ms accès: requête (concat plusieurs enregistrements en un seul)

Question

est ici un aperçu de la table d'origine:

Occurrence Number Occurrence Date 1 0 Preanalytical (Before Testing) Cup Type 2 0 Analytical (Testing Phase) 2 0 Area 3 0 Postanalytical (After Testing) 4 0 Other Practice Code Comments 
1477 2/5/2010 1.1 Specimen Mislabeled      PURSLEY 
1476 2/5/2010 1.1 Specimen Mislabeled      HPMR  
1475 2/5/2010 1.1 Specimen Mislabeled      ACCIM N008710 
1474 2/5/2010 1.1 Specimen Mislabeled      ACCIM N008636 
1473 2/5/2010 1.3 QNS-Quantity Not Sufficient      SAPMC 
1472 2/5/2010 1.3 QNS-Quantity Not Sufficient      RMG 
1471 2/5/2010 1.1 Specimen Mislabeled      NMED  
1470 2/5/2010 1.9 QNS- Specimen Spilled in transit      MRPS  
1469 2/5/2010 1.9 QNS- Specimen Spilled in transit      ANESPC 
1468 2/5/2010   2.22 Instrument Problem-reinject    LAB 
1525 2/8/2010   2.5 Other - False (+) Blanks Tecan 2   LAB 
1524 2/8/2010   2.5 Other - False (+) Blanks Tecan #1   LAB Blank 019 
1523 2/8/2010   2.22 Instrument Problem, 2.5 Other Tecan   LAB 
1519 2/8/2010     3.3A Reporting Error 4.1 LIS Problem? (see LOM 1418,1520) LAB/SJC F356028 
1518 2/8/2010 1.4 Tests Missed/Wrong Test Ordered      SDPTC F316628 
1516 2/8/2010 1.6 Test Requisition Missing      TPMCF 2 specimens both unlabeled 
1515 2/8/2010 1.1 Specimen Mislabeled      PALMETTO  
1514 2/8/2010 1.1 Specimen Mislabeled      THWR  
1513 2/8/2010 1.1 Specimen Mislabeled      THWR  

Je reçois des informations de cette table en utilisant la déclaration suivante:

select mid(Lom1,1,4) as LOM, sum([Count1]) as [Count] from (

SELECT [Lab Occurrence Form].[1 0 Preanalytical (Before Testing)] as Lom1,Count([Lab Occurrence Form].[1 0 Preanalytical (Before Testing)]) AS [Count1] 
FROM [Lab Occurrence Form] 
WHERE ((([Lab Occurrence Form].[Occurrence Date]) Between [Forms]![Meeting_Reasons_Frequency]![Text4] And [Forms]![Meeting_Reasons_Frequency]![Text2])) 
GROUP BY [Lab Occurrence Form].[1 0 Preanalytical (Before Testing)] 
HAVING Count([Lab Occurrence Form].[1 0 Preanalytical (Before Testing)])<>0 

UNION SELECT [Lab Occurrence Form].[2 0 Analytical (Testing Phase)], Count([Lab Occurrence Form].[2 0 Analytical (Testing Phase)]) AS [CountOf2 0 Analytical (Testing Phase)] 
FROM [Lab Occurrence Form] 
WHERE ((([Lab Occurrence Form].[Occurrence Date]) Between [Forms]![Meeting_Reasons_Frequency]![Text4] And [Forms]![Meeting_Reasons_Frequency]![Text2])) 
GROUP BY [Lab Occurrence Form].[2 0 Analytical (Testing Phase)] 
HAVING Count([Lab Occurrence Form].[2 0 Analytical (Testing Phase)])<>0 

union 

SELECT [Lab Occurrence Form].[3 0 Postanalytical (After Testing)], Count([Lab Occurrence Form].[3 0 Postanalytical (After Testing)]) AS [CountOf3 0 Postanalytical (After Testing)] 
FROM [Lab Occurrence Form] 
WHERE ((([Lab Occurrence Form].[Occurrence Date]) Between [Forms]![Meeting_Reasons_Frequency]![Text4] And [Forms]![Meeting_Reasons_Frequency]![Text2])) 
GROUP BY [Lab Occurrence Form].[3 0 Postanalytical (After Testing)] 
HAVING Count([Lab Occurrence Form].[3 0 Postanalytical (After Testing)])<>0 

