Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] Help with Indexer Bug

Is there any noticeable impact on speed?

Also, does it add any incorrect values?  I'm not sure if the parsing context when this option is enabled will be the
same as what the compiler will see.  In particular I don't know what happens if you are editing a header then I don't
know the parsing that is triggered by this option would know about symbols that were defined earlier in the compilation
unit.

-Andrew

On 13-10-03 11:57 AM, Guy.Bonneau@xxxxxxxxxxx wrote:
> I just tried with a  complex CDT project of a co-developer that had many
> unresolved symbol...And yes symbols are now resolved. That helped a lot.
> Thanks for the suggestion.
> 
> Guy
> 
> 
> 
> From:	Andrew Eidsness <eclipse@xxxxxxxxxx>
> To:	cdt-dev@xxxxxxxxxxx,
> Date:	2013-10-03 10:54
> Subject:	Re: [cdt-dev] Help with Indexer Bug
> Sent by:	cdt-dev-bounces@xxxxxxxxxxx
> 
> 
> 
> Out of curiosity, do you run with 'Window - Preferences - C/C++' then
> "Follow unindexed header files when producing the
> outline view" enabled or disabled?
> 
> This is a preference that I recently found (when trying to "modify the
> indexer code like Joseph did").  It seems that
> enabling this causes more header files to be indexed.  Normally (i.e.,
> other than during parsing) when a header is not
> found in the index, the fallback handling is to lookup the actual source to
> reparse it.  The value of this preference
> seems to decide whether that "lookup the source" step returns the real
> source or an empty string.  The default setting
> is off (I guess to be faster), but I think I'm getting less unresolved
> errors with it enabled.
> 
> This is just my observation of behaviour, I haven't actually studied the
> difference; which is why I'm curious about your
> setting.
> 
> -Andrew


Back to the top