Reputation: 759
I have a table called tabela1512823699024883
that looks like this:
On which I run query like this:
SELECT * FROM tabela1512823699024883 WHERE `age` = 'male'
This query does not make sense, because age
column is int
type and I'm looking for string
value in my query, but MySQL
still returns no empty rows whatsoever. Here is what query returned:
So
age
row does not contains male
value in neither rows returned. How can this be possible?
Upvotes: 8
Views: 691
Reputation: 18416
Same issue will occur with casting a string to an integer type.
SELECT CAST('male' as SIGNED); #0
#or
SELECT CAST('male' as UNSIGNED); #0
However if a number is supplied within the string.
SELECT CAST('1234male' as UNSIGNED); #1234
#and
SELECT CAST('male1234' as UNSIGNED); #0
To resolve the issue use BINARY
on the column, which will also cause the textual value to become case-sensitive.
SELECT * FROM tabela1512823699024883 WHERE BINARY `age` = 'male';
Alternatively to utilize a case-insensitive BINARY
criteria CONVERT()
the value to the desired character-set and then specify a case-inventive collation. [sic]
SELECT * FROM tabela1512823699024883 WHERE CONVERT(BINARY `age` USING utf8) COLLATE utf8_general_ci = 'male';
Upvotes: 14
Reputation: 2588
The Problem here is that whenever you try to compare a string to integer column, the value will be cast to the data type of column. Same case will happen with float , big int, tiny int etc.
What ever value is passed in comparison first being cast for that particular data type by mysql engine
Here is the test result of the query casting few values.
select cast('' AS UNSIGNED), cast('anything' AS UNSIGNED), cast(NULL AS UNSIGNED), cast(35.56 AS UNSIGNED), cast(NOW() AS UNSIGNED), cast(1 AS UNSIGNED), cast('Test' AS BINARY), cast('' AS BINARY);
SELECT * FROM `test_type` WHERE flt = 'test'
the flt is of float type column.
Upvotes: 2
Reputation: 101
"male" is 0! So you keep getting records where age is 0. To really understand what's happening, just do:
select user_name, age, age='male' from tabela1512823699024883;
Your 3rd column will be 1 for the records containing '0' age, and '0' for the records containing non-zero age.
Upvotes: 3
Reputation: 72175
Try this simple query:
SELECT 'male' + 0
This seems not to make sense. However the query executes and returns a value of 0
. This is because of implicit type conversion. String value 'male'
is converted to 0
and then added to 0
and thus 0
is returned .
The same thing happens with your query. 'male'
in converted to 0
when compared with a field of type int
.
Upvotes: 11