Tcl Source Code

        apt:
          sources:
            - ubuntu-toolchain-r-test
          packages:
            - g++-7
      env:
        - BUILD_DIR=unix
    - os: linux
      dist: xenial
      compiler: gcc-7
      addons:
        apt:
          sources:
            - ubuntu-toolchain-r-test
          packages:
            - g++-7
      env:
        - BUILD_DIR=unix
        - CFGOPT=CFLAGS="-DTCL_UTF_MAX=6"
    - os: osx
      osx_image: xcode8
      env:
        - BUILD_DIR=unix
    - os: osx
      osx_image: xcode8
      env:

'\"
'\" Copyright (c) 2005 Sergey Brester aka sebres.
'\"
'\" See the file "license.terms" for information on usage and redistribution
'\" of this file, and for a DISCLAIMER OF ALL WARRANTIES.
'\"
.TH timerate n "" Tcl "Tcl Built-In Commands"
.so man.macros
.BS
'\" Note:  do not modify the .SH NAME line immediately below!
.SH NAME
timerate \- Time-related execution resp. performance measurement of a script
.SH SYNOPSIS
\fBtimerate \fIscript\fR \fI?time ?max-count??\fR
.sp
\fBtimerate \fI?-direct?\fR \fI?-overhead double?\fR \fIscript\fR \fI?time ?max-count??\fR
.sp
\fBtimerate \fI?-calibrate?\fR \fI?-direct?\fR \fIscript\fR \fI?time ?max-count??\fR
.BE
.SH DESCRIPTION
.PP
The first and second form will evaluate \fIscript\fR until the interval
\fItime\fR given in milliseconds elapses, or for 1000 milliseconds (1 second)
if \fItime\fR is not specified.
.sp
The parameter \fImax-count\fR could additionally impose a further restriction
by the maximal number of iterations to evaluate the script.
If \fImax-count\fR is specified, the evalution will stop either this count of
iterations is reached or the time is exceeded.
.sp
It will then return a canonical tcl-list of the form
.PP
.CS
\fB0.095977 \(mcs/# 52095836 # 10419167 #/sec 5000.000 nett-ms\fR
.CE
.PP
which indicates:
.IP \(bu
the average amount of time required per iteration, in microseconds ([\fBlindex\fR $result 0])
.IP \(bu
the count how many times it was executed ([\fBlindex\fR $result 2])
.IP \(bu
the estimated rate per second ([\fBlindex\fR $result 4])
.IP \(bu
the estimated real execution time without measurement overhead ([\fBlindex\fR $result 6])
.PP
Time is measured in elapsed time using the finest timer resolution as possible,
not CPU time.
This command may be used to provide information as to how well the script or a
tcl-command is performing and can help determine bottlenecks and fine-tune
application performance.
.TP
\fI-calibrate\fR
.
To measure very fast scripts as exact as posible the calibration process
may be required.

The \fI-calibrate\fR option is used to calibrate timerate, calculating the
estimated overhead of the given script as the default overhead for future 
invocations of the \fBtimerate\fR command. If the \fItime\fR parameter is not 
specified, the calibrate procedure runs for up to 10 seconds.
.TP
\fI-overhead double\fR
.
The \fI-overhead\fR parameter supplies an estimate (in microseconds) of the
measurement overhead of each iteration of the tested script. This quantity
will be subtracted from the measured time prior to reporting results.
.TP
\fI-direct\fR
.
The \fI-direct\fR option causes direct execution of the supplied script,
without compilation, in a manner similar to the \fBtime\fR command. It can be
used to measure the cost of \fBTcl_EvalObjEx\fR, of the invocation of canonical
lists, and of the uncompiled versions of bytecoded commands.
.PP
As opposed to the \fBtime\fR commmand, which runs the tested script for a fixed
number of iterations, the timerate command runs it for a fixed time.
Additionally, the compiled variant of the script will be used during the entire
measurement, as if the script were part of a compiled procedure, if the \fI-direct\fR
option is not specified. The fixed time period and possibility of compilation allow
for more precise results and prevent very long execution times by slow scripts, making
it practical for measuring scripts with highly uncertain execution times.

.SH EXAMPLE
Estimate how fast it takes for a simple Tcl \fBfor\fR loop (including
operations on variable \fIi\fR) to count to a ten:
.PP
.CS
# calibrate:
timerate -calibrate {}
# measure:
timerate { for {set i 0} {$i<10} {incr i} {} } 5000
.CE
.PP
Estimate how fast it takes for a simple Tcl \fBfor\fR loop, ignoring the
overhead for to perform ten iterations, ignoring the overhead of the management
of the variable that controls the loop:
.PP
.CS
# calibrate for overhead of variable operations:
set i 0; timerate -calibrate {expr {$i<10}; incr i} 1000
# measure:
timerate { for {set i 0} {$i<10} {incr i} {} } 5000
.CE
.PP
Estimate the speed of calculating the hour of the day using \fBclock format\fR only,
ignoring overhead of the portion of the script that prepares the time for it to
calculate:
.PP
.CS
# calibrate:
timerate -calibrate {}
# estimate overhead:
set tm 0
set ovh [lindex [timerate { incr tm [expr {24*60*60}] }] 0]
# measure using esimated overhead:
set tm 0
timerate -overhead $ovh {
    clock format $tm -format %H
    incr tm [expr {24*60*60}]; # overhead for this is ignored
} 5000
.CE
.SH "SEE ALSO"
time(n)
.SH KEYWORDS
script, timerate, time
.\" Local Variables:
.\" mode: nroff
.\" End:

    {"pwd",		Tcl_PwdObjCmd,		NULL,			NULL,	0},
    {"read",		Tcl_ReadObjCmd,		NULL,			NULL,	CMD_IS_SAFE},
    {"seek",		Tcl_SeekObjCmd,		NULL,			NULL,	CMD_IS_SAFE},
    {"socket",		Tcl_SocketObjCmd,	NULL,			NULL,	0},
    {"source",		Tcl_SourceObjCmd,	NULL,			TclNRSourceObjCmd,	0},
    {"tell",		Tcl_TellObjCmd,		NULL,			NULL,	CMD_IS_SAFE},
    {"time",		Tcl_TimeObjCmd,		NULL,			NULL,	CMD_IS_SAFE},
    {"timerate",	Tcl_TimeRateObjCmd,	NULL,			NULL,	CMD_IS_SAFE},
    {"unload",		Tcl_UnloadObjCmd,	NULL,			NULL,	0},
    {"update",		Tcl_UpdateObjCmd,	NULL,			NULL,	CMD_IS_SAFE},
    {"vwait",		Tcl_VwaitObjCmd,	NULL,			NULL,	CMD_IS_SAFE},
    {NULL,		NULL,			NULL,			NULL,	0}
};

/*

{
    Interp *iPtr;
    Tcl_Interp *interp;
    Command *cmdPtr;
    const BuiltinFuncDef *builtinFuncPtr;
    const OpCmdInfo *opcmdInfoPtr;
    const CmdInfo *cmdInfoPtr;
    Tcl_Namespace *nsPtr;
    Tcl_HashEntry *hPtr;
    int isNew;
    CancelInfo *cancelInfo;
    union {
	char c[sizeof(short)];
	short s;
    } order;

    Tcl_CreateObjCommand(interp, "::tcl::dtrace", DTraceObjCmd, NULL, NULL);
#endif /* USE_DTRACE */

    /*
     * Register the builtin math functions.
     */

    nsPtr = Tcl_CreateNamespace(interp, "::tcl::mathfunc", NULL,NULL);
    if (nsPtr == NULL) {
	Tcl_Panic("Can't create math function namespace");
    }
#define MATH_FUNC_PREFIX_LEN 17 /* == strlen("::tcl::mathfunc::") */
    memcpy(mathFuncName, "::tcl::mathfunc::", MATH_FUNC_PREFIX_LEN);
    for (builtinFuncPtr = BuiltinFuncTable; builtinFuncPtr->name != NULL;
	    builtinFuncPtr++) {
	strcpy(mathFuncName+MATH_FUNC_PREFIX_LEN, builtinFuncPtr->name);
	Tcl_CreateObjCommand(interp, mathFuncName,
		builtinFuncPtr->objCmdProc, builtinFuncPtr->clientData, NULL);
	Tcl_Export(interp, nsPtr, builtinFuncPtr->name, 0);
    }

    /*
     * Register the mathematical "operator" commands. [TIP #174]
     */

    nsPtr = Tcl_CreateNamespace(interp, "::tcl::mathop", NULL, NULL);
    if (nsPtr == NULL) {
	Tcl_Panic("can't create math operator namespace");
    }
    Tcl_Export(interp, nsPtr, "*", 1);
#define MATH_OP_PREFIX_LEN 15 /* == strlen("::tcl::mathop::") */
    memcpy(mathFuncName, "::tcl::mathop::", MATH_OP_PREFIX_LEN);
    for (opcmdInfoPtr=mathOpCmds ; opcmdInfoPtr->name!=NULL ; opcmdInfoPtr++){
	TclOpCmdClientData *occdPtr = Tcl_Alloc(sizeof(TclOpCmdClientData));

	occdPtr->op = opcmdInfoPtr->name;
	occdPtr->i.numArgs = opcmdInfoPtr->i.numArgs;

	    }
	}
	break;
    case TCL_NUMBER_BIG:
	if (Tcl_GetBignumFromObj(interp, objv[1], &big) != TCL_OK) {
	    return TCL_ERROR;
	}
	if (mp_isneg(&big)) {
	    mp_clear(&big);
	    goto negarg;
	}
	break;
    default:
	if (TclGetWideIntFromObj(interp, objv[1], &w) != TCL_OK) {
	    return TCL_ERROR;

#ifdef TCL_WIDE_CLICKS
	clicks = TclpGetWideClicks();
#else
	clicks = (Tcl_WideInt) TclpGetClicks();
#endif
	break;
    case CLICKS_MICROS:

	clicks = TclpGetMicroseconds();
	break;
    }

    Tcl_SetObjResult(interp, Tcl_NewWideIntObj(clicks));
    return TCL_OK;
}


int
ClockMicrosecondsObjCmd(
    ClientData clientData,	/* Client data is unused */
    Tcl_Interp *interp,		/* Tcl interpreter */
    int objc,			/* Parameter count */
    Tcl_Obj *const *objv)	/* Parameter values */
{


    if (objc != 1) {
	Tcl_WrongNumArgs(interp, 1, objv, NULL);
	return TCL_ERROR;
    }

    Tcl_SetObjResult(interp, Tcl_NewWideIntObj(TclpGetMicroseconds()));

    return TCL_OK;
}

/*
 *-----------------------------------------------------------------------------
 *
 * ClockParseformatargsObjCmd --

 * Copyright (c) 2003-2009 Donal K. Fellows.
 *
 * See the file "license.terms" for information on usage and redistribution of
 * this file, and for a DISCLAIMER OF ALL WARRANTIES.
 */

#include "tclInt.h"
#include "tclCompile.h"
#include "tclRegexp.h"
#include "tclStringTrim.h"

static inline Tcl_Obj *	During(Tcl_Interp *interp, int resultCode,
			    Tcl_Obj *oldOptions, Tcl_Obj *errorInfo);
static Tcl_NRPostProc	SwitchPostProc;
static Tcl_NRPostProc	TryPostBody;

    i = count;
#ifndef TCL_WIDE_CLICKS
    Tcl_GetTime(&start);
#else
    start = TclpGetWideClicks();
#endif
    while (i-- > 0) {
	result = TclEvalObjEx(interp, objPtr, 0, NULL, 0);
	if (result != TCL_OK) {
	    return result;
	}
    }
#ifndef TCL_WIDE_CLICKS
    Tcl_GetTime(&stop);
    totalMicroSec = ((double) (stop.sec - start.sec)) * 1.0e6

    TclNewLiteralStringObj(objs[1], "microseconds");
    TclNewLiteralStringObj(objs[2], "per");
    TclNewLiteralStringObj(objs[3], "iteration");
    Tcl_SetObjResult(interp, Tcl_NewListObj(4, objs));

    return TCL_OK;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_TimeRateObjCmd --
 *
 *	This object-based procedure is invoked to process the "timerate" Tcl
 *	command. 
 *	This is similar to command "time", except the execution limited by 
 *	given time (in milliseconds) instead of repetition count.
 *
 * Example:
 *	timerate {after 5} 1000 ; # equivalent for `time {after 5} [expr 1000/5]`
 *
 * Results:
 *	A standard Tcl object result.
 *
 * Side effects:
 *	See the user documentation.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_TimeRateObjCmd(
    ClientData dummy,		/* Not used. */
    Tcl_Interp *interp,		/* Current interpreter. */
    int objc,			/* Number of arguments. */
    Tcl_Obj *const objv[])	/* Argument objects. */
{
    static 
    double measureOverhead = 0; /* global measure-overhead */
    double overhead = -1;	/* given measure-overhead */
    register Tcl_Obj *objPtr;
    register int result, i;
    Tcl_Obj *calibrate = NULL, *direct = NULL;
    Tcl_WideUInt count = 0;	/* Holds repetition count */
    Tcl_WideInt  maxms  = WIDE_MIN;
				/* Maximal running time (in milliseconds) */
    Tcl_WideUInt maxcnt = WIDE_MAX;
				/* Maximal count of iterations. */
    Tcl_WideUInt threshold = 1;	/* Current threshold for check time (faster
				 * repeat count without time check) */
    Tcl_WideUInt maxIterTm = 1;	/* Max time of some iteration as max threshold
				 * additionally avoid divide to zero (never < 1) */
    unsigned short factor = 50;	/* Factor (4..50) limiting threshold to avoid
				 * growth of execution time. */
    register Tcl_WideInt start, middle, stop;
#ifndef TCL_WIDE_CLICKS
    Tcl_Time now;
#endif

    static const char *const options[] = {
	"-direct",	"-overhead",	"-calibrate",	"--",	NULL
    };
    enum options {
	TMRT_EV_DIRECT,	TMRT_OVERHEAD,	TMRT_CALIBRATE,	TMRT_LAST
    };

    NRE_callback *rootPtr;
    ByteCode	 *codePtr = NULL;

    for (i = 1; i < objc - 1; i++) {
    	int index;
	if (Tcl_GetIndexFromObj(NULL, objv[i], options, "option", TCL_EXACT,
		&index) != TCL_OK) {
	    break;
	}
	if (index == TMRT_LAST) {
	    i++;
	    break;
	}
	switch (index) {
	case TMRT_EV_DIRECT:
	    direct = objv[i];
	    break;
	case TMRT_OVERHEAD:
	    if (++i >= objc - 1) {
		goto usage;
	    }
	    if (Tcl_GetDoubleFromObj(interp, objv[i], &overhead) != TCL_OK) {
		return TCL_ERROR;
	    }
	    break;
	case TMRT_CALIBRATE:
	    calibrate = objv[i];
	    break;
	}
    }

    if (i >= objc || i < objc-3) {
usage:
	Tcl_WrongNumArgs(interp, 1, objv, "?-direct? ?-calibrate? ?-overhead double? command ?time ?max-count??");
	return TCL_ERROR;
    }
    objPtr = objv[i++];
    if (i < objc) {	/* max-time */
	result = Tcl_GetWideIntFromObj(interp, objv[i++], &maxms);
	if (result != TCL_OK) {
	    return result;
	}
	if (i < objc) {	/* max-count*/
	    Tcl_WideInt v;
	    result = Tcl_GetWideIntFromObj(interp, objv[i], &v);
	    if (result != TCL_OK) {
		return result;
	    }
	    maxcnt = (v > 0) ? v : 0;
	}
    }

