[ros-users] dump of ros-wiki?

Eric Perko wisesage5001 at gmail.com
Wed Jan 5 02:55:48 UTC 2011


On Mon, Jan 3, 2011 at 1:59 PM, Brian Gerkey <gerkey at willowgarage.com>wrote:

> On Mon, Dec 27, 2010 at 8:24 AM, Brian Gerkey <gerkey at willowgarage.com>
> wrote:
> > On Tue, Dec 21, 2010 at 10:48 PM, Mischa Schaub <mischa.schaub at fhnw.ch>
> wrote:
> >>  I will be offline for several days but would like to work during this
> time
> >> with ros.org/wiki – how could I get a local dump of your great site on
> my
> >> laptop?
>
> > Anybody know a good way to archive a large
> > MoinMoin wiki?  If you can prescribe a way to do it (maybe it's just
> > tarring up the MoinMoin directory hierarchy on the server?), we can
> > bring up a cron job to do it regularly.
>
> We have a first cut at an offline archive of the ros.org wiki:
>  http://www.ros.org/roswiki.tar.gz
>  http://www.ros.org/doc.tar.gz
> The first tarball contains the wiki, rendered into HTML by MoinMoin's
> export tool.  Unpack it, then open `roswiki/index.html` in a web
> browser.
>

This is an excellent idea!


> The second tarball contains the auto-generated documentation pages (we
> separated it because it's quite big). If you unpack that tarball
> inside the `roswiki` directory, then links to autogenerated docs
> (e.g., API documentation produced by Doxygen, message/service
> documentation) should work.
>
> There are still some rough edges (e.g., ending up in a directory where
> you have to manually click on `index.html`).   Please have a look at
> it and provide feedback.  Once we have an archiving step that we're
> happy with, we'll start producing a new archive regularly.
>

Things seem to be okay for a first cut. Some issues I found: There are some
broken links to images (the one that I noticed so far is to the logo, but
that's not really a big deal). Also, it might be nice to include the CSS in
the dump as well. Some things might be harder to read (for example, I can
see the code examples in the tutorials being trickier). Might also be good
to include any required javascript as well, since it seems like the Ajax
that makes the version switcher works isn't available either. For example,
in the camera1394 docs, clicking between cturtle and diamondback doesn't
change the page. I'm assuming that anything not served from MoinMoin didn't
get exported, which would be why things like the CSS or JS didn't get
exported.

Another thing that would be really nice is to have this available somewhere
in _not_ just a tarball. I can see myself wanting to connect up a computer,
sync my local copy of the docs/wiki really quick, then disconnect from the
internet and get back to work. That would probably go quicker if we had
something that could do incremental changes and only transmit the docs that
actually changed. Perhaps stick the docs in a git repository so that people
that are updating only need to pull changed files or just have a readonly
rsync setup that one could pull from. Actually... putting it in a git repo
hosted on github might be a good idea anyways so that users (at least those
that know it exists) could still access it whenever ros.org goes down
(taking the main wiki with it).

It could also be nifty to see a ros package that could make it really easy
to set this stuff up and make it more useful. For example, a package could
include a facility for updating (or just always include the latest version)
of the offline docs, as well as a script that would startup a simple web
server serving content out of the offline docs directories. This would make
it possible to do something like... 'rosrun roswiki docs_server.py' and then
go to http://robot_hostname:8080 in your browser and get the offline docs
anytime you are connected to a robot and have the docs server running.

So that's my two cents on some places to go with this whole offline docs
idea.

All in all though, just having the dump as it is currently is gonna come in
handy for me. Thanks Brian.

- Eric


>
>        brian.
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20110104/57169631/attachment-0003.html>


More information about the ros-users mailing list