Creating cte statement to traverse a tree created with a bridge table - sql

Question

I will preface this by saying that I am completely new to cte statements.
I have 2 tables an items table and a linker table. the linker table containing the parent id and child id of the relationship.
I have seen lots of examples on how to traverse a tree with what I believe is called an associative system, where the parent id is stored on the child record instead of in a bridge table. But I can not seem to extrapolate this out to my scenario.
I believe I need to have a bridge table like this because any single item could have multiple parents and each parent will most likely have multiple children.
I hacked together my own way of traversing the tree but it was within another language, xojo, and I really did not like how it turned out. Essentially I was using recursive functions to dig down into each tree and queried the database each time I needed a child.
Now I am trying to create a cte statement that does the same thing. Keeping the descendants ordered below the parents is not a huge deal
So I have created a sample database and other materials to describe my issue:
This diagram shows visually what the relationships are:
diagram
This is what I would like returned from the database (some items show up in multiple places:
1 : Audio
2 : Speaker
3 : Microphone
4 : Mic Pack
3 : Microphone
5 : Di
6 : Passive Di
11 : Rapco Di
13 : Dbx Di
7 : Lighting
9 : Safety
12 : Small Safety
8 : Rigging
10 : Light Rigging
9 : Safety
12 : Small Safety
An example table:
CREATE TABLE items ( id INTEGER, name TEXT, PRIMARY KEY(id) );
CREATE TABLE `linker` ( `parent` INTEGER, `child` INTEGER, PRIMARY KEY(`parent`,`child`) );
Insert Into items(id, name) Values(1, 'Audio');
Insert Into items(id, name) Values(2, 'Speaker');
Insert Into items(id, name) Values(3, 'Microphone');
Insert Into items(id, name) Values(4, 'Mic Pack');
Insert Into items(id, name) Values(5, 'Di');
Insert Into items(id, name) Values(6, 'Passive Di');
Insert Into items(id, name) Values(7, 'Lighting');
Insert Into items(id, name) Values(8, 'Rigging');
Insert Into items(id, name) Values(9, 'Safety');
Insert Into items(id, name) Values(10, 'Lighting Rigging');
Insert Into items(id, name) Values(11, 'Rapco Di');
Insert Into items(id, name) Values(12, 'Small Safety');
Insert Into items(id, name) Values(13, 'Dbx Di');
Insert Into linker(parent, child) Values(1, 2);
Insert Into linker(parent, child) Values(1, 4);
Insert Into linker(parent, child) Values(1, 3);
Insert Into linker(parent, child) Values(4, 3);
Insert Into linker(parent, child) Values(4, 5);
Insert Into linker(parent, child) Values(5, 6);
Insert Into linker(parent, child) Values(6, 11);
Insert Into linker(parent, child) Values(6, 13);
Insert Into linker(parent, child) Values(7, 9);
Insert Into linker(parent, child) Values(9, 12);
Insert Into linker(parent, child) Values(8, 10);
Insert Into linker(parent, child) Values(10, 9);
This is the cte that I came up with that I believe came the closest, but its probably still pretty far off:
with cte As
(
Select
id,
name,
0 as level,
Cast(name as varchar(255) as sort
From items i
Left outer Join
linker li
On i.id = li.child
And li.parent is Null
Union All
Select
id,
name,
cte.level + 1,
Cast(cte.sort + '.' + i.name As Varchar(255)) as sort
From cte
Left Outer Join linker li
on li.child = cte.id
Inner Join items i
On li.parent = i.id
)
Select
id,
name,
level,
sort
From cte
Order By Sort;
Thanks for any help in advance. I am very open to the idea that everything I am doing from the data structure up is wrong, so keep that in mind when you are answering.
Edit: It is probably worth noting that the results don't need to be in order. I plan on creating a ancestry path field in the cte statement, and using that patb to populate my tree.
Edit: oops, I copied and pasted the wrong bit of cte code. I am on mobile so I did my best to change it to what I was doing on my desktop. Once I have a chance I will double check the cte statement against my notes.

Answer 1

Alright, seems like my brain just needed some rest. I have figured out a solution to my issue with much less frustration than yesterday.
My cte statement was:
with cte As(
Select
id,
name,
li.parent,
li.child,
Cast(name as Varchar(255)) as ancestory
From items i
Left Outer Join linker li
On i.id = li.child
Where li.parent is null
Union All
Select
i.id,
i.name,
li.parent,
li.child,
Cast(cte.ancestory || "." || i.name as Varchar(100)) as ancestory
From cte
Left join linker li
On cte.id = li.parent
Inner Join items i
On li.child = i.id
)
select * from cte
The results come back as:
id name parent child ancestory
1 Audio Audio
7 Lighting Lighting
8 Rigging Rigging
2 Speaker 1 2 Audio.Speaker
3 Microphone 1 3 Audio.Microphone
4 Mic Pack 1 4 Audio.Mic Pack
9 Safety 7 9 Lighting.Safety
10 Lighting 8 10 Rigging.Lighting Rigging
Rigging
3 Microphone 4 3 Audio.Mic Pack.Microphone
5 Di 4 5 Audio.Mic Pack.Di
12 Small 9 12 Lighting.Safety.Small Safety
Safety
9 Safety 10 9 Rigging.Lighting Rigging.Safety
6 Passive Di 5 6 Audio.Mic Pack.Di.Passive Di
12 Small Safety 9 12 Rigging.Lighting Rigging.Safety.Small Safety
11 Rapco Di 6 11 Audio.Mic Pack.Di.Passive Di.Rapco Di
13 Dbx Di 6 13 Audio.Mic Pack.Di.Passive Di.Dbx Di