1

We have our own intranet server with wikis on it. It is accessible from the outside, but we don't host our webpage on it. We keep most of our data in sync there, it's great for collaboration.

We marked some of our documents public and we want them accessible to the world, on our "hosting" based website.

We want to "mirror" the public articles (the ones that are visible when nobody is logged in) from this server in our office to our website somewhere on the internet.

Note: It would be nice to have some kind of cron job or something to have the hosting pull public data from the server every 4-8 hours or so to a browsable copy on the hosting service. There the website can link to it and display it in a different manner (we use different css, etc)

Neanderthal solution: I was thinking of cloning the public intranet site with something like "Teleport pro" (an old 1997 tool that still works wonders:) and then copy it to the website.

unom
  • 281
  • 5
  • 13
  • 1
    what wiki are you using? – gravyface Nov 01 '12 at 12:29
  • Apple's Mountain Lion Server Wiki... I need something like an RSS feed of all the content from the server OR something similar to google cache. Just get the content you can from the page and display it. Server is down just keep displaying the old data. Server comes up again... reload everything. – unom Nov 01 '12 at 14:27
  • I can't agree on the "not constructive" clause below. I ended up getting my server dynamic dns on the web and using iFrames to the server from the main site... this offline to online sync/publish thing is rather hard to do... I think JavaScript would be able to grab web content BUT it has to be on the same domain name. Another solution would be to have the server run a kind of script and convert web data to JSON or XML and upload it to the website and have the website read it and display it. Too bad Wiki 4.0 on Mountain Lion Server does not publish project data via RSS or similar. – unom Nov 08 '12 at 12:20

2 Answers2

1

You could try wget which works great in such cases and is easy to put into a cron job. It's available in every Linux distribution.

Sven
  • 98,649
  • 14
  • 180
  • 226
  • Tried. It doesn't play well. My url look like example.com/wiki/projects and there is some javascript in there to boot... so it has problems with that too. – unom Nov 01 '12 at 14:28
  • Have you tried the `-k` / `--convert-links` switch with wget to convert the links in the downloaded content? – James Yale Nov 01 '12 at 15:49
0

If you're using MediaWiki the DumpHTML extension pushes the whole thing out to static files: http://www.mediawiki.org/wiki/Extension:DumpHTML

edit: Not sure how this would handle pages that require a logon though.

James Yale
  • 5,182
  • 1
  • 17
  • 20