Replace vera++ with clang-format (#4518)
JerryScript-DCO-1.0-Signed-off-by: Robert Fancsik robert.fancsik@h-lab.eu
This commit is contained in:
@@ -17,6 +17,7 @@
|
||||
|
||||
#include "ecma-globals.h"
|
||||
#include "ecma-helpers.h"
|
||||
|
||||
#include "jrt-libc-includes.h"
|
||||
#include "lit-char-helpers.h"
|
||||
#include "lit-magic-strings.h"
|
||||
@@ -55,8 +56,7 @@ ecma_round_high_to_uint64 (ecma_uint128_t *num_p)
|
||||
uint64_t masked_lo = num_p->lo & ~(1ULL << 63u);
|
||||
uint64_t masked_hi = num_p->hi & 0x1;
|
||||
|
||||
if ((num_p->lo >> 63u != 0)
|
||||
&& (masked_lo > 0 || masked_hi != 0))
|
||||
if ((num_p->lo >> 63u != 0) && (masked_lo > 0 || masked_hi != 0))
|
||||
{
|
||||
return (num_p->hi + 1);
|
||||
}
|
||||
@@ -66,53 +66,52 @@ ecma_round_high_to_uint64 (ecma_uint128_t *num_p)
|
||||
/**
|
||||
* Check if 128-bit integer is zero
|
||||
*/
|
||||
#define ECMA_UINT128_IS_ZERO(name) \
|
||||
(name.hi == 0 && name.lo == 0)
|
||||
#define ECMA_UINT128_IS_ZERO(name) (name.hi == 0 && name.lo == 0)
|
||||
|
||||
/**
|
||||
* Left shift 128-bit integer by max 63 bits
|
||||
*/
|
||||
#define ECMA_UINT128_LEFT_SHIFT_MAX63(name, shift) \
|
||||
{ \
|
||||
name.hi = (name.hi << (shift)) | (name.lo >> (64 - (shift))); \
|
||||
name.lo <<= (shift); \
|
||||
}
|
||||
#define ECMA_UINT128_LEFT_SHIFT_MAX63(name, shift) \
|
||||
{ \
|
||||
name.hi = (name.hi << (shift)) | (name.lo >> (64 - (shift))); \
|
||||
name.lo <<= (shift); \
|
||||
}
|
||||
|
||||
/**
|
||||
* Right shift 128-bit integer by max 63 bits
|
||||
*/
|
||||
#define ECMA_UINT128_RIGHT_SHIFT_MAX63(name, shift) \
|
||||
{ \
|
||||
name.lo = (name.lo >> (shift)) | (name.hi << (64 - (shift))); \
|
||||
name.hi >>= (shift); \
|
||||
}
|
||||
#define ECMA_UINT128_RIGHT_SHIFT_MAX63(name, shift) \
|
||||
{ \
|
||||
name.lo = (name.lo >> (shift)) | (name.hi << (64 - (shift))); \
|
||||
name.hi >>= (shift); \
|
||||
}
|
||||
|
||||
/**
|
||||
* Add 128-bit integer
|
||||
*/
|
||||
#define ECMA_UINT128_ADD(name_add_to, name_to_add) \
|
||||
{ \
|
||||
name_add_to.hi += name_to_add.hi; \
|
||||
name_add_to.lo += name_to_add.lo; \
|
||||
if (name_add_to.lo < name_to_add.lo) \
|
||||
{ \
|
||||
name_add_to.hi++; \
|
||||
} \
|
||||
}
|
||||
{ \
|
||||
name_add_to.hi += name_to_add.hi; \
|
||||
name_add_to.lo += name_to_add.lo; \
|
||||
if (name_add_to.lo < name_to_add.lo) \
|
||||
{ \
|
||||
name_add_to.hi++; \
|
||||
} \
|
||||
}
|
||||
|
||||
/**
|
||||
* Multiply 128-bit integer by 10
|
||||
*/
|
||||
#define ECMA_UINT128_MUL10(name) \
