Date entry evaluation

[DanJC]

Folks,

There are probably better ways than this but I hope this will spark some ideas.

If you want to test a users date input for validity, try using this to define your years!!

Struct Year
{
struct month
{
int no_of_days;
char *name; //optional field if you want month names
} months[12];
} leapyear, nonleapyear; //define year data in memory;

void main()
{
int i;
for(i = 0;i<12;i++)
{
switch(i+1)
{
case 1: nonleapyear.months.no_of_days = 31;
leapyear.months.no_of_days = 31;
break;
case 2: nonleapyear.months.no_of_days = 28;
leapyear.months.no_of_days = 29;
break;
.....
}
}

if you now get a user entry then you can compare string sections to the year variables month data, e.g. if month is greater than 12 then reject entry or (to put something in code) if day value is invalid, reject entry.

if((<compare value> > nonleapyear.months[month_id].no_of_days) ¦¦ (<compare value> < 1)
{
<error routine>
}
else
{
<accepted routine>
}

If anyone has any other ideas on this then please put them on this thread because dates are pretty complicated and I don't think I know enough to get the best out of them.

&quot;Remember.. it's all gibberish until someone puts it in terms YOU understand...&quot;

DanJC

 

[Zyrenthian]

leapyear can be defined by

year%4 == 0

Just a side note

Matt

 

[michaelkrauklis]

Actually leap year is a bit more complicated than that:

   if (year mod 4 != 0)
       {!leap yaer}
   else if  (year mod 400 == 0)
       {leap year}
   else if (year mod 100 == 0)
       {!leap year}
   else
       {leap year}

Also, approx every three years a leap second is declared to make things just a bit more accurate... but I don't think this part of the calculation is really necessary.
MY[red]enigma[/red]SELF:-9
myenigmaself@yahoo.com

http://myenigmaself.gaiden.com

 

[DanJC]

Worth noting that 100 and 400 are both factors of 4 so you do only need to use x % 4.

You can expand the struct to include different month name formats.

 

[DanJC]

Worth noting that 100 and 400 are both factors of 4 so you do only need to use x % 4.

You can expand the struct to include different month name formats and if you're brave use binary tree search

 

[michaelkrauklis]

If you look at my algorithm closer you will note that if a year is divisible by 400 it is a leap year, but then if it is divisible by 100 it is not. I repeat, if it is divisible by 100 and not 400 it is not a leap year. You didn't look at the algorithm close enough. The year 1000 was not a leap year. It is divisible by 100 but not 400. The year 2000 is a leap year. It is divisible by 100 and 400. As you can see, I hope, you cannot just use %4. MY[red]enigma[/red]SELF:-9
myenigmaself@yahoo.com

http://myenigmaself.gaiden.com