"
COPYRIGHT (c) 1988 by Claus Gittinger
All Rights Reserved
This software is furnished under a license and may be used
only in accordance with the terms of that license and with the
inclusion of the above copyright notice. This software may not
be provided or otherwise made available to, or used by, any
other person. No title to or ownership of the software is
hereby transferred.
"
Integer subclass:#SmallInteger
instanceVariableNames:''
classVariableNames:''
poolDictionaries:''
category:'Magnitude-Numbers'
!
SmallInteger comment:'
COPYRIGHT (c) 1988 by Claus Gittinger
All Rights Reserved
$Header: /cvs/stx/stx/libbasic/Attic/SmallInt.st,v 1.17 1994-10-10 00:28:25 claus Exp $
'!
!SmallInteger class methodsFor:'documentation'!
copyright
"
COPYRIGHT (c) 1988 by Claus Gittinger
All Rights Reserved
This software is furnished under a license and may be used
only in accordance with the terms of that license and with the
inclusion of the above copyright notice. This software may not
be provided or otherwise made available to, or used by, any
other person. No title to or ownership of the software is
hereby transferred.
"
!
version
"
$Header: /cvs/stx/stx/libbasic/Attic/SmallInt.st,v 1.17 1994-10-10 00:28:25 claus Exp $
"
!
documentation
"
SmallIntegers are Integers in the range of at least +/- 2^30
(i.e. 31 bits, which is not guaranteed; it could be more on 64 bit cpus).
These are no real objects - they have no instances (not even storage !!)
and cannot be subclassed.
The reason is to save both storage and runtime by not collecting
SmallIntegers in the system. SmallInts are marked by having the TAG_INT
bit set, in contrast to all other objects which do not.
Since this knowledge is hardwired into the system (and there is no
class-field stored with SmallIntegers) there can be no subclass of
SmallInteger (sorry).
If you really need this kind of thing, create a subclass of Integer,
with an instance variable holding the value.
"
! !
!SmallInteger class methodsFor:'instance creation'!
basicNew
"catch instance creation
- SmallIntegers cannot be created with new"
self error:'instances of SmallInteger cannot be created with new'
!
basicNew:size
"catch instance creation
- SmallIntegers cannot be created with new"
self error:'instances of SmallInteger cannot be created with new'
! !
!SmallInteger class methodsFor:'constants'!
maxBits
"return the number of bits in instances of me.
For very special uses only - not constant across implementations"
%{ /* NOCONTEXT */
RETURN ( _MKSMALLINT(N_INT_BITS) );
%}
"SmallInteger maxBits"
!
maxBytes
"return the number of bytes in instances of me.
For very special uses only - not constant across implementations"
%{ /* NOCONTEXT */
RETURN ( _MKSMALLINT(N_INT_BITS / 8 + 1) );
%}
"SmallInteger maxBytes"
!
minVal
"return the smallest Integer representable as SmallInteger.
For very special uses only - not constant across implementations"
%{ /* NOCONTEXT */
RETURN ( _MKSMALLINT(_MIN_INT) );
%}
"SmallInteger minVal"
!
maxVal
"return the largest Integer representable as SmallInteger.
For very special uses only - not constant across implementations"
%{ /* NOCONTEXT */
RETURN ( _MKSMALLINT(_MAX_INT) );
%}
"SmallInteger maxVal"
! !
!SmallInteger class methodsFor:'queries'!
isBuiltInClass
"this class is known by the run-time-system"
^ true
!
canBeSubclassed
"return true, if its allowed to create subclasses of the receiver.
Return nil here - since it is NOT possible for SmallInteger"
^ false
! !
!SmallInteger methodsFor:'catching messages'!
basicAt:index
"catch indexed access - report an error
defined here since basicAt: in Object ommits the SmallInteger check."
self notIndexed
!
basicAt:index put:anObject
"catch indexed access - report an error
defined here since basicAt:put: in Object ommits the SmallInteger check."
self notIndexed
!
size
"return the number of indexed instvars - SmallIntegers have none."
^ 0
!
basicSize
"return the number of indexed instvars - SmallIntegers have none.
Defined here since basicSize in Object ommits the SmallInteger check."
^ 0
! !
!SmallInteger methodsFor:'copying'!
shallowCopy
"return a shallow copy of myself
- reimplemented here since smallintegers are unique"
^ self
!
simpleDeepCopy
"return a deep copy of myself
- reimplemented here since smallintegers are unique"
^ self
!
deepCopyUsing:aDictionary
"return a deep copy of myself
- reimplemented here since smallintegers are unique"
^ self
!
deepCopy
"return a deep copy of myself
- reimplemented here since smallintegers are unique"
^ self
! !
!SmallInteger methodsFor:'comparing'!
= aNumber
"return true, if the arguments value is equal to mine"
%{ /* NOCONTEXT */
if (aNumber == self) {
RETURN ( true );
}
if (! _isNonNilObject(aNumber)) {
/* a smallint or nil */
RETURN ( false );
}
if (_qClass(aNumber) == Float) {
RETURN ( ((double)_intVal(self) == _floatVal(aNumber)) ? true : false );
}
%}
.
aNumber respondsToArithmetic ifFalse:[^ false].
^ self retry:#= coercing:aNumber
!
