Why do you want to create a service for the table printer. You can
create it with "new". As service typically makes only sense when the
initialization is complicated or you want to hide the implementation.
Both is not the
case for the tables.
So on trunk you can just use it and we can also backport to 2.2.x if
necessary.
Christian
Am 05.04.2012 20:21, schrieb Reuben Garrett:
thanks, everyone, for your feedback!
it seems that to partially solve my immediate problem for the Camel
Karaf commands, i should just assume a certain console width (e.g. 80
chars) and (statically) optimize the columns based on the width of
expected content - e.g. "status" columns will be shorter due to
limited enumerated values, and "name" columns will consume the
remaining width due to arbitrary user-dependent values.
then, for the long-term general solution (benefiting other projects
than just Camel), we could pursue publishing as a service the new
table printer that Christian has been working on. i'm not sure if
this is immediately within my abilities, but it's an endeavor i would
certainly follow.
would it be agreeable to open a Karaf JIRA requesting this kind of
modularization for the table printer somewhere down the line?
~ Reuben
--
Christian Schneider
http://www.liquid-reality.de
Open Source Architect
Talend Application Integration Division http://www.talend.com
