Permalink
Fetching contributors…
Cannot retrieve contributors at this time
4712 lines (4012 sloc) 160 KB
//Written in the D programming language
/++
Module containing core time functionality, such as $(LREF Duration) (which
represents a duration of time) or $(LREF MonoTime) (which represents a
timestamp of the system's monotonic clock).
Various functions take a string (or strings) to represent a unit of time
(e.g. $(D convert!("days", "hours")(numDays))). The valid strings to use
with such functions are "years", "months", "weeks", "days", "hours",
"minutes", "seconds", "msecs" (milliseconds), "usecs" (microseconds),
"hnsecs" (hecto-nanoseconds - i.e. 100 ns) or some subset thereof. There
are a few functions that also allow "nsecs", but very little actually
has precision greater than hnsecs.
$(BOOKTABLE Cheat Sheet,
$(TR $(TH Symbol) $(TH Description))
$(LEADINGROW Types)
$(TR $(TDNW $(LREF Duration)) $(TD Represents a duration of time of weeks
or less (kept internally as hnsecs). (e.g. 22 days or 700 seconds).))
$(TR $(TDNW $(LREF TickDuration)) $(TD Represents a duration of time in
system clock ticks, using the highest precision that the system provides.))
$(TR $(TDNW $(LREF MonoTime)) $(TD Represents a monotonic timestamp in
system clock ticks, using the highest precision that the system provides.))
$(TR $(TDNW $(LREF FracSec)) $(TD Represents fractional seconds
(portions of time smaller than a second).))
$(LEADINGROW Functions)
$(TR $(TDNW $(LREF convert)) $(TD Generic way of converting between two time
units.))
$(TR $(TDNW $(LREF dur)) $(TD Allows constructing a $(LREF Duration) from
the given time units with the given length.))
$(TR $(TDNW $(LREF weeks)$(NBSP)$(LREF days)$(NBSP)$(LREF hours)$(BR)
$(LREF minutes)$(NBSP)$(LREF seconds)$(NBSP)$(LREF msecs)$(BR)
$(LREF usecs)$(NBSP)$(LREF hnsecs)$(NBSP)$(LREF nsecs))
$(TD Convenience aliases for $(LREF dur).))
$(TR $(TDNW $(LREF abs)) $(TD Returns the absolute value of a duration.))
)
$(BOOKTABLE Conversions,
$(TR $(TH )
$(TH From $(LREF Duration))
$(TH From $(LREF TickDuration))
$(TH From $(LREF FracSec))
$(TH From units)
)
$(TR $(TD $(B To $(LREF Duration)))
$(TD -)
$(TD $(D tickDuration.)$(REF_SHORT to, std,conv)$(D !Duration()))
$(TD -)
$(TD $(D dur!"msecs"(5)) or $(D 5.msecs()))
)
$(TR $(TD $(B To $(LREF TickDuration)))
$(TD $(D duration.)$(REF_SHORT to, std,conv)$(D !TickDuration()))
$(TD -)
$(TD -)
$(TD $(D TickDuration.from!"msecs"(msecs)))
)
$(TR $(TD $(B To $(LREF FracSec)))
$(TD $(D duration.fracSec))
$(TD -)
$(TD -)
$(TD $(D FracSec.from!"msecs"(msecs)))
)
$(TR $(TD $(B To units))
$(TD $(D duration.total!"days"))
$(TD $(D tickDuration.msecs))
$(TD $(D fracSec.msecs))
$(TD $(D convert!("days", "msecs")(msecs)))
))
Copyright: Copyright 2010 - 2012
License: $(WEB www.boost.org/LICENSE_1_0.txt, Boost License 1.0).
Authors: Jonathan M Davis and Kato Shoichi
Source: $(DRUNTIMESRC core/_time.d)
Macros:
NBSP= 
+/
module core.time;
import core.exception;
import core.stdc.time;
import core.stdc.stdio;
import core.internal.traits : _Unqual = Unqual;
import core.internal.string;
version(Windows)
{
import core.sys.windows.windows;
}
else version(Posix)
{
import core.sys.posix.time;
import core.sys.posix.sys.time;
}
version (OSX)
version = Darwin;
else version (iOS)
version = Darwin;
else version (TVOS)
version = Darwin;
else version (WatchOS)
version = Darwin;
//This probably should be moved somewhere else in druntime which
//is Darwin-specific.
version(Darwin)
{
public import core.sys.darwin.mach.kern_return;
extern(C) nothrow @nogc
{
struct mach_timebase_info_data_t
{
uint numer;
uint denom;
}
alias mach_timebase_info_data_t* mach_timebase_info_t;
kern_return_t mach_timebase_info(mach_timebase_info_t);
ulong mach_absolute_time();
}
}
//To verify that an lvalue isn't required.
version(unittest) private T copy(T)(T t)
{
return t;
}
/++
What type of clock to use with $(LREF MonoTime) / $(LREF MonoTimeImpl) or
$(D std.datetime.Clock.currTime). They default to $(D ClockType.normal),
and most programs do not need to ever deal with the others.
The other $(D ClockType)s are provided so that other clocks provided by the
underlying C, system calls can be used with $(LREF MonoTimeImpl) or
$(D std.datetime.Clock.currTime) without having to use the C API directly.
In the case of the monotonic time, $(LREF MonoTimeImpl) is templatized on
$(D ClockType), whereas with $(D std.datetime.Clock.currTime), its a runtime
argument, since in the case of the monotonic time, the type of the clock
affects the resolution of a $(LREF MonoTimeImpl) object, whereas with
$(REF SysTime, std,datetime), its resolution is always hecto-nanoseconds
regardless of the source of the time.
$(D ClockType.normal), $(D ClockType.coarse), and $(D ClockType.precise)
work with both $(D Clock.currTime) and $(LREF MonoTimeImpl).
$(D ClockType.second) only works with $(D Clock.currTime). The others only
work with $(LREF MonoTimeImpl).
+/
version(CoreDdoc) enum ClockType
{
/++
Use the normal clock.
+/
normal = 0,
/++
$(BLUE Linux-Only)
Uses $(D CLOCK_BOOTTIME).
+/
bootTime = 1,
/++
Use the coarse clock, not the normal one (e.g. on Linux, that would be
$(D CLOCK_REALTIME_COARSE) instead of $(D CLOCK_REALTIME) for
$(D clock_gettime) if a function is using the realtime clock). It's
generally faster to get the time with the coarse clock than the normal
clock, but it's less precise (e.g. 1 msec instead of 1 usec or 1 nsec).
Howeover, it $(I is) guaranteed to still have sub-second precision
(just not as high as with $(D ClockType.normal)).
On systems which do not support a coarser clock,
$(D MonoTimeImpl!(ClockType.coarse)) will internally use the same clock
as $(D Monotime) does, and $(D Clock.currTime!(ClockType.coarse)) will
use the same clock as $(D Clock.currTime). This is because the coarse
clock is doing the same thing as the normal clock (just at lower
precision), whereas some of the other clock types
(e.g. $(D ClockType.processCPUTime)) mean something fundamentally
different. So, treating those as $(D ClockType.normal) on systems where
they weren't natively supported would give misleading results.
Most programs should not use the coarse clock, exactly because it's
less precise, and most programs don't need to get the time often
enough to care, but for those rare programs that need to get the time
extremely frequently (e.g. hundreds of thousands of times a second) but
don't care about high precision, the coarse clock might be appropriate.
Currently, only Linux and FreeBSD support a coarser clock, and on other
platforms, it's treated as $(D ClockType.normal).
+/
coarse = 2,
/++
Uses a more precise clock than the normal one (which is already very
precise), but it takes longer to get the time. Similarly to
$(D ClockType.coarse), if it's used on a system that does not support a
more precise clock than the normal one, it's treated as equivalent to
$(D ClockType.normal).
Currently, only FreeBSD supports a more precise clock, where it uses
$(D CLOCK_MONOTONIC_PRECISE) for the monotonic time and
$(D CLOCK_REALTIME_PRECISE) for the wall clock time.
+/
precise = 3,
/++
$(BLUE Linux,Solaris-Only)
Uses $(D CLOCK_PROCESS_CPUTIME_ID).
+/
processCPUTime = 4,
/++
$(BLUE Linux-Only)
Uses $(D CLOCK_MONOTONIC_RAW).
+/
raw = 5,
/++
Uses a clock that has a precision of one second (contrast to the coarse
clock, which has sub-second precision like the normal clock does).
FreeBSD is the only system which specifically has a clock set up for
this (it has $(D CLOCK_SECOND) to use with $(D clock_gettime) which
takes advantage of an in-kernel cached value), but on other systems, the
fastest function available will be used, and the resulting $(D SysTime)
will be rounded down to the second if the clock that was used gave the
time at a more precise resolution. So, it's guaranteed that the time
will be given at a precision of one second and it's likely the case that
will be faster than $(D ClockType.normal), since there tend to be
several options on a system to get the time at low resolutions, and they
tend to be faster than getting the time at high resolutions.
So, the primary difference between $(D ClockType.coarse) and
$(D ClockType.second) is that $(D ClockType.coarse) sacrifices some
precision in order to get speed but is still fairly precise, whereas
$(D ClockType.second) tries to be as fast as possible at the expense of
all sub-second precision.
+/
second = 6,
/++
$(BLUE Linux,Solaris-Only)
Uses $(D CLOCK_THREAD_CPUTIME_ID).
+/
threadCPUTime = 7,
/++
$(BLUE FreeBSD-Only)
Uses $(D CLOCK_UPTIME).
+/
uptime = 8,
/++
$(BLUE FreeBSD-Only)
Uses $(D CLOCK_UPTIME_FAST).
+/
uptimeCoarse = 9,
/++
$(BLUE FreeBSD-Only)
Uses $(D CLOCK_UPTIME_PRECISE).
+/
uptimePrecise = 10,
}
else version(Windows) enum ClockType
{
normal = 0,
coarse = 2,
precise = 3,
second = 6,
}
else version(Darwin) enum ClockType
{
normal = 0,
coarse = 2,
precise = 3,
second = 6,
}
else version(linux) enum ClockType
{
normal = 0,
bootTime = 1,
coarse = 2,
precise = 3,
processCPUTime = 4,
raw = 5,
second = 6,
threadCPUTime = 7,
}
else version(FreeBSD) enum ClockType
{
normal = 0,
coarse = 2,
precise = 3,
second = 6,
uptime = 8,
uptimeCoarse = 9,
uptimePrecise = 10,
}
else version(NetBSD) enum ClockType
{
normal = 0,
coarse = 2,
precise = 3,
second = 6,
}
else version(Solaris) enum ClockType
{
normal = 0,
coarse = 2,
precise = 3,
processCPUTime = 4,
second = 6,
threadCPUTime = 7,
}
else
{
// It needs to be decided (and implemented in an appropriate version branch
// here) which clock types new platforms are going to support. At minimum,
// the ones _not_ marked with $(D Blue Foo-Only) should be supported.
static assert(0, "What are the clock types supported by this system?");
}
// private, used to translate clock type to proper argument to clock_xxx
// functions on posix systems
version(CoreDdoc)
private int _posixClock(ClockType clockType) { return 0; }
else
version(Posix)
{
private auto _posixClock(ClockType clockType)
{
version(linux)
{
import core.sys.linux.time;
with(ClockType) final switch(clockType)
{
case bootTime: return CLOCK_BOOTTIME;
case coarse: return CLOCK_MONOTONIC_COARSE;
case normal: return CLOCK_MONOTONIC;
case precise: return CLOCK_MONOTONIC;
case processCPUTime: return CLOCK_PROCESS_CPUTIME_ID;
case raw: return CLOCK_MONOTONIC_RAW;
case threadCPUTime: return CLOCK_THREAD_CPUTIME_ID;
case second: assert(0);
}
}
else version(FreeBSD)
{
import core.sys.freebsd.time;
with(ClockType) final switch(clockType)
{
case coarse: return CLOCK_MONOTONIC_FAST;
case normal: return CLOCK_MONOTONIC;
case precise: return CLOCK_MONOTONIC_PRECISE;
case uptime: return CLOCK_UPTIME;
case uptimeCoarse: return CLOCK_UPTIME_FAST;
case uptimePrecise: return CLOCK_UPTIME_PRECISE;
case second: assert(0);
}
}
else version(NetBSD)
{
import core.sys.netbsd.time;
with(ClockType) final switch(clockType)
{
case coarse: return CLOCK_MONOTONIC;
case normal: return CLOCK_MONOTONIC;
case precise: return CLOCK_MONOTONIC;
case second: assert(0);
}
}
else version(Solaris)
{
import core.sys.solaris.time;
with(ClockType) final switch(clockType)
{
case coarse: return CLOCK_MONOTONIC;
case normal: return CLOCK_MONOTONIC;
case precise: return CLOCK_MONOTONIC;
case processCPUTime: return CLOCK_PROCESS_CPUTIME_ID;
case threadCPUTime: return CLOCK_THREAD_CPUTIME_ID;
case second: assert(0);
}
}
else
// It needs to be decided (and implemented in an appropriate
// version branch here) which clock types new platforms are going
// to support. Also, ClockType's documentation should be updated to
// mention it if a new platform uses anything that's not supported
// on all platforms..
assert(0, "What are the monotonic clock types supported by this system?");
}
}
unittest
{
// Make sure that the values are the same across platforms.
static if(is(typeof(ClockType.normal))) static assert(ClockType.normal == 0);
static if(is(typeof(ClockType.bootTime))) static assert(ClockType.bootTime == 1);
static if(is(typeof(ClockType.coarse))) static assert(ClockType.coarse == 2);
static if(is(typeof(ClockType.precise))) static assert(ClockType.precise == 3);
static if(is(typeof(ClockType.processCPUTime))) static assert(ClockType.processCPUTime == 4);
static if(is(typeof(ClockType.raw))) static assert(ClockType.raw == 5);
static if(is(typeof(ClockType.second))) static assert(ClockType.second == 6);
static if(is(typeof(ClockType.threadCPUTime))) static assert(ClockType.threadCPUTime == 7);
static if(is(typeof(ClockType.uptime))) static assert(ClockType.uptime == 8);
static if(is(typeof(ClockType.uptimeCoarse))) static assert(ClockType.uptimeCoarse == 9);
static if(is(typeof(ClockType.uptimePrecise))) static assert(ClockType.uptimePrecise == 10);
}
/++
Represents a duration of time of weeks or less (kept internally as hnsecs).
