Hi guys,
first of all: I hope this is the right place to ask this question, if
not, please give me a hint where to go and bear with me.
In our company, we are currently running an Eclipse 3.4 installation
for several hundred software developers on Unix workstations. The
installation is shared from a network location via NFS. The needs of
the different developers are very different, some are C++ guys, some
are into Java or other stuff.
To provide the different groups an uncluttered IDE, I did NOT use p2 up
to now, I removed it manually as described on the Wiki. But now, as we
want to move on to 3.5, there is no way without p2, right?
To give a better understanding of our setup, this is what we did up to
now: There is a main installation directory for the platform. Then,
there is a separate directory for CDT, and one for JDT, and some more.
The individual user shall NOT fiddle with these install directories,
they are read-only for him. In our startup script, for each user a
special Eclipse configuration directory inside the user's homedirectory
is created, with a platform.xml that contains references to all
the extension locations that this particular user needs. This way, one
user only gets the CDT features, while another one only gets the Java
IDE, all with the same platform install.
Bottom line: The user must not configure anything manually. Our
launcher script analyzes the user's environment and creates the
corresponding Eclipse configuration (platform.xml) on the fly, without
any user interaction (and, of course, without any GUI). Only if that is
done, Eclipse itself gets launched.
Is there a way to achieve the same with Eclipse 3.5?
I'd be very thankful for any pointers or hints.
-Achim
|