|
||||
{ \
|
||||
ECMA_UINT128_LEFT_SHIFT_MAX63 (name, 1u); \
|
||||
\
|
||||
ecma_uint128_t name ## _tmp = name; \
|
||||
\
|
||||
ECMA_UINT128_LEFT_SHIFT_MAX63 (name ## _tmp, 2u); \
|
||||
\
|
||||
ECMA_UINT128_ADD (name, name ## _tmp); \
|
||||
}
|
||||
#define ECMA_UINT128_MUL10(name) \
|
||||
{ \
|
||||
ECMA_UINT128_LEFT_SHIFT_MAX63 (name, 1u); \
|
||||
\
|
||||
ecma_uint128_t name##_tmp = name; \
|
||||
\
|
||||
ECMA_UINT128_LEFT_SHIFT_MAX63 (name##_tmp, 2u); \
|
||||
\
|
||||
ECMA_UINT128_ADD (name, name##_tmp); \
|
||||
}
|
||||
|
||||
/**
|
||||
* Divide 128-bit integer by 10
|
||||
@@ -129,84 +128,82 @@ ecma_round_high_to_uint64 (ecma_uint128_t *num_p)
|
||||
*
|
||||
* Q = Q3 *2^96 + Q2 *2^64 + Q1 *2^32 + Q0 *2^0 // 128-bit quotient
|
||||
*/
|
||||
#define ECMA_UINT128_DIV10(name) \
|
||||
{ \
|
||||
/* estimation of reciprocal of 10, 128 bits right of the binary point (T1 == T2) */ \
|
||||
const uint64_t tenth_l = 0x9999999aul; \
|
||||
const uint64_t tenth_m = 0x99999999ul; \
|
||||
const uint64_t tenth_h = 0x19999999ul; \
|
||||
\
|
||||
uint64_t l0 = ((uint32_t) name.lo) * tenth_l; \
|
||||
uint64_t l1 = (name.lo >> 32u) * tenth_l; \
|
||||
uint64_t l2 = ((uint32_t) name.hi) * tenth_l; \
|
||||
uint64_t l3 = (name.hi >> 32u) * tenth_l; \
|
||||
uint64_t m0 = ((uint32_t) name.lo) * tenth_m; \
|
||||
uint64_t m1 = (name.lo >> 32u) * tenth_m; \
|
||||
uint64_t m2 = ((uint32_t) name.hi) * tenth_m; \
|
||||
uint64_t m3 = (name.hi >> 32u) * tenth_m; \
|
||||
uint64_t h0 = ((uint32_t) name.lo) * tenth_h; \
|
||||
uint64_t h1 = (name.lo >> 32u) * tenth_h; \
|
||||
uint64_t h2 = ((uint32_t) name.hi) * tenth_h; \
|
||||
uint64_t h3 = (name.hi >> 32u) * tenth_h; \
|
||||
\
|
||||
uint64_t q0 = l0 >> 32u; \
|
||||
q0 += (uint32_t) l1; \
|
||||
q0 += (uint32_t) m0; \
|
||||
\
|
||||
q0 >>= 32u; \
|
||||
q0 += l1 >> 32u; \
|
||||
q0 += m0 >> 32u; \
|
||||
q0 += (uint32_t) l2; \
|
||||
q0 += (uint32_t) m1; \
|
||||
q0 += (uint32_t) m0; \
|
||||
\
|
||||
q0 >>= 32u; \
|
||||
q0 += l2 >> 32u; \
|
||||
q0 += m1 >> 32u; \
|
||||
q0 += m0 >> 32u; \
|
||||
q0 += (uint32_t) l3; \
|
||||
q0 += (uint32_t) m2; \
|
||||
q0 += (uint32_t) m1; \
|
||||
q0 += (uint32_t) h0; \
|
||||
\
|
||||
q0 >>=32u; \
|
||||
q0 += l3 >> 32u; \
|
||||
q0 += m2 >> 32u; \
|
||||
q0 += m1 >> 32u; \
|
||||
q0 += h0 >> 32u; \
|
||||
q0 += (uint32_t) m3; \
|
||||
q0 += (uint32_t) m2; \
|
||||
q0 += (uint32_t) h1; \
|
||||
\
|
||||
uint64_t q1 = q0 >> 32u; \
|
||||
q1 += m3 >> 32u; \
|
||||
q1 += m2 >> 32u; \
|
||||