(e.g. 22 days or 700 seconds).
It is used when representing a duration of time - such as how long to
sleep with $(REF Thread.sleep, core,thread).
In std.datetime, it is also used as the result of various arithmetic
operations on time points.
Use the $(LREF dur) function or one of its non-generic aliases to create
$(D Duration)s.
It's not possible to create a Duration of months or years, because the
variable number of days in a month or year makes it impossible to convert
between months or years and smaller units without a specific date. So,
nothing uses $(D Duration)s when dealing with months or years. Rather,
functions specific to months and years are defined. For instance,
$(REF Date, std,datetime) has $(D add!"years") and $(D add!"months") for adding
years and months rather than creating a Duration of years or months and
adding that to a $(REF Date, std,datetime). But Duration is used when dealing
with weeks or smaller.
Examples:
--------------------
assert(dur!"days"(12) == dur!"hnsecs"(10_368_000_000_000L));
assert(dur!"hnsecs"(27) == dur!"hnsecs"(27));
assert(std.datetime.Date(2010, 9, 7) + dur!"days"(5) ==
std.datetime.Date(2010, 9, 12));
assert(days(-12) == dur!"hnsecs"(-10_368_000_000_000L));
assert(hnsecs(-27) == dur!"hnsecs"(-27));
assert(std.datetime.Date(2010, 9, 7) - std.datetime.Date(2010, 10, 3) ==
days(-26));
--------------------
+/
struct Duration
{
@safe pure:
public:
/++
A $(D Duration) of $(D 0). It's shorter than doing something like
$(D dur!"seconds"(0)) and more explicit than $(D Duration.init).
+/
static @property nothrow @nogc Duration zero() { return Duration(0); }
/++
Largest $(D Duration) possible.
+/
static @property nothrow @nogc Duration max() { return Duration(long.max); }
/++
Most negative $(D Duration) possible.
+/
static @property nothrow @nogc Duration min() { return Duration(long.min); }
unittest
{
assert(zero == dur!"seconds"(0));
assert(Duration.max == Duration(long.max));
assert(Duration.min == Duration(long.min));
assert(Duration.min < Duration.zero);
assert(Duration.zero < Duration.max);
assert(Duration.min < Duration.max);
assert(Duration.min - dur!"hnsecs"(1) == Duration.max);
assert(Duration.max + dur!"hnsecs"(1) == Duration.min);
}
/++
Compares this $(D Duration) with the given $(D Duration).
Returns:
$(TABLE
$(TR $(TD this &lt; rhs) $(TD &lt; 0))
$(TR $(TD this == rhs) $(TD 0))
$(TR $(TD this &gt; rhs) $(TD &gt; 0))
)
+/
int opCmp(Duration rhs) const nothrow @nogc
{
if(_hnsecs < rhs._hnsecs)
return -1;
if(_hnsecs > rhs._hnsecs)
return 1;
return 0;
}
unittest
{
foreach(T; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(U; _TypeTuple!(Duration, const Duration, immutable Duration))
{
T t = 42;
U u = t;
assert(t == u);
assert(copy(t) == u);
assert(t == copy(u));
}
}
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(E; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert((cast(D)Duration(12)).opCmp(cast(E)Duration(12)) == 0);
assert((cast(D)Duration(-12)).opCmp(cast(E)Duration(-12)) == 0);
assert((cast(D)Duration(10)).opCmp(cast(E)Duration(12)) < 0);
assert((cast(D)Duration(-12)).opCmp(cast(E)Duration(12)) < 0);
assert((cast(D)Duration(12)).opCmp(cast(E)Duration(10)) > 0);
assert((cast(D)Duration(12)).opCmp(cast(E)Duration(-12)) > 0);
assert(copy(cast(D)Duration(12)).opCmp(cast(E)Duration(12)) == 0);
assert(copy(cast(D)Duration(-12)).opCmp(cast(E)Duration(-12)) == 0);
assert(copy(cast(D)Duration(10)).opCmp(cast(E)Duration(12)) < 0);
assert(copy(cast(D)Duration(-12)).opCmp(cast(E)Duration(12)) < 0);
assert(copy(cast(D)Duration(12)).opCmp(cast(E)Duration(10)) > 0);
assert(copy(cast(D)Duration(12)).opCmp(cast(E)Duration(-12)) > 0);
assert((cast(D)Duration(12)).opCmp(copy(cast(E)Duration(12))) == 0);
assert((cast(D)Duration(-12)).opCmp(copy(cast(E)Duration(-12))) == 0);
assert((cast(D)Duration(10)).opCmp(copy(cast(E)Duration(12))) < 0);
assert((cast(D)Duration(-12)).opCmp(copy(cast(E)Duration(12))) < 0);
assert((cast(D)Duration(12)).opCmp(copy(cast(E)Duration(10))) > 0);
assert((cast(D)Duration(12)).opCmp(copy(cast(E)Duration(-12))) > 0);
}
}
}
/++
Adds, subtracts or calculates the modulo of two durations.
The legal types of arithmetic for $(D Duration) using this operator are
$(TABLE
$(TR $(TD Duration) $(TD +) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD -) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD %) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD +) $(TD TickDuration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD -) $(TD TickDuration) $(TD -->) $(TD Duration))
)
Params:
rhs = The duration to add to or subtract from this $(D Duration).
+/
Duration opBinary(string op, D)(D rhs) const nothrow @nogc
if(((op == "+" || op == "-" || op == "%") && is(_Unqual!D == Duration)) ||
((op == "+" || op == "-") && is(_Unqual!D == TickDuration)))
{
static if(is(_Unqual!D == Duration))
return Duration(mixin("_hnsecs " ~ op ~ " rhs._hnsecs"));
else if(is(_Unqual!D == TickDuration))
return Duration(mixin("_hnsecs " ~ op ~ " rhs.hnsecs"));
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(E; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert((cast(D)Duration(5)) + (cast(E)Duration(7)) == Duration(12));
assert((cast(D)Duration(5)) - (cast(E)Duration(7)) == Duration(-2));
assert((cast(D)Duration(5)) % (cast(E)Duration(7)) == Duration(5));
assert((cast(D)Duration(7)) + (cast(E)Duration(5)) == Duration(12));
assert((cast(D)Duration(7)) - (cast(E)Duration(5)) == Duration(2));
assert((cast(D)Duration(7)) % (cast(E)Duration(5)) == Duration(2));
assert((cast(D)Duration(5)) + (cast(E)Duration(-7)) == Duration(-2));
assert((cast(D)Duration(5)) - (cast(E)Duration(-7)) == Duration(12));
assert((cast(D)Duration(5)) % (cast(E)Duration(-7)) == Duration(5));
assert((cast(D)Duration(7)) + (cast(E)Duration(-5)) == Duration(2));
assert((cast(D)Duration(7)) - (cast(E)Duration(-5)) == Duration(12));
assert((cast(D)Duration(7)) % (cast(E)Duration(-5)) == Duration(2));
assert((cast(D)Duration(-5)) + (cast(E)Duration(7)) == Duration(2));
assert((cast(D)Duration(-5)) - (cast(E)Duration(7)) == Duration(-12));
assert((cast(D)Duration(-5)) % (cast(E)Duration(7)) == Duration(-5));
assert((cast(D)Duration(-7)) + (cast(E)Duration(5)) == Duration(-2));
assert((cast(D)Duration(-7)) - (cast(E)Duration(5)) == Duration(-12));
assert((cast(D)Duration(-7)) % (cast(E)Duration(5)) == Duration(-2));
assert((cast(D)Duration(-5)) + (cast(E)Duration(-7)) == Duration(-12));
assert((cast(D)Duration(-5)) - (cast(E)Duration(-7)) == Duration(2));
assert((cast(D)Duration(-5)) % (cast(E)Duration(7)) == Duration(-5));
assert((cast(D)Duration(-7)) + (cast(E)Duration(-5)) == Duration(-12));
assert((cast(D)Duration(-7)) - (cast(E)Duration(-5)) == Duration(-2));
assert((cast(D)Duration(-7)) % (cast(E)Duration(5)) == Duration(-2));
}
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
assertApprox((cast(D)Duration(5)) + cast(T)TickDuration.from!"usecs"(7), Duration(70), Duration(80));
assertApprox((cast(D)Duration(5)) - cast(T)TickDuration.from!"usecs"(7), Duration(-70), Duration(-60));
assertApprox((cast(D)Duration(7)) + cast(T)TickDuration.from!"usecs"(5), Duration(52), Duration(62));
assertApprox((cast(D)Duration(7)) - cast(T)TickDuration.from!"usecs"(5), Duration(-48), Duration(-38));
assertApprox((cast(D)Duration(5)) + cast(T)TickDuration.from!"usecs"(-7), Duration(-70), Duration(-60));
assertApprox((cast(D)Duration(5)) - cast(T)TickDuration.from!"usecs"(-7), Duration(70), Duration(80));
assertApprox((cast(D)Duration(7)) + cast(T)TickDuration.from!"usecs"(-5), Duration(-48), Duration(-38));
assertApprox((cast(D)Duration(7)) - cast(T)TickDuration.from!"usecs"(-5), Duration(52), Duration(62));
assertApprox((cast(D)Duration(-5)) + cast(T)TickDuration.from!"usecs"(7), Duration(60), Duration(70));
assertApprox((cast(D)Duration(-5)) - cast(T)TickDuration.from!"usecs"(7), Duration(-80), Duration(-70));
assertApprox((cast(D)Duration(-7)) + cast(T)TickDuration.from!"usecs"(5), Duration(38), Duration(48));
assertApprox((cast(D)Duration(-7)) - cast(T)TickDuration.from!"usecs"(5), Duration(-62), Duration(-52));
assertApprox((cast(D)Duration(-5)) + cast(T)TickDuration.from!"usecs"(-7), Duration(-80), Duration(-70));
assertApprox((cast(D)Duration(-5)) - cast(T)TickDuration.from!"usecs"(-7), Duration(60), Duration(70));
assertApprox((cast(D)Duration(-7)) + cast(T)TickDuration.from!"usecs"(-5), Duration(-62), Duration(-52));
assertApprox((cast(D)Duration(-7)) - cast(T)TickDuration.from!"usecs"(-5), Duration(38), Duration(48));
}
}
}
/++
Adds or subtracts two durations.
The legal types of arithmetic for $(D Duration) using this operator are
$(TABLE
$(TR $(TD TickDuration) $(TD +) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD TickDuration) $(TD -) $(TD Duration) $(TD -->) $(TD Duration))
)
Params:
lhs = The $(D TickDuration) to add to this $(D Duration) or to
subtract this $(D Duration) from.
+/
Duration opBinaryRight(string op, D)(D lhs) const nothrow @nogc
if((op == "+" || op == "-") &&
is(_Unqual!D == TickDuration))
{
return Duration(mixin("lhs.hnsecs " ~ op ~ " _hnsecs"));
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
assertApprox((cast(T)TickDuration.from!"usecs"(7)) + cast(D)Duration(5), Duration(70), Duration(80));
assertApprox((cast(T)TickDuration.from!"usecs"(7)) - cast(D)Duration(5), Duration(60), Duration(70));
assertApprox((cast(T)TickDuration.from!"usecs"(5)) + cast(D)Duration(7), Duration(52), Duration(62));
assertApprox((cast(T)TickDuration.from!"usecs"(5)) - cast(D)Duration(7), Duration(38), Duration(48));
assertApprox((cast(T)TickDuration.from!"usecs"(-7)) + cast(D)Duration(5), Duration(-70), Duration(-60));
assertApprox((cast(T)TickDuration.from!"usecs"(-7)) - cast(D)Duration(5), Duration(-80), Duration(-70));
assertApprox((cast(T)TickDuration.from!"usecs"(-5)) + cast(D)Duration(7), Duration(-48), Duration(-38));
assertApprox((cast(T)TickDuration.from!"usecs"(-5)) - cast(D)Duration(7), Duration(-62), Duration(-52));
assertApprox((cast(T)TickDuration.from!"usecs"(7)) + (cast(D)Duration(-5)), Duration(60), Duration(70));
assertApprox((cast(T)TickDuration.from!"usecs"(7)) - (cast(D)Duration(-5)), Duration(70), Duration(80));
assertApprox((cast(T)TickDuration.from!"usecs"(5)) + (cast(D)Duration(-7)), Duration(38), Duration(48));
assertApprox((cast(T)TickDuration.from!"usecs"(5)) - (cast(D)Duration(-7)), Duration(52), Duration(62));
assertApprox((cast(T)TickDuration.from!"usecs"(-7)) + cast(D)Duration(-5), Duration(-80), Duration(-70));
assertApprox((cast(T)TickDuration.from!"usecs"(-7)) - cast(D)Duration(-5), Duration(-70), Duration(-60));
assertApprox((cast(T)TickDuration.from!"usecs"(-5)) + cast(D)Duration(-7), Duration(-62), Duration(-52));
assertApprox((cast(T)TickDuration.from!"usecs"(-5)) - cast(D)Duration(-7), Duration(-48), Duration(-38));
}
}
}
/++
Adds, subtracts or calculates the modulo of two durations as well as
assigning the result to this $(D Duration).
