We Keep Coding

Copy document content (including formatting and page format) to another using Word Interop in c# with 100% fidelity

I want to copy the content of a document created by the user to an existing document. The existing document content must be an exact mirror to the document created by the user.
I cannot simply copy the file using System.IO or saving a copy of the document created by the user using SaveAs methods in Word Interop. This is because the existing document is a document that is generated from a webserver and has VBA modules for uploading it back to the server.
The document generated by the webserver (existing document) is a Word 2003 document, but the document created by the user is either a Word 2003 document or Word 2007+.
Having these limitations in mind, I first created the following method:
string tempsave = //location of user created document;
string savelocation = //location of existing document;
Word.Application objWordOpen = new Word.Application();
Document doclocal = objWordOpen.Documents.Open(tempsave);
Document d1 = objWordOpen.Documents.Open(savelocation);
Word.Range oRange = doclocal.Content;
oRange.Copy();
d1.Activate();
d1.UpdateStyles();
d1.ActiveWindow.Selection.WholeStory();
d1.ActiveWindow.Selection.PasteAndFormat(Word.WdRecoveryType.wdFormatOriginalFormatting);
This is generally working. However, the tables are messed up.
Also, if there is a Page Break, the output is different.
The user created document:
The output - existing document:
Also, at the end of the document a paragraph mark is added, as follows:
The user created document:
The output - existing document:
The page format is also messed up, the output having mirror margins set up.
The user created document:
The output - existing document:
I have also tried using Range.Insert() method and setting the range without copying as described here https://stackoverflow.com/a/54500605/10468231, but I am still having these issues.
I have also tried adding the VBA modules to the document, but there are also Document Variables and other custom properties and I don't want to mess with the file being uploaded to the server.
How do I handle these issues? Both the documents are based on Normal template.
I am open to another suggestion regarding this topic, but I know that .doc files are not handled as easily as .docx format, this is why I think I am stuck with COM Interop.
Thank you.
UPDATE
Based on Macropod code posted by Charles Kenyon, I have managed to copy more of the formatting from the source to target. Still, there is the difference at the page break - the paragraph mark is places on the new page, instead on the same page.
Also, the text is slightly larger, even though the Font Size is the same.
Word.Range oRange;
oRange = Source.Content;
Target.Content.FormattedText = oRange.FormattedText;
LayoutTransfer(Source, Target);
LayoutTransfer method:
private void LayoutTransfer(Document source, Document target)
{
float sPageHght;
float sPageWdth;
float sHeaderDist;
float sFooterDist;
float sTMargin;
float sBMargin;
float sLMargin;
float sRMargin;
float sGutter;
WdGutterStyle sGutterPos;
WdPaperSize lPaperSize;
WdGutterStyleOld lGutterStyle;
int lMirrorMargins;
WdVerticalAlignment lVerticalAlignment;
WdSectionStart lScnStart;
WdSectionDirection lScnDir;
int lOddEvenHdFt;
int lDiffFirstHdFt;
bool bTwoPagesOnOne;
bool bBkFldPrnt;
int bBkFldPrnShts;
bool bBkFldRevPrnt;
WdOrientation lOrientation;
foreach (Word.Section section in source.Sections)
{
lPaperSize = section.PageSetup.PaperSize;
lGutterStyle = section.PageSetup.GutterStyle;
lOrientation = section.PageSetup.Orientation;
lMirrorMargins = section.PageSetup.MirrorMargins;
lScnStart = section.PageSetup.SectionStart;
lScnDir = section.PageSetup.SectionDirection;
lOddEvenHdFt = section.PageSetup.OddAndEvenPagesHeaderFooter;
lDiffFirstHdFt = section.PageSetup.DifferentFirstPageHeaderFooter;
lVerticalAlignment = section.PageSetup.VerticalAlignment;
sPageHght = section.PageSetup.PageHeight;
sPageWdth = section.PageSetup.PageWidth;
sTMargin = section.PageSetup.TopMargin;
sBMargin = section.PageSetup.BottomMargin;
sLMargin = section.PageSetup.LeftMargin;
sRMargin = section.PageSetup.RightMargin;
sGutter = section.PageSetup.Gutter;
sGutterPos = section.PageSetup.GutterPos;
sHeaderDist = section.PageSetup.HeaderDistance;
sFooterDist = section.PageSetup.FooterDistance;
bTwoPagesOnOne = section.PageSetup.TwoPagesOnOne;
bBkFldPrnt = section.PageSetup.BookFoldPrinting;
bBkFldPrnShts = section.PageSetup.BookFoldPrintingSheets;
bBkFldRevPrnt = section.PageSetup.BookFoldRevPrinting;
var index = section.Index;
target.Sections[index].PageSetup.PaperSize = lPaperSize;
target.Sections[index].PageSetup.GutterStyle = lGutterStyle;
target.Sections[index].PageSetup.Orientation = lOrientation;
target.Sections[index].PageSetup.MirrorMargins = lMirrorMargins;
target.Sections[index].PageSetup.SectionStart = lScnStart;
target.Sections[index].PageSetup.SectionDirection = lScnDir;
target.Sections[index].PageSetup.OddAndEvenPagesHeaderFooter = lOddEvenHdFt;
target.Sections[index].PageSetup.DifferentFirstPageHeaderFooter = lDiffFirstHdFt;
target.Sections[index].PageSetup.VerticalAlignment = lVerticalAlignment;
target.Sections[index].PageSetup.PageHeight = sPageHght;
target.Sections[index].PageSetup.PageWidth = sPageWdth;
target.Sections[index].PageSetup.TopMargin = sTMargin;
target.Sections[index].PageSetup.BottomMargin = sBMargin;
target.Sections[index].PageSetup.LeftMargin = sLMargin;
target.Sections[index].PageSetup.RightMargin = sRMargin;
target.Sections[index].PageSetup.Gutter = sGutter;
target.Sections[index].PageSetup.GutterPos = sGutterPos;
target.Sections[index].PageSetup.HeaderDistance = sHeaderDist;
target.Sections[index].PageSetup.FooterDistance = sFooterDist;
target.Sections[index].PageSetup.TwoPagesOnOne = bTwoPagesOnOne;
target.Sections[index].PageSetup.BookFoldPrinting = bBkFldPrnt;
target.Sections[index].PageSetup.BookFoldPrintingSheets = bBkFldPrnShts;
target.Sections[index].PageSetup.BookFoldRevPrinting = bBkFldRevPrnt;
}
}
UPDATE 2
Actually, the page break not remaining in line with paragraph format is not an issue of copying fidelity, but rather an issue of conversion from .doc to .docx. (https://support.microsoft.com/en-us/help/923183/the-layout-of-a-document-that-contains-a-page-break-may-be-different-i)
Maybe someone thought of a method to overcome this.

