Date Validation Logic and Error message

I need help with writing a date validater in matlab without using built in time and date functions. It must check if a year is a leap year, and if so if february has no more than 29 days, and the other months satisfy their specific day maximums. If it isnt a leap year, then it checks if february has no more than 28 days, and the other months satisfy their specific day maximums. I have questions regarding checking if a year is a leap year and a new error message i have been getting.

1) Is my logic for checking for leap years correct? I used variables year1, year2, and year3 as assignments for the values of the year intputed divided by 4, 100, and 400, respectively. I then check if these values are evenly divided with the isfloat function. What would the correct logic look like?

2) I have been getting the error message "Output argument "valid" (and maybe others) not assigned during call to "valid_date"." What sources in my code could spark this error? I suspect i am overwriting the variable somehwere in my logic. I have done research to determine what this means and possible problems in the code and how to fix them, but I cannot for my life figure out what is causing the problem in my code.

I appreciate any comments. Please help because if have listened to every AC/DC album twice over while working on this code and i shouldnt have gray hair at 16 years old. I apoligize in advance for the quite messy code.

Thanks

function valid = valid_date(year, month, day)
year1 = year/4;
year2 = year/100;
year3 = year/400;
if isscalar(year) && year>=1 && isscalar(month) && month>=1 && month<=12 && isscalar(day) && day>=1 &&day<=31
    if (~isfloat(year1) && ~isfloat(year2)  && ~isfloat(year3)) || (~isfloat(year1) && isfloat(year2))
       if month == 2 
           if day<=29
               valid = true;
           else
               valid = false;
           end
       elseif month == 1 || month==3 || month==5 || month==7 || month==8 || month==10 || month==12
           if day<=31
               valid = true;
           else
               valid=false;
           end
       elseif month == 4 || month==6 || month==9 || month==11
                    if day <=30
                        valid=true;
                    else
                        valid=false; 
                    end
       else
           valid=false;
                
       end

    else
        if month <= 12
            if month == 2 && day <=28
                valid=true;
        
            elseif month==1 || month==3 || month==5 || month==7 || month==8 || month==10 || month==12
                    if day > 31
                        valid = false;
                    else
                        valid=true;
                    end
            else
                if month == 4 || month==6 || month==9 || month==11
                    if day >30
                        valid=false;
                    else
                        valid=true; 
                    end
                end
            end
        else
            valid=false;
        end
    end
else 
    valid = false;
end

end     

[Walter Roberson]

The second place you test for month 4, you do not have an "else" for that test.

[Walter Roberson]

isfloat() does not test whether a variable is integer valued: instead it tests to see whether the data type for the variable is single() or double(). So while 1920./4. = 480. is an integer value, it is stored in a floating point data type, the same as 1921./4. = 480.25 is.

See floor() and fix() and ceil() and round()

See also mod()

[Manish chouhan]

%here s is variable used for finding whether year is leap year or not 
function valid = valid_date(year, month, day)
year1 = year/4;
p = year1==round(year1);
year2 = year/100;
q = year2==round(year2);
year3 = year/400;
r = year3==round(year3);

if (r)
     s=1;
elseif ( q)
     s=0;
elseif ( p )
     s=1;
else
    s=0;
end 
s

if isscalar(year) && year>=1 && isscalar(month) && month>=1 && month<=12 && isscalar(day) && day>=1 &&day<=31
    if s
        if month == 2 
           valid= day<=29;
        elseif month == 1 || month==3 || month==5 || month==7 || month==8 || month==10 || month==12
           valid = day<=31;
        else
           valid=day <=30;
        end
    else
        if month == 2 
           valid= day<=28;
        elseif month == 1 || month==3 || month==5 || month==7 || month==8 || month==10 || month==12
           valid = day<=31;
        else
           valid=day <=30;
        end
     
    end
else 
    valid = false;
end
end

[Walter Roberson]

You should not really be processing year to determine leap-year status until you have validated that year is a scalar positive integer.

[akhil soni]

function valid = valid_date(year,month,day)
if (isscalar(year)==1 && isscalar(month)==1 && isscalar(day)==1)
    if (month>12 || date>31)
        valid=false;
    else 
    if ((month==1||month==3||month==5||month==7||month==9||month==11) && (day>0 && day<32))
        valid=true;
    elseif ((month==4||month==6||month==8||month==10||month==12) && (day>0 && day<31))
        valid=true;
    elseif (month==2 && (year/400==fix(year/400)) && (day>0 && day<30))
        valid=true;
    elseif (month==2 && ((year/100==fix(year/100))&&(year/400~=fix(year/400))) && (day>0 && day<29))
        valid=true;
    elseif (month==2 && (year/100~=fix(year/100)) && (year/400~=fix(year/400)) && (year/400==fix(year/400)) && (day>0 && day<30))
        valid=true;
    else valid=false;
    end
    end
else valid=false;
end
end
  % I have done this but error is shown in line 3 stating:-   Operands to the || and && operators must be convertible to logical scalar values.
  %what does this means?

