A free tool, DokuWiki2HtmlHelp.exe, is available to export a complete DokuWiki website into HTML Help. The Delphi source code is available under the GPL license from the same location.

1. uses ?do=index to get the first page of the index
2. parses the index pages to build a list of content pages
3. parses the the content pages for css and images
4. strips out the stuff that is not Html Help friendly (scripts, forms, etc.)
5. converts the links in the page to local links (if they are in the wiki)
6. saves the pages locally in folders to replicate the namespaces
7. builds the Html Help project file along with all of its supporting files (table of contents, index, mapping, etc.)
8. downloads and renames all needed images and css
9. calls Microsoft Html Help Compiler and generates the Compiled Html Help file (.chm)

Microsoft's Html Help Workshop (available free from Microsoft) is required for the last operation.

I have only tried it on my own DokuWiki sites.

Right now it only works if the wiki is using the ':' namespace delimiter.

You need to disable url rewriting in the wiki configuration:

$conf['userewrite'] = 0; 

If the application hangs while downloading images, you most probably referenced an image in your page that doesn't exist. (The app tries up to 3 times to fetch any given file before continuing on to the next one in the list. It will iterate the list up to 3 times trying to fetch any files that failed, so ultimately it will try up to a total of nine times for each image. –bjm) So, if you're quick you can check the webserver logs and fix the problem while the app waits.

Version 2008-05-22 or later is required for use with dokuwiki-2008-05-05 and later.

It does not work on https protocol.

The author, Bruce J. Miller, can be reached at: Rules of Thumb,inc.

While using the program I encountered the following error: I input the path for the Html Project and my dokuwiki installation: “J:\HTML Export\” and “https://myserver.com/dokuwiki/doku.php” When pressing next this error message is displayed: “Cannot open file “J:\HTML Export\Index.htm”. File not found”

url rewriting is disabled and the namespace delimiter is set to ”:”

Anybody has some suggestiosn what i should do next?

maybe you can try with a foldername that does not contains spaces ?

The spaces in the file path should not matter. It's more likely to be an issue using secure sockets (https). Did any files download, or did you get the error message immediately? –bjm

Thanks for your help! The error occurs immediately, but there is one file written into the export folder, a html.ini with this content:

[PendingIndexPages]
?do=index

[ProcessedIndexPages]

[PendingContentPages]

[ProcessedContentPages]

[PendingImages]

[RetrievedImages]

[PendingStylesheets]

[RetrievedStylesheets]

[css]

[images]

I think the ?do=index command is not working, and so there are no pages parsed at all. Probably this is because of the secure socket and the lack of access rights to the server. I will try the program on another server and see if it works without the secure transmission protocoll, at least as soon as I get the server running. -Chris

Creating an empty file named “index.htm” in the same folder as the output is being directed to seems to solve the problem.

I've encountered the same problem with the Cannot open file “G:\foo\Index.htm”. File not found I created a index page in the wiki and then the export went past this error.

After that I encountered problems when parsing scaled pictures. The first 21 pictures pass without problem and on picture 22 it stalls. I am not sure if the program is still doing something. As I cant see the whole path I don't know which picture causes the trouble.

Are missing pictures a problem I assume not, but who knows

seems like I should have read the top part on how it works. In the past I moved some pages and then moved the pictures to a different namespace and now the pictures are not liked correct on some pages. Is there a way on how I can find links to images pointing to a non existing picture — Sebastian Spiess 2008/05/21 03:12 Yes, there is an .ini file in the main export folder. It should contain a list of missing images in the [PendingImages] section. –bjm

→ See also related page export

Olivier Meunier wrote a PHP class for a conversion from wiki to XHTML: wiki2xhtml

There is an example available.

The above links are dead.

As you can see this class doesn't work out of the box for dokuwiki: the syntax is not the same, main difference are headers:

!!!title versus === title ===

But it could be a good starting point…

uhm do you know about ?do=export_html ?

no I didn't… Thanks for the tip

This could be a good place to start to write an HTML export script (using do=export_html) for the whole site (so links are rewritten to point to static pages instead of the doku.php script.)

If I do a static dump of my wiki with a website spider and use it offline without php, the DokuWiki based search is not working anymore. Does anybody has an idea how to create an offline (client-side) search? Probably java-script based?

Answer my own question, I found:

• Zoom ($49-$99) (my first pick)
• HTTrack: open source web-site copier and browser; runs on Windows, Mac OS X and Linux(free)
• Search Maker Pro ($59)
• House Spider (free)
• KSearch Client Side (free)
• The Webside Utility ($10)
• Search engine studio ($500++)
• namazu (free)

— Konrad Bauckmeier 2005-12-21

even if you just want to mirror the website, i tried httrack with default options, it worked perfect on my dokuwiki website! even with rewriting on and a totally custom design. great!

when you have a local installation/html files you could use a desktop search engine like Google Desktop, Beagle, Tracker depending on you OS… works fine with me. — seb

I wrote a program allowing to export a complete dokuwiki to XHTML; http://dokukiwix.mandragor.org/ It works as a plugin (although it needs to add 2 lines of code to dokuwiki's core :() , and you can customize the export HTML just by editing a special template. This plugin also retrieves external images, keeps their size, etc.

Let me know if you use it! yann point hamon at gmail :)