    /* if calibrate */
    if (calibrate) {

	/* if no time specified for the calibration */
	if (maxms == WIDE_MIN) {
	    Tcl_Obj *clobjv[6];
	    Tcl_WideInt maxCalTime = 5000;
	    double lastMeasureOverhead = measureOverhead;
	    
	    clobjv[0] = objv[0]; 
	    i = 1;
	    if (direct) {
	    	clobjv[i++] = direct;
	    }
	    clobjv[i++] = objPtr; 

	    /* reset last measurement overhead */
	    measureOverhead = (double)0;

	    /* self-call with 100 milliseconds to warm-up,
	     * before entering the calibration cycle */
	    TclNewIntObj(clobjv[i], 100);
	    Tcl_IncrRefCount(clobjv[i]);
	    result = Tcl_TimeRateObjCmd(dummy, interp, i+1, clobjv);
	    Tcl_DecrRefCount(clobjv[i]);
	    if (result != TCL_OK) {
		return result;
	    }

	    i--;
	    clobjv[i++] = calibrate;
	    clobjv[i++] = objPtr; 

	    /* set last measurement overhead to max */
	    measureOverhead = (double)UWIDE_MAX;

	    /* calibration cycle until it'll be preciser */
	    maxms = -1000;
	    do {
		lastMeasureOverhead = measureOverhead;
		TclNewIntObj(clobjv[i], (int)maxms);
		Tcl_IncrRefCount(clobjv[i]);
		result = Tcl_TimeRateObjCmd(dummy, interp, i+1, clobjv);
		Tcl_DecrRefCount(clobjv[i]);
		if (result != TCL_OK) {
		    return result;
		}
		maxCalTime += maxms;
		/* increase maxms for preciser calibration */
		maxms -= (-maxms / 4);
		/* as long as new value more as 0.05% better */
	    } while ( (measureOverhead >= lastMeasureOverhead
		    || measureOverhead / lastMeasureOverhead <= 0.9995)
		    && maxCalTime > 0
	    );

	    return result;
	}
	if (maxms == 0) {
	    /* reset last measurement overhead */
	    measureOverhead = 0;
	    Tcl_SetObjResult(interp, Tcl_NewLongObj(0));
	    return TCL_OK;
	}

	/* if time is negative - make current overhead more precise */
	if (maxms > 0) {
	    /* set last measurement overhead to max */
	    measureOverhead = (double)UWIDE_MAX;
	} else {
	    maxms = -maxms;
	}

    }

    if (maxms == WIDE_MIN) {
    	maxms = 1000;
    }
    if (overhead == -1) {
	overhead = measureOverhead;
    }

    /* be sure that resetting of result will not smudge the further measurement */
    Tcl_ResetResult(interp);

    /* compile object */
    if (!direct) {
	if (TclInterpReady(interp) != TCL_OK) {
	    return TCL_ERROR;
	}
	codePtr = TclCompileObj(interp, objPtr, NULL, 0);
	TclPreserveByteCode(codePtr);
    }

    /* get start and stop time */
#ifdef TCL_WIDE_CLICKS
    start = middle = TclpGetWideClicks();
    /* time to stop execution (in wide clicks) */
    stop = start + (maxms * 1000 / TclpWideClickInMicrosec());
#else
    Tcl_GetTime(&now);
    start = now.sec; start *= 1000000; start += now.usec;
    middle = start;
    /* time to stop execution (in microsecs) */
    stop = start + maxms * 1000;
#endif

    /* start measurement */
    if (maxcnt > 0)
    while (1) {
    	/* eval single iteration */
    	count++;

	if (!direct) {
	    /* precompiled */
	    rootPtr = TOP_CB(interp);
	    result = TclNRExecuteByteCode(interp, codePtr);
	    result = TclNRRunCallbacks(interp, result, rootPtr);
	} else {
	    /* eval */
	    result = TclEvalObjEx(interp, objPtr, 0, NULL, 0);
	}
	if (result != TCL_OK) {
	    /* allow break from measurement cycle (used for conditional stop) */
	    if (result != TCL_BREAK) {
		goto done;
	    }
	    /* force stop immediately */
	    threshold = 1;
	    maxcnt = 0;
	    result = TCL_OK;
	}
	
	/* don't check time up to threshold */
	if (--threshold > 0) continue;

	/* check stop time reached, estimate new threshold */
    #ifdef TCL_WIDE_CLICKS
	middle = TclpGetWideClicks();
    #else
	Tcl_GetTime(&now);
	middle = now.sec; middle *= 1000000; middle += now.usec;
    #endif
	if (middle >= stop || count >= maxcnt) {
	    break;
	}

	/* don't calculate threshold by few iterations, because sometimes first
	 * iteration(s) can be too fast or slow (cached, delayed clean up, etc) */
	if (count < 10) {
	   threshold = 1; continue;
	}

	/* average iteration time in microsecs */
	threshold = (middle - start) / count;
	if (threshold > maxIterTm) {
	    maxIterTm = threshold;
	    /* interations seems to be longer */
	    if (threshold > (maxIterTm * 2)) {
		if ((factor *= 2) > 50) factor = 50;
	    } else {
		if (factor < 50) factor++;
	    }
	} else if (factor > 4) {
	    /* interations seems to be shorter */
	    if (threshold < (maxIterTm / 2)) {
		if ((factor /= 2) < 4) factor = 4;
	    } else {
		factor--;
	    }
	}
	/* as relation between remaining time and time since last check,
	 * maximal some % of time (by factor), so avoid growing of the execution time
	 * if iterations are not consistent, e. g. wax continuously on time) */
	threshold = ((stop - middle) / maxIterTm) / factor + 1;
	if (threshold > 100000) {	    /* fix for too large threshold */
	    threshold = 100000;
	}
	/* consider max-count */
	if (threshold > maxcnt - count) {
	    threshold = maxcnt - count;
	}
    }

    {
	Tcl_Obj *objarr[8], **objs = objarr;
	Tcl_WideInt val;
	const char *fmt;

	middle -= start;		     /* execution time in microsecs */

    #ifdef TCL_WIDE_CLICKS
	/* convert execution time in wide clicks to microsecs */
	middle *= TclpWideClickInMicrosec();
    #endif

	/* if not calibrate */
	if (!calibrate) {
	    /* minimize influence of measurement overhead */
	    if (overhead > 0) {
		/* estimate the time of overhead (microsecs) */
		Tcl_WideUInt curOverhead = overhead * count;
		if (middle > (Tcl_WideInt)curOverhead) {
		    middle -= curOverhead;
		} else {
		    middle = 0;
		}
	    }
	} else {
	    /* calibration - obtaining new measurement overhead */
	    if (measureOverhead > (double)middle / count) {
		measureOverhead = (double)middle / count;
	    }
	    objs[0] = Tcl_NewDoubleObj(measureOverhead);
	    TclNewLiteralStringObj(objs[1], "\xC2\xB5s/#-overhead"); /* mics */
	    objs += 2;
	}

	val = middle / count;		     /* microsecs per iteration */
	if (val >= 1000000) {
	    objs[0] = Tcl_NewWideIntObj(val);
	} else {
	    if (val < 10)    { fmt = "%.6f"; } else
	    if (val < 100)   { fmt = "%.4f"; } else
	    if (val < 1000)  { fmt = "%.3f"; } else
	    if (val < 10000) { fmt = "%.2f"; } else
			     { fmt = "%.1f"; };
	    objs[0] = Tcl_ObjPrintf(fmt, ((double)middle)/count);
	}

	objs[2] = Tcl_NewWideIntObj(count); /* iterations */
	
	/* calculate speed as rate (count) per sec */
	if (!middle) middle++; /* +1 ms, just to avoid divide by zero */
	if (count < (WIDE_MAX / 1000000)) {
	    val = (count * 1000000) / middle;
	    if (val < 100000) {
		if (val < 100)	{ fmt = "%.3f"; } else
		if (val < 1000) { fmt = "%.2f"; } else
				{ fmt = "%.1f"; };
		objs[4] = Tcl_ObjPrintf(fmt, ((double)(count * 1000000)) / middle);
	    } else {
		objs[4] = Tcl_NewWideIntObj(val);
	    }
	} else {
	    objs[4] = Tcl_NewWideIntObj((count / middle) * 1000000);
	}

	/* estimated net execution time (in millisecs) */
	if (!calibrate) {
	    objs[6] = Tcl_ObjPrintf("%.3f", (double)middle / 1000);
	    TclNewLiteralStringObj(objs[7], "nett-ms");
	}

	/*
	* Construct the result as a list because many programs have always parsed
	* as such (extracting the first element, typically).
	*/

	TclNewLiteralStringObj(objs[1], "\xC2\xB5s/#"); /* mics/# */
	TclNewLiteralStringObj(objs[3], "#");
	TclNewLiteralStringObj(objs[5], "#/sec");
	Tcl_SetObjResult(interp, Tcl_NewListObj(8, objarr));
    }

done:

    if (codePtr != NULL) {
	TclReleaseByteCode(codePtr);
    }

    return result;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_TryObjCmd, TclNRTryObjCmd --
 *
 *	This procedure is invoked to process the "try" Tcl command. See the

				/* Procedure that unloads a loaded module */
};

/* Flags for conversion of doubles to digit strings */

#define TCL_DD_SHORTEST 		0x4
				/* Use the shortest possible string */


#define TCL_DD_E_FORMAT 		0x2
				/* Use a fixed-length string of digits,
				 * suitable for E format*/
#define TCL_DD_F_FORMAT 		0x3
				/* Use a fixed number of digits after the
				 * decimal point, suitable for F format */

#define TCL_DD_SHORTEN_FLAG 		0x4
				/* Allow return of a shorter digit string
				 * if it converts losslessly */
#define TCL_DD_NO_QUICK 		0x8
				/* Debug flag: forbid quick FP conversion */

#define TCL_DD_CONVERSION_TYPE_MASK	0x3
				/* Mask to isolate the conversion type */





/*
 *----------------------------------------------------------------
 * Procedures shared among Tcl modules but not used by the outside world:
 *----------------------------------------------------------------
 */

MODULE_SCOPE int	TclpLoadMemory(Tcl_Interp *interp, void *buffer,
			    int size, int codeSize, Tcl_LoadHandle *loadHandle,
			    Tcl_FSUnloadFileProc **unloadProcPtr, int flags);
#endif
MODULE_SCOPE void	TclInitThreadStorage(void);
MODULE_SCOPE void	TclFinalizeThreadDataThread(void);
MODULE_SCOPE void	TclFinalizeThreadStorage(void);

#ifdef TCL_WIDE_CLICKS
MODULE_SCOPE Tcl_WideInt TclpGetWideClicks(void);
MODULE_SCOPE double	TclpWideClicksToNanoseconds(Tcl_WideInt clicks);
MODULE_SCOPE double	TclpWideClickInMicrosec(void);
#else
#   ifdef _WIN32
#	define TCL_WIDE_CLICKS 1
MODULE_SCOPE Tcl_WideInt TclpGetWideClicks(void);
MODULE_SCOPE double	TclpWideClickInMicrosec(void);
#	define		TclpWideClicksToNanoseconds(clicks) \
				((double)(clicks) * TclpWideClickInMicrosec() * 1000)
#   endif
#endif
MODULE_SCOPE Tcl_WideInt TclpGetMicroseconds(void);

MODULE_SCOPE int	TclZlibInit(Tcl_Interp *interp);
MODULE_SCOPE void *	TclpThreadCreateKey(void);
MODULE_SCOPE void	TclpThreadDeleteKey(void *keyPtr);
MODULE_SCOPE void	TclpThreadSetMasterTSD(void *tsdKeyPtr, void *ptr);
MODULE_SCOPE void *	TclpThreadGetMasterTSD(void *tsdKeyPtr);
MODULE_SCOPE void	TclErrorStackResetIf(Tcl_Interp *interp,
			    const char *msg, size_t length);

MODULE_SCOPE int	Tcl_TellObjCmd(void *clientData,
			    Tcl_Interp *interp, int objc,
			    Tcl_Obj *const objv[]);
MODULE_SCOPE int	Tcl_ThrowObjCmd(void *dummy, Tcl_Interp *interp,
			    int objc, Tcl_Obj *const objv[]);
MODULE_SCOPE int	Tcl_TimeObjCmd(void *clientData,
			    Tcl_Interp *interp, int objc,
			    Tcl_Obj *const objv[]);
MODULE_SCOPE int	Tcl_TimeRateObjCmd(void *clientData,
			    Tcl_Interp *interp, int objc,
			    Tcl_Obj *const objv[]);
MODULE_SCOPE int	Tcl_TraceObjCmd(void *clientData,
			    Tcl_Interp *interp, int objc,
			    Tcl_Obj *const objv[]);
MODULE_SCOPE int	Tcl_TryObjCmd(void *clientData,
			    Tcl_Interp *interp, int objc,
			    Tcl_Obj *const objv[]);

	     * Must check for those bignum values that can fit in a long, even
	     * when auto-narrowing is enabled. Only those values in the signed
	     * long range get auto-narrowed to tclIntType, while all the
	     * values in the unsigned long range will fit in a long.
	     */

	    mp_int big;




	    unsigned long scratch, value = 0, numBytes = sizeof(unsigned long);
	    unsigned char *bytes = (unsigned char *) &scratch;

	    UNPACK_BIGNUM(objPtr, big);
	    if (mp_to_unsigned_bin_n(&big, bytes, &numBytes) == MP_OKAY) {
		while (numBytes-- > 0) {
			value = (value << CHAR_BIT) | *bytes++;
		}
		if (big.sign) {
		    if (value <= 1 + (unsigned long)LONG_MAX) {
			*longPtr = - (long) value;
			return TCL_OK;
		    }
		} else {
		    if (value <= (unsigned long)ULONG_MAX) {
			*longPtr = (long) value;
			return TCL_OK;

		    }
		}
	    }
#ifndef TCL_WIDE_INT_IS_LONG
	tooLarge:
#endif
	    if (interp != NULL) {

	if (objPtr->typePtr == &tclBignumType) {
	    /*
	     * Must check for those bignum values that can fit in a
	     * Tcl_WideInt, even when auto-narrowing is enabled.
	     */

	    mp_int big;




	    Tcl_WideUInt value = 0;
	    unsigned long numBytes = sizeof(Tcl_WideInt);
	    Tcl_WideInt scratch;
	    unsigned char *bytes = (unsigned char *) &scratch;