Answer 1

[Neelkant Newra]

function valid = valid_date(year, month, day)
th31=[1,3,5,7,8,10,12];
th30=[4,6,9,11];
if isscalar(year) && year>=1 && isscalar(month) && month>=1 && month<=12 && isscalar(day) && day>=1 &&day<=31
    if month == 2
        if year/4 == fix(year/4)
            if year/100 == fix(year/100) && year/400 == fix(year/400)
                if day<=29
                    valid=true;
                else 
                    valid=false;
                end 
            elseif year/100 == fix(year/100) && year/400 ~=fix(year/400)
                if day <=28
                    valid = true;
                else
                    valid = false;
                end
            else 
                if day <=29
                    valid = true;
                else  
                    valid =false;
                end 
            end 
        else 
            if day<=28
                valid= true;
            else  
                valid=false;
            end 
        end  
    elseif ismember(month,th31)
        if day<=31
            valid= true;
        else 
            valid = false;
        end 
    elseif ismember(month,th30)
        if day<=30
            valid = true;
        else 
            valid = false;
        end 
    end 
else 
    valid= false;
end

[Walter Roberson]

No point in figuring out whether you are dealing with a leap year unless you have been asked to validate 29th day of 2nd month.

Answer 2

[Rohan Singla]

function isvalid = valid_date(y, m, d)
% Check if the inputs are valid  %
Check that they are scalars
if ~(isscalar(y) && isscalar(m) && isscalar(d))
isvalid = false;
 % Check that inputs are positive
elseif ~all([y, m, d] > 0)
isvalid = false;
 % Check that inputs are integers (not the data type)
 elseif any(rem([y, m, d], 1))
isvalid = false;
% Check that m and d are below the max possible
elseif (m > 12) || (d > 31) isvalid = false;
% The inputs could be a valid date, let's see if they actually are 
else
% Vector of the number of days for each month
daysInMonth = [31 28 31 30 31 30 31 31 30 31 30 31];
% If leap year, change days in Feb
if isequal(rem(y, 4), 0) && (~isequal(rem(y, 100), 0) || isequal(rem(y, 400), 0))
daysInMonth(2) = 29;
end
maxDay = daysInMonth(m);
if d > maxDay
isvalid = false;
else
isvalid = true;
end
end
end

[Rohan Singla]

solutin given by matlab

Answer 3

[Arafat Roney]

function valid=valid_date(year,month,day)
if((isscalar(year))&&(isscalar(month))&&(isscalar(day)))

    if((year==abs(fix(year)))&&(isscalar(year)))
    y=true;
    else y=false;
    end
    if((month<=12)&&(month==abs(fix(month)))&&(month>0)&&(isscalar(month)))
    m=true;
    else m=false;
    end
    if((rem(year,4)==0)&&(rem(year,100)~=0)&&(rem(year,400)~=0)||(rem(year,100)==0)&&...
        (rem(year,400)==0))
    feb=29;
    else
        feb=28;
    end
    if(((month==4)||(month==6)||(month==9)||(month==11))&&(day<=30)&&(day>0)&&(day==abs(fix(day)))&&(isscalar(day)))
     d=true;
        elseif((((month==1)||(month==3)||(month==5)||(month==7))||(month==8)||...
         (month==10)||(month==12))&&(day<=31)&&(day>0)&&(day==abs(fix(day)))&&(isscalar(day)))
          d=true;
        elseif((feb==28)&&(month==2)&&(day<=28)&&(day>0)&&(day==abs(fix(day)))&&(isscalar(day)))
          d=true;
        elseif((feb==29)&&(month==2)&&(day<=29)&&(day>0)&&(day==abs(fix(day)))&&(isscalar(day)))
          d=true;
    else d=false;
    end
 
    if((y==true)&&(m==true)&&(d==true))
         valid=true;
    else
         valid=false;
    end
 
else
    valid=false;
 end
end

[Walter Roberson]

If the year is non-scalar, then (year==abs(fix(year)) will be non-scalar, and you will have non-scalar && scalar which is an error. You need to test for scalar first.

Answer 4

[Tahsin Oishee]

function valid = valid_date(y,m,d)
if ~isscalar(y) && ~isscalar(m) && ~isscalar(d)
    valid=false;  
end
if y>0 && m>0 && m<12 && d>0 && d<31
    valid=true;
else valid=false;
end
if m==1 && d==31
    valid=true;
elseif   m==3 && d==31
    valid=true;
elseif m==4 && d>30
    valid=false;
elseif m==5 && d==31
    valid=true;
elseif m==6 && d>30
    valid=false;
elseif m==7 && d==31
    valid=true;
elseif m==8 && d==31
    valid=true;
elseif m==9 && d>30
    valid=false;
elseif m==10 && d==31
    valid=true;
elseif m==11 && d>30
    valid=false;
elseif m==12 && d==31
    valid=true;
end
if m==2 && d==30 || d==31
    valid=false;
end
if m==2 && rem(y,4)==0 && rem(y,100)~=0 && rem(y,400)~=0 && d==29
    valid=true;

elseif m==2 && rem(y,100)==0 && rem(y,400)==0 && d==29
    valid=true;
else valid=false;
end

end

[Walter Roberson]

Suppose y is not scalar but m and d are scalar. ~isscalar(y) would be true, but ~isscalar(m) would be false, and true && false is false, so you would not set valid=false.

Now suppose that y, m, and d are all not scalar, then ~isscalar(y) && ~isscalar(m) && ~isscalar(d) is true, so you set valid=false. But then regardless you go on to test y>0 && m>0 and so on, which would be testing the non-scalar values and using && between the tests, which is going to be an error.

Now suppose y is -7 and m = 1 and d = 31, then y > 0 is false so y>0 && m>0 && m<12 && d>0 && d<31 is false so you set valid=false. But then you go into the if/elseif tree and m==1 && d==31 is true so you set valid=true even though the year is invalid.

Generally speaking, you should set up valid=true at the beginning, and then do tests rejecting cases by setting valid=false, and as soon as you get one valid=false then you can skip the rest of the tests.