I am trying to find a quicker way to calculate my medians in Access. You can see the code below where it queries one item code at a time, sorts, and then calculates the median. Sometimes there are 600 item codes and those items can each have a 1000+ bases associated with it. The particular table I am working with has 150,000 total records for example and it is going really slow. Is there a better way to calculate each records median all at once as opposed to one item code at a time.
Function FIncPercentile(ByVal posCode As Single, ByVal k As Single, ByVal tbl As String) As Variant
Dim rstRec As Recordset
Dim db As Database
Dim n As Integer
Dim i As Single
Dim res, d1, d2 As Currency
' Create recordset from query
Set db = CurrentDb
Set rstRec = db.OpenRecordset("SELECT Co, Base " & _
"FROM " & tbl & " " & _
"WHERE Code = " & pos & " " & _
"ORDER BY Base ASC;")
' Skip if there are no matches
If IsNull(rstRec!base) Or rstRec.RecordCount = 0 Then
FBasePercentile = Null
Exit Function
End If
' Count records
rstRec.MoveLast
n = rstRec.RecordCount
rstRec.MoveFirst
' Calculate the index where k is the percentile
i = n * k
' Test the decimal and find value accordingly
If i = Int(i) Then
rstRec.Move i - 1
d1 = rstRec!base
rstRec.MoveNext
d2 = rstRec!base
FIncPercentile = (d1 + d2) / 2
Else
i = Round(i + 0.5, 0)
rstRec.Move i - 1
FIncPercentile = rstRec!base
End If
End Function
There is no Median function in Access. Excel has one but I believe it's limited to 30 numbers, so even if you wanted to try using an Automation function, I don't believe it would work for your case.
I think you may see a noticeable speed increase by fine tuning your function and by letting Microsoft's Jet Engine pre-compile your query.
Make sure you have the Base and Code fields indexed in your table(s)
Create a Parameter Query with Code having the criteria parameter
[What Code]
Optimize your function with Recordset WITH construct, declared
variables and matched field types (Code = Long Integer???)
Time it before and after all these changes and see if there's any noticeable difference
I corrected a couple typos that may not be typos - and I make an assumption that CODE is a long integer - which again I may be wrong. Also my changes are prefaced by '***************
CREATE Precompiled Parameter Query
Create New Query called "qdfPrepMedian"
Copy/Paste SQL >> PARAMETERS [What Code] Long; SELECT Co, Base FROM <YourTableName> WHERE Code = [What Code] ORDER BY Base ASC;
Save the query
Adjusted Function
Option Explicit
'***********************
' changed posCode to Long
'***********************
Function FIncPercentile(ByVal posCode As Long, ByVal k As Single, ByVal tbl As String) As Variant
'***********************
' CREATE/USE Precompiled Parameter Query
' Create New Query called "qdfPrepMedian"
' Copy/Paste SQL >> PARAMETERS [What Code] Long; SELECT Co, Base FROM <YourTableName> WHERE Code = [What Code] ORDER BY Base ASC;
Const QRY_BY_CODES As String = "qdfPrepMedian"
Dim qdf As QueryDef
'
'***********************
Dim rstRec As Recordset
Dim db As Database
Dim n As Integer
Dim i As Single
' Declare all Currency variables on separate lines
' Otherwise they will be variants
Dim res As Currency
Dim d1 As Currency
Dim d2 As Currency
Set db = CurrentDb
'***********************
' Create readonly recordset from querydef
Set qdf = db.QueryDefs(QRY_BY_CODES)
qdf.Parameters("What Code") = posCode ' matches LONG variable passed to function
Set rstRec = qdf.OpenRecordset(dbOpenSnapshot, dbReadOnly) ' Readonly is sometimes faster
'***********************
' Use WITH rstRec
With rstRec
' Skip if there are no matches
If IsNull(!base) Or .RecordCount = 0 Then
'*** Is this a type ***
' FBasePercentile = Null
' Should it BE
FIncPercentile = Null
Exit Function
End If
' Count records
.MoveLast
n = .RecordCount
.MoveFirst
' Calculate the index where k is the percentile
i = n * k
' Test the decimal and find value accordingly
If i = Int(i) Then
.Move i - 1
d1 = !base
.MoveNext
d2 = !base
FIncPercentile = (d1 + d2) / 2
Else
i = Round(i + 0.5, 0)
.Move i - 1
FIncPercentile = !base
End If
End With
End Function
Related
I have a SQL Server connected to an Access frontend. Some text fields are set to NVARCHAR(MAX) and are working fine when I use direct table access to them.
