Encoding question

Wesley Shields wxs at FreeBSD.org
Fri Jan 20 13:32:04 UTC 2012

On Thu, Jan 19, 2012 at 08:43:14AM -0900, rflynn at acsalaska.net wrote:
> Hi,
> > I'm trying to compile a C++ software on FreeBSD. While compiling, this
> > error shows up:
> >
> > error: stray '\357' in program
> > error: stray '\273' in program
> > error: stray '\277' in program
> >
> > This file is reported (by file[1]) to be "UTF-8 Unicode (with BOM) C
> > program text, with CRLF line terminators" while the rest of the files
> > in the package are "ASCII C program text, with CRLF line terminators".
> > While I can convert the file with iconv -c -f utf-8 -t ascii file >
> > new_file in the post extract stage, I wonder if there is a more
> > suitable  way for achieving the same thing. Also I would like to avoid
> > this software from depending on iconv.
> You have three options:
> - have it fixed upstream;
> - post process on extract like above;
> - post process releases and roll your own tarball which you host yourself.

Be careful doing this third one. Often times (but not always) upstream
locations use mirrors to make sure the distfile is always available. If
you roll your own, just to fix something which can be so easily patched,
and you don't provide similar mirror functionality then you are
introducing a single point of failure we should try to avoid.

Not to mention you're now going to have to do extra work everytime there
is a new release. Either fix it upstream or patch it.

-- WXS

More information about the freebsd-ports mailing list