# Re: 12 hour clock and offset problem

Kenetic wrote:

Purl Gurl wrote:

(prior code example snipped)

Let me know if you find any glitches.

It looks like it works particularly with 12 am and 12 pm, and since
you have a handle on the script (there are some things in here I
haven't used extensively--just a novice at perl), I wonder if you can
modify it to include, .5 of an hour. The offset can be -3 or -3.5 for
3 hours and 30 minutes. Also to be considered is how :30 past the hour
would then wrap around to the next hour, which is easy enough
(although when wrapping past 12 am or pm might pose a challenge)

Rather than complete code for your task, I have opted for discussion.

You are, again, running into convention problems. Convention is to
adjust time by hours, only, to determine a local time. This is clearly
based on longitude location. Adding a problem is daylight savings time
which is used in some regions, not used in other regions.

Adjusting local time by minutes violates all conventions but does
not violate precision in time based on precise longitude location.
Use of minutes for time adjustment is truly stating degrees of
longitude in hours and minutes, which becomes rather complex, and

My suggestion is you look at longitude and latitude scripts with
giving direct attention to longitude formulas. This is a serious
suggestion which you should follow; you will find lots of coding
methods to incorporate seconds.

Right off, if your minutes will always be " x.5 " notation, this
is, always 30 minutes, I would simply add 30 minutes to your original
minutes, then adjust if this new minutes amount exceeds 60 minutes.

if (index (\$add_hours, ".5") > -1)
{ add 30 minutes and perform calculations here }

Following is example code for dealing with fractions. This example
does display working with a fractional hour and displays how precision
is lessened by use of the int() function within perl core. Use of int()
discards any fractional part and leaves only a whole number. Clearly
this reduces accuracy precision of results,

..5 equals 30 minutes
..33 equals 19.8 minutes (int return is 19)

However, if you are dealing with only tenths,

..1 .2 .3 .4 .5 .6 .7 .8 .9

there is no need for the int() function; returns are always whole numbers
without the int() function.

Although there is no need for int() in my example code, I have included
this function to demonstrate to readers how precision is reduced and to
demonstrate how code like this, how dealing with time, can become very
times beyond tenths, 3.53 hours or 3.67 hours and such, to compare code
results with hand calculated results; a fractional minute will be created
without int() functions included.

Find in my example,

This is simply a quick and dirty method to extract "3"

Find in my example,

\$add_minutes = int (\$hour_fraction * 60);

This is where precision is reduced for more complex
calculations. The int() function is NOT needed when
you deal only with tenths, 3.2 3.5 3.7 and such. This
int() can be removed,

However, if you introduce the hundredth decimal place,
you either must use int() and reduce precision or add
more code to maintain higher precision.

Again, look at longitude code examples to learn how to
create very high precision in time adjustment. You are
truly dealing with longitude calculations expressed as
time; hours, minutes and seconds. Give good attention
to the International Date Line to note how time can
be radically altered when you meet or exceed 180 degrees
longitude or, in time, plus 24 hours exactly or more.

I am not serving you up code on a silver platter. This
example code provides you the basic concept of how to
deal with time adjustments which include a fractional
incorporate this example into whatever final code you
elect to use; learning is the point, not easy answers.

No "walk through" this time. No full and complete code.
You have been provided enough help by participants here,
have been given enough code examples. You will learn a lot
more by writing your own code. Do so.

If you encounter problems after writing your own code,
then return and ask for help.

#!perl

\$time = "11:30 pm";

\$old_hour = substr (\$time, 0, index (\$time, ":"));
\$old_minutes = substr (\$time, -5, 2);

\$add_minutes = int (\$hour_fraction * 60);

if (\$new_minutes >= 60)
{
\$new_minutes = \$new_minutes - 60;
}

print "New Hour is \$new_hour";
print "\n\n";
print "New Minutes are \$new_minutes";

PRINTED RESULTS:

New Hour is 15

New Minutes are 0

---

Thanks for all the help so far, it's been overwhelming.

I must introduce something off-topic, must masturbate my
ego through rant. Research will disclose almost all within
this discussion group, almost all within our Perl Community,
for years, claim me to be an blithering idiot of a troll,
and do so, almost on a daily basis.

This amuses me, greatly.

--
Purl Gurl
--
"Then again what can you expect from a fat-assed, champagne swilling,
half-breed just off the Rez?"
- Joe Kline
.

## Relevant Pages

• Re: double to int conversion yields strange results
... giving me two different values for the int. ... the question raised is not about the precision of the ... is done in extended precision. ... It is the conversion to 'double' before the conversion to 'int' that ...
(comp.lang.c)
• Re: byte + byte -> int
... is in range for an int, but allows the line ... overflow, and the rules specifically address overflow. ... it thinks precision might be lost in a down-conversion. ... had one nice feature that our sleek fast modern machines have lost: ...
(comp.lang.java.programmer)
• Re: My scripting language - any suggestions?
... there will be a problem with precision. ... anything from Float it also does the property of Float being an ... Now what will happen if one argument is int and another ... conversion to int will not work ...
(comp.compilers)
• Re: Bezier lengths w/o lots of resources
... double qgauss(double (*func), double a, double b) ... constants need high precision. ... L = int(len(t)) dt ... So as long as the length along the curve goes smooth with t, ...
(comp.arch.embedded)
• Re: Can a double always represent an int exactly?
... and like all such has a precision of some fixed number of significant ... and int is also 64 bits wide ... then there are some values of type int which can not be exactly ... just about the guaranteed precise representation ...
(comp.lang.c)