	    UNPACK_BIGNUM(objPtr, big);
	    if (mp_to_unsigned_bin_n(&big, bytes, &numBytes) == MP_OKAY) {
		while (numBytes-- > 0) {
		    value = (value << CHAR_BIT) | *bytes++;
		}
		if (big.sign) {
		    if (value <= 1 + ~(Tcl_WideUInt)WIDE_MIN) {
			*wideIntPtr = - (Tcl_WideInt) value;
			return TCL_OK;
		    }
		} else {
		    if (value <= (Tcl_WideUInt)WIDE_MAX) {
			*wideIntPtr = (Tcl_WideInt) value;
			return TCL_OK;

		    }
		}
	    }
	    if (interp != NULL) {
		const char *s = "integer value too large to represent";
		Tcl_Obj *msg = Tcl_NewStringObj(s, -1);

 */

void
Tcl_SetBignumObj(
    Tcl_Obj *objPtr,		/* Object to set */
    mp_int *bignumValue)	/* Value to store */
{





    Tcl_WideUInt value = 0;
    unsigned long numBytes = sizeof(Tcl_WideUInt);
    Tcl_WideUInt scratch;
    unsigned char *bytes = (unsigned char *) &scratch;

    if (Tcl_IsShared(objPtr)) {
	Tcl_Panic("%s called with shared object", "Tcl_SetBignumObj");
    }
    if (mp_to_unsigned_bin_n(bignumValue, bytes, &numBytes) != MP_OKAY) {
	goto tooLargeForWide;
    }
    while (numBytes-- > 0) {
	value = (value << CHAR_BIT) | *bytes++;
    }
    if (value > ((Tcl_WideUInt)WIDE_MAX + bignumValue->sign)) {
	goto tooLargeForWide;
    }
    if (bignumValue->sign) {
	TclSetIntObj(objPtr, -(Tcl_WideInt)value);
    } else {
	TclSetIntObj(objPtr, (Tcl_WideInt)value);
    }
    mp_clear(bignumValue);
    return;

  tooLargeForWide:
    TclInvalidateStringRep(objPtr);
    TclFreeIntRep(objPtr);
    TclSetBignumIntRep(objPtr, bignumValue);
}

/*

#   include "tclWinPort.h"
#else
#   include "tclUnixPort.h"
#endif
#include "tcl.h"

#define UWIDE_MAX ((Tcl_WideUInt)-1)
#define WIDE_MAX ((Tcl_WideInt)(UWIDE_MAX >> 1))
#define WIDE_MIN ((Tcl_WideInt)((Tcl_WideUInt)WIDE_MAX+1))
#define UWIDE_MAX ((Tcl_WideUInt)-1)
#define WIDE_MAX ((Tcl_WideInt)(UWIDE_MAX >> 1))
#define WIDE_MIN ((Tcl_WideInt)((Tcl_WideUInt)WIDE_MAX+1))

#endif /* _TCLPORT */

static char *		ShorteningQuickFormat(double, int, int, double,
			    char *, int *);
static char *		StrictQuickFormat(double, int, int, double,
			    char *, int *);
static char *		QuickConversion(double, int, int, int, int, int, int,
			    int *, char **);
static void		CastOutPowersOf2(int *, int *, int *);
static char *		ShorteningInt64Conversion(Double *, Tcl_WideUInt,
			    int, int, int, int, int, int, int, int, int,
			    int, int, int *, char **);
static char *		StrictInt64Conversion(Double *, Tcl_WideUInt,
			    int, int, int, int, int, int,
			    int, int, int *, char **);
static int		ShouldBankerRoundUpPowD(mp_int *, int, int);
static int		ShouldBankerRoundUpToNextPowD(mp_int *, mp_int *,
			    int, int, mp_int *);
static char *		ShorteningBignumConversionPowD(Double *dPtr,
			    Tcl_WideUInt bw, int b2, int b5,
			    int m2plus, int m2minus, int m5,
			    int sd, int k, int len,
			    int ilim, int ilim1, int *decpt,
			    char **endPtr);
static char *		StrictBignumConversionPowD(Double *dPtr,
			    Tcl_WideUInt bw, int b2, int b5,
			    int sd, int k, int len,
			    int ilim, int ilim1, int *decpt,
			    char **endPtr);
static int		ShouldBankerRoundUp(mp_int *, mp_int *, int);
static int		ShouldBankerRoundUpToNext(mp_int *, mp_int *,
			    mp_int *, int);
static char *		ShorteningBignumConversion(Double *dPtr,
			    Tcl_WideUInt bw, int b2,
			    int m2plus, int m2minus,
			    int s2, int s5, int k, int len,
			    int ilim, int ilim1, int *decpt,
			    char **endPtr);
static char *		StrictBignumConversion(Double *dPtr,
			    Tcl_WideUInt bw, int b2,
			    int s2, int s5, int k, int len,
			    int ilim, int ilim1, int *decpt,
			    char **endPtr);
static double		BignumToBiasedFrExp(const mp_int *big, int *machexp);
static double		Pow10TimesFrExp(int exponent, double fraction,
			    int *machexp);

 *	one too high.
 *
 *----------------------------------------------------------------------
 */

static inline void
SetPrecisionLimits(
    int flags,		/* Type of conversion: TCL_DD_SHORTEST,
				 * TCL_DD_E_FMT, TCL_DD_F_FMT. */

    int k,			/* Floor(log10(number to convert)) */
    int *ndigitsPtr,		/* IN/OUT: Number of digits requested (will be
				 *         adjusted if needed). */
    int *iPtr,			/* OUT: Maximum number of digits to return. */
    int *iLimPtr,		/* OUT: Number of digits of significance if
				 *      the bignum method is used.*/
    int *iLim1Ptr)		/* OUT: Number of digits of significance if
				 *      the quick method is used. */
{
    switch (flags & TCL_DD_CONVERSION_TYPE_MASK) {






    case TCL_DD_E_FORMAT:
	if (*ndigitsPtr <= 0) {
	    *ndigitsPtr = 1;
	}
	*iLimPtr = *iLim1Ptr = *iPtr = *ndigitsPtr;
	break;
    case TCL_DD_F_FORMAT:
	*iPtr = *ndigitsPtr + k + 1;
	*iLimPtr = *iPtr;
	*iLim1Ptr = *iPtr - 1;
	if (*iPtr <= 0) {
	    *iPtr = 1;
	}
	break;
    default:
	*iLimPtr = *iLim1Ptr = -1;
	*iPtr = 18;
	*ndigitsPtr = 0;
	break;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * BumpUp --

 *
 *----------------------------------------------------------------------
 */

static inline char *
ShorteningInt64Conversion(
    Double *dPtr,		/* Original number to convert. */


    Tcl_WideUInt bw,		/* Integer significand. */
    int b2, int b5,		/* Scale factor for the significand in the
				 * numerator. */
    int m2plus, int m2minus, int m5,
				/* Scale factors for 1/2 ulp in the numerator
				 * (will be different if bw == 1. */
    int s2, int s5,		/* Scale factors for the denominator. */

	/*
	 * Does the current digit put us on the low side of the exact value
	 * but within within roundoff of being exact?
	 */

	if (b < mplus || (b == mplus
		&& (dPtr->w.word1 & 1) == 0)) {
	    /*
	     * Make sure we shouldn't be rounding *up* instead, in case the
	     * next number above is closer.
	     */

	    if (2 * b > S || (2 * b == S && (digit & 1) != 0)) {
		++digit;

	/*
	 * Does one plus the current digit put us within roundoff of the
	 * number?
	 */

	if (b > S - mminus || (b == S - mminus
		&& (dPtr->w.word1 & 1) == 0)) {
	    if (digit == 9) {
		*s++ = '9';
		s = BumpUp(s, retval, &k);
		break;
	    }
	    ++digit;
	    *s++ = '0' + digit;

 *
 *----------------------------------------------------------------------
 */

static inline char *
StrictInt64Conversion(
    Double *dPtr,		/* Original number to convert. */


    Tcl_WideUInt bw,		/* Integer significand. */
    int b2, int b5,		/* Scale factor for the significand in the
				 * numerator. */
    int s2, int s5,		/* Scale factors for the denominator. */
    int k,			/* Number of output digits before the decimal
				 * point. */
    int len,			/* Number of digits to allocate. */

static inline int
ShouldBankerRoundUpPowD(
    mp_int *b,			/* Numerator of the fraction. */
    int sd,			/* Denominator is 2**(sd*DIGIT_BIT). */
    int isodd)			/* 1 if the digit is odd, 0 if even. */
{
    int i;
    static const mp_digit topbit = ((mp_digit)1) << (DIGIT_BIT - 1);

    if (b->used < sd || (b->dp[sd-1] & topbit) == 0) {
	return 0;
    }
    if (b->dp[sd-1] != topbit) {
	return 1;
    }

 */

static inline int
ShouldBankerRoundUpToNextPowD(
    mp_int *b,			/* Numerator of the fraction. */
    mp_int *m,			/* Numerator of the rounding tolerance. */
    int sd,			/* Common denominator is 2**(sd*DIGIT_BIT). */



    int isodd,			/* 1 if the integer significand is odd. */
    mp_int *temp)		/* Work area for the calculation. */
{
    int i;

    /*
     * Compare B and S-m - which is the same as comparing B+m and S - which we

    }
    for (i = sd-1; i >= 0; --i) {
				/* Check for ==s */
	if (temp->dp[i] != 0) {	/* > s */
	    return 1;
	}
    }




    return isodd;
}

/*
 *----------------------------------------------------------------------
 *
 * ShorteningBignumConversionPowD --

 *
 *----------------------------------------------------------------------
 */

static inline char *
ShorteningBignumConversionPowD(
    Double *dPtr,		/* Original number to convert. */


    Tcl_WideUInt bw,		/* Integer significand. */
    int b2, int b5,		/* Scale factor for the significand in the
				 * numerator. */
    int m2plus, int m2minus, int m5,
				/* Scale factors for 1/2 ulp in the numerator
				 * (will be different if bw == 1). */
    int sd,			/* Scale factor for the denominator. */

	/*
	 * Does the current digit put us on the low side of the exact value
	 * but within within roundoff of being exact?
	 */

	r1 = mp_cmp_mag(&b, (m2plus > m2minus)? &mplus : &mminus);
	if (r1 == MP_LT || (r1 == MP_EQ
		&& (dPtr->w.word1 & 1) == 0)) {
	    /*
	     * Make sure we shouldn't be rounding *up* instead, in case the
	     * next number above is closer.
	     */

	    if (ShouldBankerRoundUpPowD(&b, sd, digit&1)) {
		++digit;

	}

	/*
	 * Does one plus the current digit put us within roundoff of the
	 * number?
	 */

	if (ShouldBankerRoundUpToNextPowD(&b, &mminus, sd,
		dPtr->w.word1 & 1, &temp)) {
	    if (digit == 9) {
		*s++ = '9';
		s = BumpUp(s, retval, &k);
		break;
	    }
	    ++digit;

 *
 *----------------------------------------------------------------------
 */

static inline char *
StrictBignumConversionPowD(
    Double *dPtr,		/* Original number to convert. */


    Tcl_WideUInt bw,		/* Integer significand. */
    int b2, int b5,		/* Scale factor for the significand in the
				 * numerator. */
    int sd,			/* Scale factor for the denominator. */
    int k,			/* Number of output digits before the decimal
				 * point. */
    int len,			/* Number of digits to allocate. */
    int ilim,			/* Number of digits to convert if b >= s */
    int ilim1,			/* Number of digits to convert if b < s */
    int *decpt,			/* OUTPUT: Position of the decimal point. */
    char **endPtr)		/* OUTPUT: Position of the terminal '\0' at
				 *	   the end of the returned string. */
{
    char *retval = Tcl_Alloc(len + 1);
				/* Output buffer. */
    mp_int b;			/* Numerator of the fraction being
				 * converted. */
    mp_digit digit;		/* Current output digit. */
    char *s = retval;		/* Cursor in the output buffer. */
    int i;			/* Index in the output buffer. */


    /*
     * b = bw * 2**b2 * 5**b5
     */

    TclInitBignumFromWideUInt(&b, bw);
    MulPow5(&b, b5, &b);
    mp_mul_2d(&b, b2, &b);

    /*
     * Adjust if the logarithm was guessed wrong.
     */

    if (b.used <= sd) {
	mp_mul_d(&b, 10, &b);
	ilim = ilim1;
	--k;
    }


    /*
     * Loop through the digits. Do division and mod by s == 2**(sd*DIGIT_BIT)
     * by mp_digit extraction.
     */

    i = 1;

    }

    /*
     * Endgame - store the location of the decimal point and the end of the
     * string.
     */

    mp_clear(&b);
    *s = '\0';
    *decpt = k;
    if (endPtr) {
	*endPtr = s;
    }
    return retval;
}

static inline int
ShouldBankerRoundUpToNext(
    mp_int *b,			/* Remainder from the division that produced
				 * the last digit. */
    mp_int *m,			/* Numerator of the rounding tolerance. */
    mp_int *S,			/* Denominator. */



    int isodd)			/* 1 if the integer significand is odd. */

{
    int r;
    mp_int temp;

    /*
     * Compare b and S-m: this is the same as comparing B+m and S.
     */

    mp_init(&temp);
    mp_add(b, m, &temp);
    r = mp_cmp_mag(&temp, S);
    mp_clear(&temp);
    switch(r) {
    case MP_LT:
	return 0;
    case MP_EQ:



	return isodd;

    case MP_GT:
	return 1;
    }
    Tcl_Panic("in ShouldBankerRoundUpToNext, trichotomy fails!");
    return 0;
}


 *
 *----------------------------------------------------------------------
 */

static inline char *
ShorteningBignumConversion(
    Double *dPtr,		/* Original number being converted. */

    Tcl_WideUInt bw,		/* Integer significand and exponent. */
    int b2,			/* Scale factor for the significand. */
    int m2plus, int m2minus,	/* Scale factors for 1/2 ulp in numerator. */
    int s2, int s5,		/* Scale factors for denominator. */
    int k,			/* Guessed position of the decimal point. */
    int len,			/* Size of the digit buffer to allocate. */
    int ilim,			/* Number of digits to convert if b >= s */
    int ilim1,			/* Number of digits to convert if b < s */
    int *decpt,			/* OUTPUT: Position of the decimal point. */
    char **endPtr)		/* OUTPUT: Pointer to the end of the number */
{
    char *retval = Tcl_Alloc(len+1);
				/* Buffer of digits to return. */
    char *s = retval;		/* Cursor in the return value. */
    mp_int b;			/* Numerator of the result. */
    mp_int mminus;		/* 1/2 ulp below the result. */
    mp_int mplus;		/* 1/2 ulp above the result. */
    mp_int S;			/* Denominator of the result. */
    mp_int dig;			/* Current digit of the result. */
    int digit;			/* Current digit of the result. */

    int minit = 1;		/* Fudge factor for when we misguess k. */
    int i;
    int r1;