The legal types of arithmetic for $(D Duration) using this operator are
$(TABLE
$(TR $(TD Duration) $(TD +) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD -) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD %) $(TD Duration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD +) $(TD TickDuration) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD -) $(TD TickDuration) $(TD -->) $(TD Duration))
)
Params:
rhs = The duration to add to or subtract from this $(D Duration).
+/
ref Duration opOpAssign(string op, D)(in D rhs) nothrow @nogc
if(((op == "+" || op == "-" || op == "%") && is(_Unqual!D == Duration)) ||
((op == "+" || op == "-") && is(_Unqual!D == TickDuration)))
{
static if(is(_Unqual!D == Duration))
mixin("_hnsecs " ~ op ~ "= rhs._hnsecs;");
else if(is(_Unqual!D == TickDuration))
mixin("_hnsecs " ~ op ~ "= rhs.hnsecs;");
return this;
}
unittest
{
static void test1(string op, E)(Duration actual, in E rhs, Duration expected, size_t line = __LINE__)
{
if(mixin("actual " ~ op ~ " rhs") != expected)
throw new AssertError("op failed", __FILE__, line);
if(actual != expected)
throw new AssertError("op assign failed", __FILE__, line);
}
static void test2(string op, E)
(Duration actual, in E rhs, Duration lower, Duration upper, size_t line = __LINE__)
{
assertApprox(mixin("actual " ~ op ~ " rhs"), lower, upper, "op failed", line);
assertApprox(actual, lower, upper, "op assign failed", line);
}
foreach(E; _TypeTuple!(Duration, const Duration, immutable Duration))
{
test1!"+="(Duration(5), (cast(E)Duration(7)), Duration(12));
test1!"-="(Duration(5), (cast(E)Duration(7)), Duration(-2));
test1!"%="(Duration(5), (cast(E)Duration(7)), Duration(5));
test1!"+="(Duration(7), (cast(E)Duration(5)), Duration(12));
test1!"-="(Duration(7), (cast(E)Duration(5)), Duration(2));
test1!"%="(Duration(7), (cast(E)Duration(5)), Duration(2));
test1!"+="(Duration(5), (cast(E)Duration(-7)), Duration(-2));
test1!"-="(Duration(5), (cast(E)Duration(-7)), Duration(12));
test1!"%="(Duration(5), (cast(E)Duration(-7)), Duration(5));
test1!"+="(Duration(7), (cast(E)Duration(-5)), Duration(2));
test1!"-="(Duration(7), (cast(E)Duration(-5)), Duration(12));
test1!"%="(Duration(7), (cast(E)Duration(-5)), Duration(2));
test1!"+="(Duration(-5), (cast(E)Duration(7)), Duration(2));
test1!"-="(Duration(-5), (cast(E)Duration(7)), Duration(-12));
test1!"%="(Duration(-5), (cast(E)Duration(7)), Duration(-5));
test1!"+="(Duration(-7), (cast(E)Duration(5)), Duration(-2));
test1!"-="(Duration(-7), (cast(E)Duration(5)), Duration(-12));
test1!"%="(Duration(-7), (cast(E)Duration(5)), Duration(-2));
test1!"+="(Duration(-5), (cast(E)Duration(-7)), Duration(-12));
test1!"-="(Duration(-5), (cast(E)Duration(-7)), Duration(2));
test1!"%="(Duration(-5), (cast(E)Duration(-7)), Duration(-5));
test1!"+="(Duration(-7), (cast(E)Duration(-5)), Duration(-12));
test1!"-="(Duration(-7), (cast(E)Duration(-5)), Duration(-2));
test1!"%="(Duration(-7), (cast(E)Duration(-5)), Duration(-2));
}
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
test2!"+="(Duration(5), cast(T)TickDuration.from!"usecs"(7), Duration(70), Duration(80));
test2!"-="(Duration(5), cast(T)TickDuration.from!"usecs"(7), Duration(-70), Duration(-60));
test2!"+="(Duration(7), cast(T)TickDuration.from!"usecs"(5), Duration(52), Duration(62));
test2!"-="(Duration(7), cast(T)TickDuration.from!"usecs"(5), Duration(-48), Duration(-38));
test2!"+="(Duration(5), cast(T)TickDuration.from!"usecs"(-7), Duration(-70), Duration(-60));
test2!"-="(Duration(5), cast(T)TickDuration.from!"usecs"(-7), Duration(70), Duration(80));
test2!"+="(Duration(7), cast(T)TickDuration.from!"usecs"(-5), Duration(-48), Duration(-38));
test2!"-="(Duration(7), cast(T)TickDuration.from!"usecs"(-5), Duration(52), Duration(62));
test2!"+="(Duration(-5), cast(T)TickDuration.from!"usecs"(7), Duration(60), Duration(70));
test2!"-="(Duration(-5), cast(T)TickDuration.from!"usecs"(7), Duration(-80), Duration(-70));
test2!"+="(Duration(-7), cast(T)TickDuration.from!"usecs"(5), Duration(38), Duration(48));
test2!"-="(Duration(-7), cast(T)TickDuration.from!"usecs"(5), Duration(-62), Duration(-52));
test2!"+="(Duration(-5), cast(T)TickDuration.from!"usecs"(-7), Duration(-80), Duration(-70));
test2!"-="(Duration(-5), cast(T)TickDuration.from!"usecs"(-7), Duration(60), Duration(70));
test2!"+="(Duration(-7), cast(T)TickDuration.from!"usecs"(-5), Duration(-62), Duration(-52));
test2!"-="(Duration(-7), cast(T)TickDuration.from!"usecs"(-5), Duration(38), Duration(48));
}
foreach(D; _TypeTuple!(const Duration, immutable Duration))
{
foreach(E; _TypeTuple!(Duration, const Duration, immutable Duration,
TickDuration, const TickDuration, immutable TickDuration))
{
D lhs = D(120);
E rhs = E(120);
static assert(!__traits(compiles, lhs += rhs), D.stringof ~ " " ~ E.stringof);
}
}
}
/++
Multiplies or divides the duration by an integer value.
The legal types of arithmetic for $(D Duration) using this operator
overload are
$(TABLE
$(TR $(TD Duration) $(TD *) $(TD long) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD /) $(TD long) $(TD -->) $(TD Duration))
)
Params:
value = The value to multiply this $(D Duration) by.
+/
Duration opBinary(string op)(long value) const nothrow @nogc
if(op == "*" || op == "/")
{
mixin("return Duration(_hnsecs " ~ op ~ " value);");
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert((cast(D)Duration(5)) * 7 == Duration(35));
assert((cast(D)Duration(7)) * 5 == Duration(35));
assert((cast(D)Duration(5)) * -7 == Duration(-35));
assert((cast(D)Duration(7)) * -5 == Duration(-35));
assert((cast(D)Duration(-5)) * 7 == Duration(-35));
assert((cast(D)Duration(-7)) * 5 == Duration(-35));
assert((cast(D)Duration(-5)) * -7 == Duration(35));
assert((cast(D)Duration(-7)) * -5 == Duration(35));
assert((cast(D)Duration(5)) * 0 == Duration(0));
assert((cast(D)Duration(-5)) * 0 == Duration(0));
}
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert((cast(D)Duration(5)) / 7 == Duration(0));
assert((cast(D)Duration(7)) / 5 == Duration(1));
assert((cast(D)Duration(5)) / -7 == Duration(0));
assert((cast(D)Duration(7)) / -5 == Duration(-1));
assert((cast(D)Duration(-5)) / 7 == Duration(0));
assert((cast(D)Duration(-7)) / 5 == Duration(-1));
assert((cast(D)Duration(-5)) / -7 == Duration(0));
assert((cast(D)Duration(-7)) / -5 == Duration(1));
}
}
/++
Multiplies/Divides the duration by an integer value as well as
assigning the result to this $(D Duration).
The legal types of arithmetic for $(D Duration) using this operator
overload are
$(TABLE
$(TR $(TD Duration) $(TD *) $(TD long) $(TD -->) $(TD Duration))
$(TR $(TD Duration) $(TD /) $(TD long) $(TD -->) $(TD Duration))
)
Params:
value = The value to multiply/divide this $(D Duration) by.
+/
ref Duration opOpAssign(string op)(long value) nothrow @nogc
if(op == "*" || op == "/")
{
mixin("_hnsecs " ~ op ~ "= value;");
return this;
}
unittest
{
static void test(D)(D actual, long value, Duration expected, size_t line = __LINE__)
{
if((actual *= value) != expected)
throw new AssertError("op failed", __FILE__, line);
if(actual != expected)
throw new AssertError("op assign failed", __FILE__, line);
}
test(Duration(5), 7, Duration(35));
test(Duration(7), 5, Duration(35));
test(Duration(5), -7, Duration(-35));
test(Duration(7), -5, Duration(-35));
test(Duration(-5), 7, Duration(-35));
test(Duration(-7), 5, Duration(-35));
test(Duration(-5), -7, Duration(35));
test(Duration(-7), -5, Duration(35));
test(Duration(5), 0, Duration(0));
test(Duration(-5), 0, Duration(0));
const cdur = Duration(12);
immutable idur = Duration(12);
static assert(!__traits(compiles, cdur *= 12));
static assert(!__traits(compiles, idur *= 12));
}
unittest
{
static void test(Duration actual, long value, Duration expected, size_t line = __LINE__)
{
if((actual /= value) != expected)
throw new AssertError("op failed", __FILE__, line);
if(actual != expected)
throw new AssertError("op assign failed", __FILE__, line);
}
test(Duration(5), 7, Duration(0));
test(Duration(7), 5, Duration(1));
test(Duration(5), -7, Duration(0));
test(Duration(7), -5, Duration(-1));
test(Duration(-5), 7, Duration(0));
test(Duration(-7), 5, Duration(-1));
test(Duration(-5), -7, Duration(0));
test(Duration(-7), -5, Duration(1));
const cdur = Duration(12);
immutable idur = Duration(12);
static assert(!__traits(compiles, cdur /= 12));
static assert(!__traits(compiles, idur /= 12));
}
/++
Divides two durations.
The legal types of arithmetic for $(D Duration) using this operator are
$(TABLE
$(TR $(TD Duration) $(TD /) $(TD Duration) $(TD -->) $(TD long))
)
Params:
rhs = The duration to divide this $(D Duration) by.
+/
long opBinary(string op)(Duration rhs) const nothrow @nogc
if(op == "/")
{
return _hnsecs / rhs._hnsecs;
}
unittest
{
assert(Duration(5) / Duration(7) == 0);
assert(Duration(7) / Duration(5) == 1);
assert(Duration(8) / Duration(4) == 2);
assert(Duration(5) / Duration(-7) == 0);
assert(Duration(7) / Duration(-5) == -1);
assert(Duration(8) / Duration(-4) == -2);
assert(Duration(-5) / Duration(7) == 0);
assert(Duration(-7) / Duration(5) == -1);
assert(Duration(-8) / Duration(4) == -2);
assert(Duration(-5) / Duration(-7) == 0);
assert(Duration(-7) / Duration(-5) == 1);
assert(Duration(-8) / Duration(-4) == 2);
}
/++
Multiplies an integral value and a $(D Duration).
The legal types of arithmetic for $(D Duration) using this operator
overload are
$(TABLE
$(TR $(TD long) $(TD *) $(TD Duration) $(TD -->) $(TD Duration))
)
Params:
value = The number of units to multiply this $(D Duration) by.
+/
Duration opBinaryRight(string op)(long value) const nothrow @nogc
if(op == "*")
{
return opBinary!op(value);
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert(5 * cast(D)Duration(7) == Duration(35));
assert(7 * cast(D)Duration(5) == Duration(35));
assert(5 * cast(D)Duration(-7) == Duration(-35));
assert(7 * cast(D)Duration(-5) == Duration(-35));
assert(-5 * cast(D)Duration(7) == Duration(-35));
assert(-7 * cast(D)Duration(5) == Duration(-35));
assert(-5 * cast(D)Duration(-7) == Duration(35));
assert(-7 * cast(D)Duration(-5) == Duration(35));
assert(0 * cast(D)Duration(-5) == Duration(0));
assert(0 * cast(D)Duration(5) == Duration(0));
}
}
/++
Returns the negation of this $(D Duration).
+/
Duration opUnary(string op)() const nothrow @nogc
if(op == "-")
{
return Duration(-_hnsecs);
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert(-(cast(D)Duration(7)) == Duration(-7));
assert(-(cast(D)Duration(5)) == Duration(-5));
assert(-(cast(D)Duration(-7)) == Duration(7));
assert(-(cast(D)Duration(-5)) == Duration(5));
assert(-(cast(D)Duration(0)) == Duration(0));
}
}
/++
Returns a $(LREF TickDuration) with the same number of hnsecs as this
$(D Duration).
Note that the conventional way to convert between $(D Duration) and
$(D TickDuration) is using $(REF to, std,conv), e.g.:
$(D duration.to!TickDuration())
+/
TickDuration opCast(T)() const nothrow @nogc
if(is(_Unqual!T == TickDuration))
{
return TickDuration.from!"hnsecs"(_hnsecs);
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(units; _TypeTuple!("seconds", "msecs", "usecs", "hnsecs"))
{
enum unitsPerSec = convert!("seconds", units)(1);
if(TickDuration.ticksPerSec >= unitsPerSec)
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
auto t = TickDuration.from!units(1);
assertApprox(cast(T)cast(D)dur!units(1), t - TickDuration(1), t + TickDuration(1), units);
t = TickDuration.from!units(2);
assertApprox(cast(T)cast(D)dur!units(2), t - TickDuration(1), t + TickDuration(1), units);
}
}
else
{
auto t = TickDuration.from!units(1);
assert(t.to!(units, long)() == 0, units);
t = TickDuration.from!units(1_000_000);
assert(t.to!(units, long)() >= 900_000, units);
assert(t.to!(units, long)() <= 1_100_000, units);
}
}
}
}
/++
Allow Duration to be used as a boolean.
