Pivot on Multiple Fields and Export from Access

Pivot on multiple fields and export from Access

In your question you say that

I have a way to generate a report that looks like the following

and then list the data as

Batch  Zone  Value1  Value2
-----  ----  ------  ------
   25     1       5      15
   25     2      12      31
   26     1       6      14
   26     2      10      32

Now perhaps the data may already be in "un-pivoted" form somewhere (with different Values in separate rows), but if not then you would use something like the following query to achieve that

SELECT 
    [Batch],
    "Zone" & [Zone] & "_" & "Value1" AS [ValueID],
    [Value1] AS [ValueValue]
FROM BatchDataByZone
UNION ALL
SELECT 
    [Batch],
    "Zone" & [Zone] & "_" & "Value2" AS [ValueID],
    [Value2] AS [ValueValue]
FROM BatchDataByZone

...returning:

Batch  ValueID       ValueValue
-----  ------------  ----------
   25  Zone1_Value1           5
   25  Zone2_Value1          12
   26  Zone1_Value1           6
   26  Zone2_Value1          10
   25  Zone1_Value2          15
   25  Zone2_Value2          31
   26  Zone1_Value2          14
   26  Zone2_Value2          32

However you get to that point, if you save that query as [BatchDataUnpivoted] then you could use a simple Crosstab Query to "string out" the values for each batch...

TRANSFORM Sum(BatchDataUnpivoted.[ValueValue]) AS SumOfValueValue
SELECT BatchDataUnpivoted.[Batch]
FROM BatchDataUnpivoted
GROUP BY BatchDataUnpivoted.[Batch]
PIVOT BatchDataUnpivoted.[ValueID];

...returning...

Batch  Zone1_Value1  Zone1_Value2  Zone2_Value1  Zone2_Value2
-----  ------------  ------------  ------------  ------------
   25             5            15            12            31
   26             6            14            10            32

Access 2013: Pivot multiple columns

TRANSFORM can only occur at the start of an SQL statement.

First unpivot all the years and only then pivot the metric. Now, as MS Access apparently
has its limits when it comes to complex queries (many unions), it is best to pass via
an intermediate table:

SELECT  FileIDRunID, Metric, Year, Value
FROM    (
            SELECT      FileIDRunID, Metric, 1999 As Year, [1999] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2000 As Year, [2000] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2001 As Year, [2001] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2002 As Year, [2002] As Value
            FROM        DataTable
        )
INTO myTempTable;

I have union-ed 4 years, but you should check how far you can go before MS Access refuses the statement for being too complex.
Then, add some more years, with the same number of unions:

INSERT INTO myTempTable (FileIDRunID, Metric, Year, Value)
SELECT  FileIDRunID, Metric, Year, Value
FROM    (
            SELECT      FileIDRunID, Metric, 2003 As Year, [2003] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2004 As Year, [2004] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2005 As Year, [2005] As Value
            FROM        DataTable
            UNION ALL
            SELECT      FileIDRunID, Metric, 2006 As Year, [2006] As Value
            FROM        DataTable
        );

etc... to finally do this:

TRANSFORM FIRST(Value)
SELECT   FileIDRunID,
         Year
FROM     myTempTable
GROUP BY FileIDRunID,
         Year
PIVOT    Metric

And then drop the temporary table and compress the database to get the original size back. If possible, redesign the database to use the completely unpivoted structure as in the temporary table.

Transform multiple rows into columns with unique key

Indoors today waiting out the hurricane, so I figured I'd create this custom solution. Answers for all of these steps are found elsewhere, but it was not simple to sort through all of the contrived solutions so I hope this answer is more useful all around.

The basic answer for changing the rows to columns is here. But unlike the this question's data, the sample data for that answer was already properly sequenced. As long as the column values [ID] and [Email] form unique pairs and have NO NULL values, then its possible to generate the proper sequencing using an aggregate subquery or a call to an Access aggregate function. To reduce the overall number of queries, I went ahead and generated the transformed column names in the same query. (If the values are not unique or have null values, the sequencing will be off and the final results will be missing some data.)

