Comment analyser un fichier XML à l'aide de vba

Question

Je travaille en VBA et je souhaite analyser une chaîne de caractères, par exemple

<PointN xsi:type='typens:PointN' 
xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' 
xmlns:xs='http://www.w3.org/2001/XMLSchema'>
    <X>24.365</X>
    <Y>78.63</Y>
</PointN>

et obtenir les valeurs X et Y dans deux variables entières distinctes.

Je suis un novice en matière de XML, car je suis coincé dans VB6 et VBA, en raison du domaine dans lequel je travaille.

Comment faire ?

Answer 1

Merci pour ces conseils.

Je ne sais pas si c'est la meilleure approche du problème ou non, mais voici comment j'ai réussi à le faire fonctionner. J'ai référencé la dll Microsoft XML, v2.6 dans mon VBA, et l'extrait de code suivant me donne les valeurs requises

Dim objXML As MSXML2.DOMDocument

Set objXML = New MSXML2.DOMDocument

If Not objXML.loadXML(strXML) Then  'strXML is the string with XML'
    Err.Raise objXML.parseError.ErrorCode, , objXML.parseError.reason
End If

Dim point As IXMLDOMNode
Set point = objXML.firstChild

Debug.Print point.selectSingleNode("X").Text
Debug.Print point.selectSingleNode("Y").Text

Answer 2

C'est une question un peu compliquée, mais il semble que la voie la plus directe soit de charger le document XML ou la chaîne XML via MSXML2.DOMDocument, ce qui vous permettra ensuite d'accéder aux nœuds XML.

Vous pouvez trouver plus d'informations sur MSXML2.DOMDocument sur les sites suivants :

• Manipuler des fichiers XML avec Excel VBA & Xpath
• MSXML - http://msdn.microsoft.com/en-us/library/ms763742(VS.85).aspx
• Aperçu de MSXML 4.0

Answer 3

Il s'agit d'un exemple d'analyseur OPML fonctionnant avec les fichiers opml de FeedDemon :

Sub debugPrintOPML()

' http://msdn.microsoft.com/en-us/library/ms763720(v=VS.85).aspx
' http://msdn.microsoft.com/en-us/library/system.xml.xmlnode.selectnodes.aspx
' http://msdn.microsoft.com/en-us/library/ms256086(v=VS.85).aspx ' expressions
' References: Microsoft XML

Dim xmldoc As New DOMDocument60
Dim oNodeList As IXMLDOMSelection
Dim oNodeList2 As IXMLDOMSelection
Dim curNode As IXMLDOMNode
Dim n As Long, n2 As Long, x As Long

Dim strXPathQuery As String
Dim attrLength As Byte
Dim FilePath As String

FilePath = "rss.opml"

xmldoc.Load CurrentProject.Path & "\" & FilePath

strXPathQuery = "opml/body/outline"
Set oNodeList = xmldoc.selectNodes(strXPathQuery)

For n = 0 To (oNodeList.length - 1)
    Set curNode = oNodeList.Item(n)
    attrLength = curNode.Attributes.length
    If attrLength > 1 Then ' or 2 or 3
        Call processNode(curNode)
    Else
        Call processNode(curNode)
        strXPathQuery = "opml/body/outline[position() = " & n + 1 & "]/outline"
        Set oNodeList2 = xmldoc.selectNodes(strXPathQuery)
        For n2 = 0 To (oNodeList2.length - 1)
            Set curNode = oNodeList2.Item(n2)
            Call processNode(curNode)
        Next
    End If
        Debug.Print "----------------------"
Next

Set xmldoc = Nothing

End Sub

Sub processNode(curNode As IXMLDOMNode)

Dim sAttrName As String
Dim sAttrValue As String
Dim attrLength As Byte
Dim x As Long

attrLength = curNode.Attributes.length

For x = 0 To (attrLength - 1)
    sAttrName = curNode.Attributes.Item(x).nodeName
    sAttrValue = curNode.Attributes.Item(x).nodeValue
    Debug.Print sAttrName & " = " & sAttrValue
Next
    Debug.Print "-----------"

End Sub

Celui-ci prend des arbres de dossiers à plusieurs niveaux (Awasu, NewzCrawler) :

...
Call xmldocOpen4
Call debugPrintOPML4(Null)
...

