Re: #include <> vs. ""

From: williamkempf_at_[hidden]
Date: 2001-07-14 07:55:09

--- In boost_at_y..., Neil <neil_at_e...> wrote:
> One more thing ... the difference (as I explained it) is for 
command line
> usage.
> That way, there is no question what the current directory is.
> Neil

There most definately is.  Read this entire thread and read the 
respective Standards documents.  Again, different compilers *DO* 
handle this stuff differently, and "command line usage" has nothing 
at all to do with it.

> ----- Original Message -----
> From: <williamkempf_at_h...>
> To: <boost_at_y...>
> Sent: Friday, July 13, 2001 6:55 PM
> Subject: [boost] Re: #include <> vs. ""
> 
> 
> > --- In boost_at_y..., Neil <neil_at_e...> wrote:
> > > Hmmm. Seems to be a problem with "" vs <> with include 
statements.
> > >
> > > As an "old-time" C programmer, the difference is as follows:
> > > #include <abcd.h> (compiler looks for abcd.h in the INCLUDE 
path)
> > > #include "abcd.h" (compiler looks in the current directory 
first,
> > then in
> > > the INCLUDE path).
> >
> > This is the behavior of your C compiler, not the behavior defined 
by
> > either the C standard or C++ standard.  That's the crux of
> > the "problem", both standards leave the actual behavior up to the
> > implementation with only a few very minor requirements/differences
> > between the two.  For instance, most compilers I'm aware of use
> > behavior similar to what you list above, but each defines "current
> > directory" in it's own way, some meaning the directory the 
compiler
> > was executed from, some meaning the directory the source file is 
in,
> > and some meaning the directory in which the file currently being
> > parsed is in.  The exact behavior can make a lot of difference 
for us.
> >
> > Bill Kempf
> >
> >
> > Info: http://www.boost.org  Unsubscribe:
> <mailto:boost-unsubscribe_at_y...>
> >
> > Your use of Yahoo! Groups is subject to 
http://docs.yahoo.com/info/terms/
> >
> >
> >