For a query I needeed a LEFT JOIN to get access to a WHERE clause on fields of the other table so I created something like this:
strSQL = "SELECT DISTINCT dbo_tbl_Changes.Change_Nr, dbo_tbl_Changes.Title, dbo_tbl_Changes.ChangeContent, dbo_tbl_Changes.Date
FROM dbo_tbl_Changes LEFT JOIN dbo_tbl_Parts ON dbo_tbl_Changes.Change_Nr = dbo_tbl_Parts.Change_Nr
WHERE dbo_tbl_Parts.PartType = 'bolt'
ORDER BY dbo_tbl_Changes.Date DESC "
Forms![frm_ChangeOverview].RecordSource = strSQL
Forms![frm_ChangeOverview].Requery
With this, the NVARCHAR(MAX) field ChangeContent is getting cut off and not showing all its content.
If I leave out the DISTINCT keyword, the NVARCHAR(MAX) is correctly working within the form in a RichText field.
But then I will get more results than I wawnt (doubled entries) as I need the DISTINCT keyword in that query to remove those doubled (or even more) results.
One example: The ChangeContent has 1441 characters and is truncated to 157 characters but only when I use the DISTINCT keyword.
I'm using the native SQL ODBC driver as no other can be installed in my environment.
What could be the issue?
This is a known issue and is because Access reads the field as a Long Text (memo) field. However, Access can only apply Distinct on 255 characters or less.
You may have to look up the full record afterwards, for example from this old example:
Public Function LookupMemo( _
ByVal strSource As String, _
ByVal strFieldID As String, _
ByVal strFieldMemo As String, _
ByRef lngID As Long, _
ByRef varMemo As Variant) _
As String
' Extracts without truncation to 32768 characters the
' content of a memo field in a query.
'
' Assumes proper wrapping of table/field names containing spaces
' like "[My field name]" and a single field unique numeric key.
'
' Typical usage (SQL):
'
' SELECT
' ID,
' LookupMemo("Table1", "ID", "MemoField", [ID], [MemoField]) AS FullMemo
' FROM
' Table1;
'
' 2003-12-29. Cactus Data ApS, CPH.
' Maximum length of string from memo field when retrieved in a query.
Const clngStrLen As Long = &H8000&
Dim strExpr As String
Dim strDomain As String
Dim strCriteria As String
Dim strMemo As String
Dim lngLen As Long
On Error GoTo Exit_LookupMemo
If Not IsNull(varMemo) Then
lngLen = Len(varMemo)
If lngLen < clngStrLen Then
' The memo field is not truncated.
strMemo = varMemo
ElseIf Len(strSource) > 0 And Len(strFieldID) > 0 And Len(strFieldMemo) > 0 Then
' The memo is probably truncated by the query.
' Lookup the full memo in strSource.
strExpr = strFieldMemo
strDomain = strSource
strCriteria = strFieldID & " = " & lngID & ""
strMemo = vbNullString & DLookup(strExpr, strDomain, strCriteria)
End If
Else
' Return empty string.
End If
LookupMemo = strMemo
Exit_LookupMemo:
Exit Function
Err_LookupMemo:
' Return empty string.
Resume Exit_LookupMemo
End Function
This my first attempt to pose a question to stackoverflow so please forgive if it is submitted incorrectly or is otherwise out-of-line
As a novice programmer of VBA, I am attempting to create and run a query to execute a User-created function to create a query with a Median() result.
I shamelessly plagiarized a Stack Overflow suggested answer for the VBA function.
how to calculate median in Access query using function in VBa
See below for my version.
The only changes to that function were to change the fieldname, the input query name and removal of the Optional function statements as it did not apply to my dataset.
The SQL view of the query "qryMedian" is:
SELECT tblNewOhioAllMedian.AOU2020, tblNewOhioAllMedian.number1, acbdMedian([Number1],"qryMedian") AS MedianResults
FROM tblNewOhioAllMedian
ORDER BY tblNewOhioAllMedian.AOU2020, tblNewOhioAllMedian.number1;
tblNewOhioAllMedian is the originating table
AOU2020 is defined as a double number
Number1 is an integer number
acbdMedian is the user-defined function
Both the Query and Function compiled correctly, but either the query or function is not setup right and/or the results are not being passed between the two.
Any help in resolving my issue would be much appreciated.
