Announcement

**Ray** · Jan-25-2007, 12:02 AM

I'm not exactly sure what you are asking for. Outputing the index data (which is all in a proprietary internal format for use with Zoom's search scripts) in the .CHM format does not make much sense on its own.

I presume what you are actually asking for, is the ability to use Zoom to update the internal search function in your CHM (aka Windows HTML Help) files. This is not possible, and does not seem like a simple task at all. CHM files are a binary format created by HTML Help Authoring applications, and has its own internal format for providing the search functionality AFAIK.

I am vaguely aware of the ability to add scripting to CHM files. I think it uses VBScript. Zoom currently provides an ASP script option, which also uses VBScript. Although the script is in the same language, I am sure that there are significant differences that would require a more intensive development port. Some functionality may be absent all together when running in the CHM enviornment as opposed to ASP, and this could prove to be a very difficult endeavour. I am also unsure if the scripting functionality allows us to override the normal search function built into CHM files.

Also, I am not sure what problems you are having with the built-in search functionality that is created by Microsoft's utility for creating CHM files?

Combining the facts that: there has been no demand for us replacing the existing search functionality in CHM files, with the fact that we are unfamiliar with scripting in CHM files, it is something that seems unlikely for us to pursue.

Let me know if I have misinterpreted your question.

**Peter Marchert** · Jan-25-2007, 06:42 AM

Thank you for your reply, Ray.

The functionality I suggested, means not a complete search function for chm-files, only the creation of an index.

The index file has the extension .hhk and contains simple html tags.

Here is my actual code to create the index file:

Code:

Sub CreateIndexFile()

    Dim strHTMLRoot                 As String
    Dim strIndexFile                As String
    
    Dim strWord                     As String
    Dim strTitle                    As String
    Dim strHTMLFile                 As String
    Dim strTarget                   As String
    
    Dim lngBackSlash                As Long
    Dim lngRow                      As Long
    Dim lngRowWord                  As Long
    
    Dim bytFreeFile                 As Byte
    
    strHTMLRoot = Sheets("Files").Range("C2").Value
    
    lngBackSlash = InStrRev(strHTMLRoot, "\")
    
    strHTMLRoot = Left(strHTMLRoot, lngBackSlash - 1) & "\"
    
    strIndexFile = strHTMLRoot & "Index.hhk"
    
    Kill strIndexFile
    
    bytFreeFile = FreeFile
    
    Open strIndexFile For Output As #bytFreeFile
    
        '-------------------------------------------------------------------------------------
        ' Header
        '-------------------------------------------------------------------------------------
        Print #bytFreeFile, "<!DOCTYPE HTML PUBLIC ""-//IETF//DTD HTML//EN"">"
        Print #bytFreeFile, "<HTML>"
        Print #bytFreeFile, "  <HEAD>"
        Print #bytFreeFile, "    <meta name=""GENERATOR"" content=""Microsoft&reg; HTML Help Workshop 4.1"">"
        Print #bytFreeFile, "    <!-- Sitemap 1.0 -->"
        Print #bytFreeFile, "  </HEAD>"
        Print #bytFreeFile, "  <BODY>"
        Print #bytFreeFile, "    <UL>"
                
        For lngRow = 2 To ActiveSheet.UsedRange.Rows.Count
            
            '---------------------------------------------------------------------------------
            ' Reading data
            '---------------------------------------------------------------------------------
            strWord = Trim(Cells(lngRow, 1).Value)
            If strWord = "" Then Exit For
            strTitle = Trim(Cells(lngRow, 2).Value)
            strHTMLFile = Trim(Replace(Cells(lngRow, 3).Value, strHTMLRoot, ""))
            strTarget = Trim(Cells(lngRow, 4).Value)
            If strTarget <> "" Then strHTMLFile = strHTMLFile & "#" & strTarget
            lngRowWord = lngRow
            
            '---------------------------------------------------------------------------------
            ' Printing data
            '---------------------------------------------------------------------------------
            Print #bytFreeFile, "      <LI><OBJECT type=""text/sitemap"">"
            Print #bytFreeFile, "          <param name=""Name"" value=""" & strWord & """>"
            
            Do
                
                Print #bytFreeFile, "          <param name=""Name"" value=""" & strTitle & """>"
                Print #bytFreeFile, "          <param name=""Local"" value=""" & strHTMLFile & """>"
                
                lngRowWord = lngRowWord + 1
                
                strHTMLFile = Trim(Replace(Cells(lngRowWord, 3).Value, strHTMLRoot, ""))
                strTarget = Trim(Cells(lngRowWord, 4).Value)
                If strTarget <> "" Then strHTMLFile = strHTMLFile & "#" & strTarget
                
            Loop While Cells(lngRowWord, 1) = strWord
            
            Print #bytFreeFile, "          </OBJECT>"
            
        Next
        
        '-------------------------------------------------------------------------------------
        ' Footer
        '-------------------------------------------------------------------------------------
        Print #bytFreeFile, "    <UL>"
        Print #bytFreeFile, "  </BODY>"
        Print #bytFreeFile, "</HTML>"
    
