This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
I like the current gcc handling of trigraphs
- From: "Thomas R. Truscott" <trt at cs dot duke dot edu>
- To: gcc at gcc dot gnu dot org
- Date: Wed, 5 Jun 2002 11:59:12 -0400 (EDT)
- Subject: I like the current gcc handling of trigraphs
[Just in case someone is planning changes to gcc trigraph handling...]
Please don't change the current gcc trigraph handling,
at least not with out more discussion.
In my opinion/experience, the C trigraph feature is a total mismake:
1) It solves a problem that no one actually has,
and/or one that can be solved at least as easily in some other way.
2) It adds an unintuitive complexity to the language.
I have learned that only a tiny fraction of C programmers
know what C trigraphs are, and exactly what they do.
printf("(Is this ridiculous, or what??)\n");
3) It causes real harm in real programs. For example see:
http://groups.google.com/groups?hl=en&lr=&selm=70jbai%24l9b%241%40hal.cs.duke.edu
The most helpful and portable defense against trigraphs is a
warning that they exist in the source code.
It is good for gcc to have -Wtrigraphs, and to have it enabled by -Wall.
A good (but non-portable) defense is an option to disable
trigraph processing. gcc has one, and the company for which I work
has asked other compiler vendors to provide one as well.
For example, newer Solaris compilers support -xtrigraphs=no
We should work towards removing this blight on the language.
Enabling trigraphs in gcc would be a step in the wrong direction.
Tom Truscott