The following code by Paul Edstein (macropod) may assist you. It will at least give you an idea of the complexities you are facing.
' ============================================================================================================
' KEEP NEXT THREE TOGETHER
' ============================================================================================================
'
Sub CombineDocuments()
' Paul Edstein
' https://www.msofficeforums.com/word-vba/43339-combine-multiple-word-documents.html
'
' Users occasionally need to combine multiple documents that may of may not have the same page layouts,
' Style definitions, and so on. Consequently, combining multiple documents is often rather more complex than
' simply copying & pasting content from one document to another. Problems arise when the documents have
' different page layouts, headers, footers, page numbering, bookmarks & cross-references,
' Tables of Contents, Indexes, etc., etc., and especially when those documents have used the same Style
' names with different definitions.
'
' The following Word macro (for Windows PCs only) handles the more common issues that arise when combining
' documents; it does not attempt to resolve conflicts with paragraph auto-numbering,
' document -vs- section page numbering in 'page x of y' numbering schemes, Tables of Contents or Indexing issues.
' Neither does it attempt to deal with the effects on footnote or endnote numbering & positioning or with the
' consequences of duplicated bookmarks (only one of which can exist in the merged document) and any corresponding
' cross-references.
'
' The macro includes a folder browser. Simply select the folder to process and all documents in that folder
' will be combined into the currently-active document. Word's .doc, .docx, and .docm formats will all be processed,
' even if different formats exist in the selected folder.
'
Application.ScreenUpdating = False
Dim strFolder As String, strFile As String, strTgt As String
Dim wdDocTgt As Document, wdDocSrc As Document, HdFt As HeaderFooter
strFolder = GetFolder: If strFolder = "" Then Exit Sub
Set wdDocTgt = ActiveDocument: strTgt = ActiveDocument.fullname
strFile = Dir(strFolder & "\*.doc", vbNormal)
While strFile <> ""
If strFolder & strFile <> strTgt Then
Set wdDocSrc = Documents.Open(FileName:=strFolder & "\" & strFile, AddToRecentFiles:=False, Visible:=False)
With wdDocTgt
.Characters.Last.InsertBefore vbCr
.Characters.Last.InsertBreak (wdSectionBreakNextPage)
With .Sections.Last
For Each HdFt In .Headers
With HdFt
.LinkToPrevious = False
.range.Text = vbNullString
.PageNumbers.RestartNumberingAtSection = True
.PageNumbers.StartingNumber = wdDocSrc.Sections.First.Headers(HdFt.Index).PageNumbers.StartingNumber
End With
Next
For Each HdFt In .Footers
With HdFt
.LinkToPrevious = False
.range.Text = vbNullString
.PageNumbers.RestartNumberingAtSection = True
.PageNumbers.StartingNumber = wdDocSrc.Sections.First.Headers(HdFt.Index).PageNumbers.StartingNumber
End With
Next
End With
Call LayoutTransfer(wdDocTgt, wdDocSrc)
.range.Characters.Last.FormattedText = wdDocSrc.range.FormattedText
With .Sections.Last
For Each HdFt In .Headers
With HdFt
.range.FormattedText = wdDocSrc.Sections.Last.Headers(.Index).range.FormattedText
.range.Characters.Last.Delete
End With
Next
For Each HdFt In .Footers
With HdFt
.range.FormattedText = wdDocSrc.Sections.Last.Footers(.Index).range.FormattedText