q1 += h1 >> 32u; \
|
||||
q1 += (uint32_t) m3; \
|
||||
q1 += (uint32_t) h2; \
|
||||
\
|
||||
uint64_t q32 = q1 >> 32u; \
|
||||
q32 += m3 >> 32u; \
|
||||
q32 += h2 >> 32u; \
|
||||
q32 += h3; \
|
||||
\
|
||||
name.lo = (q1 << 32u) | ((uint32_t) q0); \
|
||||
name.hi = q32; \
|
||||
}
|
||||
#define ECMA_UINT128_DIV10(name) \
|
||||
{ \
|
||||
/* estimation of reciprocal of 10, 128 bits right of the binary point (T1 == T2) */ \
|
||||
const uint64_t tenth_l = 0x9999999aul; \
|
||||
const uint64_t tenth_m = 0x99999999ul; \
|
||||
const uint64_t tenth_h = 0x19999999ul; \
|
||||
\
|
||||
uint64_t l0 = ((uint32_t) name.lo) * tenth_l; \
|
||||
uint64_t l1 = (name.lo >> 32u) * tenth_l; \
|
||||
uint64_t l2 = ((uint32_t) name.hi) * tenth_l; \
|
||||
uint64_t l3 = (name.hi >> 32u) * tenth_l; \
|
||||
uint64_t m0 = ((uint32_t) name.lo) * tenth_m; \
|
||||
uint64_t m1 = (name.lo >> 32u) * tenth_m; \
|
||||
uint64_t m2 = ((uint32_t) name.hi) * tenth_m; \
|
||||
uint64_t m3 = (name.hi >> 32u) * tenth_m; \
|
||||
uint64_t h0 = ((uint32_t) name.lo) * tenth_h; \
|
||||
uint64_t h1 = (name.lo >> 32u) * tenth_h; \
|
||||
uint64_t h2 = ((uint32_t) name.hi) * tenth_h; \
|
||||
uint64_t h3 = (name.hi >> 32u) * tenth_h; \
|
||||
\
|
||||
uint64_t q0 = l0 >> 32u; \
|
||||
q0 += (uint32_t) l1; \
|
||||
q0 += (uint32_t) m0; \
|
||||
\
|
||||
q0 >>= 32u; \
|
||||
q0 += l1 >> 32u; \
|
||||
q0 += m0 >> 32u; \
|
||||
q0 += (uint32_t) l2; \
|
||||
q0 += (uint32_t) m1; \
|
||||
q0 += (uint32_t) m0; \
|
||||
\
|
||||
q0 >>= 32u; \
|
||||
q0 += l2 >> 32u; \
|
||||
q0 += m1 >> 32u; \
|
||||
q0 += m0 >> 32u; \
|
||||
q0 += (uint32_t) l3; \
|
||||
q0 += (uint32_t) m2; \
|
||||
q0 += (uint32_t) m1; \
|
||||
q0 += (uint32_t) h0; \
|
||||
\
|
||||
q0 >>= 32u; \
|
||||
q0 += l3 >> 32u; \
|
||||
q0 += m2 >> 32u; \
|
||||
q0 += m1 >> 32u; \
|
||||
q0 += h0 >> 32u; \
|
||||
q0 += (uint32_t) m3; \
|
||||
q0 += (uint32_t) m2; \
|
||||
q0 += (uint32_t) h1; \
|
||||
\
|
||||
uint64_t q1 = q0 >> 32u; \
|
||||
q1 += m3 >> 32u; \
|
||||
q1 += m2 >> 32u; \
|
||||
q1 += h1 >> 32u; \
|
||||
q1 += (uint32_t) m3; \
|
||||
q1 += (uint32_t) h2; \
|
||||
\
|
||||
uint64_t q32 = q1 >> 32u; \
|
||||
q32 += m3 >> 32u; \
|
||||
q32 += h2 >> 32u; \
|
||||
q32 += h3; \
|
||||
\
|
||||
name.lo = (q1 << 32u) | ((uint32_t) q0); \
|
||||
name.hi = q32; \
|
||||
}
|
||||
|
||||
#if defined (__GNUC__) || defined (__clang__)
|
||||
#if defined(__GNUC__) || defined(__clang__)
|
||||
|
||||
/**
|
||||
* Count leading zeros in the topmost 64 bits of a 128-bit integer.
