What does the first argument to MySQL's Decimal function do? - mysql

Using MySQL 5.0.27
This query:
SELECT CAST('543.21' AS DECIMAL(100,2))
returns 543.21
So does this one:
SELECT CAST('543.21' AS DECIMAL(2,2))
In fact, I am having trouble figuring out what effect the parameter has. I am using it to aggregate numeric values in a varchar column (for legacy reasons!!) and round off to 2 decimal places.
Should I just pick a high number?

It describes how many total digits a field (or variable) will be able to store.
DECIMAL(100,2) - 100 total digits, 98 before, 2 after a decimal separator
DECIMAL(2,2) 2 total digits, 0 before, 2 after a decimal separator
Explained here:
http://dev.mysql.com/doc/refman/5.1/en/numeric-types.html
[added]
For rounding just use ROUND() function.

Related

MariaDB SQL query returns truncated numbers from select statement

I am trying to pull data from Maria DB. The datatype in schema is DECIMAL (12, 8).
In my program when I query using following query. it trucates to 4 (or 3) decimal places and also round off.
select CAST(FORMAT(latitude, 100) AS FLOAT) latitude from mytable
it returns 36.173 . In DB it is stored as 36.17298200
I want it to return as 36.172982
To use the number in calculations, simply use it. No conversion, no CAST.
To dislay it to 6 decimal places, do that on SELECTing:
SELECT FORMAT(latitude, 6) ...
FORMAT(..., 100) should give you lots of decimal places.
FLOAT does not have enough precision to distinguish more than 6 or 7 significant decimal places. That is, the first and third numbers nere are the closest representable numbers in FLOAT:
x4210b122 --> 36.172981262207
36.17298200
x4210b123 --> 36.172985076904
Double:
x40421624463065f9 --> 36.1729819999999975
For Latitude and Longitude:
FLOAT has a resolution of 1.7 m or 5.6 ft -- good enough for Vehicles
DECIMAL(8,6) 16 cm 1/2 ft -- Friends in a mall
According to official MySQL documents:
The DECIMAL and NUMERIC types store exact numeric data values. These types are used when it is important to preserve exact precision, for example with monetary data. In MySQL, NUMERIC is implemented as DECIMAL, so the following remarks about DECIMAL apply equally to NUMERIC.
Floating-Point Types (Approximate Value) - FLOAT, DOUBLE
So you should be writing something similar to:
select CAST(FORMAT(latitude, 100) AS DECIMAL(8,6)) latitude from mytable
As 8 is the total number of digits and 6 is precision.

How do you round floats conditionally?

I am writing a query that is used by report generating software.
Part of this is querying for the hours needed to complete a project. We record this a 2 decimal float so that we can estimate to the quarter hour.
However, if we are using it in our report and the hour we recorded is something like 8.00, I want to query it and format it so that 8.00 is just 8. However any hours with something past the decimal, like 8.25, should remain as 8.25. How can I make this work?
hours Queried Result
====== -> My Query -> ==============
8.00 8
8.25 8.25
I am using MySQL 5.6
You can use the REPLACE() function to remove .00:
REPLACE(hours, '.00', '') AS hours
You can convert it to a string and check the rightmost 2 characters and trim those if they are '00'.
SELECT TRIM(TRAILING '.00' FROM CAST(column_name AS VARCHAR));
SELECT REPLACE(Round(8.00), '.00', ' ');
I will give more example so you can clear your Logic:
MySQL ROUND() rounds a number specified as an argument up to a number specified as another argument.
Syntax:
ROUND(N,[D]);
Where 'N' is rounded up to D decimal places.
and 'D' is indicating up to how many decimal places N will be rounded.
Example 1:-
SELECT ROUND(4.43);
Output :-
4
The above MySQL statement will round the given number 4.43. No decimal places have been defined, so the default decimal value is 0.
Example 2:-
SELECT ROUND(-4.53);
Output:-
-5
The above MySQL statement will round the given number -4.53. No decimal places have been defined, so the default decimal value is 0.

