Time and correlation type (1)-tdatetime and double

// Related functions: sysutils. floattodatetime (); sysutils. tryfloattodatetime ();

  
 

In the system unit definition, tdatetime = type double;

Tdatetime is an 8-byte 64-bit data, but a double alias.

 
{Test: tdatetime is a double} var T: tdatetime; D: Double; s: string; begin T: = encodedatetime (2009, 5, 22, 11, 22, 33,999 ); {forced conversion} D: = double (t); showmessage (floattostr (d); // 39955.4740046181 {implicit conversion} D: = T; showmessage (floattostr (d )); // 39955.4740046181 {do not need to convert} showmessage (floattostr (t); // 39955.4740046181 {treat a double as tdatetime} s: = formatdatetime ('yyyy-m-d h: m: S: Z', 39955.4740046181); showmessage (s); // 2009-5-22 11: 22: 33: 999end;

  
 

No double will be converted into a valid time. The sysutils unit returns the minimum and maximum time constants:

Const mindatetime: tdatetime =-657434.0; {01/01/0100 12:00:00. 000 am} maxdatetime: tdatetime = 2958465.99999; {12/31/9999 11:59:59. 999 pm}
 

To be safe, use floattodatetime (); or tryfloattodatetime (); to convert a number to tdatetime;
There is no need to use a function for reverse conversion.

 
{Use tryfloattodatetime} var T: tdatetime; D: Double; begin D: = 0; If tryfloattodatetime (D, T) Then showmessage (datetimetostr (t); // 1899-12-30end; {floattodatetime is used; If the conversion fails or encounters an invalid time, an exception is thrown.} var T: tdatetime; D: Double; begin D: = 0; T: = floattodatetime (d ); showmessage (datetimetostr (t); // 1899-12-30end; {when the time is 00:00:00, January 1, December 30, 1899, the time value of tdatetime is 0} var D: Double; T: tdatetime; begin T: = strtodatetime ('2017-12-30 0: 0: 0: 0'); D: = double (t); showmessage (floattostr (d )); // 0end;