(aptimelib)
We have here a highly portable 64-bit
time/date (datetime) library, with microsecond granularity, epoch
basis, alphanumeric formatting, and excellent range. Right
now, it compiles and test-runs under both MingW and Cygwin, and
it is thought that any gcc version supporting 'long long
int' should compile it very well indeed, on any platform.
This project has formally begun as of only 2/8/2006; it is being
called beta code with two
known issues of significance, latest version 0.7.
The code is intended to be used with everything Win32 and quasix, to be as compiler-transportable as possible. Very badly needed is help and specific suggestions as to rigorous testing and speed increase, and also help doing distribution-building of all kinds.
The aptimelib project was instigated while working on a project for a friend (to be released ATN), and needed a time library, and on 2/7/2006 started the research. And found several, none extremely multiplatform, none entirely portable, or transportable, or even extremely interoperable. I found somewhat promising items here and there, but nothing I could count on to work in all ways, in the extreme.
Except on David Tribble's web site. I found this page:
It has a SourceForge (
) project page:
http://sourceforge.net/projects/aptimelib
and download page:
https://sourceforge.net/project/showfiles.php?group_id=159505
and download page:
https://sourceforge.net/project/showfiles.php?group_id=159505
In addition to being at SourceForge, the current release notes are
here.
The code is intended to be used with everything Win32 and quasix, to be as compiler-transportable as possible. Very badly needed is help and specific suggestions as to rigorous testing and speed increase, and also help doing distribution-building of all kinds.
The aptimelib project was instigated while working on a project for a friend (to be released ATN), and needed a time library, and on 2/7/2006 started the research. And found several, none extremely multiplatform, none entirely portable, or transportable, or even extremely interoperable. I found somewhat promising items here and there, but nothing I could count on to work in all ways, in the extreme.
Except on David Tribble's web site. I found this page:
http://david.tribble.com/text/c0xlongtime.html
So I thought maybe the thought was good, the niche was real, the work just not quite done yet. I emailed David, and received wonderful encouragement. And here we are.
Known Issues of Significance
The first issue is that mklongtime() does not work. This was originally an oversight on Jonathan's part, for which he is apologetic; however, he does not appear to have the requisite ability to code it. Help is indicated. A skeleton exists at the end of aptimelib.c.
The other is that leap-second capability is not in yet. Leap-seconds are quite rare (thirty since 1972), so accuracy is not harmed much. There is much general effort being done to standardize a solution to the problem of handling leap-seconds (see this wikipedia entry). David is working on a solution. One of the principal aspects is the fact that the rotation of the earth is not predictable with extreme accuracy and precision in the long term.
So I thought maybe the thought was good, the niche was real, the work just not quite done yet. I emailed David, and received wonderful encouragement. And here we are.
Known Issues of Significance
The first issue is that mklongtime() does not work. This was originally an oversight on Jonathan's part, for which he is apologetic; however, he does not appear to have the requisite ability to code it. Help is indicated. A skeleton exists at the end of aptimelib.c.
The other is that leap-second capability is not in yet. Leap-seconds are quite rare (thirty since 1972), so accuracy is not harmed much. There is much general effort being done to standardize a solution to the problem of handling leap-seconds (see this wikipedia entry). David is working on a solution. One of the principal aspects is the fact that the rotation of the earth is not predictable with extreme accuracy and precision in the long term.