    Close #bytFreeFile

End Sub

After creation I get something like that for the index.hhk:

Code:

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">
<HTML>
  <HEAD>
    <meta name="GENERATOR" content="Microsoft&reg; HTML Help Workshop 4.1">
    <!-- Sitemap 1.0 -->
  </HEAD>
  <BODY>
    <UL>
      <LI><OBJECT type="text/sitemap">
          <param name="Name" value="/loadinf">
          <param name="Name" value="H&#228;ufig gestellte Fragen">
          <param name="Local" value="htm\faqs.html#question2">
          </OBJECT>
      <LI><OBJECT type="text/sitemap">
          <param name="Name" value="/saveinf">
          <param name="Name" value="H&#228;ufig gestellte Fragen">
          <param name="Local" value="htm\faqs.html#question2">
          </OBJECT>
      <LI><OBJECT type="text/sitemap">
          <param name="Name" value="/silent">
          <param name="Name" value="H&#228;ufig gestellte Fragen">
          <param name="Local" value="htm\faqs.html#question2">
          </OBJECT>
    <UL>
  </BODY>
</HTML>

That output should be possible by zoom search too, I think.

Peter

**Ray** · Jan-25-2007, 07:21 AM

The "Index" tab that you see in your HTML Help window is quite different in nature to the index files we generate for a Search function. Like the index on the back of a book, this sort of end-user index should provide only keywords and nouns - they really need to be manually picked out from the content of the pages in question - which is why you are doing so manually at this point.

The Zoom Indexer scans and indexes all text content on pages, so that the search script can perform full-text searching. This means all words are indexed, nouns, verbs, adjectives - things that would just clutter up your end-user index, you'll have entries for "that", "output", "should", "be", "possible", "think"... it's just useless in that scenario.

So the index data that Zoom generates is not what you want for the "Index" of your CHM files.

**Peter Marchert** · Jan-25-2007, 07:34 AM

Originally posted by Ray View Post

So the index data that Zoom generates is not what you want for the "Index" of your CHM files.

That is nearly right. A good index will provide words which are not included in the files.

But I think it is not wrong to provide all words of the files too. So if the user do not use the full search function he will found fast in the index for what he is searching for.

In combination with the recommended links (that include words which are not found in the files) you could create a good and complete index.

Peter

**Peter Marchert** · Jan-25-2007, 07:40 AM

Originally posted by Ray View Post

This means all words are indexed, nouns, verbs, adjectives - things that would just clutter up your end-user index, you'll have entries for "that", "output", "should", "be", "possible", "think"... it's just useless in that scenario.

For this words you need a exclusion list. For German words I have this nearly completed and in my excel macro is a "learning" function for this.

Ok, this would be some more work for you and the most people using Zoom does not need this.

Peter

**Ray** · Jan-28-2007, 11:28 PM

Originally posted by Peter Marchert View Post

But I think it is not wrong to provide all words of the files too. So if the user do not use the full search function he will found fast in the index for what he is searching for.

But this would essentially be replicating the "Search" tab functionality which is built into the Windows HTML Help / CHM file viewer, not the purpose of the "Index" - which is to provide only the relevant, and associated keywords - again, like the back of a book.

Originally posted by Peter Marchert View Post

For this words you need a exclusion list. For German words I have this nearly completed and in my excel macro is a "learning" function for this.

I think you may be underestimating the differences between a full-text index and a back-of-book-style index. For example, on one page, the use of the word "shop" may not be relevant at all to a topic, it might be part of an expression in a blog entry: "... i'm going out to shop around this weekend ..." - but if you add the word "shop" to the exclusion list, you'll also be omitting it from a page where the actual word "shop" is relevant: "Welcome to my Online Shop page!".

I do not believe you can create an exclusion list to make this more useful. And as mentioned before, if the "Index" tab contained all words from the content, your users would not only have a very large list of words to dig through, but each keyword selected will return hundreds of pages, and the user has little means of digging through the results, because essentially, they would only be able to search by one keyword.

I can imagine that, for a very small list of keywords - where the words are very unique (eg. source code class names or function names), and you only want to search for those keywords (assuming you ignore ALL other normal words in the english dictionary, which is way too much to add to an exclusion list really), it can be useful. But this is really not the designed use of Zoom, and as you realize, of very little use to the majority of our users.

I should point out that there are many HTML Help file authoring applications out there, which aid in the process of creating CHM files. Although, even they usually require you to enter the "Index" keywords manually, because again, in almost all cases, manual selection would create a far more effective and useful end-of-book-style index.

**Peter Marchert** · Jan-29-2007, 05:49 AM

Hello Ray,

thanks again for your detailed answer.

I think you are right. It was just an idea and now I think it would be better to create an index manually.

Peter

Announcement

Output index in html help code

Output index in html help code

Comment

Comment

Comment

Comment

Comment

Comment

Comment