How is the site.xml read by manager
I introduced an issue today whereby i put a site.xml file in a temp directory (that i created) in the sites folder (e.g. C:\Program Files\Latitude Geographics\Geocortex Essentials\Default\REST Elements\Sites\temp\sites.xml).
The source of this file is from a site starting with TT_sitename.
The issue i introduced was that when going back to edit my original site, it always made reference to the site.xml in the temp older. This made me think, how is manager identifying the site.xml in relation to a specific site? Does it not consider the full site path & not just the first site file with mathcing site ID that it finds? I appreciate two sites should not have the same name but is I could argue this is a slight shortcoming in the manner in which the site.xml file is read into manager without validating the site path/folder?
Brad
-
Further to this, is the site listing in manager just kept in memory (I have a vague recollection this sites directory is parsed by a service regularly?) or is it a physical file list residing somewhere?
Brad
0 -
The site list is constructed by scanning the Sites folder of the Essentials REST installation, then assembling a list of Site IDs by reading the Site.XML files that it finds within.
If you copy a site by manually copying the site folder, you will get a duplicate site because the ID of the site (in the Site.XML file) does not change. REST Manager requires that the ID of the site match its folder name. When you save a site, REST Manager will save it to a folder that matches its site ID.
We recommend that you use the import site functionality built-in to REST Manager to make copies of sites. We do have a new feature request in our system to add functionality to export and import a site - if folks would like to have this, either post a response to this forum thread or contact your account manager and let them know!
0 -
Malcolm,
I like the idea of an Import site option - but what is the difference b/w using the copy site option in the rest manager vs manually copying the site. You mention manually and I take it to mean copying in explorer - not in the manager??
At present, we just copy the sites using manager and create a new viewer and import the contents into this (manually) and alter the site uri... seems to work fine for us.
cheers
G
0 -
Hi Gareth,
Site copying will make a copy of the Site.XML file according to how REST Manager would manage it. It does not copy any ancillary linked items or files within the site, nor does it copy any viewers attached to that site. It's meant to make a fresh copy of a site so you could start a new one.
Site export/import would assemble a ZIP file that contains everything necessary to recreate your site. There would have to be some limitations, like "only the contents of the Sites folder" and "may not recognize all replacement tokens", but the idea would be that you could then take the resulting ZIP file and re-import it locally (to clone a site), or move the site into a different named instance or different computer.
It would simplify migrations and site backups, and also make it easier to share configured sites amongst the user community.
-Malcolm
0
Du måste logga in om du vill lämna en kommentar.
Kommentarer
4 kommentarer