Can't union two tables even though ADODB confirms equal field counts - vba

I'm not able to union two csvs even though ADODB confirms via .Fields.Count that they both have the same number of columns.
Here's the query that's failing:
select * from csv1.csv union select * from csv2.csv
with the error message:
The number of columns in the two selected tables or queries of a union query do not match
However, when I do select * from csv1.csv and select * from csv2.csv separately, ADODB confirms that .Fields.Count = 8 for both.
Possible key to the problem:
Do I need to create two separate connections? I'm only creating one connection (to the first csv) even though there are two csvs in the query.
I was trying to figure out how to do two separate connections for the same query and it seemed like people weren't finding that necessary - I couldn't find two connections mentioned in equivalent queries people were running against csvs.
Per #Parfait's request to see more of the code:
GetDataFromCSV
Public Function GetDataFromCSV(ByVal fileReport As Scripting.File, ByVal strQuery As String, ByVal arrSourceReports As Variant) As Boolean
Dim strRevisedQuery As String
strRevisedQuery = GetRevisedQueryWithFileAliasesReplacedWithTrueFileNames(strQuery, arrSourceReports)
Dim cnn As ADODB.Connection
Set cnn = OpenConnectionToCSV(fileReport)
If cnn Is Nothing Then
GetDataFromCSV = False
Exit Function
End If
GetDataFromCSV = QueryDataFromCSV(cnn, strRevisedQuery, fileReport.Name, fileReport.Name)
End Function
OpenConnectionToCSV
Private Function OpenConnectionToCSV(ByVal fileCSV As Scripting.File, Optional boolHeadersPresent As Boolean = True) As ADODB.Connection
Dim cnn As ADODB.Connection
Set cnn = New ADODB.Connection
cnn.ConnectionTimeout = 0
Dim strfileCSVParentFolderPath As String
strfileCSVParentFolderPath = fileCSV.ParentFolder
If Right(strfileCSVParentFolderPath, 1) <> Application.PathSeparator Then strfileCSVParentFolderPath = strfileCSVParentFolderPath & Application.PathSeparator
Dim strConn As String
If boolHeadersPresent = False Then
strConn = "Provider=Microsoft.Jet.OLEDB.4.0; Data Source=" & strfileCSVParentFolderPath & ";Extended Properties=""text;HDR=NO;FMT=Delimited"""
Else
strConn = "Provider=Microsoft.Jet.OLEDB.4.0; Data Source=" & strfileCSVParentFolderPath & ";Extended Properties=""text;HDR=YES;FMT=Delimited"""
End If
If strConn <> vbNullString Then
On Error GoTo ErrorHandler
Dim lngRetryCount As Long
lngRetryCount = 0
cnn.Open strConn
On Error GoTo 0
Set OpenConnectionToCSV = cnn
End If
Exit Function
ErrorHandler:
Select Case True
Case InStr(1, Err.Description, "Connect timeout occurred", vbTextCompare) > 0
If lngRetryCount < 30 Then
Application.Wait DateAdd("s", 1, Now)
lngRetryCount = lngRetryCount + 1
Resume
Else
MsgBox "Can't connect to " & fileCSV.Path & ". Reading this file will be skipped."
Exit Function
End If
Case Else
MsgBox "Getting data from " & fileCSV.Name & " has failed with the following error message: " & Err.Number & ": " & Err.Description
On Error GoTo 0
Resume
End Select
End Function
QueryDataFromCSV
Private Function QueryDataFromCSV(ByVal cnn As ADODB.Connection, ByVal strQuery As String, ByVal strCSVName As String, ByVal strFinalReportTitle As String) As Boolean
QueryDataFromCSV = True
Dim cmd As ADODB.Command
Set cmd = PrepareQueryCommand(cnn, strQuery)
CreateQueryDebugLog cmd.CommandText
Dim rst As ADODB.Recordset
Set rst = New ADODB.Recordset
rst.Open cmd
Dim Loop1 As Long
With rst
For Loop1 = 1 To .Fields.Count
If .Fields(Loop1 - 1).Name = "F" & Loop1 Then
If Loop1 < 4 Then
MsgBox "Can't retrieve data from " & strCSVName & " because it is formatted improperly."
Else
MsgBox "Can't retrieve data from " & strCSVName & " because it is delimited improperly. The file is most likely delimited with a comma even though it has addresses or other fields that contain commas. Ask Encounters IT to change this report's delimiter to another character, such as | (pipe), in the Tidal batch file."
End If
QueryDataFromCSV = False
Exit Function
End If
Next Loop1
End With
CopyThisCSVRecordsetToResultSheets rst, strFinalReportTitle
cnn.Close
Set rst = Nothing
Set cmd = Nothing
Set cnn = Nothing
End Function
The error is occurring at rst.Open cmd in the above function QueryDataFromCSV
Illustrating schema.ini creation for #Comintern:
GetRevisedQueryWithFileAliasesReplacedWithTrueFileNames
Private Function GetRevisedQueryWithFileAliasesReplacedWithTrueFileNames(ByVal strQuery As String, ByVal arrSourceReports As Variant) As String
Dim FSO As Scripting.FileSystemObject
Set FSO = New Scripting.FileSystemObject
Dim lngPosition As Long
lngPosition = 0
Do Until lngPosition > Len(strQuery)
Dim lngStartPosition As Long
lngStartPosition = InStr(lngPosition + 1, strQuery, "from", vbTextCompare) + 5
If lngStartPosition > lngPosition Then
Dim lngEndPosition As Long
lngEndPosition = InStr(lngStartPosition + 1, strQuery, " ", vbTextCompare)
If lngEndPosition = 0 Then lngEndPosition = Len(strQuery) + 1
Dim strSourceReportTitle As String
strSourceReportTitle = Mid(strQuery, lngStartPosition, lngEndPosition - lngStartPosition)
Dim Loop2 As Long
For Loop2 = LBound(arrSourceReports, 1) To UBound(arrSourceReports, 1)
If arrSourceReports(Loop2, 1) = strSourceReportTitle Then Exit For
Next Loop2
Dim fileSource As Scripting.File
Set fileSource = FSO.GetFile(arrSourceReports(Loop2, 3))
If arrSourceReports(Loop2, 2) = "TAB" Then arrSourceReports(Loop2, 2) = Chr(9)
CreateSchemaIni fileSource, arrSourceReports(Loop2, 2)
Dim strRevisedQuery As String
If strRevisedQuery = vbNullString Then
strRevisedQuery = Replace(strQuery, "from " & strSourceReportTitle, "from " & fileSource.Name)
Else
strRevisedQuery = Replace(strRevisedQuery, "from " & strSourceReportTitle, "from " & fileSource.Name)
End If
lngPosition = lngEndPosition
Else
lngPosition = Len(strQuery) + 1
End If
Loop
GetRevisedQueryWithFileAliasesReplacedWithTrueFileNames = strRevisedQuery
End Function
CreateSchemaIni
Private Sub CreateSchemaIni(ByVal fileReport As Scripting.File, ByVal strDelimiter As String)
Dim intSystemFileNumber As Integer
intSystemFileNumber = FreeFile()
On Error GoTo ErrorHandler
Open fileReport.ParentFolder.Path & Application.PathSeparator & "Schema.ini" For Output As #intSystemFileNumber
Print #intSystemFileNumber, "[" & fileReport.Name & "]"
Print #intSystemFileNumber, "Format=Delimited(" & strDelimiter & ")"
Close #intSystemFileNumber
Exit Sub
ErrorHandler:
Select Case True
Case InStr(1, Err.Description, "Path/File Access Error", vbTextCompare) > 0
Dim strStandardQueryDebugLogPath As String
strStandardQueryDebugLogPath = fileReport.ParentFolder.Path & Application.PathSeparator & "strQuery.txt"
MsgBox strStandardQueryDebugLogPath & " was inaccessible. Creating log in same folder where your copy of the Mass Queryer is saved instead."