|
||||
*/
|
||||
#define ECMA_UINT128_CLZ_MAX63(name) \
|
||||
__builtin_clzll (name.hi)
|
||||
#define ECMA_UINT128_CLZ_MAX63(name) __builtin_clzll (name.hi)
|
||||
|
||||
/**
|
||||
* Count leading zeros in the topmost 4 bits of a 128-bit integer.
|
||||
*/
|
||||
#define ECMA_UINT128_CLZ_MAX4(name) \
|
||||
__builtin_clzll (name.hi)
|
||||
#define ECMA_UINT128_CLZ_MAX4(name) __builtin_clzll (name.hi)
|
||||
|
||||
#else /* !__GNUC__ && !__clang__ */
|
||||
|
||||
@@ -238,14 +235,12 @@ static const uint8_t ecma_uint4_clz[] = { 4, 3, 2, 2, 1, 1, 1, 1, 0, 0, 0, 0, 0,
|
||||
/**
|
||||
* Count leading zeros in the topmost 64 bits of a 128-bit integer.
|
||||
*/
|
||||
#define ECMA_UINT128_CLZ_MAX63(name) \
|
||||
ecma_uint64_clz (name.hi)
|
||||
#define ECMA_UINT128_CLZ_MAX63(name) ecma_uint64_clz (name.hi)
|
||||
|
||||
/**
|
||||
* Count leading zeros in the topmost 4 bits of a 128-bit integer.
|
||||
*/
|
||||
#define ECMA_UINT128_CLZ_MAX4(name) \
|
||||
ecma_uint4_clz[name.hi >> 60]
|
||||
#define ECMA_UINT128_CLZ_MAX4(name) ecma_uint4_clz[name.hi >> 60]
|
||||
|
||||
#endif /* __GNUC__ || __clang__ */
|
||||
|
||||
@@ -308,7 +303,7 @@ ecma_utf8_string_to_number_by_radix (const lit_utf8_byte_t *str_p, /**< utf-8 st
|
||||
{
|
||||
lit_code_point_t upper_limit = LIT_CHAR_0 + radix;
|
||||
|
||||
for (const lit_utf8_byte_t * iter_p = str_p; iter_p <= end_p; iter_p++)
|
||||
for (const lit_utf8_byte_t *iter_p = str_p; iter_p <= end_p; iter_p++)
|
||||
{
|
||||
int32_t digit_value;
|
||||
|
||||
@@ -328,22 +323,19 @@ ecma_utf8_string_to_number_by_radix (const lit_utf8_byte_t *str_p, /**< utf-8 st
|
||||
}
|
||||
#endif /* JERRY_ESNEXT */
|
||||
|
||||
for (const lit_utf8_byte_t * iter_p = str_p; iter_p <= end_p; iter_p++)
|
||||
for (const lit_utf8_byte_t *iter_p = str_p; iter_p <= end_p; iter_p++)
|
||||
{
|
||||
int32_t digit_value;
|
||||
|
||||
if (*iter_p >= LIT_CHAR_0
|
||||
&& *iter_p <= LIT_CHAR_9)
|
||||
if (*iter_p >= LIT_CHAR_0 && *iter_p <= LIT_CHAR_9)
|
||||
{
|
||||
digit_value = (*iter_p - LIT_CHAR_0);
|
||||
}
|
||||
else if (*iter_p >= LIT_CHAR_LOWERCASE_A
|
||||
&& *iter_p <= LIT_CHAR_LOWERCASE_F)
|
||||
else if (*iter_p >= LIT_CHAR_LOWERCASE_A && *iter_p <= LIT_CHAR_LOWERCASE_F)
|
||||
{
|
||||
digit_value = 10 + (*iter_p - LIT_CHAR_LOWERCASE_A);
|
||||
}
|
||||
else if (*iter_p >= LIT_CHAR_UPPERCASE_A
|
||||
&& *iter_p <= LIT_CHAR_UPPERCASE_F)
|
||||
else if (*iter_p >= LIT_CHAR_UPPERCASE_A && *iter_p <= LIT_CHAR_UPPERCASE_F)
|
||||
{
|
||||
digit_value = 10 + (*iter_p - LIT_CHAR_UPPERCASE_A);
|
||||
}
|
||||
@@ -393,20 +385,19 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
return ECMA_NUMBER_ZERO;
|
||||
}
|
||||
|
||||
if (end_p >= str_p + 2
|
||||