The function run was the Stack Overflow suggestion modified as above with my changes:
Public Function acbdMedian( _
ByVal Number1 As String, ByVal qryMedian As String) As Variant
' Purpose:
' To calculate the median value
' for a field in a table or query.
' In:
' Number1: The field
' qryMedian: The table or query
' strCriteria: An optional WHERE clause to
' apply to the table or query
' strCriteria value removed as not applicable
' Out:
' Return value: The median, if successful;
' otherwise, an error value
Dim db As DAO.Database
Dim rstDomain As DAO.Recordset
Dim strSQL As String
Dim varMedian As Variant
Dim intFieldType As Integer
Dim intRecords As Integer
Const acbcErrAppTypeError = 3169
On Error GoTo HandleErr
Set db = CurrentDb()
' Initialize the return value.
varMedian = Null
' Build a SQL string for the recordset.
strSQL = "SELECT " & Number1
strSQL = strSQL & " FROM " & qryMedian
' Use a WHERE clause only if one is passed in.
'If Len(strCriteria) > 0 Then
'strSQL = strSQL & " WHERE " & strCriteria
'End If
strSQL = strSQL & " ORDER BY " & Number1
Set rstDomain = db.OpenRecordset(strSQL, dbOpenSnapshot)
' Check the data type of the median field.
intFieldType = rstDomain.Fields(Number1).Type
Select Case intFieldType
Case dbByte, dbInteger, dbLong, dbCurrency, dbSingle, dbDouble, dbDate
' Numeric field.
If Not rstDomain.EOF Then
rstDomain.MoveLast
intRecords = rstDomain.RecordCount
' Start from the first record.
rstDomain.MoveFirst
If (intRecords Mod 2) = 0 Then
' Even number of records. No middle record, so move
' to the record right before the middle.
rstDomain.Move ((intRecords \ 2) - 1)
varMedian = rstDomain.Fields(Number1)
' Now move to the next record, the one right after
' the middle.
rstDomain.MoveNext
' Average the two values.
varMedian = (varMedian + rstDomain.Fields(Number1)) / 2
' Make sure you return a date, even when averaging
' two dates.
If intFieldType = dbDate And Not IsNull(varMedian) Then
varMedian = CDate(varMedian)
End If
Else
' Odd number of records. Move to the middle record
' and return its value.
rstDomain.Move ((intRecords \ 2))
varMedian = rstDomain.Fields(Number1)
End If
Else
' No records; return Null.
varMedian = Null
End If
Case Else
' Nonnumeric field; raise an app error.
Err.Raise acbcErrAppTypeError
End Select
acbdMedian = varMedian
ExitHere:
On Error Resume Next
rstDomain.Close
Set rstDomain = Nothing
Exit Function
HandleErr:
' Return an error value.
acbdMedian = CVErr(Err)
Resume ExitHere
End Function
Results of running the Select Query:
AOU 2020 number1 MedianResults
10 1 #Error
10 1 #Error
10 1 #Error
10 1 #Error
ETC
This query ran very slow and would have taken hours
To complete the 240,000 records at that rate
My interpretation of the results:
Each entry is a separate record with the results of the function being #Error
No Medians were calculated.
Here is the latest test table:
AOU2020 Number1
1 1
2 1
20 2
10 2
1 2
2 2
20 3
20 3
10 3
10 3
1 3
2 3
10 50
10 50
10 50
10 50
10 50
10 50
10 50
10 50
10 50
10 50
20 60
10 60
20 100
10 100
You have field in table named Number1? Fieldname must be within quote marks in function argument.
Cannot pass same query name that function is called in.
If you want a median for each AOU2020 group, then need to pass filter criteria to function. Modify function declaration:
Public Function acbdMedian( _
ByVal Number1 As String, ByVal qryMedian As String, Optional ByVal strCriteria As String) As Variant
Uncomment 3 lines concerning criteria by removing the leading apostrophe:
If Len(strCriteria) > 0 Then
strSQL = strSQL & " WHERE " & strCriteria
End If
To return one record for each AOU2020, use aggregate query.
qryMedian:
SELECT AOU2020, acbdMedian("Number1", "tblNewOhioAllMedian", "AOU2020=" & [AOU2020]) AS MedianResults
FROM tblNewOhioAllMedian
GROUP BY AOU2020;
Results with given sample data:
AOU2020 MR
1 2
2 2
10 50
20 3
Note that because function is declared as Variant type, median value will default to string and left align.