.range.Characters.Last.Delete
End With
Next
End With
End With
wdDocSrc.Close SaveChanges:=False
End If
strFile = Dir()
Wend
With wdDocTgt
' Save & close the combined document
.SaveAs FileName:=strFolder & "Forms.docx", FileFormat:=wdFormatXMLDocument, AddToRecentFiles:=False
' and/or:
.SaveAs FileName:=strFolder & "Forms.pdf", FileFormat:=wdFormatPDF, AddToRecentFiles:=False
.Close SaveChanges:=False
End With
Set wdDocSrc = Nothing: Set wdDocTgt = Nothing
Application.ScreenUpdating = True
End Sub
' ============================================================================================================
Private Function GetFolder() As String
' used by CombineDocument macro by Paul Edstein, keep together in same module
' https://www.msofficeforums.com/word-vba/43339-combine-multiple-word-documents.html
Dim oFolder As Object
GetFolder = ""
Set oFolder = CreateObject("Shell.Application").BrowseForFolder(0, "Choose a folder", 0)
If (Not oFolder Is Nothing) Then GetFolder = oFolder.Items.Item.Path
Set oFolder = Nothing
End Function
Sub LayoutTransfer(wdDocTgt As Document, wdDocSrc As Document)
' works with previous Combine Documents macro from Paul Edstein, keep together
' https://www.msofficeforums.com/word-vba/43339-combine-multiple-word-documents.html
'
Dim sPageHght As Single, sPageWdth As Single
Dim sHeaderDist As Single, sFooterDist As Single
Dim sTMargin As Single, sBMargin As Single
Dim sLMargin As Single, sRMargin As Single
Dim sGutter As Single, sGutterPos As Single
Dim lPaperSize As Long, lGutterStyle As Long
Dim lMirrorMargins As Long, lVerticalAlignment As Long
Dim lScnStart As Long, lScnDir As Long
Dim lOddEvenHdFt As Long, lDiffFirstHdFt As Long
Dim bTwoPagesOnOne As Boolean, bBkFldPrnt As Boolean
Dim bBkFldPrnShts As Boolean, bBkFldRevPrnt As Boolean
Dim lOrientation As Long
With wdDocSrc.Sections.Last.PageSetup
lPaperSize = .PaperSize
lGutterStyle = .GutterStyle
lOrientation = .Orientation
lMirrorMargins = .MirrorMargins
lScnStart = .SectionStart
lScnDir = .SectionDirection
lOddEvenHdFt = .OddAndEvenPagesHeaderFooter
lDiffFirstHdFt = .DifferentFirstPageHeaderFooter
lVerticalAlignment = .VerticalAlignment
sPageHght = .PageHeight
sPageWdth = .PageWidth
sTMargin = .TopMargin
sBMargin = .BottomMargin
sLMargin = .LeftMargin
sRMargin = .RightMargin
sGutter = .Gutter
sGutterPos = .GutterPos
sHeaderDist = .HeaderDistance
sFooterDist = .FooterDistance
bTwoPagesOnOne = .TwoPagesOnOne
bBkFldPrnt = .BookFoldPrinting
bBkFldPrnShts = .BookFoldPrintingSheets
bBkFldRevPrnt = .BookFoldRevPrinting
End With
With wdDocTgt.Sections.Last.PageSetup
.GutterStyle = lGutterStyle
.MirrorMargins = lMirrorMargins
.SectionStart = lScnStart
.SectionDirection = lScnDir
.OddAndEvenPagesHeaderFooter = lOddEvenHdFt
.DifferentFirstPageHeaderFooter = lDiffFirstHdFt
.VerticalAlignment = lVerticalAlignment
.PageHeight = sPageHght
.PageWidth = sPageWdth
.TopMargin = sTMargin
.BottomMargin = sBMargin
.LeftMargin = sLMargin
.RightMargin = sRMargin
.Gutter = sGutter
.GutterPos = sGutterPos
.HeaderDistance = sHeaderDist
.FooterDistance = sFooterDist
.TwoPagesOnOne = bTwoPagesOnOne
.BookFoldPrinting = bBkFldPrnt
.BookFoldPrintingSheets = bBkFldPrnShts
.BookFoldRevPrinting = bBkFldRevPrnt
.PaperSize = lPaperSize
.Orientation = lOrientation
End With
End Sub
' ============================================================================================================

