Hello all,

I have a large existing codebase which I would like to try with VisualGDB.  I am using the standard gcc/make toolchain.  The code is structured as follows:

- Library1
   - Library1.h
   - Library1.cpp
- App
   - App.cpp

There are already two preexisting Makefiles: one for Library1 which creates a static lib with “ar rcs” and another one that creates an executable with gcc -o App ../Library1/Library1.a.

Basically it is almost exactly the same as the example given here:

http://visualgdb.com/tutorials/linux/libraries/

However, all of the existing code is set up to include headers like this, i.e. from App.cpp:

#include “Library1.h”

And this is compensated for by passing a parameter like -I../Library1 to the gcc command for App.

In the Linux Libraries tutorial the header #include is dealt with using a full relative path i.e. “#include ../Library1/Library.h” (in step #9), but this won’t work for me as I can’t alter the existing build system/paths/etc.

Without this, Intellisense can’t open Library1.h and highlights it with the red squiggly.  I tried adding it as an Intellisense path, which worked, *however* this seemed to mean that Library1.h was cached locally and if I updated Library1.h as part of my normal editing, my local Intellisense copy was out of sync.  (It also showed up in VS2015 as a “system header” with a light blue tab all the way to the right of my window and I don’t want that either.)

So my question basically is:  how do I tell VisualGDB for my “App” project where to look for the static library header path, without using local Intellisense caching which is out of date?

P.S.  I also tried using the “VisualGDB-maintained” Makefile for “App” and added INCLUDE_DIRS := ../Library1.  This caused the build to work, but Intellisense was still not working.

What to do?

• This topic was modified 8 years, 1 month ago by Ophidian14.