I am working on a script which selects only the needed slicer items. I tried using .SlicerItems.Selected = True / False for selecting and deselecting but I am using an OLAP data source in which case .Selected is read-only. The slicer items are in the format of YYYYWW so 7th week of 2018 would be 201807.
I recorded a macro selecting some slicer items and this is what it gave me:
Sub Macro2()
ActiveWorkbook.SlicerCaches("Slicer_YYYYWW").VisibleSlicerItemsList = Array( _
"[Results].[YYYYWW].&[201726]", "[Results].[YYYYWW].&[201727]", _
"[Results].[YYYYWW].&[201728]", "[Results].[YYYYWW].&[201729]", _
"[Results].[YYYYWW].&[201730]", "[Results].[YYYYWW].&[201731]", _
"[Results].[YYYYWW].&[201732]", "[Results].[YYYYWW].&[201733]", _
"[Results].[YYYYWW].&[201734]", "[Results].[YYYYWW].&[201735]", _
"[Results].[YYYYWW].&[201736]", "[Results].[YYYYWW].&[201737]", _
"[Results].[YYYYWW].&[201738]", "[Results].[YYYYWW].&[201739]", _
"[Results].[YYYYWW].&[201740]", "[Results].[YYYYWW].&[201741]", _
"[Results].[YYYYWW].&[201742]", "[Results].[YYYYWW].&[201743]", _
"[Results].[YYYYWW].&[201744]", "[Results].[YYYYWW].&[201745]", _
"[Results].[YYYYWW].&[201746]", "[Results].[YYYYWW].&[201747]", _
"[Results].[YYYYWW].&[201748]", "[Results].[YYYYWW].&[201749]", _
"[Results].[YYYYWW].&[201750]", "[Results].[YYYYWW].&[201751]", _
"[Results].[YYYYWW].&[201801]", "[Results].[YYYYWW].&[201802]", _
"[Results].[YYYYWW].&[201803]")
End Sub
So I tried following this template and create an array like that. This is how far I have gotten:
Sub arrayTest()
Dim startDate As Long
Dim endDate As Long
Dim n As Long
Dim i As Long
Dim strN As String
Dim sl As SlicerItem
Dim strArr As Variant
Dim dur As Long
Dim result As String
endDate = Range("C17").Value ' endDate is the last SlicerItem to be selected
startDate = Range("G17").Value ' startDate is the first SlicerItem to be selected
dur = Range("C19").Value ' duration is the the number of SlicerItems to be selected
i = 0
ReDim strArr(dur) As Variant
With ActiveWorkbook.SlicerCaches("Slicer_YYYYWW")
' .ClearManualFilter
For n = startDate To endDate
strN = CStr(n) ' convert n to string
If n = 201753 Then ' this is needed for when the year changes
strN = CStr(201801)
n = 201801
End If
strArr(i) = """[Results].[YYYYWW].&[" & strN & "]""" ' write string into array
i = i + 1
' For Each sl In .SlicerCacheLevels(1).SlicerItems
' If sl.Name = strN Then
' sl.Selected = True
' Else
' sl.Selected = False ' this is read-only for OLAP data so it's not working
' End If
' Next
Next
MsgBox Join(strArr, ", ") ' the MsgBox returns the correct string to be applied to select the right slicer items
.VisibleSlicerItemsList = Join(strArr, ", ") ' Error 13: Type mismatch
End With
End Sub
Currently, the code gives Error 13: Type mismatch on .VisibleSlicerItemsList = Join(strArr, ", "), which is also commented. So I'm guessing that either dimensioning strArr as Variant is wrong, the data is not inserted correctly into strArr or it's just impossible to do it this way. In the case of the latest one, how should I do it?
The part commented out on lines 29-35 does not work as it gives the usual error of Application-defined or object-defined error (1004) on sl.Selected = False.
I had a similar issue to overcome. Which I resolved using the following code:
Sub show_SlicerItems()
Dim sc As SlicerCache
Dim sL As SlicerCacheLevel
Dim si As SlicerItem
Dim slicerItems_Array()
Dim i As Long
Application.ScreenUpdating = False
Set sc = ActiveWorkbook.SlicerCaches("Slicer_Name")
Set sL = sc.SlicerCacheLevels(1)
ActiveWorkbook.SlicerCaches("Slicer_Name").ClearManualFilter
i = 0
For Each si In sL.SlicerItems
ReDim Preserve slicerItems_Array(i)
If si.Value <> 0 Then
slicerItems_Array(i) = si.Name
i = i + 1
End If
Next
sc.VisibleSlicerItemsList = Array(slicerItems_Array)
Application.ScreenUpdating = True
End Sub
You need to feed .VisibleSlicerItemsList an array, not a string. Ditch the Join.