~= aNumber
"return true, if the arguments value is not equal to mine"
%{ /* NOCONTEXT */
if (aNumber == self) {
RETURN ( false );
}
if (! _isNonNilObject(aNumber)) {
/* a smallint or nil */
RETURN ( true );
}
if (_qClass(aNumber) == Float) {
RETURN ( ((double)_intVal(self) == _floatVal(aNumber)) ? false : true );
}
%}
.
aNumber respondsToArithmetic ifFalse:[^ true].
^ self retry:#~= coercing:aNumber
!
< aNumber
"return true, if the argument is greater than the receiver"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) < _intVal(aNumber)) ? true : false );
#else
/* tag bit does not change ordering */
RETURN ( ((INT)self < (INT)aNumber) ? true : false );
#endif
}
if (__isFloat(aNumber)) {
RETURN ( ((double)_intVal(self) < _floatVal(aNumber)) ? true : false );
}
%}
.
^ aNumber lessFromInteger:self
"^ self retry:#< coercing:aNumber"
!
> aNumber
"return true, if the argument is less than the receiver"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) > _intVal(aNumber)) ? true : false );
#else
/* tag bit does not change ordering */
RETURN ( ((INT)self > (INT)aNumber) ? true : false );
#endif
}
if (__isFloat(aNumber)) {
RETURN ( ((double)_intVal(self) > _floatVal(aNumber)) ? true : false );
}
%}
.
^ self retry:#> coercing:aNumber
!
>= aNumber
"return true, if the argument is less or equal"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) >= _intVal(aNumber)) ? true : false );
#else
/* tag bit does not change ordering */
RETURN ( ((INT)self >= (INT)aNumber) ? true : false );
#endif
}
if (__isFloat(aNumber)) {
RETURN ( ((double)_intVal(self) >= _floatVal(aNumber)) ? true : false );
}
%}
.
^ self retry:#>= coercing:aNumber
!
<= aNumber
"return true, if the argument is greater or equal"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) <= _intVal(aNumber)) ? true : false );
#else
/* tag bit does not change ordering */
RETURN ( ((INT)self <= (INT)aNumber) ? true : false );
#endif
}
if (__isFloat(aNumber)) {
RETURN ( ((double)_intVal(self) <= _floatVal(aNumber)) ? true : false );
}
%}
.
^ self retry:#<= coercing:aNumber
!
identityHash
"return an integer useful for hashing on identity"
self >= 0 ifTrue:[^ self].
^ self negated
!
min:aNumber
"return the receiver or the argument, whichever is smaller"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
if (_intVal(self) < _intVal(aNumber)) {
#else
/* tag bit does not change ordering */
if ((INT)(self) < (INT)(aNumber)) {
#endif
RETURN ( self );
}
RETURN ( aNumber );
}
if (__isFloat(aNumber)) {
if ( (double)_intVal(self) < _floatVal(aNumber) ) {
RETURN ( self );
}
RETURN ( aNumber );
}
%}
.
(self < aNumber) ifTrue:[^ self].
^ aNumber
!
max:aNumber
"return the receiver or the argument, whichever is greater"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef POSITIVE_ADDRESSES
if (_intVal(self) > _intVal(aNumber)) {
#else
/* tag bit does not change ordering */
if ((INT)(self) > (INT)(aNumber)) {
#endif
RETURN ( self );
}
RETURN ( aNumber );
}
if (__isFloat(aNumber)) {
if ( (double)_intVal(self) > _floatVal(aNumber) ) {
RETURN ( self );
}
RETURN ( aNumber );
}
%}
.
(self > aNumber) ifTrue:[^ self].
^ aNumber
! !
!SmallInteger methodsFor:'testing'!
negative
"return true, if the receiver is less than zero
reimplemented here for speed"
%{ /* NOCONTEXT */
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) < 0) ? true : false );
#else
/* tag bit does not change sign */
RETURN ( ((INT)(self) < 0) ? true : false );
#endif
%}
!
positive
"return true, if the receiver is not negative
reimplemented here for speed"
%{ /* NOCONTEXT */
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) >= 0) ? true : false );
#else
/* tag bit does not change sign */
RETURN ( ((INT)(self) >= 0) ? true : false );
#endif
%}
!
strictlyPositive
"return true, if the receiver is greater than zero
reimplemented here for speed"
%{ /* NOCONTEXT */
#ifdef POSITIVE_ADDRESSES
RETURN ( (_intVal(self) > 0) ? true : false );
#else
/* tag bit does not change sign */
RETURN ( ((INT)(self) > 0) ? true : false );
#endif
%}
!
sign
"return the sign of the receiver
reimplemented here for speed"
%{ /* NOCONTEXT */
INT val = _intVal(self);
if (val < 0) {
RETURN ( _MKSMALLINT(-1) );
}
if (val > 0) {
RETURN ( _MKSMALLINT(1) );
}
RETURN ( _MKSMALLINT(0) );
%}
!
between:min and:max
"return true if the receiver is less than or equal to the argument max
and greater than or equal to the argument min.
- reimplemented here for speed"
%{ /* NOCONTEXT */
if (_isSmallInteger(min) && _isSmallInteger(max)) {
REGISTER INT selfVal;
selfVal = _intVal(self);
if (selfVal < _intVal(min)) {
RETURN ( false );
}
if (selfVal > _intVal(max)) {
RETURN ( false );
}
RETURN ( true );
}
%}
.
(self < min) ifTrue:[^ false].
(self > max) ifTrue:[^ false].
