Names of Tab from SQL Queries

SELECT * FROM table1;
SELECT * FROM table2;

/**
* Monthly Sales Query
* 
* @label Monthly Sales
*/
SELECT sum(*) AS total
, month(sell_date) AS month
FROM sales
GROUP BY month;
/**
* Yearly Sales Query
* 
* @label Yearly Sales
*/
SELECT sum(*) AS total
, year(sell_date) AS year
FROM sales
GROUP BY year;

I have a similar use case. I paste in a query that has 7 selects. If I can name the tabs then I can easily see what is what. Otherwise I have to check tab position against position of the select command. That's not too hard, but when mentally juggling other things it is a distraction.

I find HeidiSQL a very good tool and this would be a useful addition for me.

On HeidiSQL Help I see a solution, but my queries aren't as simple as the example.

I'm hoping I'm missing something.

Here's one of my queries: select from(select date(FROM_UNIXTIME(joinStamp-83600)) as date, count(*) from base_user group by 1 order by date DESC LIMIT 20 ) sub order by date ASC;

*I'm a new user, so I can't include the reference - so search google for: heidisql Screenshot: Multiresults

I was reading through the comments, hoping to find a solution to the problem of having multiple results tabs with the same name, though I don't see that this has been solved. So I'm commenting to show interest in a solution :)

joining in - yes, I agree this would be useful. Perhaps the app could continue as is with it's current naming but allow you to say double click a tab and give it a different name if you choose to..

So I found I way to make this work. instead of doing a query like the following:

select col1, col2 from table1
where col1 is not null;

change it up to be like this:

select col1, col2 from (select col1, col2 from table1) as 'WhateverLabel'
where col1 is not null;

and the tab will show up as WhateverLabel. It's a little redundant, but at least the tabs are labeled what you want.

Hope this helps

You would have to write a comment just to have the tab named after it. Isn't that manual work for the user for a minimal effect?

When you do an important updates of data you need to be 100% sure where goes each bit of data and from which tab you take this data. But when there are users#1 users#2 you have to remember what you need to do with #1 and #2. But if you once named your tabs you then can use them later no matter how far thought process gone.

Or use your select statement to create a temp table named resultset1 and then select * from it afterward, and tab will be named more specific

Or use your select statement to create a temp table named resultset1 and then select * from it afterward, and tab will be named more specific

Sure, but that's no good if you want to update a row.

I'd favour a comment directly before the query that HSQL uses to name the tab e.g.

# My Tab 1

select * from etc

If it's not there then use default naming, if it's too long than truncate.

Cheers Muzza

Is there any solution?

Apart from the above AS somealias workaround - no.

I assume by the "somealias" work around you mean thstolte's post, right? If so, I try that and get an error (see screenshot). I'd love a way to add a comment above the query so I could identity the query result tab name easier.

(Using Version 11.0.0.5919 (64 Bit)) Thank you!

I assume by the "somealias" work around you mean thstolte's post, right? If so, I try that and get an error (see screenshot). I'd love a way to add a comment above the query so I could identity the query result tab name easier.

(Using Version 11.0.0.5919 (64 Bit)) Thank you!

I guess the version has been updated so that you don't need the quotes around the label anymore. Try the following and it should work:

select mmcontactid from (select mmcontactid from tblnameContactCSIs) as whateverlabel
where mmcontactid is not null

Unfortunately, no luck.Tab name remains the same:

Weird. It worked for me and I'm on the same version.

Hmm. Is it 64 or 32 bit?

Weird. It worked for me and I'm on the same version. ?t=10493#p36163

SELECT table_catalog FROM (SELECT table_catalog FROM COLUMNS) AS label WHERE table_catalog IS NOT NULL;

This doesn't worked for me. Version is 11.0.0.5919 64 bit not portable installation.

Yes this would be a nice-to-have. I'd put a bounty up for it if ansgar supported bounties. thstolte's solution worked for me by the way.

Even if the tab was named after the alias of the first table in the query, that would be great. Surely that would not require a large change in the codebase?

So for:

SELECT height, giantswhocanclimb
FROM cliffs AS cliffsofinsanity
WHERE location = 'insanity'

The tab would be labeled "cliffsofinsanity" instead of 'cliffs'?

Just an idea.

Anyways, thanks a gazillion for making HeidiSQL!!!

For example in dbeaver naming of resulst tabs works like this:

-- NAME: test

SELECT * FROM table AS t;

I have tried all the ways suggested or described in this thread including using "AS" and none of them work.

Is there going to be a fix for this?

