(moved from SupportForum)

4/16/04

I use wiki to organise my life, and it works! I chose ErfurtWiki because of its PlugInterface. I want to make my website wiki-based too, and not having a host with php support, I'd like to try the WikiDump plugin. However, I seem to have trouble getting it to work. I do get a .tar.gz file, but the contents are unprocessed WikiMarkup inside a .html file. Is this all the WikiDump plugin does or is something wrong with my server? I'd like to have a static-pages export function like PWP Wiki Processor has. However I don't want to switch to a new wiki again and learn new WikiMarkup. I'd much rather continue using this.

Sorry 'bout hitting the save button without finishing what I was typing earlier.

Yours, Ewout

milky: No, that is supposedly not what it should do. Sounds like it got broken somehow. I'll check for this bug tomorrow.

But, btw, did you try tools/wiki2html which does basically the same thing. Also we already have the option of exporting all pages via WikiXmlRpc, though there is yet no fancy frontend for it (eventually some other WikiWare however has an export utility based upon that XML-RPC interface). escali high capacity bathroom

Whatever, don't worry. We get that running again ;-)

milky: I've fixed multiple bugs, an updated version is now in CVS.

4/20/2004 Ewout: I've downloaded the CVS tarball and given it a try. It doesn't seem to work properly, as it is only returning a few pages from the entire wiki. Furthermore, the file is named wikidump.tar, but it should be wikidump.tar.gz in order to work properly.

However I am getting proper results using the wiki2html tool from the tool directory. To run this file on my win32 system I made a batch file (convert.bat) with the following commandline in it:

W:\usr\local\Php\php.exe wiki2html

Note that this file needs to be put in the same directory as ewiki.php if you use flat files, otherwise you'll get an error. And the path to php.exe may be different on each system.

However I wanted it to integrate my design too in the output. But the wiki2html sourcecode told me to use wget or something similar so I will give that a try.

milky: If you want to integrate your design, you'll need to edit either the wiki2html script or the WikiDump plugin - both provide that minimal plain white theme. I know this is awful, therefore I recommended wget; but I've tried to use it myself that way and I failed with my first attempt. If you want to use 'wget' then you need: Pest Control

• a command like 'wget -v -c -x -E -r -p -k -np http://example.com/wiki/'
• set EWIKI_SCRIPT to "/ewiki/index.php/%s.html"
• explicitely enable *EWIKI_USE_PATH_INFO (which is off for Apache per default)
• later remove all the unwanted pages/files that 'wget' retrieved
• suncast outdoor sheds
• Pet Supplies
• Car Seats
• metro vacuum filters
• Knives

Happily wget works quite well even under Win32, and after all it's the best tool to do that job ("you can always wget a static version of any web site"). bissell proheat steam cleaner

For wiki2html I would recommend, that you write some automated post-proccessing script, which simply adds your design to the thrown out pages. (note to myself: It was better if I added real templating functionality to it for the future.)

In the WikiDump plugin I've changed the extension from .tar.gz to just .tar, because my Mozilla decompressed the downloaded tarball on the fly (silly). I'll change that back, but otherwise it worked for me. I had the just-a-few-pages problem earlier myself, but thought it was fixed now (probably related to wiki page flags). I'll investigate a bit further...

But meanwhile, thanks for all your effort in testing it out; helps me a lot.