^ true
!
even
"return true, if the receiver is even"
%{ /* NOCONTEXT */
#ifdef POSITIVE_ADDRESSES
RETURN ( ((INT)self & 1) ? false : true );
#else
RETURN ( ((INT)self & ((INT)_MKSMALLINT(1) & ~TAG_INT)) ? false : true );
#endif
%}
!
odd
"return true, if the receiver is odd"
%{ /* NOCONTEXT */
#ifdef POSITIVE_ADDRESSES
RETURN ( ((INT)self & 1) ? true : false );
#else
RETURN ( ((INT)self & ((INT)_MKSMALLINT(1) & ~TAG_INT)) ? true : false );
#endif
%}
! !
!SmallInteger methodsFor:'arithmetic'!
+ aNumber
"return the sum of the receivers value and the arguments value"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef _ADD_IO_IO
RETURN ( _ADD_IO_IO(self, aNumber) );
#else
REGISTER INT sum;
extern OBJ _makeLarge();
sum = _intVal(self) + _intVal(aNumber);
if ((sum >= _MIN_INT) && (sum <= _MAX_INT)) {
RETURN ( _MKSMALLINT(sum) );
}
RETURN ( _makeLarge(sum) );
#endif
}
if ((aNumber != nil) && (_qClass(aNumber) == Float)) {
OBJ newFloat;
double val;
val = _floatVal(aNumber);
_qMKFLOAT(newFloat, (double)(_intVal(self)) + val, SENDER);
RETURN ( newFloat );
}
%}
.
^ aNumber sumFromInteger:self
!
- aNumber
"return the difference of the receivers value and the arguments value"
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
#ifdef _SUB_IO_IO
RETURN ( _SUB_IO_IO(self, aNumber) );
#else
REGISTER INT diff;
extern OBJ _makeLarge();
diff = _intVal(self) - _intVal(aNumber);
if ((diff >= _MIN_INT) && (diff <= _MAX_INT)) {
RETURN ( _MKSMALLINT(diff) );
}
RETURN ( _makeLarge(diff) );
#endif
}
if ((aNumber != nil) && (_qClass(aNumber) == Float)) {
OBJ newFloat;
double val;
val = _floatVal(aNumber);
_qMKFLOAT(newFloat, (double)(_intVal(self)) - val, SENDER);
RETURN ( newFloat );
}
%}
.
^ aNumber differenceFromInteger:self
!
* aNumber
"return the product of the receivers value and the arguments value"
|aLarge|
%{ /* NOCONTEXT */
INT myValue, otherValue;
unsigned INT productLow, productHi;
int negative;
# define lowBits(foo) ((foo) & 0xFFFF)
# define hiBits(foo) ((foo) >> 16)
/*
* can we use long long arithmetic ?
*/
#if defined(__GNUC__) && (__GNUC__ >= 2)
/*
* commented, since long-long arithmetic seems to
* be buggy in some implementations (sparc) ...
* (took me a while to find this out :-(
*/
# ifdef NOTDEF
# define _LONGLONG
# endif
#endif
if (_isSmallInteger(aNumber)) {
myValue = _intVal(self);
otherValue = _intVal(aNumber);
#if defined(_LONGLONG)
{
long long product;
product = (long long)myValue * (long long)otherValue;
if ((product >= (long long)_MIN_INT)
&& (product <= (long long)_MAX_INT)) {
RETURN ( _MKSMALLINT((int)product) );
}
if (product < 0) {
negative = 1;
product = -product;
} else {
negative = 0;
}
productHi = product >> 32;
productLow = product & 0xFFFFFFFF;
}
#else
negative = 0;
if (myValue < 0) {
negative = 1;
myValue = -myValue;
}
if (otherValue < 0) {
negative = (1 - negative);
otherValue = -otherValue;
}
# if defined(__GNUC__) && defined(mc68k)
asm ("mulu%.l %3,%1:%0"
: "=d" ((unsigned long)(productLow)),
"=d" ((unsigned long)(productHi))
: "%0" ((unsigned long)(myValue)),
"dmi" ((unsigned long)(otherValue)));
# else
# if defined (__GNUC__) && defined(i386)
asm ("mull %3"
: "=a" ((unsigned long)(productLow)),
"=d" ((unsigned long)(productHi))
: "%0" ((unsigned long)(myValue)),
"rm" ((unsigned long)(otherValue)));
# else
{
unsigned INT pHH, pHL, pLH, pLL;
unsigned INT low1, low2, hi1, hi2;
unsigned INT t;
/* unsigned multiply myValue * otherValue -> productHi, productLow
*
* this is too slow:
* since most machines can do 32*32 to 64 bit multiply,
* (or at least 32*32 with Overflow check)
* - need more assembler (inline) functions here
*/
low1 = lowBits(myValue);
hi1 = hiBits(myValue);
low2 = lowBits(otherValue);
hi2 = hiBits(otherValue);
pLH = low1 * hi2;
pHL = hi1 * low2;
pLL = low1 * low2;
/*
* the common case ...