&& str_p[0] == LIT_CHAR_0)
|
||||
if (end_p >= str_p + 2 && str_p[0] == LIT_CHAR_0)
|
||||
{
|
||||
switch (LEXER_TO_ASCII_LOWERCASE (str_p[1]))
|
||||
{
|
||||
case LIT_CHAR_LOWERCASE_X :
|
||||
case LIT_CHAR_LOWERCASE_X:
|
||||
{
|
||||
return ecma_utf8_string_to_number_by_radix (str_p + 2, end_p, 16 | options);
|
||||
}
|
||||
case LIT_CHAR_LOWERCASE_O :
|
||||
case LIT_CHAR_LOWERCASE_O:
|
||||
{
|
||||
return ecma_utf8_string_to_number_by_radix (str_p + 2, end_p, 8 | options);
|
||||
}
|
||||
case LIT_CHAR_LOWERCASE_B :
|
||||
case LIT_CHAR_LOWERCASE_B:
|
||||
{
|
||||
return ecma_utf8_string_to_number_by_radix (str_p + 2, end_p, 2 | options);
|
||||
}
|
||||
@@ -455,8 +446,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
{
|
||||
int32_t digit_value;
|
||||
|
||||
if (*str_p >= LIT_CHAR_0
|
||||
&& *str_p <= LIT_CHAR_9)
|
||||
if (*str_p >= LIT_CHAR_0 && *str_p <= LIT_CHAR_9)
|
||||
{
|
||||
digit_seen = true;
|
||||
digit_value = (*str_p - LIT_CHAR_0);
|
||||
@@ -489,8 +479,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
str_p++;
|
||||
}
|
||||
|
||||
if (str_p <= end_p
|
||||
&& *str_p == LIT_CHAR_DOT)
|
||||
if (str_p <= end_p && *str_p == LIT_CHAR_DOT)
|
||||
{
|
||||
str_p++;
|
||||
|
||||
@@ -504,8 +493,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
{
|
||||
int32_t digit_value;
|
||||
|
||||
if (*str_p >= LIT_CHAR_0
|
||||
&& *str_p <= LIT_CHAR_9)
|
||||
if (*str_p >= LIT_CHAR_0 && *str_p <= LIT_CHAR_9)
|
||||
{
|
||||
digit_seen = true;
|
||||
digit_value = (*str_p - LIT_CHAR_0);
|
||||
@@ -539,9 +527,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
int32_t e_in_lit = 0;
|
||||
bool e_in_lit_sign = false;
|
||||
|
||||
if (str_p <= end_p
|
||||
&& (*str_p == LIT_CHAR_LOWERCASE_E
|
||||
|| *str_p == LIT_CHAR_UPPERCASE_E))
|
||||
if (str_p <= end_p && (*str_p == LIT_CHAR_LOWERCASE_E || *str_p == LIT_CHAR_UPPERCASE_E))
|
||||
{
|
||||
str_p++;
|
||||
|
||||
@@ -569,8 +555,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
{
|
||||
int32_t digit_value;
|
||||
|
||||
if (*str_p >= LIT_CHAR_0
|
||||
&& *str_p <= LIT_CHAR_9)
|
||||
if (*str_p >= LIT_CHAR_0 && *str_p <= LIT_CHAR_9)
|
||||
{
|
||||
digit_value = (*str_p - LIT_CHAR_0);
|
||||
}
|
||||
@@ -587,7 +572,7 @@ ecma_utf8_string_to_number (const lit_utf8_byte_t *str_p, /**< utf-8 string */
|
||||
}
|
||||
|
||||
e_in_lit = e_in_lit * 10 + digit_value;
|
||||
int32_t e_check = e + (int32_t) digits - 1 + (e_in_lit_sign ? -e_in_lit : e_in_lit);
|
||||
int32_t e_check = e + (int32_t) digits - 1 + (e_in_lit_sign ? -e_in_lit : e_in_lit);
|
||||
|
||||
if (e_check > NUMBER_MAX_DECIMAL_EXPONENT)
|
||||
{
|
||||
@@ -764,8 +749,7 @@ ecma_uint32_to_utf8_string (uint32_t value, /**< value to convert */
|
||||
buf_p--;
