revweb plug-in for linux

Sivakatirswami katir at hindu.org
Thu Oct 28 21:21:09 EDT 2010


  On 10/28/10 11:52 AM, Monte Goulding wrote:
> Is the server mounted? If so an alias may do the job.
>
> Another option is to build a standalone that loads the stackfile off the server. There are many ways to kill this cat.
>
> Cheers
"In House" "LAN"  ??

I second Monte's motion on this one. I must have a dozen programs 
running in house that are deployed by just creating little splash screen 
standalones ..

  I put the stack on a web site on the server, but no one actually views 
anything in a web browser.  (well occasionally I pass off some resource 
to the web browser) Just to be clear for my own brain, I call the 
widgets that are deployed to client station "Launchers"

Our macho OS X Server is running apache....

I put the stand alone launcher on the web server, I keep all these on a 
single web page

"Rev Stacks"

with links to the launchers, one URL per standalone

[I tried a single launcher once, that downloaded an "index" stack for 
all other stacks, but it never worked well. People are so used to having 
a single "app" for each "thing"  that now I issue a special splash 
screen for every new app...]

e.g.

Grapple_Launcher.app

has  a nice graphic on the main card  and this in the stack script

on preopenstack
    set the loc of this stack to the screenloc
    send "LaunchGrapple" to this stack in 1 seconds
end preopenstack


on LaunchGrapple
    go stack url "http://inHouse.domain.org/rev_stacks/grapple.rev"
end LaunchGrapple

By using the web server you avoid all the issues of mounting the server etc.

Now, disclaimer: most of my stacks are very light weight.  After the 
launcher launches the mainstack, if he needs more resource from the 
server, then I'll run an apple script from inside the stack as  needed 
and mount the main server and then call the "heavy" resources using the 
file system

e.g.

set the filename of player "Grapple Tutorial" to url 
("file://Volumes/LanServer/WWW/inHouse.domain./Grapple" grapple.mov")

Note that even though we are not using a web browser, I'll maintain 
various resources in the WWW area, even though we are not pulling them 
via a web browser, but I *can* then if necessary, pass something off to 
the web browser...

this works great, all the resources are modular, there is nothing on the 
client station except  a standalone with the above script. I can update 
and swap out resources on the server and every one is always up to 
date.  It's all transparent to the users. They don't even realize that 
the app they are using is actually being call runtime from the server.












More information about the use-livecode mailing list