*/
if ((pHL == 0)
&& (pLH == 0)
&& ((pLL & 0xC0000000) == 0)) {
if (negative) {
RETURN ( _MKSMALLINT(- ((INT)pLL)) );
}
RETURN ( _MKSMALLINT((INT)pLL) );
}
pHH = hi1 * hi2;
/*
* pHH |--------|--------|
* pLH |--------|--------|
* pHL |--------|--------|
* pLL |--------|--------|
*/
t = lowBits(pLH) + lowBits(pHL) + hiBits(pLL);
productLow = (t << 16) + lowBits(pLL);
productHi = pHH + hiBits(t) + hiBits(pHL) + hiBits(pLH);
}
# endif
# endif
if (productHi == 0) {
if (negative) {
if (productLow <= -(_MIN_INT)) {
RETURN ( _MKSMALLINT(-((INT)productLow)) );
}
} else {
if (productLow <= _MAX_INT) {
RETURN ( _MKSMALLINT(productLow) );
}
}
}
#endif
{
extern OBJ LargeInteger,
_sign_value16_value16_value16_value16_;
static struct inlineCache val = _ILC5;
OBJ aLarge;
aLarge = (*val.ilc_func)(LargeInteger,
@symbol(sign:value16:value16:value16:value16:),
CON_COMMA nil, &val,
negative ? _MKSMALLINT(-1) : _MKSMALLINT(1),
_MKSMALLINT(lowBits(productLow)),
_MKSMALLINT(hiBits(productLow)),
_MKSMALLINT(lowBits(productHi)),
_MKSMALLINT(hiBits(productHi)) );
RETURN(aLarge);
}
} else if ((aNumber != nil) && (_qClass(aNumber) == Float)) {
OBJ newFloat;
double val;
val = _floatVal(aNumber);
_qMKFLOAT(newFloat, (double)(_intVal(self)) * val, SENDER);
RETURN ( newFloat );
}
%}
.
^ aNumber productFromInteger:self
!
/ aNumber
"return the quotient of the receivers value and the arguments value"
%{ /* NOCONTEXT */
INT me, t, val;
double dval;
if (_isSmallInteger(aNumber)) {
val = _intVal(aNumber);
if (val != 0) {
me = _intVal(self);
t = me / val;
#ifdef GOOD_OPTIMIZER
if (me % val) {
#else
/* this is stupid - all I want is to look for a remainder ...
but most compilers are too stupid and generate an extra modulu
instruction for "if (me % val)".
Even if most divide instructions already leave the remainder in
some register.
Therefore I use a multiplication which is faster than a modulu
on most machines. Hint to GNU people :-)
*/
if ((t * val) == me) {
#endif
RETURN ( _MKSMALLINT(t) );
}
}
} else {
if (__isFloat(aNumber)) {
dval = _floatVal(aNumber);
if (dval != 0.0) {
OBJ newFloat;
me = _intVal(self);
_qMKFLOAT(newFloat, (double)me / dval, SENDER);
RETURN ( newFloat );
}
}
}
%}
.
aNumber isInteger ifTrue:[
aNumber = 0 ifTrue:[
^ DivisionByZeroSignal raise.
].
^ Fraction numerator:self denominator:aNumber
].
^ aNumber quotientFromInteger:self
!
// aNumber
"return the integer part of the quotient of the receivers value
and the arguments value"
%{ /* NOCONTEXT */
INT val;
if (_isSmallInteger(aNumber)) {
val = _intVal(aNumber);
if (val != 0) {
RETURN ( _MKSMALLINT(_intVal(self) / val) );
}
} else {
if (__isFraction(aNumber)) {
OBJ t;
INT num, den;
t = _FractionInstPtr(aNumber)->f_numerator;
if (_isSmallInteger(t)) {
num = _intVal(t);
t = _FractionInstPtr(aNumber)->f_denominator;
if (_isSmallInteger(t)) {
den = _intVal(t);
RETURN ( _MKSMALLINT(_intVal(self) * den / num ));
}
}
}
}
%}
.
(aNumber = 0) ifTrue:[
^ DivisionByZeroSignal raise.
].
^ self retry:#// coercing:aNumber
!
\\ aNumber
"return the integer rest of the receivers value
divided by the arguments value"
%{ /* NOCONTEXT */
INT mySelf, val;
if (_isSmallInteger(aNumber)) {
mySelf = _intVal(self);
if (mySelf < 0) mySelf = -mySelf;
val = _intVal(aNumber);
if (val != 0) {
if (val < 0) {
RETURN ( _MKSMALLINT(-(mySelf % -val)) );
}
RETURN ( _MKSMALLINT(mySelf % val) );
}
}
%}
.
(aNumber = 0) ifTrue:[
^ DivisionByZeroSignal raise.
].
^ self retry:#\\ coercing:aNumber
!
abs
"return the absolute value of the receiver
reimplemented here for speed"
%{ /* NOCONTEXT */
INT val = _intVal(self);
if (val >= 0) {
RETURN (self);
}
if (val != _MIN_INT) {
RETURN ( _MKSMALLINT(-val) );
}
%}
.
"only reached for minVal"
^ self negated
!
negated
"return the negative value of the receiver
reimplemented here for speed"
%{ /* NOCONTEXT */
INT val = _intVal(self);
if (val != _MIN_INT) {
RETURN ( _MKSMALLINT(- val) );
}
%}
.
^ (LargeInteger value:(SmallInteger maxVal)) + 1
! !
!SmallInteger methodsFor:'modulu arithmetic'!
times:aNumber
"return the product of the receiver and the argument, as SmallInteger.
The argument must be another SmallInteger.
If the result overflows the smallInteger range, the value modulu the
smallInteger range is returned (i.e. the low bits of the product).
This is of course not always correct, but some code does a modulu anyway
and can therefore speed things up by not going through LargeIntegers."
