Can anyone explain to me exactly how cross site linking works as there is very little definition in any of the 6.7 documentation.
I have implemented my location schemes to test for a cross site link however does a business user still have to create the Active Assembly relationship between two items making sure that they select the two check boxes in the AA item relationship search window to include the site id and folder id in the relationship table?
If so can these checkboxes default to true as it seems a little insane that a user has to specify an item is a cross site link when linking to content on another site. i.e. of course choosing to link to content on another Rhythmyx site is a cross site link or am I missing some fundamental point?
I have implemented my location schemes to test for a cross site link however does a business user still have to create the Active Assembly relationship between two items making sure that they select the two check boxes in the AA item relationship search window to include the site id and folder id in the relationship table?
Yes
If so can these checkboxes default to true as it seems a little insane that a user has to specify an item is a cross site link when linking to content on another site. i.e. of course choosing to link to content on another Rhythmyx site is a cross site link or am I missing some fundamental point?
See patch RX-16016, bug id RX-15908.
RX-15908 - The following server properties are introduced to provide the default values for Include site in the link and Include folder in the link check boxes on template selection panel and search tab of Browse Content dialog.
The absence of these properties or not a true value is treated as false. The server.properties file is not shipped with these properties. The default behavior is false. Before you set these properties to true, please check with support on the impact of this setting on your site.
Although we (my organization) may have been responsible for the initial reports with percussion, we chose NOT to implement this. Why? Because if you default them to true, the system believes every link (created from that point forward) is a cross-site link. When you attempt a Remove from Folder action on a child item, the system will not allow you.
“Error executing action <Delete> : An unexpected error occurred while processing the CMS request. The error was: Action fails because the items with contentids: [36132] participate in cross site linking and require Force Remove action.”
This means you have to manually remove the associations in your content before you can remove the child from the folder. You could use the Force Remove action, but only an Admin can do that. Although, I’d rather not intervene in the content affairs of my 1400 users on a case by case basis. I’m not a big fan of the Remove from Folder action anyway, since essentially, the content item lives on in the aether, but you should probably be aware of this issue if you intend to implement the fix for RX-15908.
I was aware of the patch fix but it was exactly the implications of ading site and folder id to the object relationship table, as alluded to but not explained in the patch “read me” file that I was interested in. I have raised a support call regarding this.
It seems a little illogical to have to choose to include the site id when a user picks content from another site, since this action in itself cannot be anything else other than a coss site link. My issue with users having to take this step is the age old user generated error.
I presume that if the boxes are not ticked Rhythmyx will generate a link referencing an item of content in another site but create the link as if the item resides in the same site. i.e. Rhythmyx is completely dumb to where the content actually is at publish time and will generate a broken link.
It seems a little illogical to have to choose to include the site id when a user picks content from another site, since this action in itself cannot be anything else other than a coss site link. My issue with users having to take this step is the age old user generated error.
Agreed. Let us know if you make any headway with TS.
Basically, every URL is presumed to be relevant to the root of the site, i.e. href=“/…”. If you check the “Include Site ID,” you’ll get href=“http://somesite/…”.
If I remember correctly (and I could be completely wrong since we don’t configure the sites this way), site definitions can be nested which further complicates the issue of reliably and programmatically determining the site id. Instead of the fix for RX-15908, maybe it’d be more helpful to have a flag indicating that the particular implementation will never nest sites and add support for “auto” cross-site linking.
Hmm indeed.
So in actual fact all my if($sys.crossSiteLink) checking is totally redundant now as the business user has to specify this and Rhythmyx no longer bothers to check this fact.
That’s a major backward step (in my oh so humble opinion)!
Technically, your location scheme is not “redundant.” $sys.crossSiteLink is the indicator of whether or not the business user has marked the check box. I should have added a note that the location scheme must also use that parameter check to be effectual.