And your strArr assignment should be like this: strArr(i) = "[Results].[YYYYWW].&[" & strN & "]" i.e. you don't need to pad it out with extra "
Edit: Out of interest, I happen to be building a commercial add-in that is effectively a Pop-up Slicer, that allows you to filter an OLAP PivotTable to show all items between a range like you are attempting to do. It also lets you filter on wildcards, crazy combinations of AND and OR, and filter on lists stored in external ranges.
Here's a screenshot of it in action. Note there is a search bar up the top that lets you use < or > together to set lower and upper limits, which is what I've done in the current Search. And you can see the result: it has correctly identified the 14 items from the PivotField that fit the bill.
All I need to do to filter the PivotTable on these is click the "Filter on selected items" option, and it does just that:
But working out how to do this - particularly given the limitations of the PivotTable object model (especially where OLAP PivotTables are concerned) was a VERY long term project, with many, many hurdles to overcome to make it work seamlessly. I can't share the code I'm afraid, as this is a commercial offering that I aim to release shortly. But I just wanted to highlight that while this is certainly possible, you are going to be biting off quite a bit if you want it to not throw errors when items don't exist.
Forget my other answer...you can use the Labels Filter to do this easily, provided the field of interest is in the PivotTable as either a Rows or Columns field. Fire up the Macro Recorder, and do the following:
...and you'll see that the PivotTable gets filtered:
...and the resulting code is pretty simple:
ActiveSheet.PivotTables("PivotTable1").PivotFields("[Table1].[YYYYWW].[YYYYWW]" _
).PivotFilters.Add2 Type:=xlCaptionIsBetween, Value1:="201726", Value2:= _
"201803"
Use this:
Sub seleciona_lojas()
Dim strArr()
Dim x As Long
Dim i As Long
For x = 2 To 262
ReDim Preserve strArr(i)
strArr(i) = "[Lojas].[Location_Cd].&[" & Planilha5.Range("B" & x).Value & "]"
i = i + 1
Next x
ActiveWorkbook.SlicerCaches("SegmentaçãodeDados_Location_Cd1").VisibleSlicerItemsList = strArr
End Sub
My aim is to add the values of certain columns using a user defined function in the actual row. The columns are given in another table. I am reading the name of rows, calculating the actual value and sum them. This function called once from excel but executed 4 times. At the end it indicates a circular reference error. There is no such error in the excel file, I checked if the udf return just 42 then there is no error. First I suspected Application.Caller, but ruled out.
Function SumColumnsWithSuffix(suffix As String, rowNumber) As Integer
'can be used only in Összesíto table
Dim myTable As Excel.ListObject
Dim mySheet As Excel.Worksheet
Dim myRow As Excel.ListRow
Set mySheet = ThisWorkbook.Worksheets("összesíto")
Set myTable = mySheet.ListObjects("Számlák")
Dim columnName As String
result = 0
For Each myRow In myTable.ListRows
columnName = Intersect(myRow.Range, myTable.ListColumns("Oszlop név").Range)
columnName = "Összesíto[" & columnName & " " & suffix & "]"
'actualRow = Application.Caller.row
'rowName = actualRow & ":" & actualRow
rowName = rowNumber & ":" & rowNumber
myRowRange = Range(rowName)
actualValue = Intersect(Range(columnName), Range(rowName))
result = result + actualValue
Next myRow
SumColumnsWithSuffix = result
End Function
myRowRange is not explicitly declared (or used, actually) so it is implicitly a Variant. That means your assignment here...
myRowRange = Range(rowName)
...is also making an implicit call to .Value. That call will evaluate the results of every single cell in Range(rowName) to populate the array of Variant that it returns. If any of those cells contains a call to SumColumnsWithSuffix, you'll get a circular reference.
Can you please let me know if it is possible to list all fields name in a MS Access table?
I work in ms access far too much.
The only way I know of to do this, would be using vba, and defining for example a recordset, and looping through the fields.
Eg:
Sub ListFields()
dim rst as new adodb.recordset
rst.open "SELECT * FROM SomeTable", CurrentProject.Connection, adOpenForwardOnly, adLockReadOnly
' Note: adOpenForwardOnly and adLockReadOnly are the default values '
' for the CursorType and LockType arguments, so they are optional here '
' and are shown only for completeness '
dim ii as integer
dim ss as string
for ii = 0 to rst.fields.count - 1
ss = ss & "," & rst.fields(ii).name
next ii
Debug.Print ss
End Sub
The string variable ss will contain a comma-delimited list of all the column names in the table named "SomeTable".