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
RETURN ( _MKSMALLINT((_intVal(self) * _intVal(aNumber)) & 0x7FFFFFFF) );
}
%}
.
self primitiveFailed
!
plus:aNumber
"return the sum of the receiver and the argument, as SmallInteger.
The argument must be another SmallInteger.
If the result overflows the smallInteger range, the value modulu the
smallInteger range is returned (i.e. the low bits of the sum).
This is of course not always correct, but some code does a modulu anyway
and can therefore speed things up by not going through LargeIntegers."
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
RETURN ( _MKSMALLINT((_intVal(self) + _intVal(aNumber)) & 0x7FFFFFFF) );
}
%}
.
self primitiveFailed
!
subtract:aNumber
"return the difference of the receiver and the argument, as SmallInteger.
The argument must be another SmallInteger.
If the result overflows the smallInteger range, the value modulu the
smallInteger range is returned (i.e. the low bits of the sum).
This is of course not always correct, but some code does a modulu anyway
and can therefore speed things up by not going through LargeIntegers."
%{ /* NOCONTEXT */
if (_isSmallInteger(aNumber)) {
RETURN ( _MKSMALLINT((_intVal(self) - _intVal(aNumber)) & 0x7FFFFFFF) );
}
%}
.
self primitiveFailed
! !
!SmallInteger class methodsFor:'bit mask constants'!
bitMaskFor:index
"return a bitmask for the index's bit (index starts at 1)"
(index between:1 and:SmallInteger maxBits) ifFalse:[
^ self error:'index out of bounds'
].
^ 1 bitShift:(index - 1)
! !
!SmallInteger methodsFor:'bit operators'!
bitAt:index
"return the value of the index's bit (index starts at 1).
Notice: the result of bitAt: on negative receivers is not
defined in the language standard (since the implementation
is free to choose any internal representation for integers)"
|mask|
(index between:1 and:SmallInteger maxBits) ifFalse:[
^ self error:'index out of bounds'
].
mask := 1 bitShift:(index - 1).
((self bitAnd:mask) == 0) ifTrue:[^ 0].
^ 1
!
allMask:anInteger
"return true if all 1-bits in anInteger are also 1 in the receiver"
^ (self bitAnd:anInteger) == anInteger
"2r00001111 allMask:2r00000001"
"2r00001111 allMask:2r00011110"
"2r00001111 allMask:2r00000000"
!
anyMask:anInteger
"return true if any 1-bits in anInteger is also 1 in the receiver.
(somewhat incorrect, if the mask is zero)"
^ (self bitAnd:anInteger) ~~ 0
"2r00001111 anyMask:2r00000001"
"2r00001111 anyMask:2r11110000"
!
noMask:anInteger
"return true if no 1-bit in anInteger is 1 in the receiver"
^ (self bitAnd:anInteger) == 0
"2r00001111 noMask:2r00000001"
"2r00001111 noMask:2r11110000"
!
highBit
"return the bitIndex of the highest bit set. The returned bitIndex
starts at 1 for the least significant bit. Returns -1 if no bit is set."
%{ /* NOCONTEXT */
INT mask, index, bits;
bits = _intVal(self);
if (bits == 0) {
RETURN ( _MKSMALLINT(-1) );
}
#ifdef alpha
mask = 0x2000000000000000;
index = 62;
#else
mask = 0x20000000;
index = 30;
#endif
while (index) {
if (bits & mask) break;
mask = mask >> 1;
index--;
}
RETURN ( _MKSMALLINT(index) );
%}
"2r000100 highBit"
"2r010100 highBit"
"2r000001 highBit"
"0 highBit"
"SmallInteger maxVal highBit"
!
lowBit
"return the bitIndex of the lowest bit set. The returned bitIndex
starts at 1 for the least significant bit. Returns -1 if no bit is set."
%{ /* NOCONTEXT */
INT mask, index, bits;
bits = _intVal(self);
if (bits == 0) {
RETURN ( _MKSMALLINT(-1) );
}
mask = 1;
index = 1;
#ifdef alpha
while (index != 63) {
#else
while (index != 31) {
#endif
if (bits & mask) {
RETURN ( _MKSMALLINT(index) );
}
mask = mask << 1;
index++;
}
RETURN ( _MKSMALLINT(-1) );
%}
"2r000100 lowBit"
"2r010010 lowBit"
"2r100001 lowBit"
"0 lowBit"
!
bitShift:shiftCount
"return the value of the receiver shifted by shiftCount bits;
leftShift if shiftCount > 0; rightShift otherwise.
Notice: the result of bitShift: on negative receivers is not
defined in the language standard (since the implementation
is free to choose any internal representation for integers)"
%{ /* NOCONTEXT */
INT bits, count;
if (_isSmallInteger(shiftCount)) {
count = _intVal(shiftCount);
bits = _intVal(self);
if (count > 0) {
#if defined(_LONGLONG)
unsigned long long result;
result = bits;
if (count <= N_INT_BITS) {
result <<= count;
if (result <= _MAX_INT) {
RETURN ( _MKSMALLINT(result) );
}
{
extern OBJ LargeInteger,
_sign_value16_value16_value16_value16_;
static struct inlineCache val = _ILC5;
OBJ aLarge;
aLarge = (*val.ilc_func)(LargeInteger,
@symbol(sign:value16:value16:value16:value16:),
CON_COMMA nil, &val,
_MKSMALLINT(1),
_MKSMALLINT(lowBits(result)),
_MKSMALLINT(hiBits(result)),
_MKSMALLINT(lowBits(result >> 32)),
_MKSMALLINT(hiBits(result >> 32)) );
RETURN(aLarge);
}
}
#else
/*
* check for overflow
*/
if (count < (N_INT_BITS-1)) {
if (! (bits >> (N_INT_BITS - 1 - count))) {
RETURN ( _MKSMALLINT(bits << count) );
}
/*
* so, there is an overflow ...