Returns: `true` if this duration is non-zero.
+/
bool opCast(T : bool)() const nothrow @nogc
{
return _hnsecs != 0;
}
unittest
{
auto d = 10.minutes;
assert(d);
assert(!(d - d));
assert(d + d);
}
//Temporary hack until bug http://d.puremagic.com/issues/show_bug.cgi?id=5747 is fixed.
Duration opCast(T)() const nothrow @nogc
if(is(_Unqual!T == Duration))
{
return this;
}
/++
Splits out the Duration into the given units.
split takes the list of time units to split out as template arguments.
The time unit strings must be given in decreasing order. How it returns
the values for those units depends on the overload used.
The overload which accepts function arguments takes integral types in
the order that the time unit strings were given, and those integers are
passed by $(D ref). split assigns the values for the units to each
corresponding integer. Any integral type may be used, but no attempt is
made to prevent integer overflow, so don't use small integral types in
circumstances where the values for those units aren't likely to fit in
an integral type that small.
The overload with no arguments returns the values for the units in a
struct with members whose names are the same as the given time unit
strings. The members are all $(D long)s. This overload will also work
with no time strings being given, in which case $(I all) of the time
units from weeks through hnsecs will be provided (but no nsecs, since it
would always be $(D 0)).
For both overloads, the entire value of the Duration is split among the
units (rather than splitting the Duration across all units and then only
providing the values for the requested units), so if only one unit is
given, the result is equivalent to $(LREF total).
$(D "nsecs") is accepted by split, but $(D "years") and $(D "months")
are not.
For negative durations, all of the split values will be negative.
+/
template split(units...)
if(allAreAcceptedUnits!("weeks", "days", "hours", "minutes", "seconds",
"msecs", "usecs", "hnsecs", "nsecs")(units) &&
unitsAreInDescendingOrder(units))
{
/++ Ditto +/
void split(Args...)(out Args args) const nothrow @nogc
if(units.length != 0 && args.length == units.length && allAreMutableIntegralTypes!Args)
{
long hnsecs = _hnsecs;
foreach(i, unit; units)
{
static if(unit == "nsecs")
args[i] = cast(Args[i])convert!("hnsecs", "nsecs")(hnsecs);
else
args[i] = cast(Args[i])splitUnitsFromHNSecs!unit(hnsecs);
}
}
/++ Ditto +/
auto split() const nothrow @nogc
{
static if(units.length == 0)
return split!("weeks", "days", "hours", "minutes", "seconds", "msecs", "usecs", "hnsecs")();
else
{
static string genMemberDecls()
{
string retval;
foreach(unit; units)
{
retval ~= "long ";
retval ~= unit;
retval ~= "; ";
}
return retval;
}
static struct SplitUnits
{
mixin(genMemberDecls());
}
static string genSplitCall()
{
auto retval = "split(";
foreach(i, unit; units)
{
retval ~= "su.";
retval ~= unit;
if(i < units.length - 1)
retval ~= ", ";
else
retval ~= ");";
}
return retval;
}
SplitUnits su = void;
mixin(genSplitCall());
return su;
}
}
/+
Whether all of the given arguments are integral types.
+/
private template allAreMutableIntegralTypes(Args...)
{
static if(Args.length == 0)
enum allAreMutableIntegralTypes = true;
else static if(!is(Args[0] == long) &&
!is(Args[0] == int) &&
!is(Args[0] == short) &&
!is(Args[0] == byte) &&
!is(Args[0] == ulong) &&
!is(Args[0] == uint) &&
!is(Args[0] == ushort) &&
!is(Args[0] == ubyte))
{
enum allAreMutableIntegralTypes = false;
}
else
enum allAreMutableIntegralTypes = allAreMutableIntegralTypes!(Args[1 .. $]);
}
unittest
{
foreach(T; _TypeTuple!(long, int, short, byte, ulong, uint, ushort, ubyte))
static assert(allAreMutableIntegralTypes!T);
foreach(T; _TypeTuple!(long, int, short, byte, ulong, uint, ushort, ubyte))
static assert(!allAreMutableIntegralTypes!(const T));
foreach(T; _TypeTuple!(char, wchar, dchar, float, double, real, string))
static assert(!allAreMutableIntegralTypes!T);
static assert(allAreMutableIntegralTypes!(long, int, short, byte));
static assert(!allAreMutableIntegralTypes!(long, int, short, char, byte));
static assert(!allAreMutableIntegralTypes!(long, int*, short));
}
}
///
unittest
{
{
auto d = dur!"days"(12) + dur!"minutes"(7) + dur!"usecs"(501223);
long days;
int seconds;
short msecs;
d.split!("days", "seconds", "msecs")(days, seconds, msecs);
assert(days == 12);
assert(seconds == 7 * 60);
assert(msecs == 501);
auto splitStruct = d.split!("days", "seconds", "msecs")();
assert(splitStruct.days == 12);
assert(splitStruct.seconds == 7 * 60);
assert(splitStruct.msecs == 501);
auto fullSplitStruct = d.split();
assert(fullSplitStruct.weeks == 1);
assert(fullSplitStruct.days == 5);
assert(fullSplitStruct.hours == 0);
assert(fullSplitStruct.minutes == 7);
assert(fullSplitStruct.seconds == 0);
assert(fullSplitStruct.msecs == 501);
assert(fullSplitStruct.usecs == 223);
assert(fullSplitStruct.hnsecs == 0);
assert(d.split!"minutes"().minutes == d.total!"minutes");
}
{
auto d = dur!"days"(12);
assert(d.split!"weeks"().weeks == 1);
assert(d.split!"days"().days == 12);
assert(d.split().weeks == 1);
assert(d.split().days == 5);
}
{
auto d = dur!"days"(7) + dur!"hnsecs"(42);
assert(d.split!("seconds", "nsecs")().nsecs == 4200);
}
{
auto d = dur!"days"(-7) + dur!"hours"(-9);
auto result = d.split!("days", "hours")();
assert(result.days == -7);
assert(result.hours == -9);
}
}
pure nothrow unittest
{
foreach(D; _TypeTuple!(const Duration, immutable Duration))
{
D d = dur!"weeks"(3) + dur!"days"(5) + dur!"hours"(19) + dur!"minutes"(7) +
dur!"seconds"(2) + dur!"hnsecs"(1234567);
byte weeks;
ubyte days;
short hours;
ushort minutes;
int seconds;
uint msecs;
long usecs;
ulong hnsecs;
long nsecs;
d.split!("weeks", "days", "hours", "minutes", "seconds", "msecs", "usecs", "hnsecs", "nsecs")
(weeks, days, hours, minutes, seconds, msecs, usecs, hnsecs, nsecs);
assert(weeks == 3);
assert(days == 5);
assert(hours == 19);
assert(minutes == 7);
assert(seconds == 2);
assert(msecs == 123);
assert(usecs == 456);
assert(hnsecs == 7);
assert(nsecs == 0);
d.split!("weeks", "days", "hours", "seconds", "usecs")(weeks, days, hours, seconds, usecs);
assert(weeks == 3);
assert(days == 5);
assert(hours == 19);
assert(seconds == 422);
assert(usecs == 123456);
d.split!("days", "minutes", "seconds", "nsecs")(days, minutes, seconds, nsecs);
assert(days == 26);
assert(minutes == 1147);
assert(seconds == 2);
assert(nsecs == 123456700);
d.split!("minutes", "msecs", "usecs", "hnsecs")(minutes, msecs, usecs, hnsecs);
assert(minutes == 38587);
assert(msecs == 2123);
assert(usecs == 456);
assert(hnsecs == 7);
{
auto result = d.split!("weeks", "days", "hours", "minutes", "seconds",
"msecs", "usecs", "hnsecs", "nsecs");
assert(result.weeks == 3);
assert(result.days == 5);
assert(result.hours == 19);
assert(result.minutes == 7);
assert(result.seconds == 2);
assert(result.msecs == 123);
assert(result.usecs == 456);
assert(result.hnsecs == 7);
assert(result.nsecs == 0);
}
{
auto result = d.split!("weeks", "days", "hours", "seconds", "usecs");
assert(result.weeks == 3);
assert(result.days == 5);
assert(result.hours == 19);
assert(result.seconds == 422);
assert(result.usecs == 123456);
}
{
auto result = d.split!("days", "minutes", "seconds", "nsecs")();
assert(result.days == 26);
assert(result.minutes == 1147);
assert(result.seconds == 2);
assert(result.nsecs == 123456700);
}
{
auto result = d.split!("minutes", "msecs", "usecs", "hnsecs")();
assert(result.minutes == 38587);
assert(result.msecs == 2123);
assert(result.usecs == 456);
assert(result.hnsecs == 7);
}
{
auto result = d.split();
assert(result.weeks == 3);
assert(result.days == 5);
assert(result.hours == 19);
assert(result.minutes == 7);
assert(result.seconds == 2);
assert(result.msecs == 123);
assert(result.usecs == 456);
assert(result.hnsecs == 7);
static assert(!is(typeof(result.nsecs)));
}
static assert(!is(typeof(d.split("seconds", "hnsecs")(seconds))));
static assert(!is(typeof(d.split("hnsecs", "seconds", "minutes")(hnsecs, seconds, minutes))));
static assert(!is(typeof(d.split("hnsecs", "seconds", "msecs")(hnsecs, seconds, msecs))));
static assert(!is(typeof(d.split("seconds", "hnecs", "msecs")(seconds, hnsecs, msecs))));
static assert(!is(typeof(d.split("seconds", "msecs", "msecs")(seconds, msecs, msecs))));
static assert(!is(typeof(d.split("hnsecs", "seconds", "minutes")())));
static assert(!is(typeof(d.split("hnsecs", "seconds", "msecs")())));
static assert(!is(typeof(d.split("seconds", "hnecs", "msecs")())));
static assert(!is(typeof(d.split("seconds", "msecs", "msecs")())));
alias _TypeTuple!("nsecs", "hnsecs", "usecs", "msecs", "seconds",
"minutes", "hours", "days", "weeks") timeStrs;
foreach(i, str; timeStrs[1 .. $])
static assert(!is(typeof(d.split!(timeStrs[i - 1], str)())));
D nd = -d;
{
auto result = nd.split();
assert(result.weeks == -3);
assert(result.days == -5);
assert(result.hours == -19);
assert(result.minutes == -7);
assert(result.seconds == -2);
assert(result.msecs == -123);
assert(result.usecs == -456);
assert(result.hnsecs == -7);
}
{
auto result = nd.split!("weeks", "days", "hours", "minutes", "seconds", "nsecs")();
assert(result.weeks == -3);
assert(result.days == -5);
assert(result.hours == -19);
assert(result.minutes == -7);
assert(result.seconds == -2);
assert(result.nsecs == -123456700);
}
}
}
/++
Returns the total number of the given units in this $(D Duration).
So, unlike $(D split), it does not strip out the larger units.
+/
@property long total(string units)() const nothrow @nogc
if(units == "weeks" ||
units == "days" ||
units == "hours" ||
units == "minutes" ||
units == "seconds" ||
units == "msecs" ||
units == "usecs" ||
units == "hnsecs" ||
units == "nsecs")
{
static if(units == "nsecs")
return convert!("hnsecs", "nsecs")(_hnsecs);
else
return getUnitsFromHNSecs!units(_hnsecs);
}
///
unittest
{
assert(dur!"weeks"(12).total!"weeks" == 12);
assert(dur!"weeks"(12).total!"days" == 84);
assert(dur!"days"(13).total!"weeks" == 1);
assert(dur!"days"(13).total!"days" == 13);
assert(dur!"hours"(49).total!"days" == 2);
assert(dur!"hours"(49).total!"hours" == 49);
assert(dur!"nsecs"(2007).total!"hnsecs" == 20);
assert(dur!"nsecs"(2007).total!"nsecs" == 2000);
}
unittest
{
foreach(D; _TypeTuple!(const Duration, immutable Duration))
{
assert((cast(D)dur!"weeks"(12)).total!"weeks" == 12);
assert((cast(D)dur!"weeks"(12)).total!"days" == 84);
assert((cast(D)dur!"days"(13)).total!"weeks" == 1);
assert((cast(D)dur!"days"(13)).total!"days" == 13);
assert((cast(D)dur!"hours"(49)).total!"days" == 2);
assert((cast(D)dur!"hours"(49)).total!"hours" == 49);
assert((cast(D)dur!"nsecs"(2007)).total!"hnsecs" == 20);
assert((cast(D)dur!"nsecs"(2007)).total!"nsecs" == 2000);
}
}
/+
Converts this $(D Duration) to a $(D string).
+/
//Due to bug http://d.puremagic.com/issues/show_bug.cgi?id=3715 , we can't
//have versions of toString() with extra modifiers, so we define one version
//with modifiers and one without.
string toString()
{
return _toStringImpl();
}
/++
Converts this $(D Duration) to a $(D string).
