For the linux kernel, there is already an EDE project that can
recognize it, and will set itself up. Since I personally don't hack on
the Linux kernel, it would be great if someone who did could use it and
recommend changes or provide patches. You can validate that EDE is
active with the command semantic-c-describe-environment which will tell
you a bout the ede project used to setup things like the CPP (C
For the purposes of getting everything parsed ahead of time, you can
use GNU Global as a sort of primitive raw back-end. Semantic will use
it to find locations while looking stuff up the first time, then it will
use it's own tables once it has looked up that file once. Use M-x
semanticdb-enable-gnu-global-databases to get that started after
creating the first database using global.
Alternately, if you have files open in a particular directory,
Semantic will start parsing them in idle time. A side effect is you
can't open Emacs in a new area and just start navigating. You can force
it with semantic-debug-idle-work-function which no one has converted to
a more utilitarian use as you describe below.
As for just looking for "function details", as long as the include
path is setup correctly, this should happen automatically with no extra
setup. It will get the info from header files which it will scan
on-demand by interpreting your #include statements. It is only the
'jump to implementation' that doesn't work that way.
Aneesh Kumar wrote:
> I am trying to configure cedet with linux kernel source. My
> expectation is to be able to replace
> gnu global as the source tagging system and start using the database
> generated by semantic.
> I am using the gnu emacs from cvs which have cedet integrated. I am
> able to get tag lookup
> work with the file i opened in emacs. But finding tag globally doesn't
> work. I think that is because
> i haven't generated semantic data base for all the files in linux kernel source.
> So my questions is do we have a set of steps that one need to follow
> to get a large source code
> like linux kernel or emacs added as projects under ede and have
> semantic do a global tag lookup
> for symbol. It would nice to have something called
> import-non-build-project that just add enough
> entries with ede that is used by just semantic (something like simple
> project), but enables to
> rescan all the project files and also allows to rebuild the semantic
> database on rescan.
> I have already looked at
> and that didn't really help me to get a working setup.
> In short what i am looking for is something similar to simple-project
> but for which i can specify
> the directories which i would like semantic to generate database for.
> Then i do a rescan project
> files and it will generate the semantic database. After that search
> tag globally will locate the
> function details. Is this possible with current CEDET ?
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast and easy
> Join now and get one step closer to millions of Verizon customers
> cedet-semantic mailing list