I used a Template and copied it several times into a new Word Document after editing it.
It worked like this
Word.Range rng = wordDocTarget.Content;
rng.Collapse(Word.WdCollapseDirection.wdCollapseEnd)
rng.FormattedText = wordDocSource.Content.FormattedText
An alternative could also be to insert a whole file to a range / document
rng = wordDoc.Range
rng.Collapse(Word.WdCollapseDirection.wdCollapseEnd)
rng.InsertFile(filepath)

Remove inline styles from innerHtml using HtmlAgilityPack

I am parsing a web page to return all the unique sentences on the page, each with a minimum of two words. It almost works. The following appears as one sentence in the page however my code is dropping the text in the <b></b> tags. How do I remove the inline style/tags to return the sentence as it appears in the browser with the text in the bold tags or any other inline style like strong tags?
Currently it returns NHL Playoffs as one line of text and then Takeaways: Sharks beat Penguins for first Stanley Cup Final win as the second sentence when it is really just one sentence.
<span class="titletext"><b>NHL Playoffs</b> Takeaways: Sharks beat Penguins for first Stanley Cup Final win</span>
Here is my asp.net vb.net code (c# solution is fine).
Public Shared Function validateIsMoreThanOneWord(input As String, numberWords As Integer) As Boolean
If String.IsNullOrEmpty(input) Then
Return False
End If
Return (input.Split(New Char() {" "c}, StringSplitOptions.RemoveEmptyEntries).Length >= numberWords)
End Function
Private Sub form1_Load(sender As Object, e As EventArgs) Handles form1.Load
Try
Dim html = New HtmlDocument()
html.LoadHtml(New WebClient().DownloadString("http://news.google.ca/nwshp?hl=en&ei=4H1UV7-NNOfCjwTAl4bABw&ved=0EKkuCAkoBw"))
Dim root = html.DocumentNode
Dim myList As New List(Of String)()
For Each node As HtmlNode In root.Descendants().Where(Function(n) n.NodeType = HtmlNodeType.Text AndAlso n.ParentNode.Name <> "script" AndAlso n.ParentNode.Name <> "style" AndAlso n.ParentNode.Name <> "css")
If Not node.HasChildNodes Then
Dim text As String = HttpUtility.HtmlDecode(node.InnerText)
If Not String.IsNullOrEmpty(text) And Not String.IsNullOrWhiteSpace(text) Then
If validateIsMoreThanOneWord(text.Trim(), 2) Then
myList.Add(text.Trim())
End If
End If
End If
Next
'remove dups from array and other stuff
Dim q As String() = myList.Distinct().ToArray()
For i As Integer = 0 To UBound(q)
Response.Write(q(i).Trim() & "<br/>")
Next
Response.Write(q.Count)
Catch ex As Exception
Response.Write(ex.Message)
End Try
End Sub
Hope you can shed some light on a solution. Thanks!