With a little reformatting of the logic you should be able to insert this data into another table if you wanted to, then query it out.
Does this help?
This version is easy to run and will paste right into Access. Add this function to a module, run with F5, and copy the result from the inputbox:
Public Function FieldNames() As String
Dim sTable As String
Dim rs As DAO.Recordset
Dim n As Long
Dim sResult As String
sTable = InputBox("Name of table?")
If sTable = "" Then
Exit Function
End If
Set rs = CurrentDb.OpenRecordset(sTable)
With rs
For n = 0 To .Fields.Count - 1
sResult = sResult & .Fields(n).Name & vbCrLf
Next 'n
.Close
End With
Set rs = Nothing
InputBox "Result:" & vbCrLf & vbCrLf _
& "Copy this text (it looks jumbled, but it has one field on each line)", _
"FieldNames()", sResult
End Function
Alternative Output:
User user1003916 supplied an alternative to the InputBox to overcome the 1024 character limit (I have not tested this yet):
Sub CopyText(Text As String)
'VBA Macro using late binding to copy text to clipboard.
'By Justin Kay, 8/15/2014
Dim MSForms_DataObject As Object
Set MSForms_DataObject = CreateObject("new:{1C3B4210-F441-11CE-B9EA-00AA006B1A69}")
MSForms_DataObject.SetText Text
MSForms_DataObject.PutInClipboard
Set MSForms_DataObject = Nothing
End Sub
UPDATE: TO USE THIS SQL QUERY YOU MUST USE A TOOL SUCH AS DBEAVER.
ACCESS CLIENT WILL NOT ALLOW YOU TO QUERY IT'S HIDDEN STRUCTURES.
YIKES! IMO: I can't imagine wanting to dive into the dark underbelly of VBA
How to get Access Table Columns by SQL
SELECT * FROM information_schema.columns
WHERE TABLE_NAME="YOUR_TABLE_NAME"
AND
TABLE_SCHEMA="PUBLIC"
PS I noticed Access called my Schema "PUBLIC"
Above used an Access 2016 and was tested over ODBC and jdbc:ucanaccess and works like a charm.
Example output
You can simply use the Documenter tool. Go to Database Tools > Database Documenter, select the table and press OK.
This SQL works in Access 2016 for Queries, not Tables, but might be useful.
SELECT MSysObjects.Name AS QueryName,
IIf(IsNull([Name1]),
Right([Expression],Len([Expression])-InStr(1,[Expression],".")),[name1])
AS FieldName
FROM MSysQueries INNER JOIN MSysObjects
ON MSysQueries.ObjectId = MSysObjects.Id
WHERE MSysQueries.Attribute=6;
A quick and dirty method involves Excel. Do the following:
Open the table in Datasheet view.
Export to an Excel file, using the particular procedure for your Access version.
Open Excel file (if not already open).
Select and copy the first row containing the headers.
Add another sheet to the workbook (if none exist).
Click A1.
Paste Special > Transpose
The Fields will be pasted in a single column. To find out their Field Index number, in Cell B1 type "0", then series fill down to the last row of the field numbers.
In addition, you can sort the column alphabetically, especially for legacy flat files involving dozens of fields. This really saves a lot of time when I'm trying to convert a flatfile to a relational model.
There are already some good answers but I decided to add my own twist. Hopefully, they are self-explanatory.
Usage:
getFieldNames(TableName:="Table1",IncludeBrackets:=True,Delimiter:=vbNewLine,CopyToClipboard:=True)
getFieldNames(TableName:="Table1",IncludeBrackets:=True,CopyToClipboard:=True)
getFieldNames(TableName:="Table1",IncludeBrackets:=True)
getFieldNames(TableName:="Table1")
I use this to build an array of field names:
Chr(34) & getFieldNames(TableName:="Table1",IncludeBrackets:=False, Delimiter:= Chr(34) & "," & Chr(34)) & Chr(34)
Function getFieldNames(ByVal TableName As String, Optional ByVal IncludeBrackets As Boolean, Optional ByVal Delimiter As String = ", ", Optional ByVal CopyToClipboard As Boolean) As String
Dim rs As DAO.Recordset
On Error Resume Next
Set rs = CurrentDb.OpenRecordset(TableName)
On Error GoTo 0
If rs Is Nothing Then Exit Function
Dim results() As String
ReDim results(rs.Fields.Count - 1)
Dim n As Long
For n = 0 To rs.Fields.Count - 1
results(n) = rs.Fields(n).Name
Next
rs.Close
Dim result As String
If IncludeBrackets Then
result = "[" & Join(results, "]" & Delimiter & "[") & "]"
Else
result = Join(results, Delimiter)
End If
If CopyToClipboard Then
With CreateObject("new:{1C3B4210-F441-11CE-B9EA-00AA006B1A69}")
.SetText result
.PutInClipboard
End With
End If
getFieldNames = result
End Function
This is not SQL, but this question is the top Google result for someone like me who just needs to list out the field names needed for a query to select since Access does not support "* except foo, bar" for getting 99% of a table.
