Re: Fine, let's see you using pathconf(3) on Windows (was: Re: Then there's PATH_MAX)

On 10 Dec 2024 15:13:40 GMT, vallor <vallor@cultnix.org> wrote in
<lrr454F1u8jU1@mid.individual.net>:

On Tue, 10 Dec 2024 14:44:12 +0000, Farley Flud
<fsquared@fsquared.linux>
wrote in <180fd7fa66284739$61502$1728$802601b3@news.usenetexpress.com>:
 

On Tue, 10 Dec 2024 13:58:17 +0000, vallor wrote:
 
 

 - - cut here %<- - -
>
>

Should be "flush here."
 
 
 

if(argv[1])
    {
    path = argv[1];
    }
    }
 
 

WTF!
 
You check for arguments like this:
 
if(argc != 2)
 
or:
 
if(argv[1] == NULL)
 
 
 

    else {
    path = "/";
>
>

WTF!
 
You are assigning a pointer with the value 0x2F.
 
If it's then passed to pathconf() then KABOOM!
 
 
 

printf("%d\n",result);
 
 

"result" is an int.  pathcomp() returns a long int.
 
You should either have declared "result" as a long int or:
 
result = (int)pathconf(path,_PC_PATH_MAX)
 
 
 
Congrats.  In a few short lines you have created a huge mess.

 
Thank you for your feedback.  I have modified my program:
 
$ diff -u use_pathconf.c~ use_pathconf.c --- use_pathconf.c~ 2024-12-10
05:52:19.000000000 -0800 +++ use_pathconf.c 2024-12-10
06:57:40.671592496 -0800 @@ -4,7 +4,7 @@
 
 int main(int argc, char * argv[], char * envp[])
 {
-int result = 0;
+long result = 0;
 char * path;
 
 if(argv[1])
@@ -23,7 +23,7 @@
     perror("pathconf() returned -1:");
     return 1;
     }
-printf("%d\n",result);
+printf("%ld\n",result);
 
 return 0;
 }
 
That's the only valid criticism I see -- but I do see that you are not
as familiar with C as you claim to be.  (Please review your knowledge of
pointers to string literals before you embarrass yourself further.)
 
CC=tcc CFLAGS=-g -O2 -std=c90 -Wall -Werror -pedantic
 
Although I'm not sure if tcc supports -Wall -Werror, so just tried:
 
gcc -g -O2 -std=c90 -Wall -Werror -pedantic    use_pathconf.c   -o
use_pathconf
 
Nope, no errors.
 
$ gdb use_pathconf (gdb) break 19 Breakpoint 1 at 0x10d7: file
use_pathconf.c, line 19.
(gdb) run Starting program: /nfs/ds/src/vallor/path_max/use_pathconf
[Thread debugging using libthread_db enabled]
Using host libthread_db library
"/lib/x86_64-linux-gnu/libthread_db.so.1".
 
Breakpoint 1, main (argc=1, argv=0x7fffffffe1d8, envp=0x7fffffffe1e8) at
use_pathconf.c:19 19 errno = 0;
(gdb) print path $1 = 0x555555556004 "/"
(gdb) print *path $2 = 47 '/'
(gdb) print *(path+1)
$4 = 0 '\000'
(gdb) print "happy?"
$5 = "happy?"
 
Meanwhile, I still haven't seen your program that would run on Windows.
 
What's the matter Colonel Sanders ... chicken?

I guess that shut his trap.

BTW, if you want to get fancy:

$ rcsdiff -u ./use_pathconf.c
===================================================================
RCS file: ./RCS/use_pathconf.c,v
retrieving revision 1.1
diff -u -r1.1 ./use_pathconf.c
--- ./use_pathconf.c 2024/12/10 14:57:40 1.1
+++ ./use_pathconf.c 2024/12/11 17:56:32
@@ -7,14 +7,7 @@
 long result = 0;
 char * path;
 
-if(argv[1])
-    {
-    path = argv[1];
-    }
-    else
-    {
-    path = "/";
-    }
+path = argv[1] ? argv[1] : "/";
 
 errno = 0;
 result = pathconf(path,_PC_PATH_MAX);

 - - cut here %<- - -

(This will, of course, confound the naysayer.)

--
-v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090 Ti
   OS: Linux 6.12.4 Release: Mint 21.3 Mem: 258G
   "I know Karate, Kung Fu, and 47 other dangerous words"