Open Left(ThisWorkbook.Path, InStrRev(ThisWorkbook.Path, Application.PathSeparator, , vbTextCompare)) & "strQuery.txt" For Output As #intSystemFileNumber
Print #intSystemFileNumber, "[" & fileReport.Name & "]"
Print #intSystemFileNumber, "Format=Delimited(" & strDelimiter & ")"
Close #intSystemFileNumber
Exit Sub
Case Else
MsgBox "Creating a query debug log has failed with the following error message: " & Err.Number & ": " & Err.Description
On Error GoTo 0
Resume
End Select
End Sub

With #Comintern's help, I was able to see that I made a silly mistake having nothing to do with the question title in actuality. You can see above that my CreateSchemaIni method was creating and then overwriting the Schema.ini file for each csv I was querying rather than creating and then appending to it. By changing that method to use Open For Append instead of Open For Output, the problem was solved.

Related

How to retrieve data from other Excel using VBA and SQL?

I have a problem with my code. I tried retrieving data from other Excel file. My code works but I received full data in one cell (A1). I'm sorry but I'm just beginner, believe that's the problem related to output, but I'm not find out why:
Sub RefreshData()
'Refresh data
Dim CreateNew As Object
Dim RunSELECT As Object
Dim Data As String
Dim SQL As String
FolderPath = ActiveWorkbook.path
path = Left(FolderPath, InStrRev(FolderPath, "\") - 1)
Set CreateNew = CreateObject("ADODB.Connection")
With CreateNew
.Provider = "Microsoft.ACE.OLEDB.12.0"
.ConnectionString = "Data Source=" & path & "\Task1.xlsm; Extended Properties=Excel 12.0 Xml;HDR=YES;IMEX=1;CorruptLoad=xlRepairFile"
.Open
End With
'Run SQL
SQL = "SELECT * FROM [tw$]"
Set RunSELECT = CreateNew.Execute(SQL)
Do
output = output & RunSELECT(0) & ";" & RunSELECT(1) & ";" & RunSELECT(2) & vbNewLine
Debug.Print RunSELECT(0); ";" & RunSELECT(1) & ";" & RunSELECT(2)
RunSELECT.Movenext
Loop Until RunSELECT.EOF
ThisWorkbook.Worksheets("Dic").Range("A1").Value = output
RunSELECT.Close
CreateNew.Close
Set CreateNew = Nothing
Set RunSELECT = Nothing
End Sub
No need to wrap recordset values wtih semicolon delimiters using a Do loop. Simply use Range.CopyFromRecordset:
SQL = "SELECT * FROM [tw$]"
Set RunSELECT = CreateNew.Execute(SQL)
ThisWorkbook.Worksheets("Dic").Range("A1").CopyFromRecordset RunSELECT
RunSELECT.Close
CreateNew.Close
Set CreateNew = Nothing
Set RunSELECT = Nothing
ADOdb to Retrieve Data From Another Workbook (Without Opening It)
While playing around with Parfait's solution combined with a few posts, I came up with the function copySheetADOdb.
Adjust the constants under Source and Target in testCopySheetADOdb to test it.
The Code
Option Explicit
Sub testCopySheetADOdb()
' Initialize error handling.
Const ProcName = "testCopySheetADOdb"
On Error GoTo clearError ' Turn on error trapping.
' Source
Const Path As String = "F:\Test"
Const FileName As String = "Test.xlsx"
' Sheet Name ('SheetName') is case-insensitive i.e. 'A = a'.
Const SheetName As String = "Sheet1"
' Target
Const tgtName As String = "Sheet1"
Const tgtCell As String = "A1"
Dim wb As Workbook: Set wb = ThisWorkbook
' Define FilePath.
Dim FilePath As String
FilePath = Path & Application.PathSeparator & FileName
' Define Target Range.
Dim rng As Range
Set rng = wb.Worksheets(tgtName).Range(tgtCell)
' Test Result.
Dim Result As Boolean
Result = copySheetADODb(rng, FilePath, SheetName)
' Of course you can do all the above in one line:
'Result = copySheetADODB(Thisworkbook.Worksheets("Sheet1").Range("A1"), _
"C:\Test\Test.xlsx", "Sheet1")
' Inform user.
If Result Then
MsgBox "Worksheet successfully copied.", vbInformation, "Success"
Else
MsgBox "Worksheet not copied.", vbExclamation, "Failure"
End If
ProcExit:
Exit Sub
clearError:
Debug.Print "'" & ProcName & "':" & vbLf & " " & "Run-time error '" _
& Err.Number & "':" & vbLf & " " & Err.Description
On Error GoTo 0 ' Turn off error trapping.
GoTo ProcExit
End Sub
Function copySheetADOdb(TargetCellRange As Range, _
ByVal SourceFilePath As String, _
Optional ByVal SourceSheetName As String = "Sheet1") _
As Boolean
' Initialize error handling.
Const ProcName = "copySheetADOdb"
On Error GoTo clearError ' Turn on error trapping.
' Test Target Cell Range ('TargetCellRange').
If TargetCellRange Is Nothing Then
GoTo NoTargetCellRange
End If
If TargetCellRange.Rows.Count > 1 Or TargetCellRange.Columns.Count > 1 Then
GoTo OneCellOnly
End If
'
' Define SQL Generic String.
Const sqlGeneric As String = "SELECT * FROM [SheetName$]"
Dim conn As Object
Set conn = CreateObject("ADODB.Connection")
Dim strErr As String
With conn
.Provider = "Microsoft.ACE.OLEDB.12.0"
' If you need the headers, HDR=NO means there are no headers
' (not: do not retrieve headers) so the complete data will be retrieved.
.ConnectionString = "Data Source='" _
& SourceFilePath _
& "';" _
& "Extended Properties='" _
& "Excel 12.0 Xml;" _
& "HDR=NO;" _
& "IMEX=1;" _
& "CorruptLoad=xlRepairFile" _
& "';"
On Error GoTo connOpenError
.Open
On Error GoTo clearError
' Run SQL.
Dim SQL As String
' Replace 'SheetName' in SQL Generic String
' with the actual sheet name ('SourceSheetName').
SQL = Replace(sqlGeneric, "SheetName", SourceSheetName)
Dim rs As Object
On Error GoTo connExecuteError
Set rs = .Execute(SQL)
On Error GoTo clearError
If Not TargetCellRange Is Nothing Then
' Copy sheet.
If Not rs.EOF Then
TargetCellRange.CopyFromRecordset rs
' Write result.
copySheetADOdb = True
Else
GoTo NoRecords
End If
End If
NoRecordsExit:
rs.Close
connExecuteExit:
.Close
End With
ProcExit:
Set rs = Nothing
connOpenExit:
Set conn = Nothing
Exit Function
NoTargetCellRange:
Debug.Print "'" & ProcName & "': " & "No Target Cell Range ('Nothing')."
GoTo ProcExit
OneCellOnly:
Debug.Print "'" & ProcName & "': " _
& "Target Cell Range has to be one cell range only."
GoTo ProcExit
NoRecords:
Debug.Print "'" & ProcName & "': No records found."
GoTo NoRecordsExit
connOpenError:
If Err.Number = "-2147467259" Then ' "-2147467259 (80004005)"
strErr = "'" & SourceFilePath & "' is not a valid path"
If Left(Err.Description, Len(strErr)) = strErr Then
Debug.Print "'" & ProcName & "': " & strErr & "..."
On Error GoTo 0 ' Turn off error trapping.
GoTo connOpenExit
End If
Else
GoTo clearError
End If
connExecuteError:
If Err.Number = "-2147467259" Then ' "-2147467259 (80004005)"
strErr = "'" & SourceSheetName & "' is not a valid name"
If Left(Err.Description, Len(strErr)) = strErr Then
Debug.Print "'" & ProcName & "': " & strErr & "..."
On Error GoTo 0 ' Turn off error trapping.
GoTo connExecuteExit
End If
Else
GoTo clearError
End If
clearError:
Debug.Print "'" & ProcName & "':" & vbLf & " " & "Run-time error '" _
& Err.Number & "':" & vbLf & " " & Err.Description
On Error GoTo 0 ' Turn off error trapping.