The second challenge here is that there are two columns that need to be transformed, but the Access SQL Transform statement (i.e. the Crosstab query) only allows one transform column per query. Creating two separate queries and then joining them is rather simple, but since the Crosstab queries generate a dynamic (i.e. undetermined) number of columns, it is not really possible to interleave the Email and Comments columns without manual effort of picking each column in order. Further, explicitly specifying which columns are in the query destroys the dynamic aspect of the crosstab queries and will either leave additional column out, or it will generate an error if the overall number of columns is reduced.

UPDATE: Just after posting the original solution (now labeled Solution 2), I realized that I could resolve the column interleaving problem by actually going backward one step... first generate more rows --one row for each email address and a separate row for each comment-- before the final transform placing them on the same row altogether again.

Solution 1

Save the following query and name it [Sequenced]. To facilitate the correct column ordering in the final output, I used the word 'Remark' instead of 'Comment' since it sorts after 'Email':

SELECT Data.ID, Data.Email, Data.Comments,
    1 + DCount("[ID]","Data","[ID]=" & [ID] & " and [Email]<'" & Replace([Email],"'","''") & "'") AS SeqNum,
    Format([SeqNum],"000") & ' Email' AS EmailColumn,
    Format([SeqNum],"000") & ' Remark' AS CommentsColumn
FROM Data
ORDER BY Data.ID, Data.Email;

Save the following query and name it [Backwards]:

SELECT ID, EmailColumn AS ColumnName, Email AS [Value]
FROM Sequenced
UNION SELECT ID, CommentsColumn AS ColumnName, Comments AS [Value]
FROM Sequenced
ORDER BY [ID], [ColumnName];

Save the following query and name it [InterleavedCrosstab]:

TRANSFORM First(Backwards.Value) AS FirstOfValue
SELECT Backwards.ID
FROM Backwards
GROUP BY Backwards.ID
ORDER BY Backwards.ID, Backwards.ColumnName
PIVOT Backwards.ColumnName;

Solution 2

Save the following query and name it [Sequenced2]:

SELECT Data.ID, Data.Email, Data.Comments,
    1 + DCount("[ID]","Data","[ID]=" & [ID] & " and [Email]<'" & Replace([Email],"'","''") & "'") AS SeqNum,
    'Email' & Format([SeqNum],"000") AS EmailColumn,
    'Comments' & Format([SeqNum],"000") AS CommentsColumn
FROM Data
ORDER BY Data.ID, Data.Email;

Save the following query and name it [EmailCrosstab]:

TRANSFORM First(Sequenced2.Email) AS FirstOfEmail
SELECT Sequenced2.ID
FROM Sequenced2
GROUP BY Sequenced2.ID
ORDER BY Sequenced2.ID
PIVOT Sequenced2.EmailColumn;

Save the following query and name it [CommentsCrosstab]:

TRANSFORM First(Sequenced2.Comments) AS FirstOfComments
SELECT Sequenced2.ID
FROM Sequenced2
GROUP BY Sequenced2.ID
ORDER BY Sequenced2.ID
PIVOT Sequenced2.CommentsColumn;

Finally, the most general result query will return ALL columns, but they will not be interleaved and there will duplicate [ID] columns:

SELECT EmailCrosstab.*,
       CommentsCrosstab.*
FROM CommentsCrosstab INNER JOIN EmailCrosstab
    ON CommentsCrosstab.ID = EmailCrosstab.ID;

Here's a prettified version but only has up to 3 email and comment columns:

SELECT EmailCrosstab.ID,
    EmailCrosstab.Email001,CommentsCrosstab.Comments001,
    EmailCrosstab.Email002,CommentsCrosstab.Comments002,
    EmailCrosstab.Email003,CommentsCrosstab.Comments003
FROM CommentsCrosstab INNER JOIN EmailCrosstab
    ON CommentsCrosstab.ID = EmailCrosstab.ID;

---

Solution 3

I had already typed up the following VBA procedure when I realized the query-only solution was rather easy, so here's a bonus alternative.

Public Sub CustomTransform()
  '* This code assumes that the field values
  '* [ID] and [Email] constitute a unique pair
  '* and that there are NO NULL values.

