Ansible...SQLParseError: PostgreSQL does not support table with more than 3 dots

I am editing and extending an ansible playbook built by third parts.

I added a task which runs a query on a DB and dumps the data into a csv file:

	- name: run query1 and dump data in csv
	  community.postgresql.postgresql_copy:
		login_host: '{{ db_host }}'
		login_user: '{{ db_username }}'
		login_password: '{{ db_password }}'
		db: '{{ db_database }}'
		port: '{{ db_database_port }}'
		src: "{{ lookup('template', 'query1.sql.j2') }}"
		copy_to: "{{ log_base_path }}/results1.csv"
		options:
		  format: csv
		  delimiter: ';'
		  header: yes

the content of query1.sql.j2 is

SELECT
	KEY,
	idint,
	barcode,
	systemLV1,
	systemLV2,
	n.attribute1 AS attribute1_master,
	c.attribute1 AS attribute1_systemLV2,
	n.attribute2 AS attribute2_master,
	c.attribute2 AS attribute2_systemLV2,
	n.attribute3   AS attribute3_master,
	c.attribute3   AS attribute3_systemLV2,
	n.attribute1 = c.attribute1 AS check_attribute1_systemLV2,
	n.attribute2 = c.attribute2 AS check_attribute2_systemLV2,
	n.attribute3   = c.attribute3   AS check_attribute3_systemLV2

FROM 
	check_items_stechsystem2_ctrl2 c LEFT JOIN check_items_stechsystem1_ctrl2 n USING (KEY, idint, barcode, systemLV1) 
WHERE 
	(c.attribute1,c.attribute2,c.attribute3) != (n.attribute1,n.attribute2,n.attribute3);

when this task is runned, the following error is raised:

An exception occurred during task execution. To see the full traceback, use -vvv. The error was: ansible_collections.community.postgresql.plugins.module_utils.database.SQLParseError: PostgreSQL does not support table with more than 3 dots

  localhost failed | msg: MODULE FAILURE

See stdout/stderr for the exact error

What is the meaning of it?

On the internet I can find little information about this error.

I read in this issue page that the error is somehow related to some non-ASCII character. If so, where should I look for a non-ASCII character which breaks my playbook ?

Both the tables in my query have the following structure:

Column  |   Type   | Collation | Nullable | Default 

---------+----------+-----------+----------+---------

key     | text     |           | not null | 

idint   | text     |           |          | 

barcode     | text     |           |          | 

systemLV1 | integer  |           |          | 

systemLV2   | text     |           | not null | 

attribute1  | numeric  |           |          | 

attribute2  | smallint |           |          | 

attribute3    | bigint   |           |          | 

But the text type columns are filled up by columns of the csv file which contain stringified numeric (eventually zero-padded) data.

When I manually run the query on the DB, I get no error, and the data don't carry any strange sign.

Substituting all the multiple spaces and new lines characters from the query solved the issue.

Yet this is very weird since in that playbook there are other tasks which run queries with new line characters.

Since the Traceback indicates the issue is a SQLParseError, I guess some "bad" new line / tab / space character got into the query string while I was formatting it on some text editor.

Also it is possible that I have copied some parts of the query from some chat apps like MS Teams, which is a well known code spoiler.