    /*
     * b = bw * 2**b2 * 5**b5
     * S = 2**s2 * 5*s5

    mp_init_set_int(&mminus, minit);
    mp_mul_2d(&mminus, m2minus, &mminus);
    if (m2plus > m2minus) {
	mp_init_copy(&mplus, &mminus);
	mp_mul_2d(&mplus, m2plus-m2minus, &mplus);
    }


    /*
     * Loop through the digits.
     */

    mp_init(&dig);
    i = 1;
    for (;;) {
	mp_div(&b, &S, &dig, &b);
	if (dig.used > 1 || dig.dp[0] >= 10) {
	    Tcl_Panic("wrong digit!");
	}
	digit = dig.dp[0];

	/*
	 * Does the current digit leave us with a remainder small enough to
	 * round to it?
	 */

	r1 = mp_cmp_mag(&b, (m2plus > m2minus)? &mplus : &mminus);
	if (r1 == MP_LT || (r1 == MP_EQ && (dPtr->w.word1 & 1) == 0)) {

	    mp_mul_2d(&b, 1, &b);
	    if (ShouldBankerRoundUp(&b, &S, digit&1)) {
		++digit;
		if (digit == 10) {
		    *s++ = '9';
		    s = BumpUp(s, retval, &k);
		    break;
		}
	    }
	    *s++ = '0' + digit;
	    break;
	}

	/*
	 * Does the current digit leave us with a remainder large enough to
	 * commit to rounding up to the next higher digit?
	 */

	if (ShouldBankerRoundUpToNext(&b, &mminus, &S,
		dPtr->w.word1 & 1)) {
	    ++digit;
	    if (digit == 10) {
		*s++ = '9';
		s = BumpUp(s, retval, &k);
		break;
	    }
	    *s++ = '0' + digit;

     * Endgame - store the location of the decimal point and the end of the
     * string.
     */

    if (m2plus > m2minus) {
	mp_clear(&mplus);
    }
    mp_clear_multi(&b, &mminus, &dig, &S, NULL);
    *s = '\0';
    *decpt = k;
    if (endPtr) {
	*endPtr = s;
    }
    return retval;
}

 *
 *----------------------------------------------------------------------
 */

static inline char *
StrictBignumConversion(
    Double *dPtr,		/* Original number being converted. */

    Tcl_WideUInt bw,		/* Integer significand and exponent. */
    int b2,			/* Scale factor for the significand. */
    int s2, int s5,		/* Scale factors for denominator. */
    int k,			/* Guessed position of the decimal point. */
    int len,			/* Size of the digit buffer to allocate. */
    int ilim,			/* Number of digits to convert if b >= s */
    int ilim1,			/* Number of digits to convert if b < s */
    int *decpt,			/* OUTPUT: Position of the decimal point. */
    char **endPtr)		/* OUTPUT: Pointer to the end of the number */
{
    char *retval = Tcl_Alloc(len+1);
				/* Buffer of digits to return. */
    char *s = retval;		/* Cursor in the return value. */
    mp_int b;			/* Numerator of the result. */
    mp_int S;			/* Denominator of the result. */
    mp_int dig;			/* Current digit of the result. */
    int digit;			/* Current digit of the result. */

    int g;			/* Size of the current digit ground. */
    int i, j;

    /*
     * b = bw * 2**b2 * 5**b5
     * S = 2**s2 * 5*s5
     */

    mp_init_multi(&dig, NULL);
    TclInitBignumFromWideUInt(&b, bw);
    mp_mul_2d(&b, b2, &b);
    mp_init_set_int(&S, 1);
    MulPow5(&S, s5, &S); mp_mul_2d(&S, s2, &S);

    /*
     * Handle the case where we guess the position of the decimal point wrong.

    ++s;

    /*
     * Endgame - store the location of the decimal point and the end of the
     * string.
     */

    mp_clear_multi(&b, &S, &dig, NULL);
    *s = '\0';
    *decpt = k;
    if (endPtr) {
	*endPtr = s;
    }
    return retval;
}

 * according to the 'flags' argument. Valid values for 'flags' include:
 *	TCL_DD_SHORTEST - This is the default for floating point conversion.
 *		It constructs the shortest string of
 *		digits that will reconvert to the given number when scanned.
 *		For floating point numbers that are exactly between two
 *		decimal numbers, it resolves using the 'round to even' rule.
 *		With this value, the 'ndigits' parameter is ignored.









 *	TCL_DD_E_FORMAT - This value is used to prepare numbers for %e format
 *		conversion. It constructs a string of at most 'ndigits' digits,
 *		choosing the one that is closest to the given number (and
 *		resolving ties with 'round to even').  It is allowed to return
 *		fewer than 'ndigits' if the number converts exactly; if the
 *		TCL_DD_E_FORMAT|TCL_DD_SHORTEN_FLAG is supplied instead, it
 *		also returns fewer digits if the shorter string will still

    int flags,			/* Conversion flags. */
    int *decpt,			/* OUTPUT: Position of the decimal point. */
    int *sign,			/* OUTPUT: 1 if the result is negative. */
    char **endPtr)		/* OUTPUT: If not NULL, receives a pointer to
				 *	   one character beyond the end of the
				 *	   returned string. */
{




    Double d;			/* Union for deconstructing doubles. */
    Tcl_WideUInt bw;		/* Integer significand. */
    int be;			/* Power of 2 by which b must be multiplied */
    int bbits;			/* Number of bits needed to represent b. */
    int denorm;			/* Flag == 1 iff the input number was
				 * denormalized. */
    int k;			/* Estimate of floor(log10(d)). */

    ComputeScale(be, k, &b2, &b5, &s2, &s5);

    /*
     * Correct an incorrect caller-supplied 'ndigits'.  Also determine:
     *	i = The maximum number of decimal digits that will be returned in the
     *      formatted string.  This is k + 1 + ndigits for F format, 18 for
     *      shortest, and ndigits for E format.
     *  ilim = The number of significant digits to convert if k has been
     *         guessed correctly. This is -1 for shortest (which
     *         stop when all significance has been lost), 'ndigits' for E
     *         format, and 'k + 1 + ndigits' for F format.
     *  ilim1 = The minimum number of significant digits to convert if k has
     *	        been guessed 1 too high. This, too, is -1 for shortest,
     *	        and 'ndigits' for E format, but it's 'ndigits-1' for F
     *	        format.
     */

    SetPrecisionLimits(flags, k, &ndigits, &i, &ilim, &ilim1);

    /*
     * Try to do low-precision conversion in floating point rather than
     * resorting to expensive multiprecision arithmetic.
     */

    if (ilim >= 0 && ilim <= QUICK_MAX && !(flags & TCL_DD_NO_QUICK)) {

	     * If 10*2**s2*5**s5 == 2**(s2+1)+5**(s5+1) fits in a 64-bit word,
	     * then all our intermediate calculations can be done using exact
	     * 64-bit arithmetic with no need for expensive multiprecision
	     * operations. (This will be true for all numbers in the range
	     * [1.0e-3 .. 1.0e+24]).
	     */

	    return ShorteningInt64Conversion(&d, bw, b2, b5, m2plus,
		    m2minus, m5, s2, s5, k, len, ilim, ilim1, decpt, endPtr);
	} else if (s5 == 0) {
	    /*
	     * The denominator is a power of 2, so we can replace division by
	     * digit shifts. First we round up s2 to a multiple of DIGIT_BIT,
	     * and adjust m2 and b2 accordingly. Then we launch into a version
	     * of the comparison that's specialized for the 'power of mp_digit
	     * in the denominator' case.
	     */

	    if (s2 % DIGIT_BIT != 0) {
		int delta = DIGIT_BIT - (s2 % DIGIT_BIT);

		b2 += delta;
		m2plus += delta;
		m2minus += delta;
		s2 += delta;
	    }
	    return ShorteningBignumConversionPowD(&d, bw, b2, b5,
		    m2plus, m2minus, m5, s2/DIGIT_BIT, k, len, ilim, ilim1,
		    decpt, endPtr);
	} else {
	    /*
	     * Alas, there's no helpful special case; use full-up bignum
	     * arithmetic for the conversion.
	     */

	    return ShorteningBignumConversion(&d, bw, b2, m2plus,
		    m2minus, s2, s5, k, len, ilim, ilim1, decpt, endPtr);
	}
    } else {
	/*
	 * Non-shortening conversion.
	 */

	    /*
	     * If 10*2**s2*5**s5 == 2**(s2+1)+5**(s5+1) fits in a 64-bit word,
	     * then all our intermediate calculations can be done using exact
	     * 64-bit arithmetic with no need for expensive multiprecision
	     * operations.
	     */

	    return StrictInt64Conversion(&d, bw, b2, b5, s2, s5, k,
		    len, ilim, ilim1, decpt, endPtr);
	} else if (s5 == 0) {
	    /*
	     * The denominator is a power of 2, so we can replace division by
	     * digit shifts. First we round up s2 to a multiple of DIGIT_BIT,
	     * and adjust m2 and b2 accordingly. Then we launch into a version
	     * of the comparison that's specialized for the 'power of mp_digit
	     * in the denominator' case.
	     */

	    if (s2 % DIGIT_BIT != 0) {
		int delta = DIGIT_BIT - (s2 % DIGIT_BIT);

		b2 += delta;
		s2 += delta;
	    }
	    return StrictBignumConversionPowD(&d, bw, b2, b5,
		    s2/DIGIT_BIT, k, len, ilim, ilim1, decpt, endPtr);
	} else {
	    /*
	     * There are no helpful special cases, but at least we know in
	     * advance how many digits we will convert. We can run the
	     * conversion in steps of DIGIT_GROUP digits, so as to have many
	     * fewer mp_int divisions.
	     */

	    return StrictBignumConversion(&d, bw, b2, s2, s5, k,
		    len, ilim, ilim1, decpt, endPtr);
	}
    }
}

/*
 *----------------------------------------------------------------------

 *
 * Usage:
 *	testdoubledigits fpval ndigits type ?shorten"
 *
 * Parameters:
 *	fpval - Floating-point value to format.
 *	ndigits - Digit count to request from Tcl_DoubleDigits
 *	type - One of 'shortest', 'e', 'f'
 *	shorten - Indicates that the 'shorten' flag should be passed in.
 *
 *-----------------------------------------------------------------------------
 */

static int
TestdoubledigitsObjCmd(void *unused,
				/* NULL */
		       Tcl_Interp* interp,
				/* Tcl interpreter */
		       int objc,
				/* Parameter count */
		       Tcl_Obj* const objv[])
				/* Parameter vector */
{
    static const char* options[] = {
	"shortest",

	"e",
	"f",
	NULL
    };
    static const int types[] = {
	TCL_DD_SHORTEST,

	TCL_DD_E_FORMAT,
	TCL_DD_F_FORMAT
    };

    const Tcl_ObjType* doubleType;
    double d;
    int status;

		 * quickly if the next char in the pattern isn't a special
		 * character
		 */

		if ((p != '[') && (p != '?') && (p != '\\')) {
		    if (nocase) {
			while (*uniStr && (p != *uniStr)
				&& (p != (Tcl_UniChar)Tcl_UniCharToLower(*uniStr))) {
			    uniStr++;
			}
		    } else {
			while (*uniStr && (p != *uniStr)) {
			    uniStr++;
			}
		    }

		 * quickly if the next char in the pattern isn't a special
		 * character.
		 */

		if ((p != '[') && (p != '?') && (p != '\\')) {
		    if (nocase) {
			while ((string < stringEnd) && (p != *string)
				&& (p != (Tcl_UniChar)Tcl_UniCharToLower(*string))) {
			    string++;
			}
		    } else {
			while ((string < stringEnd) && (p != *string)) {
			    string++;
			}
		    }

#!/usr/bin/tclsh
# ------------------------------------------------------------------------
#
# test-performance.tcl --
# 
#  This file provides common performance tests for comparison of tcl-speed
#  degradation by switching between branches.
#  (currently for clock ensemble only)
#
# ------------------------------------------------------------------------
# 
# Copyright (c) 2014 Serg G. Brester (aka sebres)
# 
# See the file "license.terms" for information on usage and redistribution
# of this file.
# 

array set in {-time 500}
if {[info exists ::argv0] && [file tail $::argv0] eq [file tail [info script]]} {
  array set in $argv
}

## common test performance framework:
if {![namespace exists ::tclTestPerf]} {
  source [file join [file dirname [info script]] test-performance.tcl]
}

namespace eval ::tclTestPerf-TclClock {

namespace path {::tclTestPerf}

## set testing defaults:
set ::env(TCL_TZ) :CET

# warm-up interpeter compiler env, clock platform-related features:

## warm-up test-related features (load clock.tcl, system zones, locales, etc.):
clock scan "" -gmt 1
clock scan ""
clock scan "" -timezone :CET
clock scan "" -format "" -locale en
clock scan "" -format "" -locale de

## ------------------------------------------

proc test-format {{reptime 1000}} {
  _test_run $reptime {
    # Format : short, week only (in gmt)
    {clock format 1482525936 -format "%u" -gmt 1}
    # Format : short, week only (system zone)
    {clock format 1482525936 -format "%u"}
    # Format : short, week only (CEST)
    {clock format 1482525936 -format "%u" -timezone :CET}
    # Format : date only (in gmt)
    {clock format 1482525936 -format "%Y-%m-%d" -gmt 1}
    # Format : date only (system zone)
    {clock format 1482525936 -format "%Y-%m-%d"}
    # Format : date only (CEST)
    {clock format 1482525936 -format "%Y-%m-%d" -timezone :CET}
    # Format : time only (in gmt)
    {clock format 1482525936 -format "%H:%M" -gmt 1}
    # Format : time only (system zone)
    {clock format 1482525936 -format "%H:%M"}
    # Format : time only (CEST)
    {clock format 1482525936 -format "%H:%M" -timezone :CET}
    # Format : time only (in gmt)
    {clock format 1482525936 -format "%H:%M:%S" -gmt 1}
    # Format : time only (system zone)
    {clock format 1482525936 -format "%H:%M:%S"}
    # Format : time only (CEST)
    {clock format 1482525936 -format "%H:%M:%S" -timezone :CET}
    # Format : default (in gmt)
    {clock format 1482525936 -gmt 1 -locale en}
    # Format : default (system zone)
    {clock format 1482525936 -locale en}
    # Format : default (CEST)
    {clock format 1482525936 -timezone :CET -locale en}
    # Format : ISO date-time (in gmt, numeric zone)
    {clock format 1246379400 -format "%Y-%m-%dT%H:%M:%S %z" -gmt 1}
    # Format : ISO date-time (system zone, CEST, numeric zone)
    {clock format 1246379400 -format "%Y-%m-%dT%H:%M:%S %z"}
    # Format : ISO date-time (CEST, numeric zone)
    {clock format 1246379400 -format "%Y-%m-%dT%H:%M:%S %z" -timezone :CET}
    # Format : ISO date-time (system zone, CEST)
    {clock format 1246379400 -format "%Y-%m-%dT%H:%M:%S %Z"}
    # Format : julian day with time (in gmt):
    {clock format 1246379415 -format "%J %H:%M:%S" -gmt 1}
    # Format : julian day with time (system zone):
    {clock format 1246379415 -format "%J %H:%M:%S"}