Dim sText4 As String

Sub debugPrintOPML4(strXPathQuery As Variant)

Dim xmldoc4 As New DOMDocument60
'Dim xmldoc4 As New MSXML2.DOMDocument60 ' ?
Dim oNodeList As IXMLDOMSelection
Dim curNode As IXMLDOMNode
Dim n4 As Long

If IsNull(strXPathQuery) Then strXPathQuery = "opml/body/outline"

' http://msdn.microsoft.com/en-us/library/ms754585(v=VS.85).aspx
xmldoc4.async = False
xmldoc4.loadXML sText4
If (xmldoc4.parseError.errorCode <> 0) Then
   Dim myErr
   Set myErr = xmldoc4.parseError
   MsgBox ("You have error " & myErr.reason)
Else
'   MsgBox xmldoc4.xml
End If

Set oNodeList = xmldoc4.selectNodes(strXPathQuery)

For n4 = 0 To (oNodeList.length - 1)
    Set curNode = oNodeList.Item(n4)
    Call processNode4(strXPathQuery, curNode, n4)
Next

Set xmldoc4 = Nothing

End Sub

Sub processNode4(strXPathQuery As Variant, curNode As IXMLDOMNode, n4 As Long)

Dim sAttrName As String
Dim sAttrValue As String
Dim x As Long

For x = 0 To (curNode.Attributes.length - 1)
    sAttrName = curNode.Attributes.Item(x).nodeName
    sAttrValue = curNode.Attributes.Item(x).nodeValue
    'If sAttrName = "text"
    Debug.Print strXPathQuery & " :: " & sAttrName & " = " & sAttrValue
    'End If
Next
    Debug.Print ""

If curNode.childNodes.length > 0 Then
    Call debugPrintOPML4(strXPathQuery & "[position() = " & n4 + 1 & "]/" & curNode.nodeName)
End If

End Sub

Sub xmldocOpen4()

Dim oFSO As New FileSystemObject ' Microsoft Scripting Runtime Reference
Dim oFS
Dim FilePath As String

FilePath = "rss_awasu.opml"
Set oFS = oFSO.OpenTextFile(CurrentProject.Path & "\" & FilePath)
sText4 = oFS.ReadAll
oFS.Close

End Sub

ou mieux :

Sub xmldocOpen4()

Dim FilePath As String

FilePath = "rss.opml"

' function ConvertUTF8File(sUTF8File):
' http://www.vbmonster.com/Uwe/Forum.aspx/vb/24947/How-to-read-UTF-8-chars-using-VBA
' loading and conversion from Utf-8 to UTF
sText8 = ConvertUTF8File(CurrentProject.Path & "\" & FilePath)

End Sub

mais je ne comprends pas pourquoi xmldoc4 doit être chargé à chaque fois.

Answer 4

Mise à jour

La procédure présentée ci-dessous donne un exemple d'analyse syntaxique de XML avec VBA en utilisant les objets XML DOM. Le code est basé sur un Guide du DOM XML pour les débutants .

Public Sub LoadDocument()
    Dim xDoc As MSXML.DOMDocument
    Set xDoc = New MSXML.DOMDocument
    xDoc.validateOnParse = False
    If xDoc.Load("C:\My Documents\sample.xml") Then
        ' The document loaded successfully.
        ' Now do something intersting.
        DisplayNode xDoc.childNodes, 0
    Else
        ' The document failed to load.
        ' See the previous listing for error information.
    End If
End Sub

Public Sub DisplayNode(ByRef Nodes As MSXML.IXMLDOMNodeList, _
   ByVal Indent As Integer)

   Dim xNode As MSXML.IXMLDOMNode
   Indent = Indent + 2

   For Each xNode In Nodes
      If xNode.nodeType = NODE_TEXT Then
         Debug.Print Space$(Indent) & xNode.parentNode.nodeName & _
            ":" & xNode.nodeValue
      End If

      If xNode.hasChildNodes Then
         DisplayNode xNode.childNodes, Indent
      End If
   Next xNode
End Sub

Nota Bene - Cette première réponse montre la chose la plus simple que j'ai pu imaginer (à l'époque, je travaillais sur un problème très spécifique). Naturellement, l'utilisation des facilités XML intégrées dans le Dom XML de VBA serait beaucoup mieux. Voir les mises à jour ci-dessus.

Réponse originale

Je sais qu'il s'agit d'un très vieil article, mais je voulais partager ma solution simple à cette question compliquée. J'ai principalement utilisé des fonctions de chaînes de caractères basiques pour accéder aux données xml.