+/
//Due to bug http://d.puremagic.com/issues/show_bug.cgi?id=3715 , we can't
//have versions of toString() with extra modifiers, so we define one version
//with modifiers and one without.
string toString() const nothrow
{
return _toStringImpl();
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert((cast(D)Duration(0)).toString() == "0 hnsecs");
assert((cast(D)Duration(1)).toString() == "1 hnsec");
assert((cast(D)Duration(7)).toString() == "7 hnsecs");
assert((cast(D)Duration(10)).toString() == "1 μs");
assert((cast(D)Duration(20)).toString() == "2 μs");
assert((cast(D)Duration(10_000)).toString() == "1 ms");
assert((cast(D)Duration(20_000)).toString() == "2 ms");
assert((cast(D)Duration(10_000_000)).toString() == "1 sec");
assert((cast(D)Duration(20_000_000)).toString() == "2 secs");
assert((cast(D)Duration(600_000_000)).toString() == "1 minute");
assert((cast(D)Duration(1_200_000_000)).toString() == "2 minutes");
assert((cast(D)Duration(36_000_000_000)).toString() == "1 hour");
assert((cast(D)Duration(72_000_000_000)).toString() == "2 hours");
assert((cast(D)Duration(864_000_000_000)).toString() == "1 day");
assert((cast(D)Duration(1_728_000_000_000)).toString() == "2 days");
assert((cast(D)Duration(6_048_000_000_000)).toString() == "1 week");
assert((cast(D)Duration(12_096_000_000_000)).toString() == "2 weeks");
assert((cast(D)Duration(12)).toString() == "1 μs and 2 hnsecs");
assert((cast(D)Duration(120_795)).toString() == "12 ms, 79 μs, and 5 hnsecs");
assert((cast(D)Duration(12_096_020_900_003)).toString() == "2 weeks, 2 secs, 90 ms, and 3 hnsecs");
assert((cast(D)Duration(-1)).toString() == "-1 hnsecs");
assert((cast(D)Duration(-7)).toString() == "-7 hnsecs");
assert((cast(D)Duration(-10)).toString() == "-1 μs");
assert((cast(D)Duration(-20)).toString() == "-2 μs");
assert((cast(D)Duration(-10_000)).toString() == "-1 ms");
assert((cast(D)Duration(-20_000)).toString() == "-2 ms");
assert((cast(D)Duration(-10_000_000)).toString() == "-1 secs");
assert((cast(D)Duration(-20_000_000)).toString() == "-2 secs");
assert((cast(D)Duration(-600_000_000)).toString() == "-1 minutes");
assert((cast(D)Duration(-1_200_000_000)).toString() == "-2 minutes");
assert((cast(D)Duration(-36_000_000_000)).toString() == "-1 hours");
assert((cast(D)Duration(-72_000_000_000)).toString() == "-2 hours");
assert((cast(D)Duration(-864_000_000_000)).toString() == "-1 days");
assert((cast(D)Duration(-1_728_000_000_000)).toString() == "-2 days");
assert((cast(D)Duration(-6_048_000_000_000)).toString() == "-1 weeks");
assert((cast(D)Duration(-12_096_000_000_000)).toString() == "-2 weeks");
assert((cast(D)Duration(-12)).toString() == "-1 μs and -2 hnsecs");
assert((cast(D)Duration(-120_795)).toString() == "-12 ms, -79 μs, and -5 hnsecs");
assert((cast(D)Duration(-12_096_020_900_003)).toString() == "-2 weeks, -2 secs, -90 ms, and -3 hnsecs");
}
}
/++
Returns whether this $(D Duration) is negative.
+/
@property bool isNegative() const nothrow @nogc
{
return _hnsecs < 0;
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert(!(cast(D)Duration(100)).isNegative);
assert(!(cast(D)Duration(1)).isNegative);
assert(!(cast(D)Duration(0)).isNegative);
assert((cast(D)Duration(-1)).isNegative);
assert((cast(D)Duration(-100)).isNegative);
}
}
private:
/+
Since we have two versions of toString, we have _toStringImpl
so that they can share implementations.
+/
string _toStringImpl() const nothrow
{
static void appListSep(ref string res, uint pos, bool last) nothrow
{
if (pos == 0)
return;
if (!last)
res ~= ", ";
else
res ~= pos == 1 ? " and " : ", and ";
}
static void appUnitVal(string units)(ref string res, long val) nothrow
{
immutable plural = val != 1;
string unit;
static if (units == "seconds")
unit = plural ? "secs" : "sec";
else static if (units == "msecs")
unit = "ms";
else static if (units == "usecs")
unit = "μs";
else
unit = plural ? units : units[0 .. $-1];
res ~= signedToTempString(val, 10);
res ~= " ";
res ~= unit;
}
if (_hnsecs == 0) return "0 hnsecs";
template TT(T...) { alias T TT; }
alias units = TT!("weeks", "days", "hours", "minutes", "seconds", "msecs", "usecs");
long hnsecs = _hnsecs; string res; uint pos;
foreach (unit; units)
{
if (auto val = splitUnitsFromHNSecs!unit(hnsecs))
{
appListSep(res, pos++, hnsecs == 0);
appUnitVal!unit(res, val);
}
if (hnsecs == 0) break;
}
if (hnsecs != 0)
{
appListSep(res, pos++, true);
appUnitVal!"hnsecs"(res, hnsecs);
}
return res;
}
/+
Params:
hnsecs = The total number of hecto-nanoseconds in this $(D Duration).
+/
this(long hnsecs) nothrow @nogc
{
_hnsecs = hnsecs;
}
long _hnsecs;
}
///
unittest
{
import core.time;
// using the dur template
auto numDays = dur!"days"(12);
// using the days function
numDays = days(12);
// alternatively using UFCS syntax
numDays = 12.days;
auto myTime = 100.msecs + 20_000.usecs + 30_000.hnsecs;
assert(myTime == 123.msecs);
}
/++
Converts a $(D TickDuration) to the given units as either an integral
value or a floating point value.
Params:
units = The units to convert to. Accepts $(D "seconds") and smaller
only.
T = The type to convert to (either an integral type or a
floating point type).
td = The TickDuration to convert
+/
T to(string units, T, D)(D td) @safe pure nothrow @nogc
if(is(_Unqual!D == TickDuration) &&
(units == "seconds" ||
units == "msecs" ||
units == "usecs" ||
units == "hnsecs" ||
units == "nsecs"))
{
static if(__traits(isIntegral, T) && T.sizeof >= 4)
{
enum unitsPerSec = convert!("seconds", units)(1);
return cast(T) (td.length / (TickDuration.ticksPerSec / cast(real) unitsPerSec));
}
else static if(__traits(isFloating, T))
{
static if(units == "seconds")
return td.length / cast(T)TickDuration.ticksPerSec;
else
{
enum unitsPerSec = convert!("seconds", units)(1);
return cast(T) (td.length /
(TickDuration.ticksPerSec / cast(real) unitsPerSec));
}
}
else
static assert(0, "Incorrect template constraint.");
}
///
unittest
{
auto t = TickDuration.from!"seconds"(1000);
long tl = to!("seconds",long)(t);
assert(tl == 1000);
double td = to!("seconds",double)(t);
assert(_abs(td - 1000) < 0.001);
}
unittest
{
void testFun(string U)() {
auto t1v = 1000;
auto t2v = 333;
auto t1 = TickDuration.from!U(t1v);
auto t2 = TickDuration.from!U(t2v);
auto _str(F)(F val)
{
static if(is(F == int) || is(F == long))
return signedToTempString(val, 10);
else
return unsignedToTempString(val, 10);
}
foreach (F; _TypeTuple!(int,uint,long,ulong,float,double,real))
{
F t1f = to!(U,F)(t1);
F t2f = to!(U,F)(t2);
auto t12d = t1 / t2v;
auto t12m = t1 - t2;
F t3f = to!(U,F)(t12d);
F t4f = to!(U,F)(t12m);
static if(is(F == float) || is(F == double) || is(F == real))
{
assert((t1f - cast(F)t1v) <= 3.0,
F.stringof ~ " " ~ U ~ " " ~ doubleToString(t1f) ~ " " ~
doubleToString(cast(F)t1v)
);
assert((t2f - cast(F)t2v) <= 3.0,
F.stringof ~ " " ~ U ~ " " ~ doubleToString(t2f) ~ " " ~
doubleToString(cast(F)t2v)
);
assert(t3f - (cast(F)t1v) / (cast(F)t2v) <= 3.0,
F.stringof ~ " " ~ U ~ " " ~ doubleToString(t3f) ~ " " ~
doubleToString((cast(F)t1v)/(cast(F)t2v))
);
assert(t4f - (cast(F)(t1v - t2v)) <= 3.0,
F.stringof ~ " " ~ U ~ " " ~ doubleToString(t4f) ~ " " ~
doubleToString(cast(F)(t1v - t2v))
);
}
else
{
// even though this should be exact math it is not as internal
// in "to" floating point is used
assert(_abs(t1f) - _abs(cast(F)t1v) <= 3,
F.stringof ~ " " ~ U ~ " " ~ _str(t1f) ~ " " ~
_str(cast(F)t1v)
);
assert(_abs(t2f) - _abs(cast(F)t2v) <= 3,
F.stringof ~ " " ~ U ~ " " ~ _str(t2f) ~ " " ~
_str(cast(F)t2v)
);
assert(_abs(t3f) - _abs((cast(F)t1v) / (cast(F)t2v)) <= 3,
F.stringof ~ " " ~ U ~ " " ~ _str(t3f) ~ " " ~
_str((cast(F)t1v) / (cast(F)t2v))
);
assert(_abs(t4f) - _abs((cast(F)t1v) - (cast(F)t2v)) <= 3,
F.stringof ~ " " ~ U ~ " " ~ _str(t4f) ~ " " ~
_str((cast(F)t1v) - (cast(F)t2v))
);
}
}
}
testFun!"seconds"();
testFun!"msecs"();
testFun!"usecs"();
}
/++
These allow you to construct a $(D Duration) from the given time units
with the given length.
You can either use the generic function $(D dur) and give it the units as
a $(D string) or use the named aliases.
The possible values for units are $(D "weeks"), $(D "days"), $(D "hours"),
$(D "minutes"), $(D "seconds"), $(D "msecs") (milliseconds), $(D "usecs"),
(microseconds), $(D "hnsecs") (hecto-nanoseconds, i.e. 100 ns), and
$(D "nsecs").
Params:
units = The time units of the $(D Duration) (e.g. $(D "days")).
length = The number of units in the $(D Duration).
+/
Duration dur(string units)(long length) @safe pure nothrow @nogc
if(units == "weeks" ||
units == "days" ||
units == "hours" ||
units == "minutes" ||
units == "seconds" ||
units == "msecs" ||
units == "usecs" ||
units == "hnsecs" ||
units == "nsecs")
{
return Duration(convert!(units, "hnsecs")(length));
}
alias weeks = dur!"weeks"; /// Ditto
alias days = dur!"days"; /// Ditto
alias hours = dur!"hours"; /// Ditto
alias minutes = dur!"minutes"; /// Ditto
alias seconds = dur!"seconds"; /// Ditto
alias msecs = dur!"msecs"; /// Ditto
alias usecs = dur!"usecs"; /// Ditto
alias hnsecs = dur!"hnsecs"; /// Ditto
alias nsecs = dur!"nsecs"; /// Ditto
///
unittest
{
// Generic
assert(dur!"weeks"(142).total!"weeks" == 142);
assert(dur!"days"(142).total!"days" == 142);
assert(dur!"hours"(142).total!"hours" == 142);
assert(dur!"minutes"(142).total!"minutes" == 142);
assert(dur!"seconds"(142).total!"seconds" == 142);
assert(dur!"msecs"(142).total!"msecs" == 142);
assert(dur!"usecs"(142).total!"usecs" == 142);
assert(dur!"hnsecs"(142).total!"hnsecs" == 142);
assert(dur!"nsecs"(142).total!"nsecs" == 100);
// Non-generic
assert(weeks(142).total!"weeks" == 142);
assert(days(142).total!"days" == 142);
assert(hours(142).total!"hours" == 142);
assert(minutes(142).total!"minutes" == 142);
assert(seconds(142).total!"seconds" == 142);
assert(msecs(142).total!"msecs" == 142);
assert(usecs(142).total!"usecs" == 142);
assert(hnsecs(142).total!"hnsecs" == 142);
assert(nsecs(142).total!"nsecs" == 100);
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
assert(dur!"weeks"(7).total!"weeks" == 7);
assert(dur!"days"(7).total!"days" == 7);
assert(dur!"hours"(7).total!"hours" == 7);
assert(dur!"minutes"(7).total!"minutes" == 7);
assert(dur!"seconds"(7).total!"seconds" == 7);
assert(dur!"msecs"(7).total!"msecs" == 7);
assert(dur!"usecs"(7).total!"usecs" == 7);
assert(dur!"hnsecs"(7).total!"hnsecs" == 7);
assert(dur!"nsecs"(7).total!"nsecs" == 0);
assert(dur!"weeks"(1007) == weeks(1007));
assert(dur!"days"(1007) == days(1007));
assert(dur!"hours"(1007) == hours(1007));
assert(dur!"minutes"(1007) == minutes(1007));
assert(dur!"seconds"(1007) == seconds(1007));
assert(dur!"msecs"(1007) == msecs(1007));
assert(dur!"usecs"(1007) == usecs(1007));
assert(dur!"hnsecs"(1007) == hnsecs(1007));
assert(dur!"nsecs"(10) == nsecs(10));
}
}
// used in MonoTimeImpl
private string _clockTypeName(ClockType clockType)
{
final switch(clockType)
{
foreach(name; __traits(allMembers, ClockType))
{
case __traits(getMember, ClockType, name):
return name;
}
}
assert(0);
}
// used in MonoTimeImpl
private size_t _clockTypeIdx(ClockType clockType)
{
final switch(clockType)
{
foreach(i, name; __traits(allMembers, ClockType))
{
case __traits(getMember, ClockType, name):
return i;
}
}
assert(0);
}
/++
alias for $(D MonoTimeImpl) instantiated with $(D ClockType.normal). This is
what most programs should use. It's also what much of $(D MonoTimeImpl) uses
in its documentation (particularly in the examples), because that's what's
going to be used in most code.
