Experimental Physics and
Industrial Control System

"Hill, Jeff" <[email protected]> · Wed, 15 May 2013 00:28:36 +0000

> I would consider adding a build configuration parameter in the 3.15 version so
> this could be enabled by those who want it, but I wouldn't want it to be the
> default, which I think you suggested would be acceptable.

That would certainly be acceptable; choose the default as you see fit, or prefer.

The code blocks IDE actually has two build message windows; one of them provides raw build output and the other is condensed for fast indexing by the user (removing long paths from the users view, but retaining them internally in its database for fast access to the source code lines via the mouse).  The Visual C++ IDE has a similar approach. This is usually a good idea but sometimes vital messages are filtered out of the repackaged-for-fast-indexing window in my experience.

> The tricky question is how far to go with using it — do the dependencies need
> to be absolute?  What about calls to flex and antelope?  Resource compiler?

After I made that change C::B works fine with C/C++ source files and also so far with all of the C/C++ header files source errors that I have clicked on. 

If the dependencies aren’t broken, then I wouldn’t change them.

Probably source file types used for large projects should be the main focus; others can perhaps be on a case-by-case basis as users complain. However, if the absolute path option isn't the default then maybe you aren’t as probable to break anything by default, and consistency will be good. Nevertheless, avoid making this into a project.

Thanks,

Jeff

> -----Original Message-----
> From: Andrew Johnson [mailto:[email protected]]
> Sent: Tuesday, May 14, 2013 5:16 PM
> To: [email protected]
> Cc: Hill, Jeff
> Subject: Re: using the EPICS build system with an IDE
> 
> Hi Jeff,
> 
> On 2013-05-14 Hill, Jeff wrote:
> > Furthermore, this simple patch to configure/RULES_BUILD appears to
> produce
> > IDE compatible compiler output, which appears to be an improvement,
> from
> > my perspective.
> 
> I can see the attraction for an IDE user, but for a regular command-line user
> this pushes the gcc error messages way over to the right and in most cases on
> to another line (note that the word-wrapping below does not match my
> terminal
> output which is more ragged):
> 
> Original:
> ../../../src/std/softIoc/softMain.cpp:9: error: ‘syntax’ does not name a type
> 
> Using $(abspath $<):
> /home/phoebus3/ANJ/epics/base/mirror-
> 3.15/src/std/softIoc/softMain.cpp:9:
> error: ‘syntax’ does not name a type
> 
> I would consider adding a build configuration parameter in the 3.15 version so
> this could be enabled by those who want it, but I wouldn't want it to be the
> default, which I think you suggested would be acceptable.
> 
> I have a simple and working solution which I will discuss with Janet tomorrow.
> The tricky question is how far to go with using it — do the dependencies need
> to be absolute?  What about calls to flex and antelope?  Resource compiler?
> 
> - Andrew
> --
> It is difficult to get a man to understand something, when his salary
> depends upon his not understanding it. -- Upton Sinclair

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System