Answer adapted from a social.msdn.com answer by Patrick Wood, Access MVP
https://social.msdn.microsoft.com/Forums/office/en-US/1fe5546b-db3f-4e17-9bf8-04f4dee233b7/how-to-list-all-the-field-names-in-a-specified-table?forum=accessdev
Change tablename to your name in the module. This Function should be at the global level:
Function ListTdfFields()
' NOT doing DIMs, since then you must enable/attach ADODB or DAO
' Dim db As ADO.Database
Set db = CurrentDb
tablename = "tblProductLicense" ' <=== YOUR TABLE NAME HERE
Set tdf = db.TableDefs(tablename)
For Each fld In tdf.Fields
Debug.Print tablename; ".["; fld.Name; "]," ;
' remove ending ; for 1 line per field
Next
Debug.Print ""
Set tdf = Nothing
Set db = Nothing
End Function
Then add a macro RunCode ListTdfFields() and run it. Output will be sent to the Immediate window of the VBA design view for the module.
I would like to share this VBA solution, which I did not write, only slightly modified (changed the tableName to use 'SourceTable'). Once it is run you can query the table it creates. It takes advantage of hidden system tables.
Sub GetField2Description()
'************************************************* *********
'Purpose: 1) Deletes and recreates a table (tblFields)
' 2) Queries table MSysObjects to return names of
' all tables in the database
' 3) Populates tblFields
'Coded by: raskew
'Inputs: From debug window:
' Call GetField2Description
'Output: See tblFields
'************************************************* *********
Dim db As DAO.Database, td As TableDef
Dim rs As Recordset, rs2 As Recordset
Dim Test As String, NameHold As String
Dim typehold As String, SizeHold As String
Dim fielddescription As String, tName As String
Dim n As Long, i As Long
Dim fld As Field, strSQL As String
n = 0
Set db = CurrentDb
' Trap for any errors.
On Error Resume Next
tName = "tblFields"
'Does table "tblFields" exist? If true, delete it;
DoCmd.SetWarnings False
DoCmd.DeleteObject acTable, "tblFields"
DoCmd.SetWarnings True
'End If
'Create new tblTable
db.Execute "CREATE TABLE tblFields(Object TEXT (55), FieldName TEXT (55), FieldType TEXT (20), FieldSize Long, FieldAttributes Long, FldDescription TEXT (20));"
strSQL = "SELECT MSysObjects.Name, MSysObjects.Type From MsysObjects WHERE"
strSQL = strSQL + "((MSysObjects.Type)=1)"
strSQL = strSQL + "ORDER BY MSysObjects.Name;"
Set rs = db.OpenRecordset(strSQL)
If Not rs.BOF Then
' Get number of records in recordset
rs.MoveLast
n = rs.RecordCount
rs.MoveFirst
End If
Set rs2 = db.OpenRecordset("tblFields")
For i = 0 To n - 1
fielddescription = " "
Set td = db.TableDefs(i)
'Skip over any MSys objects
If Left(rs!Name, 4) <> "MSys" And Left(rs!Name, 1) <> "~" Then
NameHold = rs!Name
On Error Resume Next
For Each fld In td.Fields
tableName = fld.SourceTable
fielddescription = fld.Name
typehold = FieldType(fld.Type)
SizeHold = fld.Size
rs2.AddNew
rs2!Object = tableName
rs2!FieldName = fielddescription
rs2!FieldType = typehold
rs2!FieldSize = SizeHold
rs2!FieldAttributes = fld.Attributes
rs2!FldDescription = fld.Properties("description")
rs2.Update
Next fld
Resume Next
End If
rs.MoveNext
Next i
rs.Close
rs2.Close
db.Close
End Sub
Function FieldType(intType As Integer) As String
Select Case intType
Case dbBoolean
FieldType = "dbBoolean" '1
Case dbByte
FieldType = "dbByte" '2
Case dbInteger
FieldType = "dbInteger" '3
Case dbLong
FieldType = "dbLong" '4
Case dbCurrency
FieldType = "dbCurrency" '5
Case dbSingle
FieldType = "dbSingle" '6
Case dbDouble
FieldType = "dbDouble" '7
Case dbDate
FieldType = "dbDate" '8
Case dbBinary
FieldType = "dbBinary" '9
Case dbText
FieldType = "dbText" '10
Case dbLongBinary
FieldType = "dbLongBinary" '11
Case dbMemo
FieldType = "dbMemo" '12
Case dbGUID
FieldType = "dbGUID" '15
End Select
End Function
Build query:
SELECT Table_Name.*
FROM Table_Name
WHERE False;
Export to Excel
You will have each field name in one row without any data. If you select the row and copy, you can paste special>transpose and have them all in a single column.
