如何计算所有列中的数字或出现次数并使用MySQL列出它们？

Question

I have a table looking like:

| A | B | C | ... | Z |  <- names of columns
-----------------------
| 1 | 0 | 1 | ... | 1 | 
| 0 | 1 | 1 | ... | 1 | 
| 1 | 1 | 1 | ... | 1 | 
| 0 | 1 | 1 | ... | 0 | 
| 1 | 0 | 1 | ... | 1 | 

And I would like to sum up all 1s in all the columns and list them out. How can I do that using MySQL? Number of columns is about 80, if possible I would like not to list them in the SQL call.

I would like to get a response similar to this one:

A: 3
B: 3
C: 5
...
Z: 4

Answer 1

This table has been designed in a way that makes the query you describe more difficult.

Using many columns for data values that should be compared or counted together because they're the same type of value is called repeating groups . It's a violation of database normalization rules.

The more traditional way to store this data would be over 80 rows, not 80 columns.

CREATE TABLE mytable (
  id INT PRIMARY KEY,
  label CHAR(1) NOT NULL,
  value TINYINT NOT NULL
);

INSERT INTO mytable VALUES
('A', 1), ('B', 0), ('C', 1), ...

Then you could use a simple query with an aggregate function like this:

SELECT label, SUM(value)
FROM mytable
GROUP BY label;

There are times when it's worth using a denormalized table design (like your current table), but that time is when you want to optimize for a particular query. Be careful about using denormalized designs, because they optimize for one query at the expense of all other queries you might run against the same data. The query you want to make is one of those that is made more difficult by using the denormalized design you currently have.

Answer 2

There is no easy way, you will need to explicitly list the columns. A UNION query should be what you need, like:

SELECT 'A' column_name, SUM(A) cnt FROM mytable
UNION ALL SELECT 'B', SUM(B) FROM mytable
UNION ALL SELECT 'C', SUM(C) FROM mytable
...

NB: it should be possible to generate the query programmatically using any text manipulation tool (Excel, perl, ...), or dynamically using a prepared statement.