CHMReader.Net

• Download source files - 305 Kb
• Download Installer- 659 Kb
• Download Release Cabs - 696 Kb

Introduction

In my first article for ‘The Code Project’ I developed a program to view CHM files on a Pocket PC 2003 (Using embedded Visual C++) using the open source CHMLib. This version had a number of issues, so I decided to write a C# version, and here it is.

Features

• Contents Tree
• Index List with Also See jumping.
• Search
• Uses Regular expressions to patch the HTML code extracted from the CHM
• Handles CHM files which point to other CHM Files (CHM Book Shelf. E.g. MCSD NetCore.chm)Handles navigation to Anchor points
• Full C# Source included

The New Code

The new version of the CHMReader.Net uses the following libraries

• HTML Help Library by Klaus Weisser – See http://www.codeproject.com/csharp/HtmlHelp.asp
• OpenNetCF Library for the HTML Viewer Control - See http://www.opennetcf.org
• SharpZipLib Library included (required by HtmlHelp Library) See http://www.icsharpcode.net/OpenSource/SharpZipLib/Default.aspx
• I also ported from ‘C’ into ‘C#’ the CHMLib library by Jed Wing.

A Big thank you to these guys for releasing there code so we can all benefit from it. Thanks.

There are a number of points to make here concerning the porting of the CHMLib. Firstly, when porting LZW decompression code, I had to first compile the code in ‘C’ and get the pre-processor to export the source sent to the compiler. This is because the original code used a lot of macros which made it very difficult to see how to port come sections of the code. Once I had this source I could port the code to C#. I had to convert all the buffer accesses to use .Net Streams, along with the code which read buffers. This included creating special classes with hold the data from the internal structures of the CHM File Stream, and manually reading the data from the CHM file into the data members of the class. (Using BinaryReader.ReadUInt16 etc).

Once I had a version which compiled in C#, I then had to test that the code worked. This was a case of running both versions of the library side by side in 2 instances of Visual Studio 2003 and extracting files from the same CHM file, making source that the output was the same.

Once done, I replaced the base stream handling used by the HTML Help library with the ported library, and then had a working CHM Reader which could extract files, extract help indexes & perform searches using the HTML Help library.

The CHMReader

Now that I had the basics working, I then when on to develop the Reader it’s self. Once you have the filename of the page you want to display (From the Index, Contents or search) you can then extract that file from the CHM file. This file then as to be scanned, and the images, frame pages & Style Sheets used by the HTML page need to be extracted, and the HTML patched so that they point to the new location of the files. In this case a temp directory in the filing system. Also at this point I remove XML and Script tags as they are not supported. Once this is done I can then save the file, and use the OpenNetCF library’s HTML Viewer control to view the page as intended.

There is more to it that what I have described, so look that the code. The code is not optimized as yet. This is some that may or may not get done depending on time.

If you can not read a CHM File, please debug the code and post any changes, If you really can not get the CHM file to view, then please email me with the CHM file.

I hope you enjoy this version.

History