    # Format : locale date-time (en):
    {clock format 1246379415 -format "%x %X" -locale en}
    # Format : locale date-time (de):
    {clock format 1246379415 -format "%x %X" -locale de}

    # Format : locale lookup table month:
    {clock format 1246379400 -format "%b" -locale en -gmt 1}
    # Format : locale lookup 2 tables - month and day:
    {clock format 1246379400 -format "%b %Od" -locale en -gmt 1}
    # Format : locale lookup 3 tables - week, month and day:
    {clock format 1246379400 -format "%a %b %Od" -locale en -gmt 1}
    # Format : locale lookup 4 tables - week, month, day and year:
    {clock format 1246379400 -format "%a %b %Od %Oy" -locale en -gmt 1}

    # Format : dynamic clock value (without converter caches):
    setup {set i 0}
    {clock format [incr i] -format "%Y-%m-%dT%H:%M:%S" -locale en -timezone :CET}
    cleanup {puts [clock format $i -format "%Y-%m-%dT%H:%M:%S" -locale en -timezone :CET]}
    # Format : dynamic clock value (without any converter caches, zone range overflow):
    setup {set i 0}
    {clock format [incr i 86400] -format "%Y-%m-%dT%H:%M:%S" -locale en -timezone :CET}
    cleanup {puts [clock format $i -format "%Y-%m-%dT%H:%M:%S" -locale en -timezone :CET]}

    # Format : dynamic format (cacheable)
    {clock format 1246379415 -format [string trim "%d.%m.%Y %H:%M:%S "] -gmt 1}

    # Format : all (in gmt, locale en)
    {clock format 1482525936 -format "%%a = %a | %%A = %A | %%b = %b | %%h = %h | %%B = %B | %%C = %C | %%d = %d | %%e = %e | %%g = %g | %%G = %G | %%H = %H | %%I = %I | %%j = %j | %%J = %J | %%k = %k | %%l = %l | %%m = %m | %%M = %M | %%N = %N | %%p = %p | %%P = %P | %%Q = %Q | %%s = %s | %%S = %S | %%t = %t | %%u = %u | %%U = %U | %%V = %V | %%w = %w | %%W = %W | %%y = %y | %%Y = %Y | %%z = %z | %%Z = %Z | %%n = %n | %%EE = %EE | %%EC = %EC | %%Ey = %Ey | %%n = %n | %%Od = %Od | %%Oe = %Oe | %%OH = %OH | %%Ok = %Ok | %%OI = %OI | %%Ol = %Ol | %%Om = %Om | %%OM = %OM | %%OS = %OS | %%Ou = %Ou | %%Ow = %Ow | %%Oy = %Oy" -gmt 1 -locale en}
    # Format : all (in CET, locale de)
    {clock format 1482525936 -format "%%a = %a | %%A = %A | %%b = %b | %%h = %h | %%B = %B | %%C = %C | %%d = %d | %%e = %e | %%g = %g | %%G = %G | %%H = %H | %%I = %I | %%j = %j | %%J = %J | %%k = %k | %%l = %l | %%m = %m | %%M = %M | %%N = %N | %%p = %p | %%P = %P | %%Q = %Q | %%s = %s | %%S = %S | %%t = %t | %%u = %u | %%U = %U | %%V = %V | %%w = %w | %%W = %W | %%y = %y | %%Y = %Y | %%z = %z | %%Z = %Z | %%n = %n | %%EE = %EE | %%EC = %EC | %%Ey = %Ey | %%n = %n | %%Od = %Od | %%Oe = %Oe | %%OH = %OH | %%Ok = %Ok | %%OI = %OI | %%Ol = %Ol | %%Om = %Om | %%OM = %OM | %%OS = %OS | %%Ou = %Ou | %%Ow = %Ow | %%Oy = %Oy" -timezone :CET -locale de}
  }
}

proc test-scan {{reptime 1000}} {
  _test_run $reptime {
    # Scan : date (in gmt)
    {clock scan "25.11.2015" -format "%d.%m.%Y" -base 0 -gmt 1}
    # Scan : date (system time zone, with base)
    {clock scan "25.11.2015" -format "%d.%m.%Y" -base 0}
    # Scan : date (system time zone, without base)
    {clock scan "25.11.2015" -format "%d.%m.%Y"}
    # Scan : greedy match
    {clock scan "111" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "1111" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "11111" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "111111" -format "%d%m%y" -base 0 -gmt 1}
    # Scan : greedy match (space separated)
    {clock scan "1 1 1" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "111 1" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "1 111" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "1 11 1" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "1 11 11" -format "%d%m%y" -base 0 -gmt 1}
    {clock scan "11 11 11" -format "%d%m%y" -base 0 -gmt 1}

    # Scan : time (in gmt)
    {clock scan "10:35:55" -format "%H:%M:%S" -base 1000000000 -gmt 1}
    # Scan : time (system time zone, with base)
    {clock scan "10:35:55" -format "%H:%M:%S" -base 1000000000}
    # Scan : time (gmt, without base)
    {clock scan "10:35:55" -format "%H:%M:%S" -gmt 1}
    # Scan : time (system time zone, without base)
    {clock scan "10:35:55" -format "%H:%M:%S"}

    # Scan : date-time (in gmt)
    {clock scan "25.11.2015 10:35:55" -format "%d.%m.%Y %H:%M:%S" -base 0 -gmt 1}
    # Scan : date-time (system time zone with base)
    {clock scan "25.11.2015 10:35:55" -format "%d.%m.%Y %H:%M:%S" -base 0}
    # Scan : date-time (system time zone without base)
    {clock scan "25.11.2015 10:35:55" -format "%d.%m.%Y %H:%M:%S"}

    # Scan : julian day in gmt
    {clock scan 2451545 -format %J -gmt 1}
    # Scan : julian day in system TZ
    {clock scan 2451545 -format %J}
    # Scan : julian day in other TZ
    {clock scan 2451545 -format %J -timezone +0200}
    # Scan : julian day with time:
    {clock scan "2451545 10:20:30" -format "%J %H:%M:%S"}
    # Scan : julian day with time (greedy match):
    {clock scan "2451545 102030" -format "%J%H%M%S"}

    # Scan : century, lookup table month
    {clock scan {1970 Jan 2} -format {%C%y %b %d} -locale en -gmt 1}
    # Scan : century, lookup table month and day (both entries are first)
    {clock scan {1970 Jan 01} -format {%C%y %b %Od} -locale en -gmt 1}
    # Scan : century, lookup table month and day (list scan: entries with position 12 / 31)
    {clock scan {2016 Dec 31} -format {%C%y %b %Od} -locale en -gmt 1}

    # Scan : ISO date-time (CEST)
    {clock scan "2009-06-30T18:30:00+02:00" -format "%Y-%m-%dT%H:%M:%S%z"}
    {clock scan "2009-06-30T18:30:00 CEST" -format "%Y-%m-%dT%H:%M:%S %z"}
    # Scan : ISO date-time (UTC)
    {clock scan "2009-06-30T18:30:00Z" -format "%Y-%m-%dT%H:%M:%S%z"}
    {clock scan "2009-06-30T18:30:00 UTC" -format "%Y-%m-%dT%H:%M:%S %z"}

    # Scan : locale date-time (en):
    {clock scan "06/30/2009 18:30:15" -format "%x %X" -gmt 1 -locale en}
    # Scan : locale date-time (de):
    {clock scan "30.06.2009 18:30:15" -format "%x %X" -gmt 1 -locale de}

    # Scan : dynamic format (cacheable)
    {clock scan "25.11.2015 10:35:55" -format [string trim "%d.%m.%Y %H:%M:%S "] -base 0 -gmt 1}

    break
    # # Scan : long format test (allock chain)
    # {clock scan "25.11.2015" -format "%d.%m.%Y %d.%m.%Y %d.%m.%Y %d.%m.%Y %d.%m.%Y %d.%m.%Y %d.%m.%Y %d.%m.%Y" -base 0 -gmt 1}
    # # Scan : dynamic, very long format test (create obj representation, allock chain, GC, etc):
    # {clock scan "25.11.2015" -format [string repeat "[incr i] %d.%m.%Y %d.%m.%Y" 10] -base 0 -gmt 1}
    # # Scan : again:
    # {clock scan "25.11.2015" -format [string repeat "[incr i -1] %d.%m.%Y %d.%m.%Y" 10] -base 0 -gmt 1}
  } {puts [clock format $_(r) -locale en]}
}

proc test-freescan {{reptime 1000}} {
  _test_run $reptime {
    # FreeScan : relative date
    {clock scan "5 years 18 months 385 days" -base 0 -gmt 1}
    # FreeScan : relative date with relative weekday
    {clock scan "5 years 18 months 385 days Fri" -base 0 -gmt 1}
    # FreeScan : relative date with ordinal month
    {clock scan "5 years 18 months 385 days next 1 January" -base 0 -gmt 1}
    # FreeScan : relative date with ordinal month and relative weekday
    {clock scan "5 years 18 months 385 days next January Fri" -base 0 -gmt 1}
    # FreeScan : ordinal month
    {clock scan "next January" -base 0 -gmt 1}
    # FreeScan : relative week
    {clock scan "next Fri" -base 0 -gmt 1}
    # FreeScan : relative weekday and week offset 
    {clock scan "next January + 2 week" -base 0 -gmt 1}
    # FreeScan : time only with base
    {clock scan "19:18:30" -base 148863600 -gmt 1}
    # FreeScan : time only without base, gmt
    {clock scan "19:18:30" -gmt 1}
    # FreeScan : time only without base, system
    {clock scan "19:18:30"}
    # FreeScan : date, system time zone
    {clock scan "05/08/2016 20:18:30"}
    # FreeScan : date, supplied time zone
    {clock scan "05/08/2016 20:18:30" -timezone :CET}
    # FreeScan : date, supplied gmt (equivalent -timezone :GMT)
    {clock scan "05/08/2016 20:18:30" -gmt 1}
    # FreeScan : date, supplied time zone gmt
    {clock scan "05/08/2016 20:18:30" -timezone :GMT}
    # FreeScan : time only, numeric zone in string, base time gmt (exchange zones between gmt / -0500)
    {clock scan "20:18:30 -0500" -base 148863600 -gmt 1}
    # FreeScan : time only, zone in string (exchange zones between system / gmt)
    {clock scan "19:18:30 GMT" -base 148863600}
    # FreeScan : fast switch of zones in cycle - GMT, MST, CET (system) and EST
    {clock scan "19:18:30 MST" -base 148863600 -gmt 1
     clock scan "19:18:30 EST" -base 148863600
    }
  } {puts [clock format $_(r) -locale en]}
}

proc test-add {{reptime 1000}} {
  set tests {
    # Add : years
    {clock add 1246379415 5 years -gmt 1}
    # Add : months
    {clock add 1246379415 18 months -gmt 1}
    # Add : weeks
    {clock add 1246379415 20 weeks -gmt 1}
    # Add : days
    {clock add 1246379415 385 days -gmt 1}
    # Add : weekdays
    {clock add 1246379415 3 weekdays -gmt 1}

    # Add : hours
    {clock add 1246379415 5 hours -gmt 1}
    # Add : minutes
    {clock add 1246379415 55 minutes -gmt 1}
    # Add : seconds
    {clock add 1246379415 100 seconds -gmt 1}

    # Add : +/- in gmt
    {clock add 1246379415 -5 years +21 months -20 weeks +386 days -19 hours +30 minutes -10 seconds -gmt 1}
    # Add : +/- in system timezone
    {clock add 1246379415 -5 years +21 months -20 weeks +386 days -19 hours +30 minutes -10 seconds -timezone :CET}

    # Add : gmt
    {clock add 1246379415 -5 years 18 months 366 days 5 hours 30 minutes 10 seconds -gmt 1}
    # Add : system timezone
    {clock add 1246379415 -5 years 18 months 366 days 5 hours 30 minutes 10 seconds -timezone :CET}

    # Add : all in gmt
    {clock add 1246379415 4 years 18 months 50 weeks 378 days 3 weekdays 5 hours 30 minutes 10 seconds -gmt 1}
    # Add : all in system timezone
    {clock add 1246379415 4 years 18 months 50 weeks 378 days 3 weekdays 5 hours 30 minutes 10 seconds -timezone :CET}

  }
  # if does not support add of weekdays:
  if {[catch {clock add 0 3 weekdays -gmt 1}]} {
    regsub -all {\mweekdays\M} $tests "days" tests
  }
  _test_run $reptime $tests {puts [clock format $_(r) -locale en]}
}

proc test-convert {{reptime 1000}} {
  _test_run $reptime {
    # Convert locale (en -> de):
    {clock format [clock scan "Tue May 30 2017" -format "%a %b %d %Y" -gmt 1 -locale en] -format "%a %b %d %Y" -gmt 1 -locale de}
    # Convert locale (de -> en):
    {clock format [clock scan "Di Mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale de] -format "%a %b %d %Y" -gmt 1 -locale en}

    # Convert TZ: direct
    {clock format [clock scan "19:18:30" -base 148863600 -timezone EST] -timezone MST}
    {clock format [clock scan "19:18:30" -base 148863600 -timezone MST] -timezone EST}
    # Convert TZ: included in scan string & format
    {clock format [clock scan "19:18:30 EST" -base 148863600] -format "%H:%M:%S %z" -timezone MST}
    {clock format [clock scan "19:18:30 EST" -base 148863600] -format "%H:%M:%S %z" -timezone EST}

    # Format locale 1x: comparison values
    {clock format 0 -gmt 1 -locale en} 
    {clock format 0 -gmt 1 -locale de}
    {clock format 0 -gmt 1 -locale fr}
    # Format locale 2x: without switching locale (en, en)
    {clock format 0 -gmt 1 -locale en; clock format 0 -gmt 1 -locale en}
    # Format locale 2x: with switching locale (en, de)
    {clock format 0 -gmt 1 -locale en; clock format 0 -gmt 1 -locale de}
    # Format locale 3x: without switching locale (en, en, en)
    {clock format 0 -gmt 1 -locale en; clock format 0 -gmt 1 -locale en; clock format 0 -gmt 1 -locale en}
    # Format locale 3x: with switching locale (en, de, fr)
    {clock format 0 -gmt 1 -locale en; clock format 0 -gmt 1 -locale de; clock format 0 -gmt 1 -locale fr}

    # Scan locale 2x: without switching locale (en, en) + (de, de)
    {clock scan "Tue May 30 2017" -format "%a %b %d %Y" -gmt 1 -locale en; clock scan "Tue May 30 2017" -format "%a %b %d %Y" -gmt 1 -locale en}
    {clock scan "Di Mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale de; clock scan "Di Mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale de}
    # Scan locale 2x: with switching locale (en, de)
    {clock scan "Tue May 30 2017" -format "%a %b %d %Y" -gmt 1 -locale en; clock scan "Di Mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale de}
    # Scan locale 3x: with switching locale (en, de, fr)
    {clock scan "Tue May 30 2017" -format "%a %b %d %Y" -gmt 1 -locale en; clock scan "Di Mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale de; clock scan "mar. mai 30 2017" -format "%a %b %d %Y" -gmt 1 -locale fr}