GoTo ProcExit
End Function
You have problem in this code:
ThisWorkbook.Worksheets("Dic").Range("A1").Value = output
you are yourself asking to save the output to A1 Cell.
I would suggest you use for or while loop to enter the data in cells according to your need.

Excel VBA - writing Data from SQL/Recordset very slow

I am trying to write SQL Server data to an Excel sheet but it is very slow. Is there something to optimize? Approximately, 4000 entries at 20 cColumns takes 6-7 minutes.
Database ("freigabe") Module: Connecting to Database and get RecordSet
(this works like a charm)
Private Function ConnectSQL() As ADODB.Connection
Set conn = New ADODB.Connection
conn.ConnectionString = "DRIVER={SQL Server};" _
& "SERVER=xxxxx;" _
& " DATABASE=xxxxx;" _
& "UID=xxxxxx;PWD=xxxxx; OPTION=3"
conn.Open
Set ConnectSQL = conn
End Function
Public Function load(Optional ByVal FieldName As String = "", Optional ByVal fieldValue As String = "", Optional ByVal ComparisonOperator As String = "=")
'wenn fehler return?
'-> Über errorhandler retun rs oder boolen
Dim rs As New ADODB.Recordset
Dim sql As String
Dim contition As String
contition = " "
Dim sqlfrom As String
Dim sqlto As String
On Error GoTo Fehler:
sql = "SELECT * FROM " & TBLNAME & " WHERE storno='0' AND created BETWEEN '2020-02-01' AND '2020-02-15'"
Set conn = ConnectSQL()
rs.Open sql, conn, adOpenStatic
Set load = rs
Exit Function
End If
Fehler:
load = Err.Description
End Function
Get/Write: Build a connection and retrieving recordset. The While loop is taking long. I am skipping text-rich columns (it gets faster but still too long). Showing a load-window so the person doesn't think that Excel "isn't working". After that, the data get's validated (not included).
Application.ScreenUpdating = False
Application.Calculation = xlCalculationManual
Dim rs As Recordset
Dim k As Integer
Dim i As Integer
Dim startt As Double
Dim endt As Double
Dim rngDst As Range
Set rs = freigabe.load()
Set rngDst = Worksheets("Freigaben").Range("G2")
With Worksheets("Freigaben").Range("g2:Z50000")
.ClearContents
'.CopyFromRecordset rs
End With
Count = rs.RecordCount
k = 0
gui_laden.Show
startt = Timer
With rs
If Not .BOF And Not .EOF Then
.MoveLast
.MoveFirst
While Not .EOF
For i = 0 To .Fields.Count - 1
If i <> 13 And i <> 2 And i <> 10 And i <> 5 And i <> 6 And i <> 0 Then rngDst.Offset(, i) = .Fields(i).Value 'skip unneccessary data and write
Next i
k = k + 1
Debug.Print k & "/" & Count
gui_laden.lbl_status = "Lade Daten herunter: " & k & "/" & Count
gui_laden.Repaint
.MoveNext
DoEvents 'Ensure Application doesn't freeze
Set rngDst = rngDst.Offset(1)
Wend
End If
End With
endt = Timer - startt
Debug.Print "Dauer: " & endt
What I tried:
CopyFromRecordSet -> Application freezes
Test in new workbook -> same
Thank you very much!

"Automation Error" "Unspecified Error" When Using Excel-VBA and ADODB Objects

I am creating a macro that looks through a folder, performs a SQL operation on each of the documents, and copies the results into another workbook; however, every time I debug and hit the Set rs = cn.Execute(sql) line I get an
"Automation Error" "Unspecified Error" .
The even weirder thing is that when I just run the code, I get an
"Execute of object _Connection failed" error.
I have tested the SQL code in microsoft SQL server already and each of these statements are almost verbatim with code I have previously got to work.
Option Explicit
Sub hardnessTests()
On Error GoTo ErrorHandling
Dim filename As Variant, n As Long
n = 0
Call turnOff
filename = Dir("T:\Marketing\Data Analytics\GIS Data\Water Quality Portal Data\County Summary Results\*")
While filename <> ""
Dim file As String, cn As Object, rs As Object, sql As String, hardField As ADODB.Field
file = "T:\Marketing\Data Analytics\GIS Data\Water Quality Portal Data\County Summary Results\" & filename
Set cn = CreateObject("ADODB.Connection")
With cn
.Provider = "Microsoft.ACE.OLEDB.12.0"
.connectionstring = "Data Source=" & file & ";Extended Properties=""Excel 12.0 Xml;HDR=YES;Readonly=false;IMEX=0"";"
.Open
End With
sql = "SELECT c.STATEFP, c.COUNTYFP, AVG(CAST(a.ResultMeasureValue AS NUMERIC)) AS Average_Surface_Water_Hardness, COUNT(a.ResultMeasureValue) AS Number_Of_Surface_Water_Tests, " & _
"AVG(CAST(b.ResultMeasureValue AS NUMERIC)) AS Average_Groundwater_Hardness, COUNT(b.ResultMeasureValue) AS Number_Of_Groundwater_Tests " & _
"FROM (SELECT * FROM [Sheet1$] WHERE CAST(LEFT(ActivityStartDate, 4) AS NUMERIC) >= 2010 and ActivityMediaSubdivisionName = 'Surface Water') a, " & _
"(SELECT * FROM [Sheet1$] WHERE CAST(LEFT(ActivityStartDate, 4) AS NUMERIC) >= 2010 AND ActivityMediaSubdivisionName = 'Groundwater') b, " & _
"(SELECT * FROM [Sheet1$] WHERE CAST(LEFT(ActivityStartDate, 4) AS NUMERIC) >= 2010 ) c GROUP BY c.STATEFP, c.COUNTYFP ORDER BY c.STATEFP, c.COUNTYFP"
Set rs = cn.Execute(sql)
Dim wb As Workbook, fieldCount As Long
fieldCount = 0
Set wb = Workbooks.Add
wb.SaveAs "T:\Marketing\Data Analytics\GIS Data\Water Quality Portal Data\County Summary Results\ALL_HARDNESS.xlsx"
If n = 0 Then
wb.Worksheets("Sheet1").Range("A2").CopyFromRecordset rs
Else:
wb.Worksheets("Sheet1").Range("A2").End(xlDown).CopyFromRecordset rs
End If
filename = Dir
n = n + 1
Wend
For Each hardField In rs.Fields
wb.Worksheets("Sheet1").Range("A1").Offset(0, fieldCount) = hardField.Name
fieldCount = fieldCount + 1
Next hardField
Call turnOn
Exit Sub
ErrorHandling:
MsgBox ("Source: " & Err.Source & vbNewLine & "Number: " & Err.Number & vbNewLine & "Description: " & Err.Description & vbNewLine & "Help Context: " & Err.HelpContext)
Done:
End Sub
Private Sub turnOff()
Application.Calculation = xlCalculationManual
Application.ScreenUpdating = False
End Sub
Private Sub turnOn()
Application.Calculation = xlCalculationManual
Application.ScreenUpdating = True
End Sub
Thanks for any help!

Rip 20 million rows from Power Pivot ("Item.data")

I received a workbook which contains two tables in power-pivot (one around one million rows, another 20 mill rows). I would like to rip this out (as anything really - but let's say a CSV) so that I can use it in R + PostGreSQL.
I can't export to an Excel table as there are more than 1 million rows; and copy-pasting the data only works when I select around 200,000 rows.
I tried converting the xlsx into a zip and opening the "item.data" file in notepad++, however it was encrypted.