+/
alias MonoTime = MonoTimeImpl!(ClockType.normal);
/++
Represents a timestamp of the system's monotonic clock.
A monotonic clock is one which always goes forward and never moves
backwards, unlike the system's wall clock time (as represented by
$(REF SysTime, std,datetime)). The system's wall clock time can be adjusted
by the user or by the system itself via services such as NTP, so it is
unreliable to use the wall clock time for timing. Timers which use the wall
clock time could easily end up never going off due to changes made to the
wall clock time or otherwise waiting for a different period of time than
that specified by the programmer. However, because the monotonic clock
always increases at a fixed rate and is not affected by adjustments to the
wall clock time, it is ideal for use with timers or anything which requires
high precision timing.
So, MonoTime should be used for anything involving timers and timing,
whereas $(REF SysTime, std,datetime) should be used when the wall clock time
is required.
The monotonic clock has no relation to wall clock time. Rather, it holds
its time as the number of ticks of the clock which have occurred since the
clock started (typically when the system booted up). So, to determine how
much time has passed between two points in time, one monotonic time is
subtracted from the other to determine the number of ticks which occurred
between the two points of time, and those ticks are divided by the number of
ticks that occur every second (as represented by MonoTime.ticksPerSecond)
to get a meaningful duration of time. Normally, MonoTime does these
calculations for the programmer, but the $(D ticks) and $(D ticksPerSecond)
properties are provided for those who require direct access to the system
ticks. The normal way that MonoTime would be used is
--------------------
MonoTime before = MonoTime.currTime;
// do stuff...
MonoTime after = MonoTime.currTime;
Duration timeElapsed = after - before;
--------------------
$(LREF MonoTime) is an alias to $(D MonoTimeImpl!(ClockType.normal)) and is
what most programs should use for the monotonic clock, so that's what is
used in most of $(D MonoTimeImpl)'s documentation. But $(D MonoTimeImpl)
can be instantiated with other clock types for those rare programs that need
it.
See_Also:
$(LREF ClockType)
+/
struct MonoTimeImpl(ClockType clockType)
{
private enum _clockIdx = _clockTypeIdx(clockType);
private enum _clockName = _clockTypeName(clockType);
@safe:
version(Windows)
{
static if(clockType != ClockType.coarse &&
clockType != ClockType.normal &&
clockType != ClockType.precise)
{
static assert(0, "ClockType." ~ _clockName ~
" is not supported by MonoTimeImpl on this system.");
}
}
else version(Darwin)
{
static if(clockType != ClockType.coarse &&
clockType != ClockType.normal &&
clockType != ClockType.precise)
{
static assert(0, "ClockType." ~ _clockName ~
" is not supported by MonoTimeImpl on this system.");
}
}
else version(Posix)
{
enum clockArg = _posixClock(clockType);
}
else
static assert(0, "Unsupported platform");
// POD value, test mutable/const/immutable conversion
unittest
{
MonoTimeImpl m;
const MonoTimeImpl cm = m;
immutable MonoTimeImpl im = m;
m = cm;
m = im;
}
/++
The current time of the system's monotonic clock. This has no relation
to the wall clock time, as the wall clock time can be adjusted (e.g.
by NTP), whereas the monotonic clock always moves forward. The source
of the monotonic time is system-specific.
On Windows, $(D QueryPerformanceCounter) is used. On Mac OS X,
$(D mach_absolute_time) is used, while on other POSIX systems,
$(D clock_gettime) is used.
$(RED Warning): On some systems, the monotonic clock may stop counting
when the computer goes to sleep or hibernates. So, the
monotonic clock may indicate less time than has actually
passed if that occurs. This is known to happen on
Mac OS X. It has not been tested whether it occurs on
either Windows or Linux.
+/
static @property MonoTimeImpl currTime() @trusted nothrow @nogc
{
if(ticksPerSecond == 0)
{
import core.internal.abort : abort;
abort("MonoTimeImpl!(ClockType." ~ _clockName ~
") failed to get the frequency of the system's monotonic clock.");
}
version(Windows)
{
long ticks;
if(QueryPerformanceCounter(&ticks) == 0)
{
// This probably cannot happen on Windows 95 or later
import core.internal.abort : abort;
abort("Call to QueryPerformanceCounter failed.");
}
return MonoTimeImpl(ticks);
}
else version(Darwin)
return MonoTimeImpl(mach_absolute_time());
else version(Posix)
{
timespec ts;
if(clock_gettime(clockArg, &ts) != 0)
{
import core.internal.abort : abort;
abort("Call to clock_gettime failed.");
}
return MonoTimeImpl(convClockFreq(ts.tv_sec * 1_000_000_000L + ts.tv_nsec,
1_000_000_000L,
ticksPerSecond));
}
}
static @property pure nothrow @nogc
{
/++
A $(D MonoTime) of $(D 0) ticks. It's provided to be consistent with
$(D Duration.zero), and it's more explicit than $(D MonoTime.init).
+/
MonoTimeImpl zero() { return MonoTimeImpl(0); }
/++
Largest $(D MonoTime) possible.
+/
MonoTimeImpl max() { return MonoTimeImpl(long.max); }
/++
Most negative $(D MonoTime) possible.
+/
MonoTimeImpl min() { return MonoTimeImpl(long.min); }
}
unittest
{
assert(MonoTimeImpl.zero == MonoTimeImpl(0));
assert(MonoTimeImpl.max == MonoTimeImpl(long.max));
assert(MonoTimeImpl.min == MonoTimeImpl(long.min));
assert(MonoTimeImpl.min < MonoTimeImpl.zero);
assert(MonoTimeImpl.zero < MonoTimeImpl.max);
assert(MonoTimeImpl.min < MonoTimeImpl.max);
}
/++
Compares this MonoTime with the given MonoTime.
Returns:
$(BOOKTABLE,
$(TR $(TD this &lt; rhs) $(TD &lt; 0))
$(TR $(TD this == rhs) $(TD 0))
$(TR $(TD this &gt; rhs) $(TD &gt; 0))
)
+/
int opCmp(MonoTimeImpl rhs) const pure nothrow @nogc
{
if(_ticks < rhs._ticks)
return -1;
return _ticks > rhs._ticks ? 1 : 0;
}
unittest
{
const t = MonoTimeImpl.currTime;
assert(t == copy(t));
}
unittest
{
const before = MonoTimeImpl.currTime;
auto after = MonoTimeImpl(before._ticks + 42);
assert(before < after);
assert(copy(before) <= before);
assert(copy(after) > before);
assert(after >= copy(after));
}
unittest
{
const currTime = MonoTimeImpl.currTime;
assert(MonoTimeImpl(long.max) > MonoTimeImpl(0));
assert(MonoTimeImpl(0) > MonoTimeImpl(long.min));
assert(MonoTimeImpl(long.max) > currTime);
assert(currTime > MonoTimeImpl(0));
assert(MonoTimeImpl(0) < currTime);
assert(MonoTimeImpl(0) < MonoTimeImpl(long.max));
assert(MonoTimeImpl(long.min) < MonoTimeImpl(0));
}
/++
Subtracting two MonoTimes results in a $(LREF Duration) representing
the amount of time which elapsed between them.
The primary way that programs should time how long something takes is to
do
--------------------
MonoTime before = MonoTime.currTime;
// do stuff
MonoTime after = MonoTime.currTime;
// How long it took.
Duration timeElapsed = after - before;
--------------------
or to use a wrapper (such as a stop watch type) which does that.
$(RED Warning):
Because $(LREF Duration) is in hnsecs, whereas MonoTime is in system
ticks, it's usually the case that this assertion will fail
--------------------
auto before = MonoTime.currTime;
// do stuff
auto after = MonoTime.currTime;
auto timeElapsed = after - before;
assert(before + timeElapsed == after);
--------------------
This is generally fine, and by its very nature, converting from
system ticks to any type of seconds (hnsecs, nsecs, etc.) will
introduce rounding errors, but if code needs to avoid any of the
small rounding errors introduced by conversion, then it needs to use
MonoTime's $(D ticks) property and keep all calculations in ticks
rather than using $(LREF Duration).
+/
Duration opBinary(string op)(MonoTimeImpl rhs) const pure nothrow @nogc
if(op == "-")
{
immutable diff = _ticks - rhs._ticks;
return Duration(convClockFreq(diff , ticksPerSecond, hnsecsPer!"seconds"));
}
unittest
{
const t = MonoTimeImpl.currTime;
assert(t - copy(t) == Duration.zero);
static assert(!__traits(compiles, t + t));
}
unittest
{
static void test(in MonoTimeImpl before, in MonoTimeImpl after, in Duration min)
{
immutable diff = after - before;
assert(diff >= min);
auto calcAfter = before + diff;
assertApprox(calcAfter, calcAfter - Duration(1), calcAfter + Duration(1));
assert(before - after == -diff);
}
const before = MonoTimeImpl.currTime;
test(before, MonoTimeImpl(before._ticks + 4202), Duration.zero);
test(before, MonoTimeImpl.currTime, Duration.zero);
const durLargerUnits = dur!"minutes"(7) + dur!"seconds"(22);
test(before, before + durLargerUnits + dur!"msecs"(33) + dur!"hnsecs"(571), durLargerUnits);
}
/++
Adding or subtracting a $(LREF Duration) to/from a MonoTime results in
a MonoTime which is adjusted by that amount.
+/
MonoTimeImpl opBinary(string op)(Duration rhs) const pure nothrow @nogc
if(op == "+" || op == "-")
{
immutable rhsConverted = convClockFreq(rhs._hnsecs, hnsecsPer!"seconds", ticksPerSecond);
mixin("return MonoTimeImpl(_ticks " ~ op ~ " rhsConverted);");
}
unittest
{
const t = MonoTimeImpl.currTime;
assert(t + Duration(0) == t);
assert(t - Duration(0) == t);
}
unittest
{
const t = MonoTimeImpl.currTime;
// We reassign ticks in order to get the same rounding errors
// that we should be getting with Duration (e.g. MonoTimeImpl may be
// at a higher precision than hnsecs, meaning that 7333 would be
// truncated when converting to hnsecs).
long ticks = 7333;
auto hnsecs = convClockFreq(ticks, ticksPerSecond, hnsecsPer!"seconds");
ticks = convClockFreq(hnsecs, hnsecsPer!"seconds", ticksPerSecond);
assert(t - Duration(hnsecs) == MonoTimeImpl(t._ticks - ticks));
assert(t + Duration(hnsecs) == MonoTimeImpl(t._ticks + ticks));
}
/++ Ditto +/
ref MonoTimeImpl opOpAssign(string op)(Duration rhs) pure nothrow @nogc
if(op == "+" || op == "-")
{
immutable rhsConverted = convClockFreq(rhs._hnsecs, hnsecsPer!"seconds", ticksPerSecond);
mixin("_ticks " ~ op ~ "= rhsConverted;");
return this;
}
unittest
{
auto mt = MonoTimeImpl.currTime;
const initial = mt;
mt += Duration(0);
assert(mt == initial);
mt -= Duration(0);
assert(mt == initial);
// We reassign ticks in order to get the same rounding errors
// that we should be getting with Duration (e.g. MonoTimeImpl may be
// at a higher precision than hnsecs, meaning that 7333 would be
// truncated when converting to hnsecs).
long ticks = 7333;
auto hnsecs = convClockFreq(ticks, ticksPerSecond, hnsecsPer!"seconds");
ticks = convClockFreq(hnsecs, hnsecsPer!"seconds", ticksPerSecond);
auto before = MonoTimeImpl(initial._ticks - ticks);
assert((mt -= Duration(hnsecs)) == before);
assert(mt == before);
assert((mt += Duration(hnsecs)) == initial);
assert(mt == initial);
}
/++
The number of ticks in the monotonic time.
Most programs should not use this directly, but it's exposed for those
few programs that need it.
The main reasons that a program might need to use ticks directly is if
the system clock has higher precision than hnsecs, and the program needs
that higher precision, or if the program needs to avoid the rounding
errors caused by converting to hnsecs.
+/
@property long ticks() const pure nothrow @nogc
{
return _ticks;
}
unittest
{
const mt = MonoTimeImpl.currTime;
assert(mt.ticks == mt._ticks);
}
/++
The number of ticks that MonoTime has per second - i.e. the resolution
or frequency of the system's monotonic clock.
e.g. if the system clock had a resolution of microseconds, then
ticksPerSecond would be $(D 1_000_000).
+/
static @property long ticksPerSecond() pure nothrow @nogc
{
return _ticksPerSecond[_clockIdx];
}
unittest
{
assert(MonoTimeImpl.ticksPerSecond == _ticksPerSecond[_clockIdx]);
}
///
string toString() const pure nothrow
{
static if(clockType == ClockType.normal)
return "MonoTime(" ~ signedToTempString(_ticks, 10) ~ " ticks, " ~ signedToTempString(ticksPerSecond, 10) ~ " ticks per second)";
else
return "MonoTimeImpl!(ClockType." ~ _clockName ~ ")(" ~ signedToTempString(_ticks, 10) ~ " ticks, " ~
signedToTempString(ticksPerSecond, 10) ~ " ticks per second)";
}
unittest
{
static min(T)(T a, T b) { return a < b ? a : b; }
static void eat(ref string s, const(char)[] exp)
{
assert(s[0 .. min($, exp.length)] == exp, s~" != "~exp);
s = s[exp.length .. $];
}
immutable mt = MonoTimeImpl.currTime;
auto str = mt.toString();
static if(is(typeof(this) == MonoTime))
eat(str, "MonoTime(");
else
eat(str, "MonoTimeImpl!(ClockType."~_clockName~")(");
eat(str, signedToTempString(mt._ticks, 10));
eat(str, " ticks, ");
eat(str, signedToTempString(ticksPerSecond, 10));
eat(str, " ticks per second)");
}
private:
// static immutable long _ticksPerSecond;
unittest
{
assert(_ticksPerSecond[_clockIdx]);
}
long _ticks;
}
// This is supposed to be a static variable in MonoTimeImpl with the static
// constructor being in there, but https://issues.dlang.org/show_bug.cgi?id=14517
// prevents that from working. However, moving it back to a static ctor will
// reraise issues with other systems using MonoTime, so we should leave this
// here even when that bug is fixed.
private immutable long[__traits(allMembers, ClockType).length] _ticksPerSecond;
// This is called directly from the runtime initilization function (rt_init),
// instead of using a static constructor. Other subsystems inside the runtime
// (namely, the GC) may need time functionality, but cannot wait until the
// static ctors have run. Therefore, we initialize these specially. Because
// it's a normal function, we need to do some dangerous casting PLEASE take
// care when modifying this function, and it should NOT be called except from
// the runtime init.