I will this really appreciate. I have many queries with several selects included. It will be very easy and fast to add everywhere inside suggested agreement such -- NAME: test

@ansgar, please do it. It will be very helpfull for all of us begging :-). No problems for others.

OMG Dudes! I've been using HeidiSQL for years (love it), and have just completely accidentally discovered a rather hacky way of naming result tabs. Just precede the query with a comment in the format 'from' <label>. It's ugly, but it's going to make my life slightly jollier - see my little clippet snippet... X

puppriss looks nice but your hack doesn't work for me. Even with comments like these I still get table name in tab headers.

puppriss Good joke, but does not work...

qupear Ah, that's a shame. I'm on Win10 64 Heidi v.12.2.0.6576. Since I posted, I have noticed that it doesn't work for me 100% of the time. But I assure you belanp that my screenshot is absolutely genuine and worked as shown

I've been able to get distinct (but not custom) tab names by using a semicolon after each statement. For instance:

select top(5) * from table1;
select top(5) * from table2;
select top(5) * from table3;

See also the attachment

This encapsulation should work as work arround: SELECT × FROM (SELECT columns FROM table1, table2 WHERE conditions) AS TAB_NAME

However some new "command" used in comments will be heplful and harmless for all.

'OMG Dudes! I've been using HeidiSQL for years (love it), and have just completely accidentally discovered a rather hacky way of naming result tabs. Just precede the query with a comment in the format 'from' <label>. It's ugly, but it's going to make my life slightly jollier - see my little clippet snippet... X'

This only works if UNION is present in SQL :-/

It would be great if we could change the name of the TAB in the query-windows programmatically.

I tried this in order to force the name 'TO DO' - but it does not work - it still shows 'Result #1':

-- NAME: TO_DO
SELECT -- TO_DO
    TO_DO.TO_DO
FROM (SELECT 'Please undo...' AS TO_DO) TO_DO;

Next build parses such a "-- name: xyz" comment over the actual query. That comment does not necessarily need to be the first line. Note if present then it takes precedence over a table name:

Next build parses such a "-- name: xyz" comment over the actual query. That comment does not necessarily need to be the first line. Note if present then it takes precedence over a table name:

Supergreat! Thnx

Next build parses such a "-- name: xyz" comment over the actual query. That comment does not necessarily need to be the first line. Note if present then it takes precedence over a table name:

EXCELLENT THANKS !!!

I have just tried this nice feature. In simple cases it works well, but as soon as the queries become more complicated, the tab name becomes very wide. Here some queries to test: test1 works fine, 2

-- NAME: test 1
SELECT table_schema, table_name FROM information_schema.tables a LIMIT 1;

-- name: test2
SELECT table_schema, table_name FROM information_schema.tables a 
WHERE EXISTS(SELECT 1) LIMIT 1;

-- name: test3
SELECT table_schema, table_name FROM information_schema.tables a
WHERE EXISTS(SELECT * FROM information_schema.columns b 
    WHERE a.TABLE_NAME=b.table_name AND a.table_schema=b.table_schema) LIMIT 1;

-- NAME: test 4
SELECT b.table_schema, b.TABLE_NAME, b.column_name
 FROM information_schema.tables a INNER JOIN information_schema.columns b
  on a.TABLE_NAME=b.table_name AND a.table_schema=b.table_schema
LIMIT 1;

Here is the result of running all queries in a single batch. Tab test 4 cannot be seen!

queries test 4 and test 1

Should not happen, even if there are spaces behind the "name: test2". HeidiSQL trims the result name which should remove spaces.

Anyway, could you please check whether you have trailing spaces in the comment?

Should not happen, even if there are spaces behind the "name: test2". HeidiSQL trims the result name which should remove spaces.

Anyway, could you please check whether you have trailing spaces in the comment?

I checked the statements, there are no trailing spaces. The repo in my previous note is the very same.

Confirmed using your samle code.

Problem was the regular expression used for the result name, taking several new lines into the name.

Fixed in the next build:

It works in the new build 6623 🙏

There is a problem when you use variables in DECLATE. Then SEMICOLON ";"can not be used and then all tabs have the same "FIRST" name.

Most of our scripts are using DECLARE statment, because we need more actions on the same data. Please fix this problem described above with SEMICOLON.

@belanp please post the code you have and which leads to confusion here. Your screenshot only shows the result tabs, and the code is hidden.

@belanp please post the code you have and which leads to confusion here. Your screenshot only shows the result tabs, and the code is hidden.

