OpenEuphoria

1. MySQL Question

• Posted by gwalters Nov 28, 2012
• 1228 views

I have a MySQL table with a double var. I have an sql statement which adds a value of "2.60106457" into the double item in the table. When I ask mysql to display what it has it displays 2.60106492042542. This only accurate to 6 places not the expected 16. Looks to me that mysql is wrong. Has anyone seen this or can explain?

3. Re: MySQL Question

• Posted by gwalters Nov 28, 2012
• 1170 views

I discovered that I was getting stars placed in the wrong place. I am updating what was in the my star database with coordinates RA and DEC that were from the Hipparcos database which had a lot of accuracy. The sql statement looks like this and both table entries are double precision;

sql = replace stars values RAnumeric, DECnumeric with values (2.60106457, 6.88336423)

than here is the code snipit

	a = execDirectODBC(dbId, sql) 
	if atom(a) then 
		odbcError(-a) 
		a = message_box("Can't find record " & openTbNames[t] & " " & recordKey&" to update.","",0) 
		abort(1) 
	else 
		-- OK 
	end if 
 

After this update I checked the table entry for the double var using mysql query and get the wrong value mentioned above. I am using MySql 4.00.22 and matt's odbc.e routines.It looks like what is in the database is single precision. Any suggestions?

5. Re: MySQL Question

• Posted by CoJaBo2 Dec 01, 2012
• 1024 views

This is correct behaviour; the FLOAT (and DOUBLE, which is non-standard) types do not define precision- floating-point calculations are approximate. Notably, you can get into Deep Trouble using them to compute monetary values.
If you need a specific number of digits, you should use the DECIMAL type; e.g., DECIMAL(19,16) will give you +/- 999.9999999999999999.

Note that DECIMAL support is flawed in MySQL versions up through 5.0; make sure you are running the latest version (currently, 5.5.x).