//
// NOTE: the code below SPECIFICALLY does not assert when it cannot initialize
// the ticks per second array. This allows cases where a clock is never used on
// a system that doesn't support it. See bugzilla issue
// https://issues.dlang.org/show_bug.cgi?id=14863
// The assert will occur when someone attempts to use _ticksPerSecond for that
// value.
extern(C) void _d_initMonoTime()
{
// We need a mutable pointer to the ticksPerSecond array. Although this
// would appear to break immutability, it is logically the same as a static
// ctor. So we should ONLY write these values once (we will check for 0
// values when setting to ensure this is truly only called once).
auto tps = cast(long[])_ticksPerSecond[];
// If we try to do anything with ClockType in the documentation build, it'll
// trigger the static assertions related to ClockType, since the
// documentation build defines all of the possible ClockTypes, which won't
// work when they're used in the static ifs, because no system supports them
// all.
version(CoreDdoc)
{}
else version(Windows)
{
long ticksPerSecond;
if(QueryPerformanceFrequency(&ticksPerSecond) != 0)
{
foreach(i, typeStr; __traits(allMembers, ClockType))
{
// ensure we are only writing immutable data once
if(tps[i] != 0)
// should only be called once
assert(0);
tps[i] = ticksPerSecond;
}
}
}
else version(Darwin)
{
immutable long ticksPerSecond = machTicksPerSecond();
foreach(i, typeStr; __traits(allMembers, ClockType))
{
// ensure we are only writing immutable data once
if(tps[i] != 0)
// should only be called once
assert(0);
tps[i] = ticksPerSecond;
}
}
else version(Posix)
{
timespec ts;
foreach(i, typeStr; __traits(allMembers, ClockType))
{
static if(typeStr != "second")
{
enum clockArg = _posixClock(__traits(getMember, ClockType, typeStr));
if(clock_getres(clockArg, &ts) == 0)
{
// ensure we are only writing immutable data once
if(tps[i] != 0)
// should only be called once
assert(0);
// For some reason, on some systems, clock_getres returns
// a resolution which is clearly wrong:
// - it's a millisecond or worse, but the time is updated
// much more frequently than that.
// - it's negative
// - it's zero
// In such cases, we'll just use nanosecond resolution.
tps[i] = ts.tv_sec != 0 || ts.tv_nsec <= 0 || ts.tv_nsec >= 1000
? 1_000_000_000L : 1_000_000_000L / ts.tv_nsec;
}
}
}
}
}
// Tests for MonoTimeImpl.currTime. It has to be outside, because MonoTimeImpl
// is a template. This unittest block also makes sure that MonoTimeImpl actually
// is instantiated with all of the various ClockTypes so that those types and
// their tests are compiled and run.
unittest
{
// This test is separate so that it can be tested with MonoTime and not just
// MonoTimeImpl.
auto norm1 = MonoTime.currTime;
auto norm2 = MonoTimeImpl!(ClockType.normal).currTime;
assert(norm1 <= norm2);
static bool clockSupported(ClockType c)
{
version (Linux_Pre_2639) // skip CLOCK_BOOTTIME on older linux kernels
return c != ClockType.second && c != ClockType.bootTime;
else
return c != ClockType.second; // second doesn't work with MonoTimeImpl
}
foreach(typeStr; __traits(allMembers, ClockType))
{
mixin("alias type = ClockType." ~ typeStr ~ ";");
static if (clockSupported(type))
{
auto v1 = MonoTimeImpl!type.currTime;
auto v2 = MonoTimeImpl!type.currTime;
scope(failure)
{
printf("%s: v1 %s, v2 %s, tps %s\n",
(type.stringof ~ "\0").ptr,
numToStringz(v1._ticks),
numToStringz(v2._ticks),
numToStringz(typeof(v1).ticksPerSecond));
}
assert(v1 <= v2);
foreach(otherStr; __traits(allMembers, ClockType))
{
mixin("alias other = ClockType." ~ otherStr ~ ";");
static if (clockSupported(other))
{
static assert(is(typeof({auto o1 = MonTimeImpl!other.currTime; auto b = v1 <= o1;})) ==
is(type == other));
}
}
}
}
}
/++
Converts the given time from one clock frequency/resolution to another.
See_Also:
$(LREF ticksToNSecs)
+/
long convClockFreq(long ticks, long srcTicksPerSecond, long dstTicksPerSecond) @safe pure nothrow @nogc
{
// This would be more straightforward with floating point arithmetic,
// but we avoid it here in order to avoid the rounding errors that that
// introduces. Also, by splitting out the units in this way, we're able
// to deal with much larger values before running into problems with
// integer overflow.
return ticks / srcTicksPerSecond * dstTicksPerSecond +
ticks % srcTicksPerSecond * dstTicksPerSecond / srcTicksPerSecond;
}
///
unittest
{
// one tick is one second -> one tick is a hecto-nanosecond
assert(convClockFreq(45, 1, 10_000_000) == 450_000_000);
// one tick is one microsecond -> one tick is a millisecond
assert(convClockFreq(9029, 1_000_000, 1_000) == 9);
// one tick is 1/3_515_654 of a second -> 1/1_001_010 of a second
assert(convClockFreq(912_319, 3_515_654, 1_001_010) == 259_764);
// one tick is 1/MonoTime.ticksPerSecond -> one tick is a nanosecond
// Equivalent to ticksToNSecs
auto nsecs = convClockFreq(1982, MonoTime.ticksPerSecond, 1_000_000_000);
}
unittest
{
assert(convClockFreq(99, 43, 57) == 131);
assert(convClockFreq(131, 57, 43) == 98);
assert(convClockFreq(1234567890, 10_000_000, 1_000_000_000) == 123456789000);
assert(convClockFreq(1234567890, 1_000_000_000, 10_000_000) == 12345678);
assert(convClockFreq(123456789000, 1_000_000_000, 10_000_000) == 1234567890);
assert(convClockFreq(12345678, 10_000_000, 1_000_000_000) == 1234567800);
assert(convClockFreq(13131, 3_515_654, 10_000_000) == 37350);
assert(convClockFreq(37350, 10_000_000, 3_515_654) == 13130);
assert(convClockFreq(37350, 3_515_654, 10_000_000) == 106239);
assert(convClockFreq(106239, 10_000_000, 3_515_654) == 37349);
// It would be too expensive to cover a large range of possible values for
// ticks, so we use random values in an attempt to get reasonable coverage.
import core.stdc.stdlib;
immutable seed = cast(int)time(null);
srand(seed);
scope(failure) printf("seed %d\n", seed);
enum freq1 = 5_527_551L;
enum freq2 = 10_000_000L;
enum freq3 = 1_000_000_000L;
enum freq4 = 98_123_320L;
immutable freq5 = MonoTime.ticksPerSecond;
// This makes it so that freq6 is the first multiple of 10 which is greater
// than or equal to freq5, which at one point was considered for MonoTime's
// ticksPerSecond rather than using the system's actual clock frequency, so
// it seemed like a good test case to have.
import core.stdc.math;
immutable numDigitsMinus1 = cast(int)floor(log10(freq5));
auto freq6 = cast(long)pow(10, numDigitsMinus1);
if(freq5 > freq6)
freq6 *= 10;
foreach(_; 0 .. 10_000)
{
long[2] values = [rand(), cast(long)rand() * (rand() % 16)];
foreach(i; values)
{
scope(failure) printf("i %s\n", numToStringz(i));
assertApprox(convClockFreq(convClockFreq(i, freq1, freq2), freq2, freq1), i - 10, i + 10);
assertApprox(convClockFreq(convClockFreq(i, freq2, freq1), freq1, freq2), i - 10, i + 10);
assertApprox(convClockFreq(convClockFreq(i, freq3, freq4), freq4, freq3), i - 100, i + 100);
assertApprox(convClockFreq(convClockFreq(i, freq4, freq3), freq3, freq4), i - 100, i + 100);
scope(failure) printf("sys %s mt %s\n", numToStringz(freq5), numToStringz(freq6));
assertApprox(convClockFreq(convClockFreq(i, freq5, freq6), freq6, freq5), i - 10, i + 10);
assertApprox(convClockFreq(convClockFreq(i, freq6, freq5), freq5, freq6), i - 10, i + 10);
// This is here rather than in a unittest block immediately after
// ticksToNSecs in order to avoid code duplication in the unit tests.
assert(convClockFreq(i, MonoTime.ticksPerSecond, 1_000_000_000) == ticksToNSecs(i));
}
}
}
/++
Convenience wrapper around $(LREF convClockFreq) which converts ticks at
a clock frequency of $(D MonoTime.ticksPerSecond) to nanoseconds.
It's primarily of use when $(D MonoTime.ticksPerSecond) is greater than
hecto-nanosecond resolution, and an application needs a higher precision
than hecto-nanoceconds.
See_Also:
$(LREF convClockFreq)
+/
long ticksToNSecs(long ticks) @safe pure nothrow @nogc
{
return convClockFreq(ticks, MonoTime.ticksPerSecond, 1_000_000_000);
}
///
unittest
{
auto before = MonoTime.currTime;
// do stuff
auto after = MonoTime.currTime;
auto diffInTicks = after.ticks - before.ticks;
auto diffInNSecs = ticksToNSecs(diffInTicks);
assert(diffInNSecs == convClockFreq(diffInTicks, MonoTime.ticksPerSecond, 1_000_000_000));
}
/++
The reverse of $(LREF ticksToNSecs).
+/
long nsecsToTicks(long ticks) @safe pure nothrow @nogc
{
return convClockFreq(ticks, 1_000_000_000, MonoTime.ticksPerSecond);
}
unittest
{
long ticks = 123409832717333;
auto nsecs = convClockFreq(ticks, MonoTime.ticksPerSecond, 1_000_000_000);
ticks = convClockFreq(nsecs, 1_000_000_000, MonoTime.ticksPerSecond);
assert(nsecsToTicks(nsecs) == ticks);
}
/++
$(RED Warning: TickDuration will be deprecated in the near future (once all
uses of it in Phobos have been deprecated). Please use
$(LREF MonoTime) for the cases where a monotonic timestamp is needed
and $(LREF Duration) when a duration is needed, rather than using
TickDuration. It has been decided that TickDuration is too confusing
(e.g. it conflates a monotonic timestamp and a duration in monotonic
clock ticks) and that having multiple duration types is too awkward
and confusing.)
Represents a duration of time in system clock ticks.
The system clock ticks are the ticks of the system clock at the highest
precision that the system provides.
+/
struct TickDuration
{
/++
The number of ticks that the system clock has in one second.
If $(D ticksPerSec) is $(D 0), then then $(D TickDuration) failed to
get the value of $(D ticksPerSec) on the current system, and
$(D TickDuration) is not going to work. That would be highly abnormal
though.
+/
static immutable long ticksPerSec;
/++
The tick of the system clock (as a $(D TickDuration)) when the
application started.
+/
static immutable TickDuration appOrigin;
static @property @safe pure nothrow @nogc
{
/++
It's the same as $(D TickDuration(0)), but it's provided to be
consistent with $(D Duration) and $(D FracSec), which provide $(D zero)
properties.
+/
TickDuration zero() { return TickDuration(0); }
/++
Largest $(D TickDuration) possible.
+/
TickDuration max() { return TickDuration(long.max); }
/++
Most negative $(D TickDuration) possible.
+/
TickDuration min() { return TickDuration(long.min); }
}
unittest
{
assert(zero == TickDuration(0));
assert(TickDuration.max == TickDuration(long.max));
assert(TickDuration.min == TickDuration(long.min));
assert(TickDuration.min < TickDuration.zero);
assert(TickDuration.zero < TickDuration.max);
assert(TickDuration.min < TickDuration.max);
assert(TickDuration.min - TickDuration(1) == TickDuration.max);
assert(TickDuration.max + TickDuration(1) == TickDuration.min);
}
@trusted shared static this()
{
version(Windows)
{
if(QueryPerformanceFrequency(cast(long*)&ticksPerSec) == 0)
ticksPerSec = 0;
}
else version(Darwin)
{
ticksPerSec = machTicksPerSecond();
}
else version(Posix)
{
static if(is(typeof(clock_gettime)))
{
timespec ts;
if(clock_getres(CLOCK_MONOTONIC, &ts) != 0)
ticksPerSec = 0;
else
{
//For some reason, on some systems, clock_getres returns
//a resolution which is clearly wrong (it's a millisecond
//or worse, but the time is updated much more frequently
//than that). In such cases, we'll just use nanosecond
//resolution.
ticksPerSec = ts.tv_nsec >= 1000 ? 1_000_000_000
: 1_000_000_000 / ts.tv_nsec;
}
}
else
ticksPerSec = 1_000_000;
}
if(ticksPerSec != 0)
appOrigin = TickDuration.currSystemTick;
}
unittest
{
assert(ticksPerSec);
}
/++
The number of system ticks in this $(D TickDuration).