* handle it as largeInteger
*/
/* FALL THROUGH */
}
#endif
} else {
/*
* right shifts cannot overflow
*/
if (count < 0) {
/*
* some machines ignore shifts bigger than
* the number of bits in an int ...
*/
if (count < (-N_INT_BITS-1))
RETURN (_MKSMALLINT(0));
RETURN ( _MKSMALLINT(bits >> -count) );
}
RETURN (self );
}
}
%}.
(shiftCount isMemberOf:SmallInteger) ifTrue:[
^ (LargeInteger value:self) bitShift:shiftCount
].
^ self bitShift:(shiftCount coerce:1)
!
bitOr:anInteger
"return the bitwise-or of the receiver and the argument, anInteger"
%{ /* NOCONTEXT */
/* oring the tags doesn't change it */
if (_isSmallInteger(anInteger)) {
RETURN ( ((OBJ) ((INT)self | (INT)anInteger)) );
}
%}
.
^ self retry:#bitOr coercing:anInteger
"(2r000000100 bitOr:2r00000011) radixPrintStringRadix:2"
!
bitAnd:anInteger
"return the bitwise-and of the receiver and the argument, anInteger"
%{ /* NOCONTEXT */
/* anding the tags doesn't change it */
if (_isSmallInteger(anInteger)) {
RETURN ( ((OBJ) ((INT)self & (INT)anInteger)) );
}
%}
.
^ self retry:#bitAnd coercing:anInteger
"(2r001010100 bitAnd:2r00001111) radixPrintStringRadix:2"
!
bitXor:anInteger
"return the bitwise-exclusive-or of the receiver and the argument, anInteger"
%{ /* NOCONTEXT */
/* xoring the tags turns it off - or it in again */
if (_isSmallInteger(anInteger)) {
RETURN ( (OBJ)( ((INT)self ^ (INT)anInteger) | TAG_INT) );
}
%}
.
^ self retry:#bitXor coercing:anInteger
!
bitInvert
"return the value of the receiver with all bits inverted"
%{ /* NOCONTEXT */
/* invert anything except tag bits */
RETURN ( ((OBJ) ((INT)self ^ ~TAG_MASK)) );
%}
!
bitTest:aMask
"return true, if any bit from aMask is set in the receiver"
%{ /* NOCONTEXT */
/* and all bits except tag */
if (_isSmallInteger(aMask)) {
RETURN ( ((INT)self & ((INT)aMask & ~TAG_MASK)) ? true : false );
}
%}
.
^ self retry:#bitTest coercing:aMask
! !
!SmallInteger methodsFor:'byte access'!
digitLength
"return the number bytes used by this Integer"
^ self abs highBit - 1 // 8 + 1
!
digitAt:index
"return 8 bits of value, starting at byte index"
%{ /* NOCONTEXT */
INT val;
if (_isSmallInteger(index)) {
val = _intVal(self);
if (val < 0)
val = -val;
switch (_intVal(index)) {
case 1:
RETURN ( _MKSMALLINT( val & 0xFF) );
case 2:
RETURN ( _MKSMALLINT( (val >> 8) & 0xFF) );
case 3:
RETURN ( _MKSMALLINT( (val >> 16) & 0xFF) );
case 4:
RETURN ( _MKSMALLINT( (val >> 24) & 0xFF) );
#ifdef alpha
case 5:
RETURN ( _MKSMALLINT( (val >> 32) & 0xFF) );
case 6:
RETURN ( _MKSMALLINT( (val >> 40) & 0xFF) );
case 7:
RETURN ( _MKSMALLINT( (val >> 48) & 0xFF) );
case 8:
RETURN ( _MKSMALLINT( (val >> 56) & 0xFF) );
#endif
}
}
%}.
index > 0 ifFalse:[
self primitiveFailed
].
^ 0
"(16r12345678 digitAt:1) printStringRadix:16"
"(16r12345678 digitAt:3) printStringRadix:16"
! !
!SmallInteger methodsFor:'misc math functions'!
gcd:anInteger
"return the greatest common divisor (Euclid's algorithm).
This has been redefined here for more speed since due to the
use of gcd in Fraction code, it has become time-critical for
some code. (thanx to MessageTally)"
%{ /* NOCONTEXT */
if (_isSmallInteger(anInteger)) {
INT orgArg, ttt, selfInt, temp;
ttt = orgArg = _intVal(anInteger);
if (ttt) {
selfInt = _intVal(self);
while (ttt != 0) {
temp = selfInt % ttt;
selfInt = ttt;
ttt = temp;
}
/*
* since its not defined in what the sign of
* a modulu result is when the arg is negative,
* change it explicitely here ...
*/
if (orgArg < 0) {
/* result should be negative */
if (selfInt > 0) selfInt = -selfInt;
} else {
/* result should be positive */
if (selfInt < 0) selfInt = -selfInt;
}
RETURN ( _MKSMALLINT(selfInt) );
}
}
%}
.