I came here searching for the same requirement and after refering to this thread, drafted below code for my requirement. The Field names form the source table will be added to an array and later the Field names are assigned to the second table. Just sharing it here, this migh help someone later..
Public Sub FieldName_Change()
Dim intNumberOfFields, intX As Integer
Dim txtTableName,txttmpTableName txtFieldName() As String
intNumberOfFields = GetFieldNames(txtTableName, txtFieldName())
For intX = 1 To intNumberOfFields
CurrentDb.TableDefs(txttmpTableName).Fields("F" & intX).Name = txtFieldName(intX)
Next intX
End Sub
Public Function GetFieldNames(ByVal txtTableName As String, ByRef txtFiledName() As String) As Integer
Dim rs As DAO.Recordset
Dim n As Long
Dim sResult As String
Set rs = CurrentDb.OpenRecordset(txtTableName)
ReDim txtFiledName(rs.Fields.Count)
With rs
For n = 0 To .Fields.Count - 1
txtFiledName(n + 1) = .Fields(n).Name
Next n
.Close
GetFieldNames = n
End With
Set rs = Nothing
End Function
Not sure why no-one came up with the following, I had the same issue and none of the above answers either worked or were suitable (No VB,thank you).
This is simple Java (error checking etc omitted)
String query="SELECT TOP 1 * FROM mytable";
ResultSet rs=connectionSQL.createStatement().executeQuery(query);
ResultSetMetaData rsm=rs.getMetaData();
int colcount=rsm.getColumnCount();
for(int f=1; f<=colcount; f++) {
String coltype=rsm.getColumnTypeName(f);
String colname=rsm.getColumnName(f);
.... (whatever)
}
So there's your columNames and the data types.
Works like a charm in MSACCESS
Seems like this task was easier in older days. Likely this answer is highly version-dependent. It works for me in a quick test against an Access 2007 DB:
select
Specs.SpecName AS TableName,
Columns.FieldName
from
MSysIMEXColumns Columns
inner join MSysIMEXSpecs Specs on Specs.SpecID = Columns.SpecID
order by
Specs.SpecName,
Columns.FieldName
Give this a go...
private void Button_OldDDLDump_Click(object sender, EventArgs e)
{
string _cnstr = "connectionstringhere";
OleDbConnection _cn = new OleDbConnection(_cnstr);
try
{
_cn.Open();
System.Data.DataTable _dt = null;
try
{
_dt = _cn.GetSchema("tables");
m_msghelper.AppendArray( DataTableHelper.DataTableToString(_dt) );
}
catch (Exception _ex)
{
System.Diagnostics.Debug.WriteLine(_ex.ToString());
}
finally
{
_dt.Dispose();
}
}
catch (Exception _ex)
{
System.Diagnostics.Debug.WriteLine(_ex.ToString());
}
finally
{
_cn.Close();
}
}
Helper method to dump the database structure to a string array here..
public static class DataTableHelper
{
public static string[] DataTableToString( System.Data.DataTable dt )
{
List<string> _retval = new List<string>();
foreach (System.Data.DataRow row in dt.Rows)
{
foreach (System.Data.DataColumn col in dt.Columns)
{
_retval.Add( string.Format("{0} = {1}", col.ColumnName, row[col]) );
}
_retval.Add( "============================");
}
return _retval.ToArray();
}
}
select column_name from information_schema.columns where table_name='table'