    # Format TZ 2x: comparison values
    {clock format 0 -timezone CET -format "%Y-%m-%d %H:%M:%S %z"}
    {clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"}
    # Format TZ 2x: without switching
    {clock format 0 -timezone CET -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone CET -format "%Y-%m-%d %H:%M:%S %z"}
    {clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"}
    # Format TZ 2x: with switching
    {clock format 0 -timezone CET -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"}
    # Format TZ 3x: with switching (CET, EST, MST)
    {clock format 0 -timezone CET -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone MST -format "%Y-%m-%d %H:%M:%S %z"}
    # Format TZ 3x: with switching (GMT, EST, MST)
    {clock format 0 -gmt 1 -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone EST -format "%Y-%m-%d %H:%M:%S %z"; clock format 0 -timezone MST -format "%Y-%m-%d %H:%M:%S %z"}

    # FreeScan TZ 2x (+1 system-default): without switching TZ
    {clock scan "19:18:30 MST" -base 148863600; clock scan "19:18:30 MST" -base 148863600}
    {clock scan "19:18:30 EST" -base 148863600; clock scan "19:18:30 EST" -base 148863600}
    # FreeScan TZ 2x (+1 system-default): with switching TZ
    {clock scan "19:18:30 MST" -base 148863600; clock scan "19:18:30 EST" -base 148863600}
    # FreeScan TZ 2x (+1 gmt, +1 system-default)
    {clock scan "19:18:30 MST" -base 148863600 -gmt 1; clock scan "19:18:30 EST" -base 148863600}
    
    # Scan TZ: comparison included in scan string vs. given
    {clock scan "2009-06-30T18:30:00 CEST" -format "%Y-%m-%dT%H:%M:%S %z"}
    {clock scan "2009-06-30T18:30:00 CET" -format "%Y-%m-%dT%H:%M:%S %z"}
    {clock scan "2009-06-30T18:30:00" -timezone CET -format "%Y-%m-%dT%H:%M:%S"}
  }
}

proc test-other {{reptime 1000}} {
  _test_run $reptime {
    # Bad zone
    {catch {clock scan "1 day" -timezone BAD_ZONE -locale en}}

    # Scan : julian day (overflow)
    {catch {clock scan 5373485 -format %J}}

    # Scan : test rotate of GC objects (format is dynamic, so tcl-obj removed with last reference)
    {set i 0; time { clock scan "[incr i] - 25.11.2015" -format "$i - %d.%m.%Y" -base 0 -gmt 1 } 50}
    # Scan : test reusability of GC objects (format is dynamic, so tcl-obj removed with last reference)
    {set i 50; time { clock scan "[incr i -1] - 25.11.2015" -format "$i - %d.%m.%Y" -base 0 -gmt 1 } 50}
  }
}

proc test-ensemble-perf {{reptime 1000}} {
  _test_run $reptime {
    # Clock clicks (ensemble)
    {clock clicks}
    # Clock clicks (direct)
    {::tcl::clock::clicks}
    # Clock seconds (ensemble)
    {clock seconds}
    # Clock seconds (direct)
    {::tcl::clock::seconds}
    # Clock microseconds (ensemble)
    {clock microseconds}
    # Clock microseconds (direct)
    {::tcl::clock::microseconds}
    # Clock scan (ensemble)
    {clock scan ""}
    # Clock scan (direct)
    {::tcl::clock::scan ""}
    # Clock format (ensemble)
    {clock format 0 -f %s}
    # Clock format (direct)
    {::tcl::clock::format 0 -f %s}
  }
}

proc test {{reptime 1000}} {
  puts ""
  test-ensemble-perf [expr {$reptime / 2}]; #fast enough
  test-format $reptime
  test-scan $reptime
  test-freescan $reptime
  test-add $reptime
  test-convert [expr {$reptime / 2}]; #fast enough
  test-other $reptime

  puts \n**OK**
}

}; # end of ::tclTestPerf-TclClock

# ------------------------------------------------------------------------

# if calling direct:
if {[info exists ::argv0] && [file tail $::argv0] eq [file tail [info script]]} {
  ::tclTestPerf-TclClock::test $in(-time)
}

# ------------------------------------------------------------------------
#
# test-performance.tcl --
# 
#  This file provides common performance tests for comparison of tcl-speed
#  degradation or regression by switching between branches.
#
#  To execute test case evaluate direct corresponding file "tests-perf\*.perf.tcl".
#
# ------------------------------------------------------------------------
# 
# Copyright (c) 2014 Serg G. Brester (aka sebres)
# 
# See the file "license.terms" for information on usage and redistribution
# of this file.
# 

namespace eval ::tclTestPerf {
# warm-up interpeter compiler env, calibrate timerate measurement functionality:

# if no timerate here - import from unsupported:
if {[namespace which -command timerate] eq {}} {
  namespace inscope ::tcl::unsupported {namespace export timerate}
  namespace import ::tcl::unsupported::timerate
}

# if not yet calibrated:
if {[lindex [timerate {} 10] 6] >= (10-1)} {
  puts -nonewline "Calibration ... "; flush stdout
  puts "done: [lrange \
    [timerate -calibrate {}] \
  0 1]"
}

proc {**STOP**} {args} {
  return -code error -level 4 "**STOP** in [info level [expr {[info level]-2}]] [join $args { }]" 
}

proc _test_get_commands {lst} {
  regsub -all {(?:^|\n)[ \t]*(\#[^\n]*|\msetup\M[^\n]*|\mcleanup\M[^\n]*)(?=\n\s*(?:[\{\#]|setup|cleanup|$))} $lst "\n{\\1}"
}

proc _test_out_total {} {
  upvar _ _

  set tcnt [llength $_(itm)]
  if {!$tcnt} {
    puts ""
    return
  }

  set mintm 0x7fffffff
  set maxtm 0
  set nett 0
  set wtm 0
  set wcnt 0
  set i 0
  foreach tm $_(itm) {
    if {[llength $tm] > 6} {
      set nett [expr {$nett + [lindex $tm 6]}]
    }
    set wtm [expr {$wtm + [lindex $tm 0]}]
    set wcnt [expr {$wcnt + [lindex $tm 2]}]
    set tm [lindex $tm 0]
    if {$tm > $maxtm} {set maxtm $tm; set maxi $i}
    if {$tm < $mintm} {set mintm $tm; set mini $i}
    incr i
  }

  puts [string repeat ** 40]
  set s [format "%d cases in %.2f sec." $tcnt [expr {([clock milliseconds] - $_(starttime)) / 1000.0}]]
  if {$nett > 0} {
    append s [format " (%.2f nett-sec.)" [expr {$nett / 1000.0}]]
  }
  puts "Total $s:"
  lset _(m) 0 [format %.6f $wtm]
  lset _(m) 2 $wcnt
  lset _(m) 4 [format %.3f [expr {$wcnt / (($nett ? $nett : ($tcnt * [lindex $_(reptime) 0])) / 1000.0)}]]
  if {[llength $_(m)] > 6} {
    lset _(m) 6 [format %.3f $nett]
  }
  puts $_(m)
  puts "Average:"
  lset _(m) 0 [format %.6f [expr {[lindex $_(m) 0] / $tcnt}]]
  lset _(m) 2 [expr {[lindex $_(m) 2] / $tcnt}]
  if {[llength $_(m)] > 6} {
    lset _(m) 6 [format %.3f [expr {[lindex $_(m) 6] / $tcnt}]]
    lset _(m) 4 [format %.0f [expr {[lindex $_(m) 2] / [lindex $_(m) 6] * 1000}]]
  }
  puts $_(m)
  puts "Min:"
  puts [lindex $_(itm) $mini]
  puts "Max:"
  puts [lindex $_(itm) $maxi]
  puts [string repeat ** 40]
  puts ""
}

proc _test_run {args} {
  upvar _ _
  # parse args:
  set _(out-result) 1
  if {[lindex $args 0] eq "-no-result"} {
    set _(out-result) 0
    set args [lrange $args 1 end]
  }
  if {[llength $args] < 2 || [llength $args] > 3} {
    return -code error "wrong # args: should be \"[lindex [info level [info level]] 0] ?-no-result? reptime lst ?outcmd?\""
  }
  set outcmd {puts $_(r)}
  set args [lassign $args reptime lst]
  if {[llength $args]} {
    set outcmd [lindex $args 0]
  }
  # avoid output if only once:
  if {[lindex $reptime 0] <= 1 || ([llength $reptime] > 1 && [lindex $reptime 1] == 1)} {
    set _(out-result) 0
  }
  array set _ [list itm {} reptime $reptime starttime [clock milliseconds]]

  # process measurement:
  foreach _(c) [_test_get_commands $lst] {
    puts "% [regsub -all {\n[ \t]*} $_(c) {; }]"
    if {[regexp {^\s*\#} $_(c)]} continue
    if {[regexp {^\s*(?:setup|cleanup)\s+} $_(c)]} {
      puts [if 1 [lindex $_(c) 1]]
      continue
    }
    # if output result (and not once):
    if {$_(out-result)} {
      set _(r) [if 1 $_(c)]
      if {$outcmd ne {}} $outcmd
      if {[llength $_(reptime)] > 1} { # decrement max-count
        lset _(reptime) 1 [expr {[lindex $_(reptime) 1] - 1}]
      }
    }
    puts [set _(m) [timerate $_(c) {*}$_(reptime)]]
    lappend _(itm) $_(m)
    puts ""
  }
  _test_out_total
}

}; # end of namespace ::tclTestPerf

#!/usr/bin/tclsh

# ------------------------------------------------------------------------
#
# timer-event.perf.tcl --
# 
#  This file provides performance tests for comparison of tcl-speed
#  of timer events (event-driven tcl-handling).
#
# ------------------------------------------------------------------------
# 
# Copyright (c) 2014 Serg G. Brester (aka sebres)
# 
# See the file "license.terms" for information on usage and redistribution
# of this file.
# 


if {![namespace exists ::tclTestPerf]} {
  source [file join [file dirname [info script]] test-performance.tcl]
}


namespace eval ::tclTestPerf-Timer-Event {

namespace path {::tclTestPerf}

proc test-queue {{reptime {1000 10000}}} {

  set howmuch [lindex $reptime 1]

  # because of extremely short measurement times by tests below, wait a little bit (warming-up),
  # to minimize influence of the time-gradation (just for better dispersion resp. result-comparison)
  timerate {after 0} 156

  puts "*** up to $howmuch events ***"
  # single iteration by update, so using -no-result (measure only):
  _test_run -no-result $reptime [string map [list \{*\}\$reptime $reptime \$howmuch $howmuch \\# \#] {
    # generate up to $howmuch idle-events:
    {after idle {set foo bar}}
    # update / after idle:
    {update; if {![llength [after info]]} break}
    
    # generate up to $howmuch idle-events:
    {after idle {set foo bar}}
    # update idletasks / after idle:
    {update idletasks; if {![llength [after info]]} break}

    # generate up to $howmuch immediate events:
    {after 0 {set foo bar}}
    # update / after 0:
    {update; if {![llength [after info]]} break}
    
    # generate up to $howmuch 1-ms events:
    {after 1 {set foo bar}}
    setup {after 1}
    # update / after 1:
    {update; if {![llength [after info]]} break}

    # generate up to $howmuch immediate events (+ 1 event of the second generation):
    {after 0 {after 0 {}}}
    # update / after 0 (double generation):
    {update; if {![llength [after info]]} break}

    # cancel forwards "after idle" / $howmuch idle-events in queue:
    setup {set i 0; timerate {set ev([incr i]) [after idle {set foo bar}]} {*}$reptime}
    setup {set le $i; set i 0; list 1 .. $le; # cancel up to $howmuch events}
    {after cancel $ev([incr i]); if {$i >= $le} break}
    cleanup {update; unset -nocomplain ev}
    # cancel backwards "after idle" / $howmuch idle-events in queue:
    setup {set i 0; timerate {set ev([incr i]) [after idle {set foo bar}]} {*}$reptime}
    setup {set le $i; incr i; list $le .. 1; # cancel up to $howmuch events}
    {after cancel $ev([incr i -1]); if {$i <= 1} break}
    cleanup {update; unset -nocomplain ev}

    # cancel forwards "after 0" / $howmuch timer-events in queue:
    setup {set i 0; timerate {set ev([incr i]) [after 0 {set foo bar}]} {*}$reptime}
    setup {set le $i; set i 0; list 1 .. $le; # cancel up to $howmuch events}
    {after cancel $ev([incr i]); if {$i >= $howmuch} break}
    cleanup {update; unset -nocomplain ev}
    # cancel backwards "after 0" / $howmuch timer-events in queue:
    setup {set i 0; timerate {set ev([incr i]) [after 0 {set foo bar}]} {*}$reptime}
    setup {set le $i; incr i; list $le .. 1; # cancel up to $howmuch events}
    {after cancel $ev([incr i -1]); if {$i <= 1} break}
    cleanup {update; unset -nocomplain ev}
    
    # end $howmuch events.
    cleanup {if [llength [after info]] {error "unexpected: [llength [after info]] events are still there."}}
  }]
}

proc test-access {{reptime {1000 5000}}} {
  set howmuch [lindex $reptime 1]

  _test_run $reptime [string map [list \{*\}\$reptime $reptime \$howmuch $howmuch] {
    # event random access: after idle + after info (by $howmuch events)
    setup {set i -1; timerate {set ev([incr i]) [after idle {}]} {*}$reptime}
    {after info $ev([expr {int(rand()*$i)}])}
    cleanup {update; unset -nocomplain ev}
    # event random access: after 0 + after info (by $howmuch events)
    setup {set i -1; timerate {set ev([incr i]) [after 0 {}]} {*}$reptime}
    {after info $ev([expr {int(rand()*$i)}])}
    cleanup {update; unset -nocomplain ev}

    # end $howmuch events.
    cleanup {if [llength [after info]] {error "unexpected: [llength [after info]] events are still there."}}
  }]
}

proc test-exec {{reptime 1000}} {
  _test_run $reptime {
    # after idle + after cancel
    {after cancel [after idle {set foo bar}]}
    # after 0 + after cancel
    {after cancel [after 0 {set foo bar}]}
    # after idle + update idletasks
    {after idle {set foo bar}; update idletasks}
    # after idle + update
    {after idle {set foo bar}; update}
    # immediate: after 0 + update
    {after 0 {set foo bar}; update}
    # delayed: after 1 + update
    {after 1 {set foo bar}; update}
    # empty update:
    {update}
    # empty update idle tasks:
    {update idletasks}

    # simple shortest sleep:
    {after 0}
  }
}

proc test-nrt-capability {{reptime 1000}} {
  _test_run $reptime {
    # comparison values:
    {after 0 {set a 5}; update}
    {after 0 {set a 5}; vwait a}