Cela suppose que vous avez des données xml (dans la variable temp) qui ont été renvoyées par une fonction VBA. Il est intéressant de noter que l'on peut également voir comment je me connecte à un service web xml pour récupérer la valeur. La fonction montrée dans l'image prend également une valeur de référence car cette fonction VBA Excel est accessible à partir d'une cellule en utilisant = NomFonction(valeur1, valeur2) pour renvoyer des valeurs via le service web dans une feuille de calcul.

openTag = ""
closeTag = "" 
' Locate the position of the enclosing tags
startPos = InStr(1, temp, openTag)
endPos = InStr(1, temp, closeTag)
startTagPos = InStr(startPos, temp, ">") + 1
' Parse xml for returned value
Data = Mid(temp, startTagPos, endPos - startTagPos)

Answer 5

Code d'analyse XML

Option Explicit
    Dim Path As String ' input path name
    Dim FileName As String ' input file name
    Dim intColumnCount As Integer ' column counter
    Dim intLoop As Integer ' Looping integer
    Dim objDictionary As Scripting.Dictionary ' dictionary object to store column identification for id, method, query string etc
    Dim intPrevRequest_id As Integer 'stores previous request id
    Dim intCurrRequest_id As Integer 'stores current request id

    Dim strWholeReq As String ' Full request that is ready to be written to file
    Dim strStartQuotes As String ' Placeholder which holds starting double quotes
    Dim strEndQuotes As String ' Placeholder which holds ending double quotes
    Dim strStepName As String ' First line of the Parsed_XML_Function. e.g. Parsed_XML_Function("Step5",
    'Here 5 comes from intStepNum variable

    Dim strUrl As String ' contains URL and Query string
    Dim strQueryStr As String ' Query string
    Dim strMethod As String ' Method part of request
    Dim strBody As String 'Body attributes
    Dim strMisc As String ' Misc items such as Resource, Snapshot number etc
    Dim strContentType As String ' Content type of request
    Dim intStepNum As Integer ' iterative count to identify step
    Dim objFileSys As Scripting.FileSystemObject ' file system object
    Dim objFile As Scripting.File 'file object
    Dim objTextStr As Scripting.TextStream 'text stream object
    Dim ActionFileName As String ' destination action name
'this funciton is the main function which calls other functions
Sub Main()

    Path = Worksheets(1).Cells(1, 2).Value
    FileName = Worksheets(1).Cells(2, 2).Value
    ActionFileName = Worksheets(1).Cells(3, 2).Value
    'open xml file
    Workbooks.Open FileName:=Path & "\" & FileName
    'activate the workbook
    Windows(FileName).Activate

    'delete first row
    Rows("1:1").Select
    Selection.Delete Shift:=xlUp
    Range(Selection, Selection.End(xlToRight)).Select
    ActiveSheet.Name = "PARSINGVS_XML"

    'get total columns and analyze the columns
    intColumnCount = Worksheets("PARSINGVS_XML").UsedRange.Columns.Count
    Set objDictionary = New Dictionary

    intLoop = 1
    For intLoop = 1 To intColumnCount
        If InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "Request/#id", 1) > 0 Then
            objDictionary.Add "Req_id", intLoop

         ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "Request/@Method", 1) > 0 Then
            objDictionary.Add "Req_method", intLoop

         ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "Request/@Url", 1) > 0 Then
            objDictionary.Add "Req_url", intLoop

         ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "FormPostHttpBody/@ContentType", 1) > 0 Then
            objDictionary.Add "Req_contenttype", intLoop

         ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "FormPostParameter/@Name", 1) > 0 Then
            objDictionary.Add "Req_itemdata_name", intLoop

         ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "FormPostParameter/@Value", 1) > 0 Then
            objDictionary.Add "Req_itemdata_value", intLoop

        ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "QueryStringParameter/@Name", 1) > 0 Then
            objDictionary.Add "Req_querystring_name", intLoop

        ElseIf InStr(1, Worksheets("PARSINGVS_XML").Cells(1, intLoop), "QueryStringParameter/@Value", 1) > 0 Then
            objDictionary.Add "Req_querystring_value", intLoop
        End If
    Next

    'Loop through all requests and capture querysting, itemdata, url, method, action and content type
    '-----------------------------------------------
    'Initialize variables ot default value at start
    '-----------------------------------------------
    intPrevRequest_id = 1
    intCurrRequest_id = 1
    strStartQuotes = """"
    strEndQuotes = """," & vbCrLf
    intStepNum = 1
    strQueryStr = ""
    strBody = ""