I put together some VBA which works for around 0.5 mill rows:
Public Sub CreatePowerPivotDmvInventory()
Dim conn As ADODB.Connection
Dim sheet As Excel.Worksheet
Dim wbTarget As Workbook
On Error GoTo FailureOutput
Set wbTarget = ActiveWorkbook
wbTarget.Model.Initialize
Set conn = wbTarget.Model.DataModelConnection.ModelConnection.ADOConnection
' Call function by passing the DMV name
' E.g. Partners
WriteDmvContent "Partners", conn
MsgBox "Finished"
Exit Sub
FailureOutput:
MsgBox Err.Description
End Sub
Private Sub WriteDmvContent(ByVal dmvName As String, ByRef conn As ADODB.Connection)
Dim rs As ADODB.Recordset
Dim mdx As String
Dim i As Integer
mdx = "EVALUATE " & dmvName
Set rs = New ADODB.Recordset
rs.ActiveConnection = conn
rs.Open mdx, conn, adOpenForwardOnly, adLockOptimistic
' Setup CSV file (improve this code)
Dim myFile As String
myFile = "H:\output_table_" & dmvName & ".csv"
Open myFile For Output As #1
' Output column names
For i = 0 To rs.Fields.count - 1
If i = rs.Fields.count - 1 Then
Write #1, rs.Fields(i).Name
Else
Write #1, rs.Fields(i).Name,
End If
Next i
' Output of the query results
Do Until rs.EOF
For i = 0 To rs.Fields.count - 1
If i = rs.Fields.count - 1 Then
Write #1, rs.Fields(i)
Else
Write #1, rs.Fields(i),
End If
Next i
rs.MoveNext
Loop
Close #1
rs.Close
Set rs = Nothing
Exit Sub
FailureOutput:
MsgBox Err.Description
End Sub
DAX Studio will allow you to query the data model in an Excel workbook and output to various formats, including flat files.
The query you'll need is just:
EVALUATE
<table name>
I have found a working (VBA) solution [but greggy's also works for me too!] -> my table was too big to export in one chunk so I loop over and filter by 'month'. This seems to work and produces a 1.2 gb CSV after I append all together:
Function YYYYMM(aDate As Date)
YYYYMM = year(aDate) * 100 + month(aDate)
End Function
Function NextYYYYMM(YYYYMM As Long)
If YYYYMM Mod 100 = 12 Then
NextYYYYMM = YYYYMM + 100 - 11
Else
NextYYYYMM = YYYYMM + 1
End If
End Function
Public Sub CreatePowerPivotDmvInventory()
Dim conn As ADODB.Connection
Dim tblname As String
Dim wbTarget As Workbook
On Error GoTo FailureOutput
Set wbTarget = ActiveWorkbook
wbTarget.Model.Initialize
Set conn = wbTarget.Model.DataModelConnection.ModelConnection.ADOConnection
' Call function by passing the DMV name
tblname = "table1"
WriteDmvContent tblname, conn
MsgBox "Finished"
Exit Sub
FailureOutput:
MsgBox Err.Description
End Sub
Private Sub WriteDmvContent(ByVal dmvName As String, ByRef conn As ADODB.Connection)
Dim rs As ADODB.Recordset
Dim mdx As String
Dim i As Integer
'If table small enough:
'mdx = "EVALUATE " & dmvName
'Other-wise filter:
Dim eval_field As String
Dim eval_val As Variant
'Loop through year_month
Dim CurrYM As Long, LimYM As Long
Dim String_Date As String
CurrYM = YYYYMM(#12/1/2000#)
LimYM = YYYYMM(#12/1/2015#)
Do While CurrYM <= LimYM
String_Date = CStr(Left(CurrYM, 4)) + "-" + CStr(Right(CurrYM, 2))
Debug.Print String_Date
eval_field = "yearmonth"
eval_val = String_Date
mdx = "EVALUATE(CALCULATETABLE(" & dmvName & ", " & dmvName & "[" & eval_field & "] = """ & eval_val & """))"
Debug.Print (mdx)
Set rs = New ADODB.Recordset
rs.ActiveConnection = conn
rs.Open mdx, conn, adOpenForwardOnly, adLockOptimistic
' Setup CSV file (improve this code)
Dim myFile As String
myFile = "H:\vba_tbl_" & dmvName & "_" & eval_val & ".csv"
Debug.Print (myFile)
Open myFile For Output As #1
' Output column names
For i = 0 To rs.Fields.count - 1
If i = rs.Fields.count - 1 Then
Write #1, """" & rs.Fields(i).Name & """"
Else
Write #1, """" & rs.Fields(i).Name & """",
End If
Next i
' Output of the query results
Do Until rs.EOF
For i = 0 To rs.Fields.count - 1
If i = rs.Fields.count - 1 Then
Write #1, """" & rs.Fields(i) & """"
Else
Write #1, """" & rs.Fields(i) & """",
End If
Next i
rs.MoveNext
Loop
CurrYM = NextYYYYMM(CurrYM)
i = i + 1
Close #1
rs.Close
Set rs = Nothing
Loop
Exit Sub
FailureOutput:
MsgBox Err.Description
End Sub
I've modified mptevsion script - now it saves data from the table to separate csv every n rows (100k rows by default, could be changed by changing chunk_size).
The advantage of this script is that it doesn't rely on any field in the table to separate the data, to accomplish that it is using TOPNSKIP (https://dax.guide/topnskip/).
Public Sub CreatePowerPivotDmvInventory()
ActiveWorkbook.Model.Initialize
Dim save_path As String
Dim chunk_size As Long
save_path = "H:\power pivot\csv"
tblName = "data"
chunk_size = 100000
Dim rs As ADODB.Recordset
Dim mdx As String
Dim i As Long
Dim rows_limit As Long
Dim rows_left As Long
Dim conn As ADODB.Connection
Set conn = ActiveWorkbook.Model.DataModelConnection.ModelConnection.ADOConnection
' calculating number of rows in a table
mdx = "evaluate {COUNTROWS('" & tblName & "')}"
Set rs = New ADODB.Recordset
rs.ActiveConnection = conn
rs.Open mdx, conn, adOpenForwardOnly, adLockOptimistic
rows_limit = rs.Fields(0)
rows_left = rows_limit
chunk_id = 1
Do While rows_left > 0
If rows_left < chunk_size Then
chunk_size = rows_left
End If
mdx = "define var data_table = '" & tblName & "'" & Chr(10) & _
"EVALUATE(" & Chr(10) & _
" TOPNSKIP(" & chunk_size & ", " & rows_limit - rows_left & ", data_table)" & Chr(10) & _
");"
Debug.Print (mdx)
Set rs = New ADODB.Recordset
rs.ActiveConnection = conn
rs.Open mdx, conn, adOpenForwardOnly, adLockOptimistic
' Setup CSV file (improve this code)
Dim myFile As String
myFile = save_path & "\vba_tbl_" & tblName & "_" & chunk_id & ".csv"
Debug.Print (myFile)
Open myFile For Output As #1
' Output column names
For i = 0 To rs.Fields.Count - 1
If i = rs.Fields.Count - 1 Then
Write #1, """" & rs.Fields(i).Name & """"
Else
Write #1, """" & rs.Fields(i).Name & """",
End If
Next i
' Output of the query results
Do Until rs.EOF
For i = 0 To rs.Fields.Count - 1
If i = rs.Fields.Count - 1 Then
Write #1, """" & rs.Fields(i) & """"
Else
Write #1, """" & rs.Fields(i) & """",
End If
Next i
rs.MoveNext
Loop
rows_left = rows_left - chunk_size
chunk_id = chunk_id + 1
Close #1
rs.Close
Set rs = Nothing
Loop
MsgBox "Finished"
Exit Sub
FailureOutput:
MsgBox Err.Description
End Sub

excel vba - query on a spreadsheet

if i have these 2 tables:
is there some sort of excel vba code (using ADO) that could acheive these desired results which could utilise any query i put in the SQL sheet?
Here's some VBA code that allows you to read an Excel range using the text SQL driver. It's quite a complex example, but I'm guessing that you came here because you're a fairly advanced user with a more complex problem than the examples we see on other sites.
