2009/3/11 Clancy <clanc...@cybec.com.au> > On Tue, 10 Mar 2009 15:12:57 +0000, stut...@gmail.com (Stuart) wrote: > > >Please keep the discussion on-list. > > > >2009/3/10 Clancy <clanc...@cybec.com.au> > > > >> Hi, > >> > >> After I posted my message to the group today I realised that my program > >> achieves its > >> almost infinite flexibility by loading different include files in > different > >> circumstances, > >> and from many different parts of the program. The program always > specifies > >> the location > >> of the include file by the relative path from the root directory. If > the > >> uncertainty > >> about the current working directory you worry about were really a > problem > >> this would never > >> work, but I have been doing this for well over a year, and had never had > >> any problems. > >> > > > >Just because you haven't experienced a problem with the way you do things > >yet it certainly doesn't mean the problem isn't there. You're not > specifying > >the location of an include file relative to the home directory, it's > >relative to the current working directory which is not necessarily the > >directory you expect it to be. > > The only circumstances in which my assumption could fail would be if I > changed to a new > provider with a different operating system. In the unlikely event of this > happening all I > would need to do would be to determine the actual directory, and chdir to > the desired one. > This would require at most half a dozen lines of code at the start of the > program, so it > would not be a significant hassle.
Like I said it's up to you what you decide to care about. One of the many things I choose to care about is writing portable code, an important aspect of which is not relying on absolute file locations or server configuration as far as possible. -Stuart -- http://stut.net/