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

Re: Eh, what? on build.

To: "Henry B. Hotz" <hotz@jpl.nasa.gov>
Subject: Re: Eh, what? on build.
From: Dave Love <d.love@dl.ac.uk>
Date: Fri, 06 Feb 2004 12:39:54 +0000
Cc: heimdal-discuss@sics.se, George N Karas <George.N.Karas@jpl.nasa.gov>
In-Reply-To: <p05210603bc48732f3e39@[137.78.212.225]> (Henry B. Hotz'smessage of "Thu, 5 Feb 2004 14:51:50 -0800")
References: <40150AF7.8020803@rubin.hu><p05210605bc3b3b42e074@[137.78.212.225]> <40161DE9.6000909@rubin.hu><am1xplai99.fsf@nutcracker.stacken.kth.se><p05210603bc48732f3e39@[137.78.212.225]>
Sender: owner-heimdal-discuss@sics.se
User-Agent: Gnus/5.1005 (Gnus v5.10.5) Emacs/21.2 (gnu/linux)

"Henry B. Hotz" <hotz@jpl.nasa.gov> writes:

> My first reaction was "how did this *ever* compile with the missing
> quotes?"

`Traditional' preprocessors.

> My second reaction was "how did the compiler ever notice
> the problem given the #if 0//#endif block?"  This looks like a
> compiler bug, really.

No.  Look up `preprocessing tokens' in the C89 standard (at least).
The contents of #if blocks aren't outside the normal lexical rules.

With gcc 3.2 you'd have got `warning: multi-line string literals are
deprecated', and presumably its NEWS says they would be errors in the
next release.  Using `--traditional-cpp' in CPPFLAGS for configure
probably does what you expect, though it _might_ break something in
headers.

References:
- Eh, what? on build.
  - From: "Henry B. Hotz" <hotz@jpl.nasa.gov>

Prev by Date: Re: Eh, what? on build.
Next by Date: Re: Build environment incomplete and how to handle that.
Prev by thread: Re: Eh, what? on build.
Next by thread: Re: Build environment incomplete and how to handle that.
Index(es):
- Date
- Thread