Since you are looping over all root descendant nodes which parent is not <script>, nor <style> nor css, you will indeed treat every child node from .titleText as a different piece of text.
What you want is to retrieve the InnerText of each .titletext entry.
The following is what I would do in C#, you can get the idea of what you need to do.
HtmlWeb w = new HtmlWeb();
var htmlDoc = w.Load("http://news.google.ca/nwshp?hl=en&ei=4H1UV7-NNOfCjwTAl4bABw&ved=0EKkuCAkoBw");
var textTitles = htmlDoc.DocumentNode.SelectNodes("//span[#class='titletext']");
//for testing purposes
foreach (var textTitle in textTitles)
Console.WriteLine(textTitle.InnerText);

Vb.net how to compare large text files

Hi All below code is how to compare contents in two text file and is work fine for record in files, but my issue when files have a lot line ( 80000 up) my code work very very slow and i cannot accept it. please kindly give me some idea
Public Class Form1
Const TEST1 = "D:\a.txt"
Const TEST2 = "D:\b.txt"
Public file1 As New Dictionary(Of String, String)
Public file2 As New Dictionary(Of String, String)
Public text1 As String()
Public i As Integer
Private Sub Button1_Click(ByVal sender As System.Object, ByVal e As System.EventArgs) Handles Button1.Click
'Declare two dictionaries. The key for each will be the text from the input line up to,
'but not including the first ",". The valus for each will be the entire input line.
'Dim file1 As New Dictionary(Of String, String)
'Dim file2 As New Dictionary(Of String, String)
'Dim text1 As String()
For Each line As String In System.IO.File.ReadAllLines(TEST1)
Dim part() As String = line.Split(",")
file1.Add(part(0), line)
Next
For Each line As String In System.IO.File.ReadAllLines(TEST2)
Dim part() As String = line.Split(",")
file2.Add(part(0), line)
Next
' AddText("The following lines from " & TEST2 & " are also in " & TEST1)
For Each key As String In file1.Keys
If file2.ContainsKey(key) Then
TextBox1.Text &= (file1(key)) & vbCrLf
MsgBox(file2(key))
Label1.Text = file1(key)
Else
TextBox2.Text &= (file1(key)) & vbCrLf
End If
Next
text1 = TextBox1.Lines
IO.File.WriteAllLines("D:\Same.txt", text1)
text1 = TextBox2.Lines
IO.File.WriteAllLines("D:\Differrent.txt", text1)
End Sub

The first thing I would change is the use of a Dictionary. I would use an Hashset.
See HashSet versus Dictionary
Then I would change the ReadAllLines loop. The ReadAllLines loads every line in memory before starting the loop, while ReadLines doesn't read all lines but you can start to work on your line immediately.
See What's the fastest way to read a text file line-by-line?
The third point is switching the order of the files read. First read the TEST2 file then the TEST1. This because while you load TEST1 lines you could immediately check if the file2 Hashset contains the key and Add the found line in a list of found strings while the line not found in a list of not found strings.
Dim TEST1 = "D:\temp\test3.txt"
Dim TEST2 = "D:\temp\test6.txt"
Dim file2Keys As New Hashset(Of String)
For Each line As String In System.IO.File.ReadLines(TEST2)
Dim parts = line.Split(",")
file2Keys.Add(parts(0))
Next
Dim listFound As New List(Of String)()
Dim listNFound= New List(Of String)()
For Each line As String In System.IO.File.ReadLines(TEST1)
Dim parts = line.Split(",")
If file2Keys.Contains(parts(0)) Then
listFound.Add(line)
Else
listNFound.Add(line)
End If
Next
IO.File.WriteAllText("D:\temp\Same.txt", String.Join(Environment.NewLine, listFound.ToArray()))
IO.File.WriteAllText("D:\temp\Differrent.txt", String.Join(Environment.NewLine, listNFound.ToArray()))

vb.net dataset.load xml file containing string "&"