    # conditional vwait with very brief wait-time:
    {after 1 {set a timeout}; vwait a; expr {$::a ne "timeout" ? 1 : "0[unset ::a]"}}
    {after 0 {set a timeout}; vwait a; expr {$::a ne "timeout" ? 1 : "0[unset ::a]"}}
  }
}

proc test-long {{reptime 1000}} {
  _test_run $reptime {
    # in-between important event by amount of idle events:
    {time {after idle {after 30}} 10; after 1 {set important 1}; vwait important;}
    cleanup {foreach i [after info] {after cancel $i}}
    # in-between important event (of new generation) by amount of idle events:
    {time {after idle {after 30}} 10; after 1 {after 0 {set important 1}}; vwait important;} 
    cleanup {foreach i [after info] {after cancel $i}}
  }
}

proc test {{reptime 1000}} {
  test-exec $reptime
  foreach howmuch {5000 50000} {
    test-access [list $reptime $howmuch]
  }
  test-nrt-capability $reptime
  test-long $reptime

  puts ""
  foreach howmuch { 10000 20000 40000 60000 } {
    test-queue [list $reptime $howmuch]
  }

  puts \n**OK**
}

}; # end of ::tclTestPerf-Timer-Event

# ------------------------------------------------------------------------

# if calling direct:
if {[info exists ::argv0] && [file tail $::argv0] eq [file tail [info script]]} {
  array set in {-time 500}
  array set in $argv
  ::tclTestPerf-Timer-Event::test $in(-time)
}

test cmdMZ-5.7 {Tcl_TimeObjCmd: errors generate right trace} {
    list [catch {time {error foo}} msg] $msg $::errorInfo
} {1 foo {foo
    while executing
"error foo"
    invoked from within
"time {error foo}"}}

test cmdMZ-6.1 {Tcl_TimeRateObjCmd: basic format of command} {
    list [catch {timerate} msg] $msg
} {1 {wrong # args: should be "timerate ?-direct? ?-calibrate? ?-overhead double? command ?time ?max-count??"}}
test cmdMZ-6.2.1 {Tcl_TimeRateObjCmd: basic format of command} {
    list [catch {timerate a b c d} msg] $msg
} {1 {wrong # args: should be "timerate ?-direct? ?-calibrate? ?-overhead double? command ?time ?max-count??"}}
test cmdMZ-6.2.2 {Tcl_TimeRateObjCmd: basic format of command} {
    list [catch {timerate a b c} msg] $msg
} {1 {expected integer but got "b"}}
test cmdMZ-6.2.3 {Tcl_TimeRateObjCmd: basic format of command} {
    list [catch {timerate a b} msg] $msg
} {1 {expected integer but got "b"}}
test cmdMZ-6.3 {Tcl_TimeRateObjCmd: basic format of command} {
    list [catch {timerate -overhead b {} a b} msg] $msg
} {1 {expected floating-point number but got "b"}}
test cmdMZ-6.4 {Tcl_TimeRateObjCmd: compile of script happens even with negative iteration counts} {
    list [catch {timerate "foreach a {c d e} \{" -12456} msg] $msg
} {1 {missing close-brace}}
test cmdMZ-6.5 {Tcl_TimeRateObjCmd: result format and one iteration} {
    regexp {^\d+.\d+ \ws/# 1 # \d+ #/sec \d+.\d+ nett-ms$} [timerate {} 0]
} 1
test cmdMZ-6.6 {Tcl_TimeRateObjCmd: slower commands take longer, but it remains almost the same time of measument} {
    set m1 [timerate {after 0} 20]
    set m2 [timerate {after 1} 20]
    list \
	[expr {[lindex $m1 0] < [lindex $m2 0]}] \
	[expr {[lindex $m1 0] < 100}] \
	[expr {[lindex $m2 0] >= 500}] \
	[expr {[lindex $m1 2] > 1000}] \
	[expr {[lindex $m2 2] <= 50}] \
	[expr {[lindex $m1 4] > 10000}] \
	[expr {[lindex $m2 4] < 10000}] \
	[expr {[lindex $m1 6] > 10 && [lindex $m1 6] < 50}] \
	[expr {[lindex $m2 6] > 10 && [lindex $m2 6] < 50}]
} [lrepeat 9 1]
test cmdMZ-6.7 {Tcl_TimeRateObjCmd: errors generate right trace} {
    list [catch {timerate {error foo} 1} msg] $msg $::errorInfo
} {1 foo {foo
    while executing
"error foo"
    invoked from within
"timerate {error foo} 1"}}
test cmdMZ-6.8 {Tcl_TimeRateObjCmd: allow (conditional) break from timerate} {
    set m1 [timerate {break}]
    list \
	[expr {[lindex $m1 0] < 1000}] \
	[expr {[lindex $m1 2] == 1}] \
	[expr {[lindex $m1 4] > 1000}] \
	[expr {[lindex $m1 6] < 10}]
} {1 1 1 1}
test cmdMZ-6.9 {Tcl_TimeRateObjCmd: max count of iterations} {
    set m1 [timerate {} 1000 5];	# max-count wins
    set m2 [timerate {after 20} 1 5];	# max-time wins
    list [lindex $m1 2] [lindex $m2 2]
} {5 1}
test cmdMZ-6.10 {Tcl_TimeRateObjCmd: huge overhead cause 0us result} {
    set m1 [timerate -overhead 1e6 {after 10} 100 1]
    list \
	[expr {[lindex $m1 0] == 0.0}] \
	[expr {[lindex $m1 2] == 1}] \
	[expr {[lindex $m1 4] == 1000000}] \
	[expr {[lindex $m1 6] <= 0.001}]
} {1 1 1 1}

# The tests for Tcl_WhileObjCmd are in while.test

# cleanup
cleanupTests
}
namespace delete ::tcl::test::cmdMZ

	    {\%}	{} \
	    "\\\n"	"\n" \
	    {\(+-}	"±" \
	    {\(co}	"©" \
	    {\(em}	"—" \
	    {\(en}	"–" \
	    {\(fm}	"′" \
	    {\(mc}	"µ" \
	    {\(mu}	"×" \
	    {\(mi}	"−" \
	    {\(->}	"<font size=\"+1\">&#8594;</font>" \
	    {\fP}	{\fR} \
	    {\.}	. \
	    {\(bu}	"&#8226;" \
	    ]

{
    return time(NULL);
}

/*
 *----------------------------------------------------------------------
 *
 * TclpGetMicroseconds --
 *
 *	This procedure returns the number of microseconds from the epoch.
 *	On most Unix systems the epoch is Midnight Jan 1, 1970 GMT.
 *
 * Results:
 *	Number of microseconds from the epoch.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

Tcl_WideInt
TclpGetMicroseconds(void)
{
    Tcl_Time time;

    tclGetTimeProcPtr(&time, tclTimeClientData);
    return ((Tcl_WideInt)time.sec)*1000000 + time.usec;
}

/*
 *----------------------------------------------------------------------
 *
 * TclpGetClicks --
 *
 *	This procedure returns a value that represents the highest resolution
 *	clock available on the system. There are no garantees on what the
 *	resolution will be. In Tcl we will call this value a "click". The
 *	start time is also system dependant.
 *

#else
#error Wide high-resolution clicks not implemented on this platform
#endif
    }

    return nsec;
}

/*
 *----------------------------------------------------------------------
 *
 * TclpWideClickInMicrosec --
 *
 *	This procedure return scale to convert click values from the 
 *	TclpGetWideClicks native resolution to microsecond resolution
 *	and back.
 *
 * Results:
 * 	1 click in microseconds as double.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

double
TclpWideClickInMicrosec(void)
{
    if (tclGetTimeProcPtr != NativeGetTime) {
	return 1.0;
    } else {
#ifdef MAC_OSX_TCL
	static int initialized = 0;
	static double scale = 0.0;

	if (initialized) {
	    return scale;
	} else {
	    mach_timebase_info_data_t tb;

	    mach_timebase_info(&tb);
	    /* value of tb.numer / tb.denom = 1 click in nanoseconds */
	    scale = ((double)tb.numer) / tb.denom / 1000;
	    initialized = 1;
	    return scale;
	}
#else
#error Wide high-resolution clicks not implemented on this platform
#endif
    }
}
#endif /* TCL_WIDE_CLICKS */

/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetTime --
 *

typedef struct {
    CRITICAL_SECTION cs;	/* Mutex guarding this structure. */
    int initialized;		/* Flag == 1 if this structure is
				 * initialized. */
    int perfCounterAvailable;	/* Flag == 1 if the hardware has a performance
				 * counter. */
    DWORD calibrationInterv;	/* Calibration interval in seconds (start 1 sec) */
    HANDLE calibrationThread;	/* Handle to the thread that keeps the virtual
				 * clock calibrated. */
    HANDLE readyEvent;		/* System event used to trigger the requesting
				 * thread when the clock calibration procedure
				 * is initialized for the first time. */
    HANDLE exitEvent; 		/* Event to signal out of an exit handler to
				 * tell the calibration loop to terminate. */
    LARGE_INTEGER nominalFreq;	/* Nominal frequency of the system performance
				 * counter, that is, the value returned from
				 * QueryPerformanceFrequency. */

    /*
     * The following values are used for calculating virtual time. Virtual
     * time is always equal to:
     *    lastFileTime + (current perf counter - lastCounter)
     *				* 10000000 / curCounterFreq
     * and lastFileTime and lastCounter are updated any time that virtual time
     * is returned to a caller.
     */

    ULARGE_INTEGER fileTimeLastCall;
    LARGE_INTEGER perfCounterLastCall;
    LARGE_INTEGER curCounterFreq;
    LARGE_INTEGER posixEpoch;	/* Posix epoch expressed as 100-ns ticks since
				 * the windows epoch. */

    /*
     * Data used in developing the estimate of performance counter frequency
     */

    Tcl_WideUInt fileTimeSample[SAMPLES];
				/* Last 64 samples of system time. */
    Tcl_WideInt perfCounterSample[SAMPLES];
				/* Last 64 samples of performance counter. */
    int sampleNo;		/* Current sample number. */
} TimeInfo;

static TimeInfo timeInfo = {
    { NULL, 0, 0, NULL, NULL, 0 },
    0,
    0,
    1,
    (HANDLE) NULL,
    (HANDLE) NULL,
    (HANDLE) NULL,
#ifdef HAVE_CAST_TO_UNION
    (LARGE_INTEGER) (Tcl_WideInt) 0,
    (ULARGE_INTEGER) (DWORDLONG) 0,
    (LARGE_INTEGER) (Tcl_WideInt) 0,
    (LARGE_INTEGER) (Tcl_WideInt) 0,
    (LARGE_INTEGER) (Tcl_WideInt) 0,
#else
    {0, 0},
    {0, 0},
    {0, 0},
    {0, 0},
    {0, 0},
#endif
    { 0 },
    { 0 },
    0
};

/*
 * Scale to convert wide click values from the TclpGetWideClicks native
 * resolution to microsecond resolution and back.
 */
static struct {
    int initialized;		/* 1 if initialized, 0 otherwise */
    int perfCounter;		/* 1 if performance counter usable for wide clicks */
    double microsecsScale;	/* Denominator scale between clock / microsecs */
} wideClick = {0, 0.0};


/*
 * Declarations for functions defined later in this file.
 */

static void		StopCalibration(ClientData clientData);
static DWORD WINAPI	CalibrationThread(LPVOID arg);
static void 		UpdateTimeEachSecond(void);
static void		ResetCounterSamples(Tcl_WideUInt fileTime,
			    Tcl_WideInt perfCounter, Tcl_WideInt perfFreq);
static Tcl_WideInt	AccumulateSample(Tcl_WideInt perfCounter,
			    Tcl_WideUInt fileTime);
static void		NativeScaleTime(Tcl_Time* timebuf,
			    ClientData clientData);
static Tcl_WideInt	NativeGetMicroseconds(void);
static void		NativeGetTime(Tcl_Time* timebuf,
			    ClientData clientData);

/*
 * TIP #233 (Virtualized Time): Data for the time hooks, if any.
 */

 *
 *----------------------------------------------------------------------
 */

Tcl_WideUInt
TclpGetSeconds(void)
{
    Tcl_WideInt usecSincePosixEpoch;

    /* Try to use high resolution timer */
    if ( tclGetTimeProcPtr == NativeGetTime
      && (usecSincePosixEpoch = NativeGetMicroseconds())
    ) {
	return usecSincePosixEpoch / 1000000;
    } else {
	Tcl_Time t;

	tclGetTimeProcPtr(&t, tclTimeClientData);	/* Tcl_GetTime inlined. */
	return t.sec;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * TclpGetClicks --
 *

 *
 *----------------------------------------------------------------------
 */

Tcl_WideUInt
TclpGetClicks(void)
{
    Tcl_WideInt usecSincePosixEpoch;

    /* Try to use high resolution timer */
    if ( tclGetTimeProcPtr == NativeGetTime
      && (usecSincePosixEpoch = NativeGetMicroseconds())
    ) {
	return (Tcl_WideUInt)usecSincePosixEpoch;
    } else {
	/*
	* Use the Tcl_GetTime abstraction to get the time in microseconds, as
	* nearly as we can, and return it.
	*/

	Tcl_Time now;		/* Current Tcl time */


	tclGetTimeProcPtr(&now, tclTimeClientData);	/* Tcl_GetTime inlined */
	return (Tcl_WideUInt)(now.sec * 1000000) + now.usec;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * TclpGetWideClicks --
 *
 *	This procedure returns a WideInt value that represents the highest
 *	resolution clock in microseconds available on the system.
 *
 * Results:
 *	Number of microseconds (from some start time).
 *
 * Side effects:
 *	This should be used for time-delta resp. for measurement purposes
 *	only, because on some platforms can return microseconds from some
 *	start time (not from the epoch).
 *
 *----------------------------------------------------------------------
 */

Tcl_WideInt
TclpGetWideClicks(void)
{
    LARGE_INTEGER curCounter;

    if (!wideClick.initialized) {
	LARGE_INTEGER perfCounterFreq;

	/*
	 * The frequency of the performance counter is fixed at system boot and
	 * is consistent across all processors. Therefore, the frequency need 
	 * only be queried upon application initialization.
	 */
	if (QueryPerformanceFrequency(&perfCounterFreq)) {
	    wideClick.perfCounter = 1;
	    wideClick.microsecsScale = 1000000.0 / perfCounterFreq.QuadPart;
	} else {
	    /* fallback using microseconds */
	    wideClick.perfCounter = 0;
	    wideClick.microsecsScale = 1;
	}
	
	wideClick.initialized = 1;
    }
    if (wideClick.perfCounter) {
	if (QueryPerformanceCounter(&curCounter)) {
	    return (Tcl_WideInt)curCounter.QuadPart;
	}
	/* fallback using microseconds */
	wideClick.perfCounter = 0;
	wideClick.microsecsScale = 1;
	return TclpGetMicroseconds();
    } else {
    	return TclpGetMicroseconds();
    }
}

