无法使用DISTINCT使用连接的表更新主表-访问：此记录集不可更新

Question

gurus! I'm using SQL Server linked tables in Access Forms. In MainTable I need to update and insert records, but Access won't let it, for update it says "This Recordset is not updateable". I know, it's couse DISTINCT, but it's necessary for TableType records - I need only one related name_ds from TableTypes (even first by npr) and in result just thees 7 MainTable records not 16 (without DISTINCT). Any workarounds? Simple structure -

MainTable: id, npr, name, type, datasource_fk.
TableDS: id, name_ds, something. 
TableType: id, npr, name_type, something_type.

Data - MainTable:

1;12;"Olie";"percentage";1
2;15;"Tol";"count";2
3;13;"Opp";"percentage";1
4;12;"Hypq";"count";3
5;14;"Gete";"count";1
6;;"Mour";"count";2
7;;"Ellt";"percentage";3

TableDS:

1;"City1";"q"
2;"City2";"a"
3;"State1";"z"
4;"State2";"x"

TableType:

1;12;"City1";"w"
2;15;"City1";"s"
3;13;"City1";"x"
4;14;"City2";"w"
5;14;"City1";"s"
6;13;"City3";"p"
7;12;"City1";"t"
8;12;"City1";"n"
9;12;"State1";"r"
10;15;"State1";"r"

SQL, result -

SELECT DISTINCT t3.npr AS npr_type, t1.npr, t1.id, t1.name, t2.name_ds, t1.datasource_fk, t1.types
FROM (MainTable AS t1 LEFT JOIN TableDS AS t2 ON t1.datasource_fk = t2.id) LEFT JOIN TableType AS t3 ON t1.npr = t3.npr;

---------------------------------------------------------------------------------------------------------------------------------------------
|     npr_type      |        npr        |        id         |       name        |      name_ds      |   datasource_fk   |       types       |
---------------------------------------------------------------------------------------------------------------------------------------------
|                   |                   |                 6 | Mour              | City2             |                 2 | count             |
---------------------------------------------------------------------------------------------------------------------------------------------
|                   |                   |                 7 | Ellt              | State1            |                 3 | percentage        |
---------------------------------------------------------------------------------------------------------------------------------------------
|                12 |                12 |                 1 | Olie              | City1             |                 1 | percentage        |
---------------------------------------------------------------------------------------------------------------------------------------------
|                12 |                12 |                 4 | Hypq              | State1            |                 3 | count             |
---------------------------------------------------------------------------------------------------------------------------------------------
|                13 |                13 |                 3 | Opp               | City1             |                 1 | percentage        |
---------------------------------------------------------------------------------------------------------------------------------------------
|                14 |                14 |                 5 | Gete              | City1             |                 1 | count             |
---------------------------------------------------------------------------------------------------------------------------------------------
|                15 |                15 |                 2 | Tol               | City2             |                 2 | count             |
---------------------------------------------------------------------------------------------------------------------------------------------

Answer 1

You are getting 16 matches on your joins because MainTable npr column matches multiple times with TableType npr column.

1;12;"Olie";"percentage";1

Matches to

7;12;"City1";"t"
8;12;"City1";"n"
9;12;"State1";"r"
1;12;"City1";"w"

Your best bet is to use a where clause for column TableType.somethingtype. You can try LEFT JOIN on TableDS and TableType using multiple columns but really, you may need to adjust your data. In other words, inactivate some rows. The following query will show you what you're up against:

SELECT t3.npr AS npr_type, 
        t1.npr, 
        t1.id, 
        t1.name, 
        t2.name_ds, 
        t1.datasource_fk, 
        t1.type,
        t3.something_type
FROM #MainTable t1 
      LEFT JOIN #TableDS AS t2 
      ON t1.datasource_fk = t2.id
      LEFT JOIN #TableType AS t3 
      ON t1.npr = t3.npr
ORDER BY t3.npr, 
        t1.npr, 
        t1.id, 
        t1.name, 
        t2.name_ds, 
        t1.datasource_fk, 
        t1.type,
        t3.something_type

So, after you figure out your data. Then you may be able to do something like:

SELECT t3.npr AS npr_type, 
        t1.npr, 
        t1.id, 
        t1.name, 
        t2.name_ds, 
        t1.datasource_fk, 
        t1.type,
        t3.something_type
FROM #MainTable t1 
      LEFT JOIN #TableDS AS t2 
      ON t1.datasource_fk = t2.id
      LEFT JOIN #TableType AS t3 
      ON t1.npr = t3.npr
WHERE
    (t1.npr = 12 AND t3.something_type = 'n')   
    OR
    (t1.npr = 14 AND t3.something_type = 's')
    OR
    (t1.npr = 13 AND t3.something_type = 'p')
    OR
    (t1.npr = 15 AND t3.something_type = 's')
    OR
    (t1.npr IS NULL)