|
||||
*buf_p = (lit_utf8_byte_t) ((value % 10) + LIT_CHAR_0);
|
||||
value /= 10;
|
||||
}
|
||||
while (value != 0);
|
||||
} while (value != 0);
|
||||
|
||||
JERRY_ASSERT (buf_p >= out_buffer_p);
|
||||
|
||||
@@ -807,8 +791,7 @@ ecma_number_to_uint32 (ecma_number_t num) /**< ecma-number */
|
||||
|
||||
if (abs_num >= num_2_pow_32)
|
||||
{
|
||||
num_in_uint32_range = ecma_number_calc_remainder (abs_num,
|
||||
num_2_pow_32);
|
||||
num_in_uint32_range = ecma_number_calc_remainder (abs_num, num_2_pow_32);
|
||||
}
|
||||
else
|
||||
{
|
||||
@@ -822,8 +805,7 @@ ecma_number_to_uint32 (ecma_number_t num) /**< ecma-number */
|
||||
const uint32_t ret = sign ? -uint32_num : uint32_num;
|
||||
|
||||
#ifndef JERRY_NDEBUG
|
||||
if (sign
|
||||
&& uint32_num != 0)
|
||||
if (sign && uint32_num != 0)
|
||||
{
|
||||
JERRY_ASSERT (ret == uint64_2_pow_32 - uint32_num);
|
||||
}
|
||||
@@ -885,16 +867,16 @@ ecma_number_to_int32 (ecma_number_t num) /**< ecma-number */
|
||||
} /* ecma_number_to_int32 */
|
||||
|
||||
/**
|
||||
* Perform conversion of ecma-number to decimal representation with decimal exponent.
|
||||
*
|
||||
* Note:
|
||||
* The calculated values correspond to s, n, k parameters in ECMA-262 v5, 9.8.1, item 5:
|
||||
* - parameter out_digits_p corresponds to s, the digits of the number;
|
||||
* - parameter out_decimal_exp_p corresponds to n, the decimal exponent;
|
||||
* - return value corresponds to k, the number of digits.
|
||||
*
|
||||
* @return the number of digits
|
||||
*/
|
||||
* Perform conversion of ecma-number to decimal representation with decimal exponent.
|
||||
*
|
||||
* Note:
|
||||
* The calculated values correspond to s, n, k parameters in ECMA-262 v5, 9.8.1, item 5:
|
||||
* - parameter out_digits_p corresponds to s, the digits of the number;
|
||||
* - parameter out_decimal_exp_p corresponds to n, the decimal exponent;
|
||||
* - return value corresponds to k, the number of digits.
|
||||
*
|
||||
* @return the number of digits
|
||||
*/
|
||||
lit_utf8_size_t
|
||||
ecma_number_to_decimal (ecma_number_t num, /**< ecma-number */
|
||||
lit_utf8_byte_t *out_digits_p, /**< [out] buffer to fill with digits */
|
||||
@@ -951,7 +933,8 @@ ecma_number_to_utf8_string (ecma_number_t num, /**< ecma-number */
|
||||
if (ecma_number_is_infinity (num))
|
||||
{
|
||||
/* 4. */
|
||||
dst_p = lit_copy_magic_string_to_buffer (LIT_MAGIC_STRING_INFINITY_UL, dst_p,
|
||||
dst_p = lit_copy_magic_string_to_buffer (LIT_MAGIC_STRING_INFINITY_UL,
|
||||
dst_p,
|
||||
(lit_utf8_size_t) (buffer_p + buffer_size - dst_p));
|
||||
JERRY_ASSERT (dst_p <= buffer_p + buffer_size);
|
||||
return (lit_utf8_size_t) (dst_p - buffer_p);
|
||||
|
||||
Reference in New Issue
Block a user