DECLARE @x integer

-- name: Result 1 SET @x = 1 SELECT 1 WHERE 1= @x

-- name: Result 2 SET @x = 2 SELECT 2 WHERE 2=@x

DECLARE @x integer

-- name: Result 1

SET @x = 1

SELECT 1 WHERE 1= @x

-- name: Result 2

SET @x = 2

SELECT 2 WHERE 2=@x

@belanp You need to separate the queries with a semicolon, at least after the SELECT, so the "name:xyz" is bound to the query inside:

DECLARE @x INTEGER

-- name: Result 1
SET @x = 1
SELECT 1 WHERE 1= @x;

-- name: Result 2
SET @x = 2
SELECT 2 WHERE 2=@x;

With HeidiSQL build 12.6.0.6837, connected to MariaDB Galera Cluster 10.6.17, this does not seem to work from within stored-procedures. Should this work or is it impossible to implement?

No, not within a procedure definition. Only outside, in a query tab.

@belanp You need to separate the queries with a semicolon, at least after the SELECT, so the "name:xyz" is bound to the query inside:

DECLARE @x INTEGER

-- name: Result 1
SET @x = 1
SELECT 1 WHERE 1= @x;

-- name: Result 2
SET @x = 2
SELECT 2 WHERE 2=@x;

sorry, but not working: / SQL error (137): Must declare the scalar variable "@x". /

@ferino you are mixing things up here. Be aware these "-- name: xyz" comments are supported only in the query editor. Nowhere else. Please test with your own queries. The SQL above was just an example taken out of your own comment.

@belanp You need to separate the queries with a semicolon, at least after the SELECT, so the "name:xyz" is bound to the query inside:

It is not working on MS SQL AZURE. If you use ";" semicolon there, then variable is unknown in second select a produce ERROR. Could you please use for parsing only comment "-- name: XYZ". No semicolon is needed there for parsing...

It doesn't work for me either:

-- name: name1
SELECT 1 AS n1
;
-- name: name2
SELECT 2 AS n2 
;

Please note the semicolon is required for splitting queries, and for finding the right tab name from the "name" comment.

Also, you need to activate "Send queries one by one" for it to work properly:

On the mysql server (5.5.5-10.5.9-MariaDB-1:10.5.9+maria~focal-log) it's different again: "Send queries one by one" is activated.

So I don't know where it actually works.

As said, the semicolon ist the crux. I tried to make it more clear with some color applied on the code areas which you splitted with a semicolon. Your tab names are applied on the SET.. queries only, which do not create a result:

at last, 2024 people is here

How about using the table alias (for example FROM table_name AS table_alias ) for the name of the query's tab? I am only getting "Result #1 ... Result #2...etc - which doesn't even use the name of the primary (or in my case, only) table I'm selecting from. I attached a screen shot showing this along with the overly simple bit of code I use that checks the call_log table to see how recent the data is there (since these are supposed to be real-enough-time mirror servers - but sometimes lag - and I like to check before running anything time-sensitive against them). Like a previous poster, I have been doing this positionally - remembering which SELECT statement came in what order - which isn't that difficult - but - would definitely be easier if they were named from the name of the FROM table's alias...that way we would be in FULL control of the tab's name...thanks!

Please ignore my last reply about the "...FROM table_name AS table_alias..." - the -- name: TabName works beautifully!

THANK YOU!!!

Example of multiple tabs being named using the -- name: TabName syntax - thanks Ansgar!

Semicolon "destroys" declaration of the variable, but we all use the same value of variable on several selects in the same script. So we cannot use semicolon. We need to set variable value only once at the top of script, instead of 10 times, because we always change this value manualy, before we run this script.

When we do not use semicolon Heidy stil gives us different tabs for each (main) select. But we are not able to give them meaningful names.

Just collect "List of tab names" of all tabs defined TAB names in script. Then you generate TABs with results in Heidy frontend. And then give them names from the "List of TAB names" - one by one. It's easy. Furthermore we can define the "List of names" anytime, for example at the begin of the script, if it is problem to collect them during parsing rows. Like: -- list of tab names: "aplles", "berries", "melons" We just need the way how to give names for the TABS in the comments. Nothing more. Thanks a lot for you work with Heidy.

Looks like a nice extension to that "-- name:" comment.

Next build supports a new "-- names:" comment, and please note the plural "s":

• will be splitted by comma
• will be used to populate the result tab captions, as much as it holds
• -- names: xyz, abc, ... can be placed in the first KB of SQL text
• it may be part of the first query, but does not need to. It can also appear in the second or later queries.
• -- name: xyz has still priority, when also used for a single query

Here's an example, showing some special cases:

Great, WORKS EXCELENTLY, thanks :-)