^ super gcd:anInteger
!
intlog10
"return the truncation of log10 of the receiver -
stupid implementation; used to find out the number of digits needed
to print a number/and for conversion to a LargeInteger"
self <= 0 ifTrue:[
self error:'logarithm of negative integer'
].
self < 10 ifTrue:[^ 1].
self < 100 ifTrue:[^ 2].
self < 1000 ifTrue:[^ 3].
self < 10000 ifTrue:[^ 4].
self < 100000 ifTrue:[^ 5].
self < 1000000 ifTrue:[^ 6].
self < 10000000 ifTrue:[^ 7].
self < 100000000 ifTrue:[^ 8].
self < 1000000000 ifTrue:[^ 9].
^ 10
! !
!SmallInteger methodsFor:'coercing and converting'!
coerce:aNumber
"return aNumber converted into receivers type"
^ aNumber asInteger
!
generality
"return the generality value - see ArithmeticValue>>retry:coercing:"
^ 20
!
asFloat
"return a Float with same value as receiver"
%{ /* NOCONTEXT */
OBJ newFloat;
_qMKFLOAT(newFloat, (double)_intVal(self), SENDER);
/*
_qNew(newFloat, sizeof(struct floatstruct), SENDER);
_InstPtr(newFloat)->o_class = Float;
_FloatInstPtr(newFloat)->f_floatvalue = _intVal(self);
*/
RETURN ( newFloat );
%}
!
asLargeInteger
"return a LargeInteger with same value as receiver"
^ LargeInteger value:self
!
asCharacter
"Return a character with the receiver as ascii value"
^ Character value:self
! !
!SmallInteger methodsFor:'iterators'!
timesRepeat:aBlock
"evaluate the argument, aBlock self times"
|count "{ Class: SmallInteger }" |
count := self.
[count > 0] whileTrue:[
aBlock value.
count := count - 1
]
!
to:stop do:aBlock
"evaluate aBlock for every integer between (and including) the receiver
and the argument, stop.
Reimplemented for speed"
|home|
%{
REGISTER INT tmp;
INT final;
REGISTER OBJFUNC code;
extern OBJ Block, _value_;
static struct inlineCache blockVal = _ILC1;
REGISTER OBJ rHome;
if (_isSmallInteger(stop)) {
tmp = _intVal(self);
final = _intVal(stop);
#ifdef OLD
if (__isBlock(aBlock)
#else
if (__isBlockLike(aBlock)
#endif
&& ((code = _BlockInstPtr(aBlock)->b_code) != (OBJFUNC)nil)
&& (_BlockInstPtr(aBlock)->b_nargs == _MKSMALLINT(1))) {
#ifdef NEW_BLOCK_CALL
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(aBlock, CON_COMMA _MKSMALLINT(tmp));
tmp++;
}
#else /* old BLOCK_CALL */
/*
* arg is a compiled block -
* directly call it without going through "Block-value"
*/
home = _BlockInstPtr(aBlock)->b_home;
rHome = home;
if ((rHome == nil) || (_qSpace(rHome) >= STACKSPACE)) {
/*
* home will not move - keep in in a register
*/
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(rHome, CON_COMMA _MKSMALLINT(tmp));
tmp++;
}
} else {
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(home, CON_COMMA _MKSMALLINT(tmp));
tmp++;
}
}
#endif /* NEW_BLOCK_CALL */
} else {
/*
* arg is something else - call it with value"
*/
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*blockVal.ilc_func)(aBlock,
@symbol(value:),
CON_COMMA nil, &blockVal,
_MKSMALLINT(tmp));
tmp++;
}
}
RETURN ( self );
}
%}
.
^ super to:stop do:aBlock
"1 to:10 do:[:i | i printNewline]"
!
to:stop by:incr do:aBlock
"reimplemented for speed"
|home|
%{
REGISTER INT tmp, step;
REGISTER INT final;
REGISTER OBJFUNC code;
extern OBJ Block, _value_;
static struct inlineCache blockVal = _ILC1;
REGISTER OBJ rHome;
if (_isSmallInteger(incr)
&& _isSmallInteger(stop)) {
tmp = _intVal(self);
final = _intVal(stop);
step = _intVal(incr);
#ifdef OLD
if (__isBlock(aBlock)
#else
if (__isBlockLike(aBlock)
#endif
&& ((code = _BlockInstPtr(aBlock)->b_code) != (OBJFUNC)nil)
&& (_BlockInstPtr(aBlock)->b_nargs == _MKSMALLINT(1))) {
#ifdef NEW_BLOCK_CALL
if (step < 0) {
while (tmp >= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(aBlock, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
} else {
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(aBlock, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
}
#else
/*
* arg is a compiled block -
* directly call it without going through "Block-value"
*/
home = _BlockInstPtr(aBlock)->b_home;
rHome = home;
if (step < 0) {
if ((rHome == nil) || (_qSpace(rHome) >= STACKSPACE)) {
/*
* home is on stack - will not move
*/
while (tmp >= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(rHome, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
} else {
while (tmp >= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(home, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
}
} else {
if ((rHome == nil) || (_qSpace(rHome) >= STACKSPACE)) {
/*
* home is on stack - will not move
*/
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(rHome, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
} else {
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*code)(home, CON_COMMA _MKSMALLINT(tmp));
tmp += step;
}
}
}
#endif
} else {
/*
* arg is something else - call it with value"
*/
if (step < 0) {
while (tmp >= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*blockVal.ilc_func)(aBlock,
@symbol(value:),
CON_COMMA nil, &blockVal,
_MKSMALLINT(tmp));
tmp += step;
}
} else {
while (tmp <= final) {
if (InterruptPending != nil) interrupt(CONARG);
(*blockVal.ilc_func)(aBlock,
@symbol(value:),
CON_COMMA nil, &blockVal,
_MKSMALLINT(tmp));
tmp += step;
}
}
}
RETURN ( self );
}
%}
.
