OK - we are also trying to formulate a question for our hosting company, hoping that they may know some reason why the http behavior is different than the cron behavior. But we have some gaps in our understanding of how the api works. Here is what we are intending to ask:
We are running a php script in a cron job and the results are different than when we run it in http mode (through a browser window). The script calls two variables %msglink% and %_site_% each of these variables are supposed to call the domain uri of the site (
www.domain.com
). The %site% variable is only supposed to call the domain part (
www.domain.com
), while the %msglink% variable is supposed to contain both the domain and additional path parameters, like
www.domain.com/index.php?option=com_uddeim&task=show&Itemid=100050&messageid=48557
.
Unfortunately, when we run the script from the cron job, %msglink% fails to bring the
www.domain.com
part of the url; it only brings the index.php?option=com_uddeim&task=show&Itemid=100050&messageid=48557 part of the url. When run from http, it brings the complete url.
The %_site_% variable brings the domain information in both cron and http modes (the way it is supposed to do).
The script that creates these links is as follows:
%_site_% script:
[here is what we don't know]
%msglink% script:
[here is what we don't know]
Can you think of any reason or problem with the script that is creating this %msglink% url links?
Slabbi - can you provide the missing information on the scripts?