Before I post the code in full, here's the original 'sample usage' comment in the core function, FetchXLRecordSet:
' Sample usage:
'
' Set rst = FetchXLRecordSet(SQL, "TableAccountLookup", "TableCashMap")
'
' Where the query uses two named ranges, "TableAccountLookup" and "TableCashMap"
' as shown in this SQL statement:
'
' SELECT
' B.Legal_Entity_Name, B.Status,
' SUM(A.USD_Settled) As Settled_Cash
' FROM
' [TableAccountLookup] AS A,
' [TableCashMap] AS B
' WHERE
' A.Account IS NOT NULL
' AND B.Cash_Account IS NOT NULL
' AND A.Account = B.Cash_Account
' GROUP BY
' B.Legal_Entity_Name,
' B.Status
It's clunky, forcing you to name the tables (or list the range addresses in full) when you run the query, but it simplifies the code.
Option Explicit
Option Private Module
' ADODB data retrieval functions to support Excel
' Online reference for connection strings:
' http://www.connectionstrings.com/oracle#p15
' Online reference for ADO objects & properties:
' http://msdn.microsoft.com/en-us/library/ms678086(v=VS.85).aspx
' External dependencies:
' Scripting - C:\Program files\scrrun.dll
' ADO - C:\Program files\Common\system\ado\msado27.tlb
Private m_strTempFolder As String
Private m_strConXL As String
Private m_objConnXL As ADODB.Connection
Public Property Get XLConnection() As ADODB.Connection
On Error GoTo ErrSub
' The Excel database drivers have memory problems so we use the text driver
' to read csv files in a temporary folder. We populate these files from
' ranges specified for use as tables by the FetchXLRecordSet() function.
Dim objFSO As Scripting.FileSystemObject
Set objFSO = New Scripting.FileSystemObject
Set m_objConnXL = New ADODB.Connection
' Specify and clear a temporary folder:
m_strTempFolder = objFSO.GetSpecialFolder(2).ShortPath
If Right(m_strTempFolder, 1) <> "\" Then
m_strTempFolder = m_strTempFolder & "\"
End If
m_strTempFolder = m_strTempFolder & "XLSQL"
Application.DisplayAlerts = False
If objFSO.FolderExists(m_strTempFolder) Then
objFSO.DeleteFolder m_strTempFolder
End If
If Not objFSO.FolderExists(m_strTempFolder) Then
objFSO.CreateFolder m_strTempFolder
End If
If Right(m_strTempFolder, 1) <> "\" Then
m_strTempFolder = m_strTempFolder & "\"
End If
' JET OLEDB text driver connection string:
' Provider=Microsoft.Jet.OLEDB.4.0;Data Source=c:\txtFilesFolder\;Extended Properties="text;HDR=Yes;FMT=Delimited";
' ODBC text driver connection string:
' Driver={Microsoft Text Driver (*.txt; *.csv)};Dbq=c:\txtFilesFolder\;Extensions=asc,csv,tab,txt;
m_strConXL = "Provider=Microsoft.Jet.OLEDB.4.0;Data Source=" & m_strTempFolder & ";"
m_strConXL = m_strConXL & "Extended Properties=" & Chr(34) & "text;HDR=Yes;IMEX=1" & Chr(34) & ";"
With m_objConnXL
.CursorLocation = adUseClient
.CommandTimeout = 90
.ConnectionString = m_strConXL
.Mode = adModeRead
End With
If m_objConnXL.State = adStateClosed Then
Application.StatusBar = "Connecting to the local Excel tables"
m_objConnXL.Open
End If
Set XLConnection = m_objConnXL
ExitSub:
Application.StatusBar = False
Exit Property
ErrSub:
MsgPopup "Error connecting to the Excel local data. Please contact Application Support.", vbCritical + vbApplicationModal, "Database connection failure!", 10
Resume ErrEnd
' Resume ExitSub
ErrEnd:
End ' Terminal error. Halt.
End Property
Public Sub CloseConnections()
On Error Resume Next
Set m_objConnXL = Nothing
End Sub
Public Function FetchXLRecordSet(ByVal SQL As String, ParamArray TableNames()) As ADODB.Recordset
' This allows you to retrieve data from Excel ranges using SQL. You
' need to pass additional parameters specifying each range you're using as a table
' so that the these ranges can be saved as csv files in the 'XLSQL' temporary folder
' Note that your query must use the 'table' naming conventions required by the Excel
' database drivers: http://www.connectionstrings.com/excel#20
On Error Resume Next
Dim i As Integer
Dim iFrom As Integer
Dim strRange As String
Dim j As Integer
Dim k As Integer
If IsEmpty(TableNames) Then
TableNames = Array("")
End If
If InStr(TypeName(TableNames), "(") < 1 Then
TableNames = Array(TableNames)
End If
Set FetchXLRecordSet = New ADODB.Recordset
With FetchXLRecordSet
.CacheSize = 8
Set .ActiveConnection = XLConnection
iFrom = InStr(8, SQL, "From", vbTextCompare) + 4
For i = LBound(TableNames) To UBound(TableNames)
strRange = ""
strRange = TableNames(i)
If strRange = "0" Or strRange = "" Then
j = InStr(SQL, "FROM") + 4
j = InStr(j, SQL, "[")
k = InStr(j, SQL, "]")
strRange = Mid(SQL, j + 1, k - j - 1)
End If
RangeToFile strRange
SQL = Left(SQL, iFrom) & Replace(SQL, strRange, strRange & ".csv", iFrom + 1, 1)
SQL = Replace(SQL, "$.csv", ".csv")
SQL = Replace(SQL, ".csv$", ".csv")
SQL = Replace(SQL, ".csv.csv", ".csv")
Next i
.Open SQL, , adOpenStatic, , adCmdText + adAsyncFetch
i = 0
Do While .State > 1
i = (i + 1) Mod 3
Application.StatusBar = "Connecting to the database" & String(i, ".")
Sleep 250
Loop
End With
Application.StatusBar = False
End Function
Public Function ReadRangeSQL(SQL_Range As Excel.Range) As String
' Read a range into a string.
' Each row is delimited with a carriage-return and a line break.
' Empty cells are concatenated into the string as 'Tabs' of four spaces.
'NH Feb 2018: you cannot return more than 32767 chars into a range.
Dim i As Integer
Dim j As Integer
Dim arrCells As Variant
Dim arrRows() As String
Dim arrRowX() As String
Dim strRow As String
Dim boolIndent As Boolean
Const SPACE As String * 1 = " "
Const SPACE4 As String * 4 = " "
Const MAX_LEN As Long = 32767
arrCells = SQL_Range.Value2
If InStr(TypeName(arrCells), "(") Then
ReDim arrRows(LBound(arrCells, 1) To UBound(arrCells, 1))
ReDim arrRowX(LBound(arrCells, 2) To UBound(arrCells, 2))
For i = LBound(arrCells, 1) To UBound(arrCells, 1) - 1
boolIndent = True
For j = LBound(arrCells, 2) To UBound(arrCells, 2)
If isError(arrCells(i, j)) Then
SQL_Range(i, j).Calculate
End If
If Not isError(arrCells(i, j)) Then
arrRowX(j) = arrCells(i, j)
Else
arrRowX(j) = vbNullString
End If
If boolIndent And arrRowX(j) = "" Then
arrRowX(j) = SPACE4
Else
boolIndent = False
End If
Next j
arrRows(i) = Join(arrRowX, SPACE)
If Len(Trim$(arrRows(i))) = 0 Then
arrRows(i) = vbNullString
Else
arrRows(i) = RTrim$(Join(arrRowX, SPACE))
End If
Next i
Erase arrCells
Erase arrRowX
ReadRangeSQL = Join(arrRows, vbCrLf)
Erase arrRows
ReadRangeSQL = Replace(ReadRangeSQL, vbCrLf & vbCrLf, vbCrLf)
Else
ReadRangeSQL = CStr(arrCells)
End If
If Len(ReadRangeSQL) > MAX_LEN Then
' Trip terminating spaces from each row:
Do While InStr(1, ReadRangeSQL, SPACE & vbCrLf, vbBinaryCompare) > 0
ReadRangeSQL = Replace(ReadRangeSQL, SPACE & vbCrLf, vbCrLf)
Loop
End If
If Len(ReadRangeSQL) > MAX_LEN Then
' Reduce the 'tab' size to 2 selectively, after each row's indentation
arrRows = Split(ReadRangeSQL, vbCrLf)
For i = LBound(arrRows) To UBound(arrRows)
If Len(arrRows(i)) > 16 Then
If InStr(12, arrRows(i), SPACE4) > 0 Then
arrRows(i) = Left$(arrRows(i), 12) & Replace(Right$(arrRows(i), Len(arrRows(i)) - 12), SPACE4, SPACE & SPACE)
End If
End If
Next i
ReadRangeSQL = Join(arrRows, vbCrLf)
Erase arrRows
End If
If Len(ReadRangeSQL) > MAX_LEN Then
' Reduce the 'tab' size to 2 indiscriminately. This will make your SQL illegible:
Do While InStr(1, ReadRangeSQL, SPACE4, vbBinaryCompare) > 0
ReadRangeSQL = Replace(ReadRangeSQL, SPACE4, SPACE & SPACE)
Loop
End If
End Function
Public Sub RangeToFile(ByRef strRange As String)
' Output a range to a csv file in a temporary folder created by the XLConnection function
' strRange specifies a range in the current workbook using the 'table' naming conventions
' specified for Excel OLEDB database drivers: http://www.connectionstrings.com/excel#20
' The first row of the range is assumed to be a set of column names.