^ super to:stop do:aBlock
"1 to:10 by:3 do:[:i | i printNewline]"
! !
!SmallInteger class methodsFor:'binary storage'!
binaryDefinitionFrom: stream manager: manager
"read the binary representation as stored in storeBinaryOn:"
| value |
value := stream next bitAnd: 16r7F.
value > 16r3F ifTrue: [
value := value - 16r80
].
value := (value bitShift: 8) bitOr: stream next.
value := (value bitShift: 8) bitOr: stream next.
value := (value bitShift: 8) bitOr: stream next.
^ value
! !
!SmallInteger methodsFor:'binary storage'!
hasSpecialBinaryRepresentation
"return true, if the receiver has a special binary representation"
^ true
!
storeBinaryOn: stream manager: manager
"append a binary representation onto stream.
Redefined since SmallIntegers are stored as their value with the 32nd bit
set as a tag.
To make the binary file a bit more compact, zeros and single byte ints
are stored with a more compact representation (using special type-codes)."
self == 0 ifTrue:[
stream nextPut: manager codeForZero.
^ self
].
(self between:0 and:255) ifTrue:[
stream nextPut: manager codeForByteInteger.
stream nextPut: self.
^ self
].
stream nextPut: (((self bitShift: -24) bitAnd: 16rFF) bitOr: 16r80).
stream nextPut: ((self bitShift: -16) bitAnd: 16rFF).
stream nextPut: ((self bitShift: -8) bitAnd: 16rFF).
stream nextPut: (self bitAnd: 16rFF)
! !
!SmallInteger methodsFor:'printing & storing'!
printOn:aStream
"append my printstring (base 10) to aStream."
aStream nextPutAll:(self printString)
!
printString
"return my printstring (base 10)"
"since printf knows pretty good how to do this,
here is an exception to the rule of basing printString
upon the printOn: method."
%{ /* NOCONTEXT */
char buffer[30];
OBJ newString;
#ifdef THISCONTEXT_IN_REGISTER
OBJ sav = __thisContext;
#endif
int sz;
sprintf(buffer, "%d", _intVal(self));
#ifdef THISCONTEXT_IN_REGISTER
__thisContext = sav;
#endif
sz = sizeof(struct stringheader) + strlen(buffer) + 1;
_qNew(newString, sz, SENDER);
_InstPtr(newString)->o_class = String;
strcpy(_stringVal(newString), buffer);
RETURN (newString);
%}
!
printOn:aStream base:radix
"append my printstring in any number base to aStream.
The radix argument should be between 2 and 36."
aStream nextPutAll:(self printStringRadix:radix)
!
printStringRadix:radix
"return my printstring (optimized for bases 16, 10 and 8)"
%{ /* NOCONTEXT */
char *format = (char *)0;
char buffer[30];
OBJ newString;
int sz;
if (_isSmallInteger(radix)) {
switch (_intVal(radix)) {
case 10:
format = "%d";
break;
case 16:
format = "%x";
break;
case 8:
format = "%o";
break;
}
}
if (format) {
#ifdef THISCONTEXT_IN_REGISTER
OBJ sav = __thisContext;
#endif
sprintf(buffer, format, _intVal(self));
#ifdef THISCONTEXT_IN_REGISTER
__thisContext = sav;
#endif
sz = sizeof(struct stringheader) + strlen(buffer) + 1;
_qNew(newString, sz, SENDER);
_InstPtr(newString)->o_class = String;
strcpy(_stringVal(newString), buffer);
RETURN (newString);
}
%}
.
"fall back for seldom used bases"
^ super printStringRadix:radix
"123 printStringRadix:16"
"123 printStringRadix:8"
"123 printStringRadix:2"
"123 printStringRadix:3"
"123 printStringRadix:1"
!
printfPrintString:formatString
"non-portable, but sometimes useful.
return a printed representation of the receiver
as specified by formatString, which is defined by the C-
function 'printf'.
No checking for string overrun - the resulting string
must be shorter than 256 chars or else ...
This method is NONSTANDARD and may be removed without notice."
%{ /* STACK: 400 */
char buffer[256];
if (__isString(formatString)) {
#ifdef THISCONTEXT_IN_REGISTER
OBJ sav = __thisContext;
#endif
sprintf(buffer, _stringVal(formatString), _intVal(self));
#ifdef THISCONTEXT_IN_REGISTER
__thisContext = sav;
#endif
RETURN ( _MKSTRING(buffer COMMA_SND) );
}
%}
.
self primitiveFailed
"123 printfPrintString:'%%d -> %d'"
"123 printfPrintString:'%%6d -> %6d'"
"123 printfPrintString:'%%x -> %x'"
"123 printfPrintString:'%%4x -> %4x'"
"123 printfPrintString:'%%04x -> %04x'"
! !