    Set objFileSys = New Scripting.FileSystemObject
    objFileSys.CreateTextFile (Path & "\" & ActionFileName)
    Set objFile = objFileSys.GetFile(Path & "\" & ActionFileName)
    Set objTextStr = objFile.OpenAsTextStream(ForAppending, TristateUseDefault)

    intLoop = 2 'first line is the header
    For intLoop = 2 To Worksheets("PARSINGVS_XML").UsedRange.Rows.Count
        If objDictionary.Exists("Req_id") Then
            intCurrRequest_id = Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_id")).Value)
        Else
            MsgBox "XML do nto contain Request id column"
            Exit Sub
        End If

        'if current and previous request id are not same OR we are at end of steps the write to file
        If (intPrevRequest_id <> intCurrRequest_id) Or (intLoop = Worksheets("PARSINGVS_XML").UsedRange.Rows.Count) Then
            Call WriteToFile
            'iterate to next step
            intStepNum = intStepNum + 1
            strQueryStr = ""
            strBody = ""
            intPrevRequest_id = intCurrRequest_id
        End If

        Call Write_Remaining_DESTINATIONVS_Req ' build the DESTINATIONVS request apart from Body & Query string
        Call WriteQuery_Body 'build hte body and querystring
    Next
    MsgBox "Completed"
    Set objDictionary = Nothing
    objTextStr.Close

    Set objTextStr = Nothing
    Set objFile = Nothing
    Set objFileSys = Nothing
    Windows(FileName).Close (False)
End Sub
'funciton to write contents to file
Sub WriteToFile()

    strWholeReq = strWholeReq & vbCrLf & strStepName & strUrl

    If strQueryStr <> "" Then
        strWholeReq = strWholeReq & "?" & strQueryStr
    End If

    strWholeReq = strWholeReq & strEndQuotes & strMethod & strContentType & strMisc

    If strBody <> "" Then
        strWholeReq = strWholeReq & strStartQuotes & "Body=" & strBody & strEndQuotes
    End If

    strWholeReq = strWholeReq & " LAST);" & vbCrLf

    objTextStr.WriteLine strWholeReq

    strWholeReq = ""
End Sub
'function to build the querystring and body part which are iterative
Sub WriteQuery_Body()
    If objDictionary.Exists("Req_querystring_name") Then
        If Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_querystring_name")).Value) <> "" Then
            If strQueryStr <> "" Then
                strQueryStr = strQueryStr & "&"
            End If
            'Querystring
            strQueryStr = strQueryStr & Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_querystring_name")).Value) & "=" & _
               Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_querystring_value")).Value)
        End If
    End If
    If objDictionary.Exists("Req_itemdata_name") Then
        If Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_itemdata_name")).Value) <> "" Then
            If strBody <> "" Then
                strBody = strBody & "&"
            End If
            'Body
            strBody = strBody & Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_itemdata_name")).Value) & "=" & _
               Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_itemdata_value")).Value)
        End If
    End If
End Sub
'function which creates remaining part of web_custom request other than querystring and body
Sub Write_Remaining_DESTINATIONVS_Req()
    'Name of Parsed_XML_Function("Step2",
    strStepName = "Parsed_XML_Function(" & strStartQuotes & "Step" & intStepNum & strEndQuotes

    If objDictionary.Exists("Req_url") Then
        '"URL = "
        strUrl = strStartQuotes & _
        "URL=" & Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_url")).Value)
    End If

    If objDictionary.Exists("Req_method") Then
        'Method =
        strMethod = strStartQuotes & _
        "Method=" & Trim(Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_method")).Value)) & strEndQuotes
    End If

    If objDictionary.Exists("Req_contenttype") Then
        'ContentType =
        If Trim(Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_contenttype")).Value)) <> "" Then
        strContentType = strStartQuotes & _
        "RecContentType=" & Trim(Trim(Worksheets("PARSINGVS_XML").Cells(intLoop, objDictionary("Req_contenttype")).Value)) & strEndQuotes
        Else
           strContentType = strStartQuotes & "RecContentType=text/html" & strEndQuotes
        End If
    Else
        strContentType = strStartQuotes & "RecContentType=text/html" & strEndQuotes
    End If
    'remaining all
    strMisc = strStartQuotes & "TargetFrame=" & strEndQuotes & _
         strStartQuotes & "Resource=0" & strEndQuotes & _
         strStartQuotes & "Referer=" & strEndQuotes & _
         strStartQuotes & "Mode=HTML" & strEndQuotes & _
         strStartQuotes & "Snapshot=t" & intStepNum & ".inf" & strEndQuotes
End Sub