On Error Resume Next
Dim objFSO As Scripting.FileSystemObject
Dim rng As Excel.Range
Dim strFile As String
Dim arrData As Variant
Dim iRow As Long
Dim jCol As Long
Dim strData As String
Dim strLine As String
strRange = Replace(strRange, "[", "")
strRange = Replace(strRange, "]", "")
If Right(strRange, 1) = "$" Then
strRange = Replace(strRange, "$", "")
Set rng = ThisWorkbook.Worksheets(strRange).UsedRange
Else
strRange = Replace(strRange, "$", "")
Set rng = Range(strRange)
If rng Is Nothing Then
Set rng = ThisWorkbook.Worksheets(strRange).UsedRange
End If
End If
If rng Is Nothing Then
Exit Sub
End If
Set objFSO = New Scripting.FileSystemObject
strFile = m_strTempFolder & strRange & ".csv"
If objFSO.FileExists(strFile) Then
objFSO.DeleteFile strFile, True
End If
If objFSO.FileExists(strFile) Then
Exit Sub
End If
arrData = rng.Value2
With objFSO.OpenTextFile(strFile, ForWriting, True)
' Header row:
strLine = ""
strData = ""
iRow = LBound(arrData, 1)
For jCol = LBound(arrData, 2) To UBound(arrData, 2)
strData = arrData(iRow, jCol)
strData = Replace(strData, Chr(34), Chr(39))
strData = Replace(strData, Chr(10), " ")
strData = Replace(strData, Chr(13), " ")
strData = strData & ","
strLine = strLine & strData
Next jCol
strLine = Left(strLine, Len(strLine) - 1) ' Trim trailing comma
If Len(Replace(Replace(strLine, Chr(34), ""), ",", "")) > 0 Then
.WriteLine strLine
End If
' Rest of the data
For iRow = LBound(arrData, 1) + 1 To UBound(arrData, 1)
strLine = ""
strData = ""
For jCol = LBound(arrData, 2) To UBound(arrData, 2)
If IsError(arrData(iRow, jCol)) Then
strData = "#ERROR"
Else
strData = arrData(iRow, jCol)
strData = Replace(strData, Chr(34), Chr(39))
strData = Replace(strData, Chr(10), " ")
strData = Replace(strData, Chr(13), " ")
strData = Replace(strData, Chr(9), " ")
strData = Trim(strData)
End If
strData = Chr(34) & strData & Chr(34) & "," ' Quotes to coerce all values to text
strLine = strLine & strData
Next jCol
strLine = Left(strLine, Len(strLine) - 1) ' Trim trailing comma
If Len(Replace(Replace(strLine, Chr(34), ""), ",", "")) > 0 Then
.WriteLine strLine
End If
Next iRow
.Close
End With ' textstream object from objFSO.OpenTextFile
Set objFSO = Nothing
Erase arrData
Set rng = Nothing
End Sub
And finally, Writing a Recordset to a Range - the code would be trivial if it wasn't for all the errors you have to handle:
Public Sub RecordsetToRange(rngTarget As Excel.Range, objRecordset As ADODB.Recordset, Optional FieldList As Variant, Optional ShowFieldNames As Boolean = False, Optional Orientation As Excel.XlRowCol = xlRows)
' Write an ADO Recordset to an Excel range in a single 'hit' to the sheet
' Calling function is responsible for setting the record pointer (must not be EOF!)
' The target range is resized automatically to the dimensions of the array, with the top left cell used as the start point.
On Error Resume Next
Dim OutputArray As Variant
Dim i As Integer
Dim iCol As Integer
Dim iRow As Integer
Dim varField As Variant
If objRecordset Is Nothing Then
Exit Sub
End If
If objRecordset.State <> 1 Then
Exit Sub
End If
If objRecordset.BOF And objRecordset.EOF Then
Exit Sub
End If
If Orientation = xlColumns Then
If IsEmpty(FieldList) Or IsMissing(FieldList) Then
OutputArray = objRecordset.GetRows
Else
OutputArray = objRecordset.GetRows(Fields:=FieldList)
End If
Else
If IsEmpty(FieldList) Or IsMissing(FieldList) Then
OutputArray = ArrayTranspose(objRecordset.GetRows)
Else
OutputArray = ArrayTranspose(objRecordset.GetRows(Fields:=FieldList))
End If
End If
ArrayToRange rngTarget, OutputArray
If ShowFieldNames Then
If Orientation = xlColumns Then
ReDim OutputArray(LBound(OutputArray, 1) To UBound(OutputArray, 1), 1 To 1)
iRow = LBound(OutputArray, 1)
If IsEmpty(FieldList) Or IsMissing(FieldList) Then
For i = 0 To objRecordset.Fields.Count - 1
If i > UBound(OutputArray, 1) Then
Exit For
End If
OutputArray(iRow + i, 1) = objRecordset.Fields(i).Name
Next i
Else
If InStr(TypeName(FieldList), "(") < 1 Then
FieldList = Array(FieldList)
End If
i = 0
For Each varField In FieldList
OutputArray(iRow + i, 1) = CStr(varField)
i = i = 1
Next
End If 'IsEmpty(FieldList) Or IsMissing(FieldList)
ArrayToRange rngTarget.Cells(1, 0), OutputArray
Else
ReDim OutputArray(1 To 1, LBound(OutputArray, 2) To UBound(OutputArray, 2))
iCol = LBound(OutputArray, 2)
If IsEmpty(FieldList) Or IsMissing(FieldList) Then
For i = 0 To objRecordset.Fields.Count - 1
If i > UBound(OutputArray, 2) Then
Exit For
End If
OutputArray(1, iCol + i) = objRecordset.Fields(i).Name
Next i
Else
If InStr(TypeName(FieldList), "(") < 1 Then
FieldList = Array(FieldList)
End If
i = 0
For Each varField In FieldList
OutputArray(1, iCol + i) = CStr(varField)
i = i = 1
Next
End If ' IsEmpty(FieldList) Or IsMissing(FieldList)
ArrayToRange rngTarget.Cells(0, 1), OutputArray
End If ' Orientation = xlColumns
End If 'ShowFieldNames
Erase OutputArray
End Sub
Public Function ArrayTranspose(InputArray As Variant) As Variant
' Transpose InputArray.
' Returns InputArray unchanged if it is not a 2-Dimensional Variant(x,y)
Dim iRow As Long
Dim iCol As Long
Dim iRowCount As Long
Dim iColCount As Long
Dim boolNoRows As Boolean
Dim BoolNoCols As Boolean
Dim OutputArray As Variant
If IsEmpty(InputArray) Then
ArrayTranspose = InputArray
Exit Function
End If
If InStr(1, TypeName(InputArray), "(") < 1 Then
ArrayTranspose = InputArray
Exit Function
End If
' Check that we can read the array's dimensions:
On Error Resume Next
Err.Clear
iRowCount = 0
iRowCount = UBound(InputArray, 1)
If Err.Number <> 0 Then
boolNoRows = True
End If
Err.Clear
Err.Clear
iColCount = 0
iColCount = UBound(InputArray, 2)
If Err.Number <> 0 Then
BoolNoCols = True
End If
Err.Clear
If boolNoRows Then
' ALL arrays have a defined Ubound(MyArray, 1)!