MySQL TRUNCATE(double, 4) is modifying value instead of truncating

I'm trying to truncate a double with 4 decimals, but I get the last digit rounded (and the double has 4 decimals only!)
double:
-2.5805
SELECT TRUNCATE(double, 4) FROM `table`
Result:
-2.5804
What's more interesting is that:
SELECT TRUNCATE(mycol, 4) AS col1, TRUNCATE(-2.5805, 4) AS col2
FROM mytable
returns:
col1 | col2
--------+--------
-2.5804 | -2.5805
Demo here
The double data type is a floating point type, meaning these numbers are not stored as exact numbers. As MySQL documentation on floating point types explains:
Because floating-point values are approximate and not stored as exact
values, attempts to treat them as exact in comparisons may lead to
problems. They are also subject to platform or implementation
dependencies.
This means that the number -2.5805 is not an exact number when it is stored as a double. The truncate() function, however, returns an exact number with an exact number of digits. When the -2.5805 (as double) is expanded, the 4th decimal digit is 4.
The TRUNCATE(-2.5805, 4) expression returns -2.5805 because MySQL treats numeric literals expressed with fixed number of decimal digits as decimal data type, which is an exact data type. Therefore, the truncate() function just simply returns the same number of decimal digits.

Column type for saving very different number of decimals

I need to store numbers like
21000
1.0002
0.00230235
12323235
0.2349523
This is sensordata so it is important to keep the exact value.
THere are many options.
My solution would be to multiply all values by 1 million, and store them as a bigint. Would that make sense?
That makes sense but I'd recommend that you just use the decimal datatype: https://dev.mysql.com/doc/refman/5.7/en/precision-math-decimal-characteristics.html
If you were to multiply by million and if a dataset you receive has one more decimal than you'd expect, you'd end up multiplying that number by 10 million and all other numbers by 10. Instead, using the decimal datatype will give you 30 numbers to the right of the decimal.
The declaration syntax for a DECIMAL column is DECIMAL(M,D). The
ranges of values for the arguments in MySQL 5.7 are as follows:
M is the maximum number of digits (the precision). It has a range of 1
to 65.
D is the number of digits to the right of the decimal point (the
scale). It has a range of 0 to 30 and must be no larger than M.
and
The SQL standard requires that the precision of NUMERIC(M,D) be
exactly M digits. For DECIMAL(M,D), the standard requires a precision
of at least M digits but permits more. In MySQL, DECIMAL(M,D) and
NUMERIC(M,D) are the same, and both have a precision of exactly M
digits.
For a full explanation of the internal format of DECIMAL values, see
the file strings/decimal.c in a MySQL source distribution. The format
is explained (with an example) in the decimal2bin() function.
To format your numbers, you could do formatting like this answer describes: Format number to 2 decimal places
Example
create table test (
price decimal(40,20)
);
-- all the above insertions will succeed cleanly
insert into test values (1.5), (1.66), (1.777), (1.12345678901234567890);
-- notice we have 21 digits after decimal
-- MySQL will insert data with 20 decimal and add a warning regarding data truncation
insert into test values (1.123456789012345678901);
Data
select * from test
price
1.50000000000000000000
1.66000000000000000000
1.77700000000000000000
1.12345678901234567890
1.12345678901234567890
select cast(price as decimal(40,2)) from test
price
1.50
1.66
1.78
1.12
1.12

Exact decimal division in MySQL

How to properly divide two DECIMAL values in MySQL and get a result accurate to the number of digits defined in the column type?
Example:
select cast(1/2 as decimal(4,4)),
cast(1 as decimal(4,4))/2,
cast(1 as decimal(4,4))/cast(2 as decimal(4,4));
Results:
'0.5000', '0.49995000', '1.00000000'
Why is the second result innacurate? Why is the third result not '0.5000'?
Note: I can't just use the first form, I need to perform calculations with columns stored as decimals.
The problem is that DECIMAL data type declaration requires 2 arguments, the first one being the total number of digits, including the fractional part.
According to the docs (http://dev.mysql.com/doc/refman/5.7/en/precision-math-decimal-characteristics.html):
The declaration syntax for a DECIMAL column is DECIMAL(M,D). The
ranges of values for the arguments in MySQL 5.1 are as follows:
M is the maximum number of digits (the precision). It has a range of 1
to 65. (Older versions of MySQL permitted a range of 1 to 254.)
D is the number of digits to the right of the decimal point (the
scale). It has a range of 0 to 30 and must be no larger than M.
So if you use 5 total digits instead of 4, everything is ok:
select cast(1/2 as decimal(5,4)),
cast(1 as decimal(5,4))/2,
cast(1 as decimal(5,4))/cast(2 as decimal(5,4)),
cast(cast(1 as decimal(5,4))/cast(2 as decimal(5,4)) as decimal(5,4));
Results
0.5000, 0.50000000, 0.50000000, 0.5000