Read Word Document file

Hi...
I am Beginner for Visual Basic..I am using VB6.0..
I want to read a word document & search the word document for specified words...
i am getting problem when i read word document..
*.txt is easy & fine, but if i read *.doc some problems are rising..
it' not as such ....so if anyone know the details please let me know.
Thank you

not sure what u are after,
I think its something along these lines

VB Code:

Private objWord As Word.Application
Private wd As Word.Document
 
Private Sub Command1_Click()
    Dim myRange As Range
    Dim sSearchfor As String
    sSearchfor = InputBox("What do u want to search for?")
    If sSearchfor = "" Then Exit Sub
    
    If objWord Is Nothing Then
        Set objWord = CreateObject("Word.Application")
    Else
        Set objWord = GetObject(, "Word.Application")
    End If
    DoEvents
    Set wd = objWord.Documents.Open("c:\Test.doc")
    
    Set myRange = wd.Content
    myRange.Find.Execute FindText:=sSearchfor, Forward:=True
    If myRange.Find.Found Then
        MsgBox "The document contains :'" & sSearchfor & "'", vbInformation
    Else
        MsgBox "The document do NOT contain'" & sSearchfor & "'", vbInformation
    End If
    
    If Not (wd Is Nothing) Then Set wd = Nothing
    If Not (objWord Is Nothing) Then objWord.Application.Quit
    If Not (objWord Is Nothing) Then Set objWord = Nothing
 
End Sub

I assume that you are trying to read the .doc file directly the same way you are trying to read the .txt.. this wont work because Micro$oft uses an proprietary encoding standard to encrypt (somewhat) the contents of your files. This means that you can only open and read your .doc file using micro$ofts preferred tool.. MS WORD..

you need to actually automate word to open up the file.. then you can easily assess the Find functionality built into Word

I am sure many postings have been made on this forum on how to automate word to do the above.. search around.

-mcd

and there we have it.. peet got to it before i could even submit my posting!.. thats what you need

Now he got a sample AND an explanation :)

I really should get better on commenting my code :D

Well, from what I see of your code around this forum, its clean enough that someone should be able to read it without comments..

i think comments are evil.. but you have to do it especially in production systems..

-mdc

Quote:

---

Micro$oft uses an proprietary encoding standard to encrypt (somewhat) the contents of your files. This means that you can only open and read your .doc file using micro$ofts preferred tool.. MS WORD..

you need to actually automate word to open up the file

---

Not true.

DOC files are not any more "encrypted" than RTF or HTML files. The file has to hold the formating codes for Word, and Word can do a lot of formating, so you get more "junk" than RTF or TXT or HTM (but it isn't encrypted to only work in Word like PDF files pparently are). The problem with trying to open Word files like text files is that a DOC file has a CHR(26) at the begining, just after the "ÐÏ à¡±". That will cause an input after end of file error. So you need to read binary. Then the trick is to sort out the actual document text from the formatting text. I'm working with Word 2000 and i seems to be consistent in the binary format. This should get your file text without using Word at all:

VB Code:

Private Sub Command1_Click()
    GetWordDocText
End Sub
 
Sub GetWordDocText()
 
    Dim strDocPath As String
    Dim strDocText As String
    Dim intDocTextStart As Integer
    Dim intDocTextEnd As Integer
 
    strDocPath = "C:\My Documents\BlahBlah.doc"
 
    Open strDocPath For Binary As #1
        strDocText = Space(LOF(1))
        Get #1, , strDocText
    Close #1
    
    ' Find key character to before actual text.
    intDocTextStart = InStr(1, strDocText, "Ù")
    ' Text starts after the first Ù & 3 null characters.
    intDocTextStart = intDocTextStart + 4
    
    ' Text ends with a null.
    intDocTextEnd = InStr(intDocTextStart + 1, strDocText, Chr$(0))
 
    strDocText = Mid$(strDocText, intDocTextStart, intDocTextEnd - intDocTextStart)
        
    MsgBox strDocText
 
End Sub

ps: I believe the file format is RichEdit3.0.

Excellent information you provided here WorkHorse. I didnt know that you could read a .doc without word by using a binary read. I was always under the impression that, while .docs were not actually encrypted, that they were encoded in a semi-plain text manner that you could sort of read, but not get complete accurate information unless you used word. (take for example all of this junk in the file: "ÿ ÿ ö
Ö€€€ €€€ Ö€€€").

Thanks for the information!

-mcd