On Wed, 30 May 2012 16:20:23 +0200, Juan Jose Garcia-Ripoll wrote
> On Sat, May 26, 2012 at 10:36 AM, cm <xiao.ao.jiang.hu.cm@...> wrote:
I save the file with windows notepad with default code. And load file into ecl
with :windows-cp936, it works fine!
> but when I save the file with utf-8 format in notepad, and load file with
:utf-8, it report error:
> Problem identified. Notepad inserts a BOM character (0xFEFF) at the
> beginning of the file. This is not standard, but just something that
> Windows uses for UTF-8 and which is discouraged in Unix.
No need to discuss the smartness of Microsoft's editors ...
> I cannot cope with this at the level of encoding, because that would
> break Unix applications and would make ECL work differently on Windows
> and on Unix. Instead I have hacked a simpler solution: BOM is
> interpreted by the Lisp reader as a whitespace.
... but this is not a hack - the BOM _is_ zero-width nonbreaking space and hence
should be treated as whitespace.
Cheers, Ralf Mattes
> I will upload a patch tonight or tomorrow. Thanks for providing me with the
three test cases. They are really useful!
> Instituto de Física Fundamental, CSIC
> c/ Serrano, 113b, Madrid 28006 (Spain)
R. Mattes -
Hochschule fuer Musik Freiburg