  Dim i As Integer, MaxIDRows As Integer
  Dim IDSeq As Integer
  Dim lastID As Long
  Dim IDstring As String
  Dim tbl As TableDef
  Dim idx As Index
  Dim db As Database
  Dim rsSrc As Recordset2, rsResult As Recordset2
  Const resultTable As String = "Customer Crosstab"

  Set db = CurrentDb

  MaxIDRows = db.OpenRecordset( _
      "SELECT Max(Counter.Rows) AS MaxRows" & _
      " FROM ( SELECT Count(Data.[ID]) AS [Rows]" & _
        " FROM Data GROUP BY Data.[ID]) AS Counter" _
      ).Fields(0).Value

  '* Column count <= 254 : ID + N * (Email + Comment columns)
  If MaxIDRows = 0 Then
    MsgBox "No data.", vbOKOnly Or vbExclamation, "No Data"
    Exit Sub
  ElseIf MaxIDRows >= 252 / 2 Then
    MsgBox "Maximum number of columns exceeded.", _
        vbOKOnly Or vbExclamation, "Maximum Columns Exceeded"
    Exit Sub
  End If

  On Error Resume Next
  db.TableDefs.Delete resultTable
  Err.Clear
  On Error GoTo 0

  Set tbl = db.CreateTableDef(resultTable)
  With tbl
    ' Create fields and append them to the new TableDef
    ' object. This must be done before appending the
    ' TableDef object to the TableDefs collection of the
    ' Northwind database.
    .Fields.Append .CreateField("ID", dbLong)

    For i = 1 To MaxIDRows
      IDstring = Format(i, "000")
      .Fields.Append .CreateField("Email" & IDstring, dbText, 255)
      .Fields.Append .CreateField("Comments" & IDstring, dbText, 255)
    Next

    Set idx = .CreateIndex("Primary Key")
    idx.Fields.Append idx.CreateField("ID")
    idx.Primary = True
    .Indexes.Append idx
  End With
  db.TableDefs.Append tbl

  Set rsResult = db.OpenRecordset(resultTable, dbOpenTable)
  Set rsSrc = db.OpenRecordset( _
      "SELECT ID, Email, Comments" & _
      " FROM Data" & _
      " ORDER BY ID, Email")

  lastID = -1
  Do Until rsSrc.EOF
    If rsSrc!id <> lastID Then
      If lastID <> -1 Then
        rsResult.Update
      End If

      IDSeq = 0
      rsResult.AddNew
      rsResult!id = rsSrc!id
    End If
    lastID = rsSrc!id

    IDSeq = IDSeq + 1
    IDstring = Format(IDSeq, "000")

    rsResult.Fields("Email" & IDstring) = rsSrc!email
    rsResult.Fields("Comments" & IDstring) = rsSrc!Comments

    rsSrc.MoveNext
  Loop
  rsSrc.Close

  If rsResult.EditMode <> dbEditNone Then
    rsResult.Update
  End If
  rsResult.Close

  Debug.Print "CustomTransform Done"
End Sub

Pivot Query in MS Access

Consider:

TRANSFORM First(Data.Cat) AS FirstOfCat
SELECT Data.ID, Data.Name
FROM Data
GROUP BY Data.ID, Data.Name
PIVOT "Cat" & DCount("*","Data","ID=" & [ID] & " AND Cat<'" & [Cat] & "'")+1;

Or if there is a unique record identifier field - autonumber should serve:

TRANSFORM First(Data.Cat) AS FirstOfCat
SELECT Data.ID, Data.Name
FROM Data
GROUP BY Data.ID, Data.Name
PIVOT "Cat" & DCount("*","Data","ID=" & [ID] & " AND ID_PK<" & [ID_PK])+1;

Access Union/Pivot to Swap Columns and Rows

This will have to be a two-step process to transform. First you will have to rotate the data in your current query to be in rows instead of columns, then you will have to transform the dates into columns instead of rows.

The query will be something like this:

TRANSFORM max(val) as MaxValue
SELECT col
FROM
(
  SELECT [Date], '# Insp' as Col, [# Insp] as  val 
  FROM yourQuery
  UNION ALL
  SELECT [Date], '# Passed' as Col, [# Passed] as val
  FROM yourQuery
  UNION ALL
  SELECT [Date], '# Failed' as Col, [# Failed] as val
  FROM yourQuery
  UNION ALL
  SELECT [Date], '% Acceptance' as Col, [% Acceptance] as val
  FROM yourQuery
) 
GROUP BY col
PIVOT [Date]

I am guessing the your current query is saved in your database, you will replace the yourQuery in my example with the name of your query.

I just tested this in MS Access 2003 with the values in your sample above and it produced the result you want.

---

Related Topics