I'm using a Dataset.Load statement to load a XMl File and on the file I have some tags with the "&" character and this is causing a exception. Are there any way to Load the XML to the dataset or replacing the & for another string.
I tried to do a Replace but when I use StringVar.Replace("&","e") for example when I have "ç" or "ã" strings on the file this chars are replaced for an wrong sequence of chars.
I was trying this
My.Computer.FileSystem.WriteAllText(MyFilePath, My.Computer.FileSystem.ReadAllText(MyFilePath, System.Text.Encoding.UTF8).Replace(" & ", "&"), False, System.Text.Encoding.UTF8)
but it happens that some files has "A&B" or any other combination of letters before and after the "&"
I'll be glad if anyone can help-me.
Thanks

`Hello Guys, I solved my problem. The problem was really #Blorgbeard sayd the Xml File was coming not valid.
Public Shared Function Decompress(text As String) As String
Dim bytes As Byte() = Convert.FromBase64String(text)
Using msi = New MemoryStream(bytes)
Using mso = New MemoryStream()
Using gs = New System.IO.Compression.GZipStream(msi, System.IO.Compression.CompressionMode.Decompress)
Dim bytesAux As Byte() = New Byte(4095) {}
Dim cnt As Integer
While (InlineAssignHelper(cnt, gs.Read(bytesAux, 0, bytesAux.Length))) <> 0
mso.Write(bytesAux, 0, cnt)
End While
End Using
Dim streamReader As StreamReader = New StreamReader(mso, System.Text.Encoding.UTF8, True)
Dim XmlDoc As String
mso.Seek(0, SeekOrigin.Begin)
XmlDoc = streamReader.ReadToEnd
Return XmlDoc
End Using
End Using
End Function`
this is what I did to get and return the string containing the correct XML data to be write to file.

Using A Local file path in a Streamwriter object ASP.Net

I am trying to create a csv file of some data. I have wrote a function that successfully does this....
Private Sub CreateCSVFile(ByVal dt As DataTable, ByVal strFilePath As String)
Dim sw As New StreamWriter(strFilePath, False)
''# First we will write the headers.
''#DataTable dt = m_dsProducts.Tables[0];
Dim iColCount As Integer = dt.Columns.Count
For i As Integer = 0 To iColCount - 1
sw.Write(dt.Columns(i))
If i < iColCount - 1 Then
sw.Write(",")
End If
Next
sw.Write(sw.NewLine)
''# Now write all the rows.
For Each dr As DataRow In dt.Rows
For i As Integer = 0 To iColCount - 1
If Not Convert.IsDBNull(dr(i)) Then
sw.Write(dr(i).ToString())
End If
If i < iColCount - 1 Then
sw.Write(",")
End If
Next
sw.Write(sw.NewLine)
Next
sw.Close()
End Sub
The problem is I am not using the streamwriter object correctly for what I trying to accomplish. Since this is an asp.net I need the user to pick a local filepath to put the file on. If I pass any path to this function its gonna try to write it to the directory specified on the server where the code is. I would like this to popup and let the user select a place on their local machine to put the file....
Dim exData As Byte() = File.ReadAllBytes(Server.MapPath(eio))
File.Delete(Server.MapPath(eio))
Response.AddHeader("content-disposition", String.Format("attachment; filename={0}", fn))
Response.ContentType = "application/x-msexcel"
Response.BinaryWrite(exData)
Response.Flush()
Response.End()
I am calling the first function in code like this...
Dim emplTable As DataTable = SiteAccess.DownloadEmployee_H()
CreateCSVFile(emplTable, "C:\\EmplTable.csv")
Where I dont want to have specify the file loaction (because this will put the file on the server and not on a client machine) but rather let the user select the location on their client machine.
Can someone help me put this together? Thanks in advance.
I have recreated my export function and now it lets the usr select a download path, but one column in the data being downloaded has data in the form of "Doe, John" this column is called "EPLNME" this messes up the output file because its reading the comma in the data and now the data is off by a column in the output file can someone help me stop this specific incident im not sure how I can. Here is the code...
Private Sub ExportCSV(ByVal data As DataTable, ByVal nameOfFile As String)
Dim context As HttpContext = HttpContext.Current
context.Response.Clear()
context.Response.ContentType = "text/csv"
context.Response.AddHeader("Content-Disposition", "attachment; filename=" + nameOfFile + ".csv")
''#Write column header names
For i = 0 To data.Columns.Count - 1
If (i > 0) Then
context.Response.Write(",")
End If
context.Response.Write(data.Columns(i).ColumnName)
Next
context.Response.Write(Environment.NewLine)
''#Write data
For Each row As DataRow In data.Rows
For i = 0 To data.Columns.Count - 1
If (i > 0) Then
context.Response.Write(",")
End If
context.Response.Write(row.Item(i).ToString())
Next
context.Response.Write(Environment.NewLine)
Next
context.Response.End()
End Sub

