Oracle issue in group by - sql

Question

I know this question is asked by many people. But I still couldn't figure out why this is happening. Couldn't understand this logic.
I have a table mytesttable with columns id, company_name and employee_name.
I am trying to get the employee details grouping them with respect to company name. So I used the below query:
select *
from mytesttable
group by company_name;
But I get the below issue:
ORA-00979: not a GROUP BY expression
00979. 00000 - "not a GROUP BY expression"
*Cause:
*Action:
Error at Line: 28 Column: 2
Now I have tried putting count(1) in select along with my columns, tried grouping using 2 columns etc. Still the same issue. Can anyone explain me how to achieve this?
Because this is a simple group by. Logic seems to be right, but wondering why it's not fetching me the result.

Answer 1

Your query is:
select *
from mytesttable
group by company_name;
The * expands to all the columns. So this becomes:
select company_name, col1, col2, col3, . . . -- your question doesn't specify the column names
from mytesttable
group by company_name;
When you specify the group by, you are specifying that there is one row per company_name in the result set. The other columns are normally filled with aggregation functions, such as MIN(), SUM(), or LISTAGG().
What value should be chosen for col1? In general, SQL does not attempt to answer this question. Instead, it returns a syntax error. This is not Oracle-specific. This is the definition of the language.
What you probably want is:
select company_name, count(*) as num_employees
from mytesttable
group by company_name;

Answer 2

In group by clause you have to put all selected columns. As you select all but given only one column in group by as a result it thrown error, but if you simply try like this way it will work. Actually group by used for aggregated function
select company_name, count(*) from mytesttable group by company_name;

Answer 3

I think you mis-understand what GROUP BY does?
It doesn't put all similar records next to each other, that's ORDER BY. What GROUP BY does is to collapse all similar records in to the same row, to allow aggregate calculations like SUM() and MIN() and COUNT(), etc.
So, two examples using the same input...
id | company_name | employee_name
----+--------------+---------------
1 | zzz | aaa
2 | xxx | bbb
3 | yyy | ccc
4 | zzz | ddd
5 | xxx | eee
Using ORDER BY...
SELECT * FROM mytesttable ORDER BY company_name, employee_name
id | company_name | employee_name
----+--------------+---------------
2 | xxx | bbb
5 | xxx | eee
3 | yyy | ccc
1 | zzz | aaa
4 | zzz | ddd
Using GROUP BY...
SELECT
company_name,
COUNT(*) number_of_employees,
MAX(id) highest_id_in_company
FROM
mytesttable
GROUP BY
company_name
ORDER BY
company_name
company_name | number_of_employees | highest_id_in_company
--------------+---------------------+-----------------------
xxx | 2 | 5
yyy | 1 | 3
zzz | 2 | 4

Answer 4

If you use group-by expression,you have to use aggregate functions such as max,min,average,count in select statements.
For example;
select count(*),company_name,employee_name
from mytesttable
group by company_name,employee_name
order by company_name;