[Toybox] [PATCH 2/2] pending/file: reuse int

Isaac Dunham ibid.ag at gmail.com
Sun Feb 28 17:17:20 PST 2016


On Sun, Feb 28, 2016 at 06:58:49PM -0500, Rich Felker wrote:
> On Sun, Feb 28, 2016 at 03:37:08PM -0800, Isaac Dunham wrote:
> > ---
> >  toys/pending/file.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> > 
> > diff --git a/toys/pending/file.c b/toys/pending/file.c
> > index 83e8c8e..789ef59 100644
> > --- a/toys/pending/file.c
> > +++ b/toys/pending/file.c
> > @@ -126,7 +126,7 @@ static void do_regular_file(int fd, char *name)
> >  {
> >    char *s;
> >    int len = read(fd, s = toybuf, sizeof(toybuf)-256);
> > -  int magic;
> > +  int peer;
> >  
> >    if (len<0) perror_msg("%s", name);
> >  
> > @@ -177,8 +177,8 @@ static void do_regular_file(int fd, char *name)
> >      else if (toybuf[5] == '2') cpioformat = "SVR4 with CRC";
> >      xprintf("ASCII cpio archive (%s)\n", cpioformat);
> >    }
> > -  else if (len>33 && (magic=peek(&s,2), magic==0143561 || magic==070707)) {
> > -    if (magic == 0143561) printf("byte-swapped ");
> > +  else if (len>33 && (peer=peek(&s,2), peer==0143561 || peer==070707)) {
> > +    if (peer == 0143561) printf("byte-swapped ");
> >      xprintf("cpio archive\n");
> >    }
> >    // tar archive (ustar/pax or gnu)
> > @@ -187,10 +187,10 @@ static void do_regular_file(int fd, char *name)
> >    }
> >    // zip/jar/apk archive, ODF/OOXML document, or such
> >    else if (len>5 && strstart(&s, "PK\03\04")) {
> > -    int ver = (int)(char)(toybuf[4]);
> > +    peer = (int)(char)(toybuf[4]);
> >      xprintf("Zip archive data");
> > -    if (ver)
> > -      xprintf(", requires at least v%d.%d to extract", ver/10, ver%10);
> > +    if (peer)
> > +      xprintf(", requires at least v%d.%d to extract", peer/10, peer%10);
> >      xputc('\n');
> >    }
> >    else {
> > -- 
> 
> What's the point of this patch? Does it actually change the generated
> code for the better? I would expect it to be at best the same code,
> and possibly worse.

Really, it comes out the same (at least with GCC 5.3 like Alpine uses);
no idea how lame a compiler it takes for this to be useful.

So I guess it's best to ignore it.

Thanks,
Isaac Dunham

 1456708640.0


More information about the Toybox mailing list