First, you need to overload your function like this, to allow sending your output directly to either a stream or a path:
Private Sub CreateCSVFile(ByVal dt As DataTable, ByVal strFilePath As String)
Using sw As New StreamWriter(strFilePath)
CreateCSVFile(dt, sw)
End Using
End Sub
Private Sub CreateCSVFile(ByVal dt As DataTable, ByVal outStream As TextWriter)
''# First we will write the headers.
Dim delimiter As String = String.Empty
For Each col As DataColumn in dt.Columns
outStream.Write(delimiter)
outStream.Write(col.ColumnName)
delimiter = ","
Next col
outStream.Write(outStream.NewLine)
int flushCount = 0;
''# Now write all the rows.
For Each dr As DataRow In dt.Rows
delimiter = String.Empty
For i As Integer = 0 To dt.Columns.Count -1
outStream.Write(delimiter)
If Not Convert.IsDBNull(dr(i)) Then
outStream.Write("""") ''#Wrap fields in quotes to allow for commas in field data
''# Need to escape the quotes as well
outStream.Write(dr(i).ToString().Replace("""", """"""))
outStream.Write("""")
End If
delimiter = ","
Next i
outStream.Write(outStream.NewLine)
''# Flush the buffer periodically
flushCount += 1
If flushCount > 100 Then
outStream.Flush()
flushCount = 0
End If
Next dr
End Sub
Notice that your function works pretty much exactly the same as before, but you can now write to a file or directly to a stream, and you didn't have to re-write a lot of code to make it work. Pretty much anything you write that works with files should be written this way. I made a few other improvements to the code as well, but the main thing is the method that does the actual work should always accept a TextWriter and then just add overloads if you want to be able to accept anything else like a file path.
Now what you can do is take the Content Type and Header from Ben Robinson's answer and use this new method to write directly to the asp.net response buffer:
Response.ContentType = "text/csv";
Response.AddHeader("Content-Disposition", "attachment; filename=NameOfFile");
CreateCSVFile(SiteAccess.DownloadEmployee_H(), Response.Output)
Response.Flush()
Response.End()

You, don't really nead a streamwriter, that is for creating files on the machine where the code is running. Use a StringBuilder to build up the string that represents the CSV file then do the following:
Response.ContentType = "text/csv";
Response.AddHeader("Content-Disposition", "attachment; filename=NameOfFile");
Response.Write(MyStringBuilder.ToString());
If you do need to create a file because you need to store it on the server and also transmit it to the user. Create the file as you are doing and the replace the last line with
Response.TransmitFile("filePath");

You can use a MemoryStream to hold the binary data on the server, instead of writing them to a file.
1) Write the contents youn want to put in the CSV into the memory stream
2) Read from the MemoryStream into the Response when required.
Hope it helps!