Communities

Writing
Writing
Codidact Meta
Codidact Meta
The Great Outdoors
The Great Outdoors
Photography & Video
Photography & Video
Scientific Speculation
Scientific Speculation
Cooking
Cooking
Electrical Engineering
Electrical Engineering
Judaism
Judaism
Languages & Linguistics
Languages & Linguistics
Software Development
Software Development
Mathematics
Mathematics
Christianity
Christianity
Code Golf
Code Golf
Music
Music
Physics
Physics
Linux Systems
Linux Systems
Power Users
Power Users
Tabletop RPGs
Tabletop RPGs
Community Proposals
Community Proposals
tag:snake search within a tag
answers:0 unanswered questions
user:xxxx search by author id
score:0.5 posts with 0.5+ score
"snake oil" exact phrase
votes:4 posts with 4+ votes
created:<1w created < 1 week ago
post_type:xxxx type of post
Search help
Notifications
Mark all as read See all your notifications »
Q&A

Welcome to Software Development on Codidact!

Will you help us build our independent community of developers helping developers? We're small and trying to grow. We welcome questions about all aspects of software development, from design to code to QA and more. Got questions? Got answers? Got code you'd like someone to review? Please join us.

Comments on What is the purpose of having underscored names and then defining a non-underscored alias to it?

Parent

What is the purpose of having underscored names and then defining a non-underscored alias to it?

+5
−0

In a C implementation in <stdio.h> on Linux I saw something like:

extern FILE *__stdinp;
extern FILE *__stdoutp;
extern FILE *__stderrp;

And then:

#define stdin __stdinp
#define stdout __stdoutp
#define stderr __stderrp

My question is, for what reason would an implementation provide the __name items and then create aliases to them instead of just providing the canonical name to begin with?

History
Why does this post require moderator attention?
You might want to add some details to your flag.
Why should this post be closed?

1 comment thread

Can't really be answered (1 comment)
Post
+1
−0

Quoting here the 1999 C standard, as being close enough to "this century" while also being hella old.

Source: ISO/IEC 9899:1999

7.19 Input/output <stdio.h>

7.19.1 Introduction

1 The header <stdio.h> declares three types, several macros, and many functions for performing input and output.
...
3 The macros are NULL (described in 7.17);

... [[much elided]] ...;

stderr
stdin
stdout
which are expressions of type ‘‘pointer to FILE’’ that point to the FILE objects associated, respectively, with the standard error, input, and output streams.

Note that the C standard here explicitly labels the symbols stdin, stdout, and stderr as being macros. There is no flexibility, no implementation-dependency, no per-thread or per-CPU optional components. These symbols are macros because the standard says they are.

If a library provides you with only extern FILE * stdin (&c.) then they are out of compliance, and your C environment is not "standard C" and cannot ever be "standard C." (You may not actually care about this. Many people, including me, do not. But still...)

Note that macros are required not to be recursive on expansion, meaning it is possible to do something like:

#define stdin stdin
extern FILE* stdin;

which both conforms to the letter of the standard and also flips off whatever original intent was present. This is why we can't have nice things.

History
Why does this post require moderator attention?
You might want to add some details to your flag.

1 comment thread

Macro naming (6 comments)
Macro naming
Lundin‭ wrote 5 months ago

As per your last code example, it would be fine for the library to name the variable extern FILE *stdin; and then do #define stdin stdin. This would still enable stuff like #ifdef stdin - it doesn't "flip off" any original intent. Why they didn't do that in the unknown standard lib mentioned, I have no idea. Perhaps a naming convention that all variables inside the library should always use __.

Lundin‭ wrote 5 months ago

On the other hand I don't understand the C rationale to make these macros either. If the intention was to allow evil things like #ifdef stdin #undef stdin #endif #define stdin something_else then it wouldn't be wise to use stdin as the name for the internal variable.

aghast‭ wrote 5 months ago

I suspect that if you go far enough back, stdin was probably an expression macro, like &Global_file_structs[0] or some such. So this is probably the case of some early library corrupting the standard. They probably could have done a better job of allowing the name to be a macro without requiring the name to be a macro, but then they might also have been using #ifdef stdin as a guard for stdio.h.

Too many layers of "might have" and not enough evidence. The "Rationale" document has entries for other macros defined in this section, but doesn't bother calling out the std* names. I interpret this to mean that they considered using macros for the names to be "existing practice" and not worth mentioning.

alx‭ wrote 4 months ago

Lundin‭ A minor nitpick on your comment: #undef doesn't need to be guarded by #ifdef. #undef is defined by ISO C to be a no-op if there's no such macro. http://port70.net/~nsz/c/c11/n1570.html#6.10.3.5p2

Lundin‭ wrote 4 months ago

alx‭ stdin is not necessarily a macro. C17 7.21.1: "stdin ... "which are expressions of type ‘‘pointer to FILE’’". You can't #undef expressions.

alx‭ wrote 4 months ago · edited 4 months ago

Lundin‭ You can't, in the sense that the expression won't be undefined (because #undef is interpreted by the preprocessor, there are still no identifiers that aren't macros; it's just text), but you can, in the sense that your program will be valid.

#undef stdin
#define stdin something_else

The above is as evil as your suggested evil code. It will do the same exact thing, and is equally compliant.

The place I often use #undef is in test programs, where I write (as the very last #include):

#undef NDEBUG
#include <assert.h>

And I don't care if NDEBUG was defined previously or not; I want to make sure it's not in my test program, to use asserts in it.

Skipping 1 deleted comment.