Re: [PATCH v2] tests: Trying fixes test-replication.c on msys2.

[Thomas Huth <thuth@redhat.com>]

On 05/09/2020 05.11, 罗勇刚(Yonggang Luo) wrote:
> 
> 
> On Fri, Sep 4, 2020 at 9:07 PM Thomas Huth <thuth@redhat.com
> <mailto:thuth@redhat.com>> wrote:
> 
>     On 04/09/2020 00.06, Yonggang Luo wrote:
>     > Signed-off-by: Yonggang Luo <luoyonggang@gmail.com
>     <mailto:luoyonggang@gmail.com>>
>     > ---
>     >  tests/test-replication.c | 17 +++++++++++++----
>     >  1 file changed, 13 insertions(+), 4 deletions(-)
>     >
>     > diff --git a/tests/test-replication.c b/tests/test-replication.c
>     > index 9ab3666a90..d0e06f8d77 100644
>     > --- a/tests/test-replication.c
>     > +++ b/tests/test-replication.c
>     > @@ -23,14 +23,18 @@
>     > 
>     >  /* primary */
>     >  #define P_ID "primary-id"
>     > -static char p_local_disk[] = "/tmp/p_local_disk.XXXXXX";
>     > +#define P_LOCAL_DISK "%s/p_local_disk.XXXXXX"
>     > +static char p_local_disk[PATH_MAX];
>     > 
>     >  /* secondary */
>     >  #define S_ID "secondary-id"
>     >  #define S_LOCAL_DISK_ID "secondary-local-disk-id"
>     > -static char s_local_disk[] = "/tmp/s_local_disk.XXXXXX";
>     > -static char s_active_disk[] = "/tmp/s_active_disk.XXXXXX";
>     > -static char s_hidden_disk[] = "/tmp/s_hidden_disk.XXXXXX";
>     > +#define S_LOCAL_DISK "%s/s_local_disk.XXXXXX"
>     > +static char s_local_disk[PATH_MAX];
>     > +#define S_ACTIVE_DISK "%s/s_active_disk.XXXXXX"
>     > +static char s_active_disk[PATH_MAX];
>     > +#define S_HIDDEN_DISK "%s/s_hidden_disk.XXXXXX"
>     > +static char s_hidden_disk[PATH_MAX];
>     > 
>     >  /* FIXME: steal from blockdev.c */
>     >  QemuOptsList qemu_drive_opts = {
>     > @@ -571,7 +575,12 @@ static void setup_sigabrt_handler(void)
>     >  int main(int argc, char **argv)
>     >  {
>     >      int ret;
>     > +    const char *tmpdir = g_get_tmp_dir();
>     >      qemu_init_main_loop(&error_fatal);
>     > +    sprintf(p_local_disk, P_LOCAL_DISK, tmpdir);
>     > +    sprintf(s_local_disk, S_LOCAL_DISK, tmpdir);
>     > +    sprintf(s_active_disk, S_ACTIVE_DISK, tmpdir);
>     > +    sprintf(s_hidden_disk, S_HIDDEN_DISK, tmpdir);
> 
>     Sounds like the right way to go, but I think I'd do it without the
>     #defines and simply use the strings directly here, what do you think?
> 
> I place them at the same place by define is for easily readable, if I
> directly place at sprintf, then the code are harder to read 

IMHO it's easier to read the code the other way round: For understanding
the sprintf and its arguments, you have to know the format string, e.g.
will the "tmpdir" be handled via "%s", or "%p" or maybe something
completely different? If you then have to look up a macro first, it is a
cumbersome indirection. #defines are certainly fine for things that are
used multiple times, but here the strings are only used once, so the
indirection is really not needed.

 Thomas