' This variant's dimensions cannot be determined
OutputArray = InputArray
ElseIf BoolNoCols Then
' It's a vector. Strictly speaking, a vector cannot be 'transposed', as
' calling the ordinal a 'row' or a 'column' is arbitrary or meaningless.
' But... By convention, Excel users regard a vector as an array of 1 to n
' rows and 1 column. So we'll 'transpose' it into a Variant(1 to 1, 1 to n)
ReDim OutputArray(1 To 1, LBound(InputArray, 1) To UBound(InputArray, 1))
For iRow = LBound(InputArray, 1) To UBound(InputArray, 1)
OutputArray(1, iRow) = InputArray(iRow)
Next iRow
Else
ReDim OutputArray(LBound(InputArray, 2) To UBound(InputArray, 2), LBound(InputArray, 1) To UBound(InputArray, 1))
If IsEmpty(OutputArray) Then
ArrayTranspose = InputArray
Exit Function
End If
If InStr(1, TypeName(OutputArray), "(") < 1 Then
ArrayTranspose = InputArray
Exit Function
End If
For iRow = LBound(InputArray, 1) To UBound(InputArray, 1)
For iCol = LBound(InputArray, 2) To UBound(InputArray, 2)
OutputArray(iCol, iRow) = InputArray(iRow, iCol)
Next iCol
Next iRow
End If
ExitFunction:
ArrayTranspose = OutputArray
Erase OutputArray
End Function
Postscript: Running SQL on Excel 'Table' Objects
For completeness, here's the code for a barebones 'read Excel Table objects with SQL' function that handles all the text-file hacking in the background.
I'm posting it now, a while after my original answer went up, because everyone's using the rich 'table' object for tabulated data in Excel:
' Run a JOIN query on your tables, and write the field names and data to Sheet1:
SaveTable "Table1"
SaveTable "Table2"
SQL= SQL & "SELECT * "
SQL= SQL & " FROM Table1 "
SQL= SQL & " LEFT JOIN Table2 "
SQL= SQL & " ON Table1.Client = Table2.Client"
RunSQL SQL, Sheet1.Range("A1")
...And the full listing (give or take a couple of functions in the previous code dump) is:
Public Function RunSQL(SQL As String, TargetRange As Excel.Range, Optional DataSetName As String)
' Run SQL against table files in the local ExcelSQL folder and write the results to a target range
' The full implementation of ExcelSQL provides a fully-featured UI on a control sheet
' This is a cut-down version which runs everything automatically, without audit & error-reporting
' SQL can be read from ranges using the ReadRangeSQL function
' If no target range object is passed in, and a Data set name is specified, the recordset will be
' saved as [DataSetName].csv in the local Excel SQL folder for subsequent SQL queries
' If no target range is specified and no Data set name specified, returns the recordet object
Dim rst As ADODB.Recordset
If Left(SQL, 4) = "SQL_" Then
SQL = ReadRangeSQL(ThisWorkbook.Names(SQL).RefersToRange)
End If
Set rst = FetchTextRecordset(SQL)
If TargetRange Is Nothing Then
If DataSetName = "" Then
Set RunSQL = rst
Else
RecordsetToCSV rst, DataSetName, , , , , , , False
Set rst = Nothing
End If
Else
RecordsetToRange rst, TargetRange, True
Set rst = Nothing
End If
End Function
Public Function FetchTextRecordset(SQL As String) As ADODB.Recordset
' Fetch records from the saved text files in the Temp SQL Folder:
On Error Resume Next
Dim i As Integer
Dim iFrom As Integer
If InStr(1, connText, "IMEX=1", vbTextCompare) > 0 Then SetSchema
Set FetchTextRecordset = New ADODB.Recordset
With FetchTextRecordset
.CacheSize = 8
Set .ActiveConnection = connText
On Error GoTo ERR_ADO
.Open SQL, , adOpenStatic, , adCmdText + adAsyncFetch
i = 0
Do While .State > 1
i = (i + 1) Mod 3
Application.StatusBar = "Waiting for data" & String(i, ".")
Application.Wait Now + (0.25 / 24 / 3600)
Loop
End With
Application.StatusBar = False
ExitSub:
Exit Function
ERR_ADO:
Dim strMsg
strMsg = vbCrLf & vbCrLf & "If this is a 'file' error, someone's got one of the source data files open: try again in a few minutes." & vbCrLf & vbCrLf & "Otherwise, please make a note of this error message and contact the developer, or " & SUPPORT & "."
If Verbose Then
MsgBox "Error &H" & Hex(Err.Number) & ": " & Err.Description & strMsg, vbCritical + vbMsgBoxHelpButton, "Data retrieval error:", Err.HelpFile, Err.HelpContext
End If
Resume ExitSub
Exit Function
' Try this if SQL is too big to debug in the immediate window:
' FSO.OpenTextFile("C:\Temp\SQL.txt",ForWriting,True).Write SQL
' Shell "Notepad.exe C:\Temp\SQL.txt", vbNormalFocus
'Resume
End Function
Private Property Get connText() As ADODB.Connection
On Error GoTo ErrSub
Dim strTempFolder
If m_objConnText Is Nothing Then
Set m_objConnText = New ADODB.Connection
strTempFolder = TempSQLFolder ' this will test whether the folder permits SQL READ operations
Application.DisplayAlerts = False
' MS-Access ACE OLEDB Provider
m_strConnText = "Provider=Microsoft.ACE.OLEDB.12.0;Data Source=" & Chr(34) & strTempFolder & Chr(34) & ";Persist Security Info=True;"
m_strConnText = m_strConnText & "Extended Properties=" & Chr(34) & "text;CharacterSet=UNICODE;HDR=Yes;HDR=Yes;IMEX=1;MaxScanRows=1" & Chr(34) & ";"
End If
If Not m_objConnText Is Nothing Then
With m_objConnText
If .State = adStateClosed Then
Application.StatusBar = "Connecting to the local Excel tables"
.CursorLocation = adUseClient
.CommandTimeout = 90
.ConnectionString = m_strConnText
.Mode = adModeRead
.Open
End If
End With
If m_objConnText.State = adStateClosed Then
Set m_objConnText = Nothing
End If
End If
Set connText = m_objConnText
ExitSub:
Application.StatusBar = False
Exit Property
ErrSub:
MsgBox "Error connecting to the Excel local data. Please contact " & SUPPORT & ".", vbCritical + vbApplicationModal, "Database connection failure!", 10
Resume ErrEnd
' Resume ExitSub
ErrEnd:
End ' Terminal error. Halt.
End Property
Public Sub CloseConnections()
On Error Resume Next
Set m_objConnText = Nothing
End Sub
Public Function TempSQLFolder() As String
Application.Volatile False
' Location of temporary table files used by the SQL text data functions
' Also runs a background process to clear out files over 7 days old
' The best location is a named subfolder in the user's temp folder. The
' user local 'temp' folder is discoverable on all Windows systems using
' GetObject("Scripting.FileSystemObject").GetSpecialFolder(2).ShortPath
' and will usually be C:\Users\[User Name]\AppData\Local\Temp
' Dependencies:
' Object Property FSO (Returns Scripting.FilesystemObject)
'
Dim strCMD As String
Dim strMsg As String
Dim strNamedFolder As String
Static strTempFolder As String ' Cache it
Dim iRetry As Integer
Dim i As Long
' If we've already found a usable temp folder, use the static value
' without querying the file system and testing write privileges again:
If strTempFolder <> "" Then
TempSQLFolder = strTempFolder
Exit Function
End If
On Error Resume Next
strTempFolder = GetObject("Scripting.FileSystemObject").GetSpecialFolder(2).ShortPath
If Right(strTempFolder, 1) <> "\" Then
strTempFolder = strTempFolder & "\"
End If
strTempFolder = strTempFolder & "XLSQL"
If Not FSO.FolderExists(strTempFolder) Then
FSO.CreateFolder strTempFolder
End If
i = 1
Do Until FSO.FolderExists(strTempFolder) Or i > 6
Sleep i * 250
Application.StatusBar = "Waiting for SQL cache folder" & String(i Mod 4, ".")
