 @@ static cJSON_bool parse_number(cJSON * const item, parse_buffer * const input_bu
case '9':
case '+':
case '-':
number_string_length++;
break;

case 'e':
case 'E':
number_string_length++;
has_exponent = true;
break;

case '.':
Expand All @@ -353,11 +432,19 @@ static cJSON_bool parse_number(cJSON * const item, parse_buffer * const input_bu
}
}
loop_end:
/* malloc for temporary buffer, add 1 for '\0' */
number_c_string = (unsigned char *) input_buffer->hooks.allocate(number_string_length + 1);
if (number_c_string == NULL)
/* Use stack buffer if possible, fall back to heap for unusually long numbers */
if (number_string_length < sizeof(number_c_string_stack))
{
return false; /* allocation failure */
number_c_string = number_c_string_stack;
}
else
{
number_c_string = (unsigned char *) input_buffer->hooks.allocate(number_string_length + 1);
if (number_c_string == NULL)
{
return false; /* allocation failure */
}
number_c_string_allocated = true;
}

memcpy(number_c_string, buffer_at_offset(input_buffer), number_string_length);
Expand All @@ -375,11 +462,15 @@ static cJSON_bool parse_number(cJSON * const item, parse_buffer * const input_bu
}
}

(void)has_exponent; /* used only to detect fast path exclusion above */

number = strtod((const char*)number_c_string, (char**)&after_end);
if (number_c_string == after_end)
{
/* free the temporary buffer */
input_buffer->hooks.deallocate(number_c_string);
if (number_c_string_allocated)
{
input_buffer->hooks.deallocate(number_c_string);
}
return false; /* parse_error */
}

Expand All @@ -402,8 +493,10 @@ static cJSON_bool parse_number(cJSON * const item, parse_buffer * const input_bu
item->type = cJSON_Number;

input_buffer->offset += (size_t)(after_end - number_c_string);
/* free the temporary buffer */
input_buffer->hooks.deallocate(number_c_string);
if (number_c_string_allocated)
{
input_buffer->hooks.deallocate(number_c_string);
}
return true;
}

Expand Down Expand Up @@ -1094,9 +1187,14 @@ static parse_buffer *buffer_skip_whitespace(parse_buffer * const buffer)
return buffer;
}

while (can_access_at_index(buffer, 0) && (buffer_at_offset(buffer)[0] <= 32))
while (can_access_at_index(buffer, 0))
{
buffer->offset++;
unsigned char c = buffer_at_offset(buffer)[0];
if (c != ' ' && c != '\t' && c != '\r' && c != '\n')
{
break;
}
buffer->offset++;
}

if (buffer->offset == buffer->length)
Expand Down Expand Up @@ -1367,48 +1465,73 @@ static cJSON_bool parse_value(cJSON * const item, parse_buffer * const input_buf
return false; /* no input */
}

/* parse the different types of values */
/* null */
if (can_read(input_buffer, 4) && (strncmp((const char*)buffer_at_offset(input_buffer), "null", 4) == 0))
{
item->type = cJSON_NULL;
input_buffer->offset += 4;
return true;
}
/* false */
if (can_read(input_buffer, 5) && (strncmp((const char*)buffer_at_offset(input_buffer), "false", 5) == 0))
{
item->type = cJSON_False;
input_buffer->offset += 5;
return true;
}
/* true */
if (can_read(input_buffer, 4) && (strncmp((const char*)buffer_at_offset(input_buffer), "true", 4) == 0))
{
item->type = cJSON_True;
item->valueint = 1;
input_buffer->offset += 4;
return true;
}
/* string */
if (can_access_at_index(input_buffer, 0) && (buffer_at_offset(input_buffer)[0] == '\"'))
{
return parse_string(item, input_buffer);
}
/* number */
if (can_access_at_index(input_buffer, 0) && ((buffer_at_offset(input_buffer)[0] == '-') || ((buffer_at_offset(input_buffer)[0] >= '0') && (buffer_at_offset(input_buffer)[0] <= '9'))))
{
return parse_number(item, input_buffer);
}
/* array */
if (can_access_at_index(input_buffer, 0) && (buffer_at_offset(input_buffer)[0] == '['))
/* Dispatch based on first character for faster type detection */
if (cannot_access_at_index(input_buffer, 0))
{
return parse_array(item, input_buffer);
return false;
}
/* object */
if (can_access_at_index(input_buffer, 0) && (buffer_at_offset(input_buffer)[0] == '{'))
{
return parse_object(item, input_buffer);

switch (buffer_at_offset(input_buffer)[0])
{
/* string */
case '\"':
return parse_string(item, input_buffer);

/* number */
case '-':
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
return parse_number(item, input_buffer);

/* array */
case '[':
return parse_array(item, input_buffer);

/* object */
case '{':
return parse_object(item, input_buffer);

/* null */
case 'n':
if (can_read(input_buffer, 4) && (strncmp((const char*)buffer_at_offset(input_buffer), "null", 4) == 0))
{
item->type = cJSON_NULL;
input_buffer->offset += 4;
return true;
}
break;

/* true */
case 't':
if (can_read(input_buffer, 4) && (strncmp((const char*)buffer_at_offset(input_buffer), "true", 4) == 0))
{
item->type = cJSON_True;
item->valueint = 1;
input_buffer->offset += 4;
return true;
}
break;

/* false */
case 'f':
if (can_read(input_buffer, 5) && (strncmp((const char*)buffer_at_offset(input_buffer), "false", 5) == 0))
{
item->type = cJSON_False;
input_buffer->offset += 5;
return true;
}
break;

default:
break;
}

return false;
Expand Down Expand Up @@ -1798,16 +1921,12 @@ static cJSON_bool print_object(const cJSON * const item, printbuffer * const out
{
if (output_buffer->format)
{
size_t i;
output_pointer = ensure(output_buffer, output_buffer->depth);
if (output_pointer == NULL)
{
return false;
}
for (i = 0; i < output_buffer->depth; i++)
{
*output_pointer++ = '\t';
}
memset(output_pointer, '\t', output_buffer->depth);
output_buffer->offset += output_buffer->depth;
}

Expand Down Expand Up @@ -1867,11 +1986,9 @@ static cJSON_bool print_object(const cJSON * const item, printbuffer * const out
}
if (output_buffer->format)
{
size_t i;
for (i = 0; i < (output_buffer->depth - 1); i++)
{
*output_pointer++ = '\t';
}
size_t indent = output_buffer->depth - 1;
memset(output_pointer, '\t', indent);
output_pointer += indent;
}
*output_pointer++ = '}';
*output_pointer = '\0';
Expand Down