UNION SELECT [Lab Occurrence Form].[4 0 Other], Count([Lab Occurrence Form].[4 0 Other]) AS [CountOf4 0 Other] 
FROM [Lab Occurrence Form] 
WHERE ((([Lab Occurrence Form].[Occurrence Date]) Between [Forms]![Meeting_Reasons_Frequency]![Text4] And [Forms]![Meeting_Reasons_Frequency]![Text2])) 
GROUP BY [Lab Occurrence Form].[4 0 Other] 
HAVING Count([Lab Occurrence Form].[4 0 Other])<>0 
ORDER BY 1, 2) 

group by mid(Lom1,1,4); 

c'est ce que la requête renvoie:

LOM Count 
1.1  231 
1.11 21 
1.3  103 
1.4  6 
1.5  1 
1.6  25 
1.8  2 
1.9  88 
2.1  8 
2.22 5 
2.24 1 
2.3  1 
2.4  1 
2.5  29 
3.2  13 
3.3  8 
3.3A 4 
4.1  2 
4.6  1 
4.8  7 

J'ai besoin d'ajouter une autre colonne ici. Disons que c'est column3

c'est la sortie qui ont besoin:

LOM Count column3 
1.1  231 everything from original table where LOM LIKE *1.1* separated by "," 
1.11 21 everything from original table where LOM=1.11 separated by "," 
1.3  103 everything from original table where LOM=1.3 separated by "," 
1.4  6 everything from original table where LOM=1.4 separated by "," 
1.5  1 everything from original table where LOM=1.5 separated by "," 
1.6  25 
1.8  2 
1.9  88 
2.1  8 
2.22 5 
2.24 1 
2.3  1 
2.4  1 
2.5  29 
3.2  13 
3.3  8 
3.3A 4 
4.1  2 
4.6  1 
4.8  7 
prac 1 

qui signifierait le premier élément dans la colonne 3 serait "something1, something2, etc ... somethingelse231"

i excuses si cette explication est horrible, s'il vous plaît laissez-moi savoir si je peux préciser quoi que ce soit

Answer 1

Voici une solution que je trouve: http://www.access-programmers.co.uk/forums/showpost.php?p=272455&postcount=2

Il nécessite l'écriture d'une fonction VBA. Je ne connais pas un moyen de le faire avec SQL directement dans Access.

Public Function Conc(Fieldx, Identity, Value, Source) As Variant 
    Dim cnn As ADODB.Connection 
    Dim rs As ADODB.Recordset 
    Dim SQL As String 
    Dim vFld As Variant 

    Set cnn = CurrentProject.Connection 
    Set rs = New ADODB.Recordset 
    vFld = Null 

    SQL = "SELECT [" & Fieldx & "] as Fld" & _ 
     " FROM [" & Source & "]" & _ 
     " WHERE [" & Identity & "]=" & Value 

    ' open recordset. 
    rs.Open SQL, cnn, adOpenForwardOnly, adLockReadOnly 

    ' concatenate the field. 
    Do While Not rs.EOF 
    If Not IsNull(rs!Fld) Then 
     vFld = vFld & ", " & rs!Fld 
    End If 
    rs.MoveNext 
    Loop 
    ' remove leading comma and space. 
    vFld = Mid(vFld, 3) 

    Set cnn = Nothing 
    Set rs = Nothing 

    ' return concatenated string. 
    Conc = vFld 
End Function 

Vous pouvez alors l'utiliser dans une requête comme ceci:

SELECT [tblData].[ID], 
     Conc("Field1","ID",[ID],"tblData") AS Field1, 
     Conc("Field2","ID",[ID],"tblData") AS Field2 
FROM tblData 
GROUP BY [tblData].[ID]; 

Modifier Donc, votre première requête ressemblerait à quelque chose comme ceci:

SELECT [Lab Occurrence Form].[1 0 Preanalytical (Before Testing)] as Lom1, 
     Conc("NameOfTheFieldToConcatenate", 
      "[Lab Occurrence Form].[1 0 Preanalytical (Before Testing)]", 
      [Lab Occurrence Form].[1 0 Preanalytical (Before Testing)], 
      "[Lab Occurrence Form]"), 
     Count([Lab Occurrence Form].[1 0 Preanalytical (Before Testing)]) AS [Count1] 
FROM [Lab Occurrence Form] 
WHERE ((([Lab Occurrence Form].[Occurrence Date]) Between [Forms]![Meeting_Reasons_Frequency]![Text4] And [Forms]![Meeting_Reasons_Frequency]![Text2])) 
GROUP BY [Lab Occurrence Form].[1 0 Preanalytical (Before Testing)] 
HAVING Count([Lab Occurrence Form].[1 0 Preanalytical (Before Testing)])<>0 

Notez que vous pouvez Vous devez modifier un peu la fonction Conc() pour obtenir la comparaison WildSwable au lieu d'une correspondance exacte dans le champ LOM.