Loop
If Not FSO.FolderExists(strTempFolder) Then
GoTo Retry
End If
If Right(strTempFolder, 1) <> "\" Then
strTempFolder = strTempFolder & "\"
End If
TempSQLFolder = strTempFolder
Application.StatusBar = False
End Function
Public Property Get FSO() As Scripting.FileSystemObject '
' Return a File System Object
On Error Resume Next
If m_objFSO Is Nothing Then
Set m_objFSO = CreateObject("Scripting.FileSystemObject") ' New Scripting.FileSystemObject
End If
If m_objFSO Is Nothing Then
Shell "Regsvr32.exe /s scrrun.dll", vbHide
Set m_objFSO = CreateObject("Scripting.FileSystemObject")
End If
Set FSO = m_objFSO
End Property
Public Sub SaveTable(Optional TableName As String = "*")
' Export a Table object to the local SQL Folder as a csv file
' If no name is specified, all tables are exported asynchronously
' This step is essential for running SQL on the tables
Dim wks As Excel.Worksheet
Dim oList As Excel.ListObject
Dim sFile As String
Dim bAsync As Boolean
If TableName = "*" Then
bAsync = True
Else
bAsync = False
End If
For Each wks In ThisWorkbook.Worksheets
For Each oList In wks.ListObjects
If oList.Name Like TableName Then
sFile = oList.Name
ArrayToCSV oList.Range.Value2, sFile, , , , , , , , bAsync
'Debug.Print "[" & sFile & ".csv] "
End If
Next oList
Next wks
SetSchema
End Sub
Public Sub RemoveTable(Optional TableName As String = "*")
On Error Resume Next
' Clear up the temporary 'Table' files in the user local temp folder:
Dim wks As Excel.Worksheet
Dim oList As Excel.ListObject
Dim sFile As String
Dim sFolder As String
sFolder = TempSQLFolder
For Each wks In ThisWorkbook.Worksheets
For Each oList In wks.ListObjects
If oList.Name Like TableName Then
sFile = oList.Name & ".csv"
If Len(Dir(sFile)) > 0 Then
Shell "CMD /c DEL " & Chr(34) & sFolder & sFile & Chr(34), vbHide ' asynchronous deletion
End If
End If
Next oList
Next wks
End Sub
Share and enjoy: this is all a horrible hack, but it gives you a stable SQL platform.
And we still don't have a stable 'native' platform for SQL on Excel: the Microsoft.ACE.OLEDB.14.0 Excel data provider still has the same memory leak as Microsoft.Jet.OLEDB.4.0 and the Excel ODBC driver that preceded it, twenty years ago.
Some notes:
sFullName = ActiveWorkbook.FullName
sSheet = ActiveSheet.Name
Set cn = CreateObject("adodb.connection")
scn = "Provider=Microsoft.Jet.OLEDB.4.0;Data Source=" _
& sFullName _
& ";Extended Properties=""Excel 8.0;HDR=Yes;IMEX=1"";"
cn.Open scn
Set rs = CreateObject("adodb.recordset")
For Each c In Sheet4.UsedRange
sSQL = sSQL & c.Value & " "
Next
rs.Open sSQL, cn
Sheet5.Range("a10").CopyFromRecordset rs
There is an ODBC driver for Excel.
See: http://support.microsoft.com/kb/178717
And: http://msdn.microsoft.com/en-us/library/ms711711%28v=vs.85%29.aspx
In order to get data out of a database and into Excel you do the following steps.
Record a macro
Import external data, choose a new source, select DSN ODBC as the type of source.
Now choose Excel-file as the type of ODBC source.
Pick the Excel sheet you want to query.
Every table needs to be in a named range, leave the option select a table checked, Excel will not allow us to insert a query just yet.
Follow the wizard and save the .odc file. Open it again and choose edit query. Now you can insert your select statement.
Stop recording and edit the recorded macro to suit your needs.
It looks like source and target are odbc queries. You need to parse the table name out of those queries and replace SoureTable and TargetTable in your query with the right table names.
Sub ExecuteSQL()
Dim sSql As String
Dim rCell As Range
Dim adConn As ADODB.Connection
Dim adRs As ADODB.Recordset
Dim lWherePos As Long
Const sSOURCE As String = "SourceTable"
Const sTARGET As String = "TargetTable"
Const sODBC As String = "ODBC;"
'Buld the sql statement
For Each rCell In Intersect(wshSql.UsedRange, wshSql.Columns(1)).Cells
If Not IsEmpty(rCell.Value) Then
sSql = sSql & rCell.Value & Space(1)
End If
Next rCell
'replace the table names
sSql = Replace(sSql, sSOURCE, GetTableName(wshSource.QueryTables(1).CommandText), 1, 1)
sSql = Replace(sSql, sTARGET, GetTableName(wshTarget.QueryTables(1).CommandText), 1, 1)
'execute the query
Set adConn = New ADODB.Connection
adConn.Open Replace(wshSource.QueryTables(1).Connection, sODBC, "")
Set adRs = adConn.Execute(sSql)
'copy the results
wshResults.Range("A1").CopyFromRecordset adRs
adRs.Close
adConn.Close
Set adRs = Nothing
Set adConn = Nothing
End Sub
Function GetTableName(sSql As String) As String
Dim lFromStart As Long
Dim lFromEnd As Long
Dim sReturn As String
Const sFROM As String = "FROM "
Const sWHERE As String = "WHERE "
'find where FROM starts and ends
'I'm looking for WHERE as the end, but you'll need to look for everything possible, like ORDER BY etc.
lFromStart = InStr(1, sSql, sFROM)
lFromEnd = InStr(lFromStart, sSql, sWHERE)
If lFromEnd = 0 Then
sReturn = Mid$(sSql, lFromStart + Len(sFROM), Len(sSql))
Else
sReturn = Mid$(sSql, lFromStart + Len(sFROM), lFromEnd - lFromStart - Len(sFROM) - 1)
End If
GetTableName = sReturn
End Function
Another problem that you might run into is the way Excel (or MSQuery) constructs the SQL statements in an external data query. If you leave it as the default, you'll likely get something like this
SELECT * FROM `C:\somepath\myfile.mdb`.tblTable1 tblTable1 WHERE ...
I have no idea why it does it that way, but you can change it to
SELECT * FROM tblTable1 WHERE ...
and the above code should work. Parsing SQL statements sucks, so don't expect this to be easy. Once you think you have all the possibilities, another will pop up.
Finally, you should get the error "Too few parameters, expected 1" or something similar. In SourceTable, the first field is emp_no, but you have emp_id in your SQL. Make sure your SQL in the SQL sheet is correct. It can be frustrating trying to track down those errors.
I'm using very simple code which helps me to query worksheet range :
Sub hello_jet()
Dim rs As ADODB.Recordset
Dim cn As ADODB.Connection
Dim strQuery As String
Set cn = New ADODB.Connection
With cn
.Provider = "Microsoft.ACE.OLEDB.12.0"
.ConnectionString = "Data Source=C:\yourPath\ADO_test.xls " & _
";Extended Properties=""Excel 8.0;HDR=Yes;"""
.Open
End With
'Microsoft.ACE.OLEDB.12.0 for database engine built in Windows 7 64
strQuery = "SELECT a,sum(c) FROM [Sheet1$A1:C6] GROUP BY a;"
''if range [Sheet1$A1:C6] is named as namedRange you can you its name directly in query:
'strQuery = "SELECT a,sum(c) FROM namedRange GROUP BY a;"
Set rs = cn.Execute(strQuery)
ActiveCell.CopyFromRecordset rs 'useful method
rs.Close
End Sub