You can convert this $(D length) into the number of seconds by dividing
it by $(D ticksPerSec) (or using one the appropriate property function
to do it).
+/
long length;
/++
Returns the total number of seconds in this $(D TickDuration).
+/
@property long seconds() @safe const pure nothrow @nogc
{
return this.to!("seconds", long)();
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
assert((cast(T)TickDuration(ticksPerSec)).seconds == 1);
assert((cast(T)TickDuration(ticksPerSec - 1)).seconds == 0);
assert((cast(T)TickDuration(ticksPerSec * 2)).seconds == 2);
assert((cast(T)TickDuration(ticksPerSec * 2 - 1)).seconds == 1);
assert((cast(T)TickDuration(-1)).seconds == 0);
assert((cast(T)TickDuration(-ticksPerSec - 1)).seconds == -1);
assert((cast(T)TickDuration(-ticksPerSec)).seconds == -1);
}
}
/++
Returns the total number of milliseconds in this $(D TickDuration).
+/
@property long msecs() @safe const pure nothrow @nogc
{
return this.to!("msecs", long)();
}
/++
Returns the total number of microseconds in this $(D TickDuration).
+/
@property long usecs() @safe const pure nothrow @nogc
{
return this.to!("usecs", long)();
}
/++
Returns the total number of hecto-nanoseconds in this $(D TickDuration).
+/
@property long hnsecs() @safe const pure nothrow @nogc
{
return this.to!("hnsecs", long)();
}
/++
Returns the total number of nanoseconds in this $(D TickDuration).
+/
@property long nsecs() @safe const pure nothrow @nogc
{
return this.to!("nsecs", long)();
}
/++
This allows you to construct a $(D TickDuration) from the given time
units with the given length.
Params:
units = The time units of the $(D TickDuration) (e.g. $(D "msecs")).
length = The number of units in the $(D TickDuration).
+/
static TickDuration from(string units)(long length) @safe pure nothrow @nogc
if(units == "seconds" ||
units == "msecs" ||
units == "usecs" ||
units == "hnsecs" ||
units == "nsecs")
{
enum unitsPerSec = convert!("seconds", units)(1);
return TickDuration(cast(long)(length * (ticksPerSec / cast(real)unitsPerSec)));
}
unittest
{
foreach(units; _TypeTuple!("seconds", "msecs", "usecs", "nsecs"))
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
assertApprox((cast(T)TickDuration.from!units(1000)).to!(units, long)(),
500, 1500, units);
assertApprox((cast(T)TickDuration.from!units(1_000_000)).to!(units, long)(),
900_000, 1_100_000, units);
assertApprox((cast(T)TickDuration.from!units(2_000_000)).to!(units, long)(),
1_900_000, 2_100_000, units);
}
}
}
/++
Returns a $(LREF Duration) with the same number of hnsecs as this
$(D TickDuration).
Note that the conventional way to convert between $(D TickDuration)
and $(D Duration) is using $(REF to, std,conv), e.g.:
$(D tickDuration.to!Duration())
+/
Duration opCast(T)() @safe const pure nothrow @nogc
if(is(_Unqual!T == Duration))
{
return Duration(hnsecs);
}
unittest
{
foreach(D; _TypeTuple!(Duration, const Duration, immutable Duration))
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
auto expected = dur!"seconds"(1);
assert(cast(D)cast(T)TickDuration.from!"seconds"(1) == expected);
foreach(units; _TypeTuple!("msecs", "usecs", "hnsecs"))
{
D actual = cast(D)cast(T)TickDuration.from!units(1_000_000);
assertApprox(actual, dur!units(900_000), dur!units(1_100_000));
}
}
}
}
//Temporary hack until bug http://d.puremagic.com/issues/show_bug.cgi?id=5747 is fixed.
TickDuration opCast(T)() @safe const pure nothrow @nogc
if(is(_Unqual!T == TickDuration))
{
return this;
}
/++
Adds or subtracts two $(D TickDuration)s as well as assigning the result
to this $(D TickDuration).
The legal types of arithmetic for $(D TickDuration) using this operator
are
$(TABLE
$(TR $(TD TickDuration) $(TD +=) $(TD TickDuration) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD -=) $(TD TickDuration) $(TD -->) $(TD TickDuration))
)
Params:
rhs = The $(D TickDuration) to add to or subtract from this
$(D $(D TickDuration)).
+/
ref TickDuration opOpAssign(string op)(TickDuration rhs) @safe pure nothrow @nogc
if(op == "+" || op == "-")
{
mixin("length " ~ op ~ "= rhs.length;");
return this;
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
auto a = TickDuration.currSystemTick;
auto result = a += cast(T)TickDuration.currSystemTick;
assert(a == result);
assert(a.to!("seconds", real)() >= 0);
auto b = TickDuration.currSystemTick;
result = b -= cast(T)TickDuration.currSystemTick;
assert(b == result);
assert(b.to!("seconds", real)() <= 0);
foreach(U; _TypeTuple!(const TickDuration, immutable TickDuration))
{
U u = TickDuration(12);
static assert(!__traits(compiles, u += cast(T)TickDuration.currSystemTick));
static assert(!__traits(compiles, u -= cast(T)TickDuration.currSystemTick));
}
}
}
/++
Adds or subtracts two $(D TickDuration)s.
The legal types of arithmetic for $(D TickDuration) using this operator
are
$(TABLE
$(TR $(TD TickDuration) $(TD +) $(TD TickDuration) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD -) $(TD TickDuration) $(TD -->) $(TD TickDuration))
)
Params:
rhs = The $(D TickDuration) to add to or subtract from this
$(D TickDuration).
+/
TickDuration opBinary(string op)(TickDuration rhs) @safe const pure nothrow @nogc
if(op == "+" || op == "-")
{
return TickDuration(mixin("length " ~ op ~ " rhs.length"));
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
T a = TickDuration.currSystemTick;
T b = TickDuration.currSystemTick;
assert((a + b).seconds > 0);
assert((a - b).seconds <= 0);
}
}
/++
Returns the negation of this $(D TickDuration).
+/
TickDuration opUnary(string op)() @safe const pure nothrow @nogc
if(op == "-")
{
return TickDuration(-length);
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
assert(-(cast(T)TickDuration(7)) == TickDuration(-7));
assert(-(cast(T)TickDuration(5)) == TickDuration(-5));
assert(-(cast(T)TickDuration(-7)) == TickDuration(7));
assert(-(cast(T)TickDuration(-5)) == TickDuration(5));
assert(-(cast(T)TickDuration(0)) == TickDuration(0));
}
}
/++
operator overloading "<, >, <=, >="
+/
int opCmp(TickDuration rhs) @safe const pure nothrow @nogc
{
return length < rhs.length ? -1 : (length == rhs.length ? 0 : 1);
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
foreach(U; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
T t = TickDuration.currSystemTick;
U u = t;
assert(t == u);
assert(copy(t) == u);
assert(t == copy(u));
}
}
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
foreach(U; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
T t = TickDuration.currSystemTick;
U u = t + t;
assert(t < u);
assert(t <= t);
assert(u > t);
assert(u >= u);
assert(copy(t) < u);
assert(copy(t) <= t);
assert(copy(u) > t);
assert(copy(u) >= u);
assert(t < copy(u));
assert(t <= copy(t));
assert(u > copy(t));
assert(u >= copy(u));
}
}
}
/++
The legal types of arithmetic for $(D TickDuration) using this operator
overload are
$(TABLE
$(TR $(TD TickDuration) $(TD *) $(TD long) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD *) $(TD floating point) $(TD -->) $(TD TickDuration))
)
Params:
value = The value to divide from this duration.
+/
void opOpAssign(string op, T)(T value) @safe pure nothrow @nogc
if(op == "*" &&
(__traits(isIntegral, T) || __traits(isFloating, T)))
{
length = cast(long)(length * value);
}
unittest
{
immutable curr = TickDuration.currSystemTick;
TickDuration t1 = curr;
immutable t2 = curr + curr;
t1 *= 2;
assert(t1 == t2);
t1 = curr;
t1 *= 2.0;
immutable tol = TickDuration(cast(long)(_abs(t1.length) * double.epsilon * 2.0));
assertApprox(t1, t2 - tol, t2 + tol);
t1 = curr;
t1 *= 2.1;
assert(t1 > t2);
foreach(T; _TypeTuple!(const TickDuration, immutable TickDuration))
{
T t = TickDuration.currSystemTick;
assert(!__traits(compiles, t *= 12));
assert(!__traits(compiles, t *= 12.0));
}
}
/++
The legal types of arithmetic for $(D TickDuration) using this operator
overload are
$(TABLE
$(TR $(TD TickDuration) $(TD /) $(TD long) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD /) $(TD floating point) $(TD -->) $(TD TickDuration))
)
Params:
value = The value to divide from this $(D TickDuration).
Throws:
$(D TimeException) if an attempt to divide by $(D 0) is made.
+/
void opOpAssign(string op, T)(T value) @safe pure
if(op == "/" &&
(__traits(isIntegral, T) || __traits(isFloating, T)))
{
if(value == 0)
throw new TimeException("Attempted division by 0.");
length = cast(long)(length / value);
}
unittest
{
immutable curr = TickDuration.currSystemTick;
immutable t1 = curr;
TickDuration t2 = curr + curr;
t2 /= 2;
assert(t1 == t2);
t2 = curr + curr;
t2 /= 2.0;
immutable tol = TickDuration(cast(long)(_abs(t2.length) * double.epsilon / 2.0));
assertApprox(t1, t2 - tol, t2 + tol);
t2 = curr + curr;
t2 /= 2.1;
assert(t1 > t2);
_assertThrown!TimeException(t2 /= 0);
foreach(T; _TypeTuple!(const TickDuration, immutable TickDuration))
{
T t = TickDuration.currSystemTick;
assert(!__traits(compiles, t /= 12));
assert(!__traits(compiles, t /= 12.0));
}
}
/++
The legal types of arithmetic for $(D TickDuration) using this operator
overload are
$(TABLE
$(TR $(TD TickDuration) $(TD *) $(TD long) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD *) $(TD floating point) $(TD -->) $(TD TickDuration))
)
Params:
value = The value to divide from this $(D TickDuration).
+/
TickDuration opBinary(string op, T)(T value) @safe const pure nothrow @nogc
if(op == "*" &&
(__traits(isIntegral, T) || __traits(isFloating, T)))
{
return TickDuration(cast(long)(length * value));
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
T t1 = TickDuration.currSystemTick;
T t2 = t1 + t1;
assert(t1 * 2 == t2);
immutable tol = TickDuration(cast(long)(_abs(t1.length) * double.epsilon * 2.0));
assertApprox(t1 * 2.0, t2 - tol, t2 + tol);
assert(t1 * 2.1 > t2);
}
}
/++
The legal types of arithmetic for $(D TickDuration) using this operator
overload are
$(TABLE
$(TR $(TD TickDuration) $(TD /) $(TD long) $(TD -->) $(TD TickDuration))
$(TR $(TD TickDuration) $(TD /) $(TD floating point) $(TD -->) $(TD TickDuration))
)
Params:
value = The value to divide from this $(D TickDuration).
Throws:
$(D TimeException) if an attempt to divide by $(D 0) is made.
+/
TickDuration opBinary(string op, T)(T value) @safe const pure
if(op == "/" &&
(__traits(isIntegral, T) || __traits(isFloating, T)))
{
if(value == 0)
throw new TimeException("Attempted division by 0.");
return TickDuration(cast(long)(length / value));
}
unittest
{
foreach(T; _TypeTuple!(TickDuration, const TickDuration, immutable TickDuration))
{
T t1 = TickDuration.currSystemTick;
T t2 = t1 + t1;
assert(t2 / 2 == t1);
immutable tol = TickDuration(cast(long)(_abs(t2.length) * double.epsilon / 2.0));
assertApprox(t2 / 2.0, t1 - tol, t1 + tol);
assert(t2 / 2.1 < t1);
_assertThrown!TimeException(t2 / 0);
}
}
/++
Params:
ticks = The number of ticks in the TickDuration.
+/
@safe pure nothrow @nogc this(long ticks)
{
this.length = ticks;
}
unittest
{
foreach(i; [-42, 0, 42])
assert(TickDuration(i).length == i);
}
/++
The current system tick. The number of ticks per second varies from
system to system. $(D currSystemTick) uses a monotonic clock, so it's
intended for precision timing by comparing relative time values, not for
getting the current system time.
On Windows, $(D QueryPerformanceCounter) is used. On Mac OS X,
$(D mach_absolute_time) is used, while on other Posix systems,
$(D clock_gettime) is used. If $(D mach_absolute_time) or
$(D clock_gettime) is unavailable, then Posix systems use
$(D gettimeofday) (the decision is made when $(D TickDuration) is
compiled), which unfortunately, is not monotonic, but if
$(D mach_absolute_time) and $(D clock_gettime) aren't available, then
$(D gettimeofday) is the the best that there is.
$(RED Warning):
On some systems, the monotonic clock may stop counting when
the computer goes to sleep or hibernates. So, the monotonic
clock could be off if that occurs. This is known to happen
on Mac OS X. It has not been tested whether it occurs on
either Windows or on Linux.
Throws:
$(D TimeException) if it fails to get the time.
+/
static @property TickDuration currSystemTick() @trusted nothrow @nogc
{
import core.internal.abort : abort;
version(Windows)
{
ulong ticks;
if(QueryPerformanceCounter(cast(long*)&ticks) == 0)
abort("Failed in QueryPerformanceCounter().");
return TickDuration(ticks);
}
else version(Darwin)
{