/*
 *----------------------------------------------------------------------
 *
 * TclpWideClickInMicrosec --
 *
 *	This procedure return scale to convert wide click values from the 
 *	TclpGetWideClicks native resolution to microsecond resolution
 *	and back.
 *
 * Results:
 * 	1 click in microseconds as double.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

double
TclpWideClickInMicrosec(void)
{
    if (!wideClick.initialized) {
    	(void)TclpGetWideClicks();	/* initialize */
    }
    return wideClick.microsecsScale;
}

/*
 *----------------------------------------------------------------------
 *
 * TclpGetMicroseconds --
 *
 *	This procedure returns a WideInt value that represents the highest
 *	resolution clock in microseconds available on the system.
 *
 * Results:
 *	Number of microseconds (from the epoch).
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

Tcl_WideInt 
TclpGetMicroseconds(void)
{
    Tcl_WideInt usecSincePosixEpoch;

    /* Try to use high resolution timer */
    if ( tclGetTimeProcPtr == NativeGetTime
      && (usecSincePosixEpoch = NativeGetMicroseconds())
    ) {
	return usecSincePosixEpoch;
    } else {
	/*
	* Use the Tcl_GetTime abstraction to get the time in microseconds, as
	* nearly as we can, and return it.
	*/

	Tcl_Time now;

	tclGetTimeProcPtr(&now, tclTimeClientData);	/* Tcl_GetTime inlined */
	return (((Tcl_WideInt)now.sec) * 1000000) + now.usec;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetTime --
 *

 *----------------------------------------------------------------------
 */

void
Tcl_GetTime(
    Tcl_Time *timePtr)		/* Location to store time information. */
{
    Tcl_WideInt usecSincePosixEpoch;

    /* Try to use high resolution timer */
    if ( tclGetTimeProcPtr == NativeGetTime
      && (usecSincePosixEpoch = NativeGetMicroseconds())
    ) {
	timePtr->sec = (long) (usecSincePosixEpoch / 1000000);
	timePtr->usec = (unsigned long) (usecSincePosixEpoch % 1000000);
    } else {
    	tclGetTimeProcPtr(timePtr, tclTimeClientData);
    }
}

/*
 *----------------------------------------------------------------------
 *
 * NativeScaleTime --
 *

     * Native scale is 1:1. Nothing is done.
     */
}

/*
 *----------------------------------------------------------------------
 *
 * NativeGetMicroseconds --
 *
 *	Gets the current system time in microseconds since the beginning
 *	of the epoch: 00:00 UCT, January 1, 1970.
 *
 * Results:
 *	Returns the wide integer with number of microseconds from the epoch, or
 *	0 if high resolution timer is not available.
 *
 * Side effects:
 *	On the first call, initializes a set of static variables to keep track
 *	of the base value of the performance counter, the corresponding wall
 *	clock (obtained through ftime) and the frequency of the performance
 *	counter. Also spins a thread whose function is to wake up periodically
 *	and monitor these values, adjusting them as necessary to correct for
 *	drift in the performance counter's oscillator.
 *
 *----------------------------------------------------------------------
 */

static inline Tcl_WideInt
NativeCalc100NsTicks(
    ULONGLONG fileTimeLastCall,
    LONGLONG perfCounterLastCall,
    LONGLONG curCounterFreq,
    LONGLONG curCounter
) {
    return fileTimeLastCall + 
	((curCounter - perfCounterLastCall) * 10000000 / curCounterFreq);
}

static Tcl_WideInt
NativeGetMicroseconds(void)
{
    /*
     * Initialize static storage on the first trip through.
     *
     * Note: Outer check for 'initialized' is a performance win since it
     * avoids an extra mutex lock in the common case.
     */

    if (!timeInfo.initialized) {
	TclpInitLock();
	if (!timeInfo.initialized) {

	    timeInfo.posixEpoch.LowPart = 0xD53E8000;
	    timeInfo.posixEpoch.HighPart = 0x019DB1DE;

	    timeInfo.perfCounterAvailable =
		    QueryPerformanceFrequency(&timeInfo.nominalFreq);

	    /*
	     * Some hardware abstraction layers use the CPU clock in place of
	     * the real-time clock as a performance counter reference. This
	     * results in:

    if (timeInfo.perfCounterAvailable && timeInfo.curCounterFreq.QuadPart!=0) {
	/*
	 * Query the performance counter and use it to calculate the current
	 * time.
	 */

	ULONGLONG fileTimeLastCall;
	LONGLONG perfCounterLastCall, curCounterFreq;
				/* Copy with current data of calibration cycle */

	LARGE_INTEGER curCounter;
				/* Current performance counter. */











	QueryPerformanceCounter(&curCounter);

	/*
	 * Hold time section locked as short as possible
	 */
	EnterCriticalSection(&timeInfo.cs);

	fileTimeLastCall = timeInfo.fileTimeLastCall.QuadPart;
	perfCounterLastCall = timeInfo.perfCounterLastCall.QuadPart;
	curCounterFreq = timeInfo.curCounterFreq.QuadPart;

	LeaveCriticalSection(&timeInfo.cs);

	/*
	 * If calibration cycle occurred after we get curCounter
	 */
	if (curCounter.QuadPart <= perfCounterLastCall) {
	    /* Calibrated file-time is saved from posix in 100-ns ticks */



	    return fileTimeLastCall / 10;
	}

	/*
	 * If it appears to be more than 1.1 seconds since the last trip
	 * through the calibration loop, the performance counter may have
	 * jumped forward. (See MSDN Knowledge Base article Q274323 for a
	 * description of the hardware problem that makes this test
	 * necessary.) If the counter jumps, we don't want to use it directly.
	 * Instead, we must return system time. Eventually, the calibration
	 * loop should recover.
	 */

	if (curCounter.QuadPart - perfCounterLastCall <
		11 * curCounterFreq * timeInfo.calibrationInterv / 10
	) {
	    /* Calibrated file-time is saved from posix in 100-ns ticks */
	    return NativeCalc100NsTicks(fileTimeLastCall,
		perfCounterLastCall, curCounterFreq, curCounter.QuadPart) / 10;

	}
    }



    /*
     * High resolution timer is not available.
     */
    return 0;
}

/*
 *----------------------------------------------------------------------
 *
 * NativeGetTime --
 *
 *	TIP #233: Gets the current system time in seconds and microseconds
 *	since the beginning of the epoch: 00:00 UCT, January 1, 1970.
 *
 * Results:
 *	Returns the current time in timePtr.
 *
 * Side effects:
 *	See NativeGetMicroseconds for more information.
 *
 *----------------------------------------------------------------------
 */

static void
NativeGetTime(
    Tcl_Time *timePtr,
    ClientData clientData)
{
    Tcl_WideInt usecSincePosixEpoch;

    /*
     * Try to use high resolution timer.
     */
    if ( (usecSincePosixEpoch = NativeGetMicroseconds()) ) {
	timePtr->sec = (long) (usecSincePosixEpoch / 1000000);
	timePtr->usec = (unsigned long) (usecSincePosixEpoch % 1000000);
    } else {
	/*
	* High resolution timer is not available. Just use ftime.
	*/

	struct _timeb t;

	_ftime(&t);
	timePtr->sec = (long)t.time;
	timePtr->usec = t.millitm * 1000;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * StopCalibration --
 *
 *	Turns off the calibration thread in preparation for exiting the
 *	process.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Sets the 'exitEvent' event in the 'timeInfo' structure to ask the
 *	thread in question to exit, and waits for it to do so.
 *
 *----------------------------------------------------------------------
 */

void TclWinResetTimerResolution(void);

static void
StopCalibration(
    ClientData unused)		/* Client data is unused */
{
    SetEvent(timeInfo.exitEvent);

     */

    GetSystemTimeAsFileTime(&curFileTime);
    QueryPerformanceCounter(&timeInfo.perfCounterLastCall);
    QueryPerformanceFrequency(&timeInfo.curCounterFreq);
    timeInfo.fileTimeLastCall.LowPart = curFileTime.dwLowDateTime;
    timeInfo.fileTimeLastCall.HighPart = curFileTime.dwHighDateTime;
    /* Calibrated file-time will be saved from posix in 100-ns ticks */
    timeInfo.fileTimeLastCall.QuadPart -= timeInfo.posixEpoch.QuadPart;

    ResetCounterSamples(timeInfo.fileTimeLastCall.QuadPart,
	    timeInfo.perfCounterLastCall.QuadPart,
	    timeInfo.curCounterFreq.QuadPart);

    /*
     * Wake up the calling thread. When it wakes up, it will release the

static void
UpdateTimeEachSecond(void)
{
    LARGE_INTEGER curPerfCounter;
				/* Current value returned from
				 * QueryPerformanceCounter. */
    FILETIME curSysTime;	/* Current system time. */
    static LARGE_INTEGER lastFileTime; /* File time of the previous calibration */
    LARGE_INTEGER curFileTime;	/* File time at the time this callback was
				 * scheduled. */
    Tcl_WideInt estFreq;	/* Estimated perf counter frequency. */
    Tcl_WideInt vt0;		/* Tcl time right now. */
    Tcl_WideInt vt1;		/* Tcl time one second from now. */
    Tcl_WideInt tdiff;		/* Difference between system clock and Tcl
				 * time. */
    Tcl_WideInt driftFreq;	/* Frequency needed to drift virtual time into
				 * step over 1 second. */

    /*
     * Sample performance counter and system time (from posix epoch).
     */


    GetSystemTimeAsFileTime(&curSysTime);
    curFileTime.LowPart = curSysTime.dwLowDateTime;
    curFileTime.HighPart = curSysTime.dwHighDateTime;
    curFileTime.QuadPart -= timeInfo.posixEpoch.QuadPart;
    /* If calibration still not needed (check for possible time switch) */
    if ( curFileTime.QuadPart > lastFileTime.QuadPart
      && curFileTime.QuadPart < lastFileTime.QuadPart +
      				    (timeInfo.calibrationInterv * 10000000)
    ) {
    	/* again in next one second */
	return;
    }
    QueryPerformanceCounter(&curPerfCounter);
    
    lastFileTime.QuadPart = curFileTime.QuadPart;

    /*
     * We devide by timeInfo.curCounterFreq.QuadPart in several places. That
     * value should always be positive on a correctly functioning system. But
     * it is good to be defensive about such matters. So if something goes
     * wrong and the value does goes to zero, we clear the
     * timeInfo.perfCounterAvailable in order to cause the calibration thread
     * to shut itself down, then return without additional processing.
     */

    if (timeInfo.curCounterFreq.QuadPart == 0){

	timeInfo.perfCounterAvailable = 0;
	return;
    }

    /*
     * Several things may have gone wrong here that have to be checked for.
     *  (1) The performance counter may have jumped.

     *
     * vt1 = 20000000 + curFileTime
     *
     * The frequency that we need to use to drift the counter back into place
     * is estFreq * 20000000 / (vt1 - vt0)
     */


    vt0 = NativeCalc100NsTicks(timeInfo.fileTimeLastCall.QuadPart,
	    timeInfo.perfCounterLastCall.QuadPart, timeInfo.curCounterFreq.QuadPart,


	    curPerfCounter.QuadPart);
    /*
     * If we've gotten more than a second away from system time, then drifting
     * the clock is going to be pretty hopeless. Just let it jump. Otherwise,
     * compute the drift frequency and fill in everything.
     */

    tdiff = vt0 - curFileTime.QuadPart;
    if (tdiff > 10000000 || tdiff < -10000000) {
    	/* jump to current system time, use curent estimated frequency */
    	vt0 = curFileTime.QuadPart;

    } else {
    	/* calculate new frequency and estimate drift to the next second */
	vt1 = 20000000 + curFileTime.QuadPart;
	driftFreq = (estFreq * 20000000 / (vt1 - vt0));
	/* 
	 * Avoid too large drifts (only half of the current difference),
	 * that allows also be more accurate (aspire to the smallest tdiff),
	 * so then we can prolong calibration interval by tdiff < 100000
	 */
	driftFreq = timeInfo.curCounterFreq.QuadPart +
		(driftFreq - timeInfo.curCounterFreq.QuadPart) / 2;

	/* 
	 * Average between estimated, 2 current and 5 drifted frequencies,
	 * (do the soft drifting as possible)
	 */
	estFreq = (estFreq + 2 * timeInfo.curCounterFreq.QuadPart + 5 * driftFreq) / 8;
    }
    
    /* Avoid too large discrepancy from nominal frequency */
    if (estFreq > 1003*timeInfo.nominalFreq.QuadPart/1000) {
	estFreq = 1003*timeInfo.nominalFreq.QuadPart/1000;
	vt0 = curFileTime.QuadPart;
    } else if (estFreq < 997*timeInfo.nominalFreq.QuadPart/1000) {
	estFreq = 997*timeInfo.nominalFreq.QuadPart/1000;
	vt0 = curFileTime.QuadPart;
    } else if (vt0 != curFileTime.QuadPart) {
	/* 
	 * Be sure the clock ticks never backwards (avoid it by negative drifting)
	 * just compare native time (in 100-ns) before and hereafter using 
	 * new calibrated values) and do a small adjustment (short time freeze)
	 */
	LARGE_INTEGER newPerfCounter;
	Tcl_WideInt nt0, nt1;

	QueryPerformanceCounter(&newPerfCounter);
	nt0 = NativeCalc100NsTicks(timeInfo.fileTimeLastCall.QuadPart,
		timeInfo.perfCounterLastCall.QuadPart, timeInfo.curCounterFreq.QuadPart,
		newPerfCounter.QuadPart);
	nt1 = NativeCalc100NsTicks(vt0,
		curPerfCounter.QuadPart, estFreq,
		newPerfCounter.QuadPart);
	if (nt0 > nt1) { /* drifted backwards, try to compensate with new base */
	    /* first adjust with a micro jump (short frozen time is acceptable) */
	    vt0 += nt0 - nt1;
	    /* if drift unavoidable (e. g. we had a time switch), then reset it */
	    vt1 = vt0 - curFileTime.QuadPart;
	    if (vt1 > 10000000 || vt1 < -10000000) {
	    	/* larger jump resp. shift relative new file-time */
	    	vt0 = curFileTime.QuadPart;
	    }
	}
    }

    /* In lock commit new values to timeInfo (hold lock as short as possible) */
    EnterCriticalSection(&timeInfo.cs);

    /* grow calibration interval up to 10 seconds (if still precise enough) */
    if (tdiff < -100000 || tdiff > 100000) {
	/* too long drift - reset calibration interval to 1000 second */
	timeInfo.calibrationInterv = 1;
    } else if (timeInfo.calibrationInterv < 10) {
	timeInfo.calibrationInterv++;
    }

    timeInfo.fileTimeLastCall.QuadPart = vt0;
    timeInfo.curCounterFreq.QuadPart = estFreq;
    timeInfo.perfCounterLastCall.QuadPart = curPerfCounter.QuadPart;

    LeaveCriticalSection(&timeInfo.cs);
}

/*
 *----------------------------------------------------------------------