Whilst updating the Provisioner 3PC for revo RC2 release I managed to hack together its back end a bit to produce the screen shot below as a proof of concept idea.
The screen shot shows a resource tree with with some dummy data in it as produced from an Evolution install, NOT a Revolution one. The data is dummied at the mo but the principle seems to be OK. Basically we create an assets/connectors folder in Evolution, populate this with a hierarchy of getnode/list/id processors a la Revolution, point Provisioner at it, read the Evolution database using SQL, not XPDO, massage and return the data in JSON as expected by Revo and off you go.
Ok, theres no contexts etc. but we should be able to do basic data imports between Revolution and Evolution of resources, elements, user data and the like, i.e common stuff. This may help people convert Evolution installs to Revolution ones.
There’s lots more to do here so this won’t be done for Revo RC2, also you will need to install the processors in to any existing Evolution install you want to import into revo, its not automatic like revo to revo, evo needs the extra support. There’s also security considerations here of data visibilty but I’ll leave this initially to get something up and working. It should work with most evo installs, even older 0.9.xx stuff.
Use MODx, or the cat gets it!
Very cool. Nice work.
Ok, we’ve now got an Evolution resource tree in the Provisioner 3PC. The attached screenshot shows the tree of the example site content install of an Evolution 1.0.3 site. Notice also that there are two menu entries on the resources now, ’Import’, as the original provisioner and also ’Import Convert Tags’, this imports the resource but runs it through the tag conversion script first. Hopefully this will help with tag breakage coming from Evo to Revo, at least in simple cases.
The mechanics are the same as for a revo site, first you have to log in with a manager account, but now you select the type of the remote site, either revo or Evo, see screenshot2. The controller in the remote evo site checks for a manager login before allowing any data transfer, much the same as revo does this.
We also create a dummy context called ’Evolution’. This sits at the top of the tree with the Evolution resources under it. There will only be one of course here and its not a ’real’ one, it just keeps the interface the same.
It should just be a matter now of implementing the various processors needed for elements/files etc. Hopefully I’ll be able to release this in the next few weeks.
Use MODx, or the cat gets it!
wow! shamblett, looking really nice..!!
Website: www.mercologia.com
MODX Revo Tutorials: www.modxperience.com
MODX Professional Partner
Nice work. Can´t wait for it to appear.
OK, I’ve released this into extras as version 1.4(may take a little time to appear). Upgrade as normal and install the gateway connector pack into any Evolution site you wish to provision from. Please have a quick look at the user guide, I’ve attached it here until the package becomes available.
Use MODx, or the cat gets it!
