>>> =?ISO-8859-15?Q?Nathan_H=FCsken?= <semantic@...> seems to think that:
>Hello dear community,
>I am just getting used to semantic, and I have a few problems:
>- Whenever I call "semantic-complete-analyze-inline" it takes like
>forever (well probably more like 5-10 seconds) to show me the list. Is
>there a way to speed this function up?
The first time you run the completion engine on a file, it may take a
while to load in the tags for all the header files that it can find.
If your total collection of tag databases currently loaded is large
for your project, it is likely that semantic is not scaling well in
>- When I am in a class-methode, type this-> and call
>"semantic-complete-analyze-inline" it often gives me list like this:
>(TAB for more)
>Is this a know problem? Is there a solution?
Heh. That's funny. I'm totally not surprised, but neither had I run
the completion engine on classes with overloaded operators before.
The solution is for someone to add a filter on the tail of the
completion engine to allow C++ to strip operators out of the
All the other methods should be there too, so once you type an alpha
character, that ought to be cleared up.
I'll add it to my todo list.
>- How can I make all the functions of packages installed on my system
>known to semantic?
The current tool to use `semanticdb-create-ebrowse-database' against
directories you have in your include path.
This is much faster than the previous mechanism, but is also pretty
new and may exhibit problems that need to be addressed.
Due to the wide range of subtle problems in this area, I've been
working on a debugging system for it. Hopefully when I get it
working, it can be used to provide the detailed info needed to fix
problems that don't scale into a small piece of sample code.
Eric Ludlam: zappo@..., eric@...
Home: http://www.ludlam.net Siege: http://www.siege-engine.com
Emacs: http://cedet.sourceforge.net GNU: http://www.gnu.org