[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Difference between GCC 2.95 and GCC3.1

From: Richard Frith-Macdonald
Subject: Re: Difference between GCC 2.95 and GCC3.1
Date: Fri, 28 Jun 2002 07:45:44 +0100

On Friday, June 28, 2002, at 06:29 AM, Richard Frith-Macdonald wrote:

On Thursday, June 27, 2002, at 11:23 PM, Tom Hageman wrote:

2. The change is a bugfix ... the old version was incorrect, quoting
character which
should not be quoted.

Are you sure it was a bug? On OPENSTEP 4.2 the output of the above test
program is:

Jun 27 23:49:59 gnustep-plist-test[4564] The Plist is {EX = "-16"; }

No ... I don't have an old OPENSTEP system to play with, and have never
found detailed documentation on exactly which characters should be
quoted.  The existing code quotes only those characters that need to
be quoted to ensure that the string can be parsed unambiguously.  The
old code failed to parse some strings I was told were legal ... it
was quoting (and expecting to be quoted) characters it shouldn't.

If you can find some detailed documentation on exactly what characters
are quoted in OPENSTEP, I'd be very interested.  Ideally I think we
should generate quoted strings in the same manner as OPENSTEP, but read
and parse strings more forgivingly as we do now.

PS. I've hacked in a change to ensure that almost anything non-alphanumeric gets quoted on output, but I'd still like to know how OPENSTEP quotes or doesn't quote all characters. Perhaps a test using all the Latin1 character set and
a few unicode characters not in that set?

reply via email to

[Prev in Thread] Current Thread [Next in Thread]