Finding Connected Subgraphs in an Undirected Graph

Problem:

Given an undirected graph represented by two columns, 'Identifier1' and 'Identifier2', how do we group identifiers that are related to each other and assign them unique group IDs?

Solution:

This problem can be solved by treating the data as edges in a graph and traversing all edges recursively.

Recursive Algorithm:

1. Create a table containing all unique identifiers from both columns.
2. Create a table containing all edges (pairs of identifiers) in both directions.
3. Define a recursive query that traverses the graph starting from each identifier and builds a path of traversed identifiers.
4. Group the results by the starting identifier (anchor identifier) to identify connected components.
5. Assign a unique group ID to each connected component based on the anchor identifier.

Example Query (SQL):

WITH
CTE_Idents AS (
    SELECT Ident1 AS Ident
    FROM @T

    UNION

    SELECT Ident2 AS Ident
    FROM @T
),
CTE_Pairs AS (
    SELECT Ident1, Ident2
    FROM @T
    WHERE Ident1 <> Ident2

    UNION

    SELECT Ident2 AS Ident1, Ident1 AS Ident2
    FROM @T
    WHERE Ident1 <> Ident2
),
CTE_Recursive AS (
    SELECT
        CAST(CTE_Idents.Ident AS varchar(8000)) AS AnchorIdent 
        , Ident1
        , Ident2
        , CAST(','   Ident1   ','   Ident2   ',' AS varchar(8000)) AS IdentPath
        , 1 AS Lvl
    FROM 
        CTE_Pairs
        INNER JOIN CTE_Idents ON CTE_Idents.Ident = CTE_Pairs.Ident1

    UNION ALL

    SELECT 
        CTE_Recursive.AnchorIdent 
        , CTE_Pairs.Ident1
        , CTE_Pairs.Ident2
        , CAST(CTE_Recursive.IdentPath   CTE_Pairs.Ident2   ',' AS varchar(8000)) AS IdentPath
        , CTE_Recursive.Lvl   1 AS Lvl
    FROM
        CTE_Pairs
        INNER JOIN CTE_Recursive ON CTE_Recursive.Ident2 = CTE_Pairs.Ident1
    WHERE
        CTE_Recursive.IdentPath NOT LIKE CAST('%,'   CTE_Pairs.Ident2   ',%' AS varchar(8000))
),
CTE_RecursionResult AS (
    SELECT AnchorIdent, Ident1, Ident2
    FROM CTE_Recursive
),
CTE_CleanResult AS (
    SELECT AnchorIdent, Ident1 AS Ident
    FROM CTE_RecursionResult

    UNION

    SELECT AnchorIdent, Ident2 AS Ident
    FROM CTE_RecursionResult
)
SELECT
    CTE_Idents.Ident
    ,CASE WHEN CA_Data.XML_Value IS NULL 
    THEN CTE_Idents.Ident ELSE CA_Data.XML_Value END AS GroupMembers
    ,DENSE_RANK() OVER(ORDER BY 
        CASE WHEN CA_Data.XML_Value IS NULL 
        THEN CTE_Idents.Ident ELSE CA_Data.XML_Value END
    ) AS GroupID
FROM
    CTE_Idents
    CROSS APPLY
    (
        SELECT CTE_CleanResult.Ident ','
        FROM CTE_CleanResult
        WHERE CTE_CleanResult.AnchorIdent = CTE_Idents.Ident
        ORDER BY CTE_CleanResult.Ident FOR XML PATH(''), TYPE
    ) AS CA_XML(XML_Value)
    CROSS APPLY
    (
        SELECT CA_XML.XML_Value.value('.', 'NVARCHAR(MAX)')
    ) AS CA_Data(XML_Value)
WHERE
    CTE_Idents.Ident IS NOT NULL
ORDER BY Ident;

Key Points:

• The recursive CTE (Common Table Expression) traverses the graph and builds connected components.
• The final SELECT statement assigns group IDs and generates output in the desired format.
• This solution is optimized to avoid redundant calculations and provide efficient results.