[libvirt] [PATCH] internal: Use g_strcmp0 in STR(N)EQ_NULLABLE
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL) -#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0) #define NUL_TERMINATE(buf) do { (buf)[sizeof(buf)-1] = '\0'; } while (0) -- 2.21.0
On Thu, Oct 24, 2019 at 04:12:00PM +0200, Peter Krempa wrote:
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL)
-#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0)
Why not use g_strcmp0 directly? Pavel
On Thu, Oct 24, 2019 at 04:22:09PM +0200, Pavel Hrdina wrote:
On Thu, Oct 24, 2019 at 04:12:00PM +0200, Peter Krempa wrote:
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL)
-#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0)
Why not use g_strcmp0 directly?
The same reason we never used strcmp() in the first place - it is easier to read STREQ vs strcmp == 0 / strcmp != 0. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Thu, Oct 24, 2019 at 03:29:27PM +0100, Daniel P. Berrangé wrote:
On Thu, Oct 24, 2019 at 04:22:09PM +0200, Pavel Hrdina wrote:
On Thu, Oct 24, 2019 at 04:12:00PM +0200, Peter Krempa wrote:
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL)
-#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0)
Why not use g_strcmp0 directly?
The same reason we never used strcmp() in the first place - it is easier to read STREQ vs strcmp == 0 / strcmp != 0.
Right, I was looking at g_str_equal() which returns gboolean and for some reason didn't realized that g_strcmp0 returns int. Only if there would be g_str_equal0(). Pavel
On Thu, Oct 24, 2019 at 16:22:09 +0200, Pavel Hrdina wrote:
On Thu, Oct 24, 2019 at 04:12:00PM +0200, Peter Krempa wrote:
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL)
-#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0)
Why not use g_strcmp0 directly?
Because I don't really care about replacing everyting.
On 10/24/19 10:31 AM, Peter Krempa wrote:
On Thu, Oct 24, 2019 at 16:22:09 +0200, Pavel Hrdina wrote:
On Thu, Oct 24, 2019 at 04:12:00PM +0200, Peter Krempa wrote:
Signed-off-by: Peter Krempa <pkrempa@redhat.com> --- src/internal.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/internal.h b/src/internal.h index 5b0a2335f5..0ff9f496ac 100644 --- a/src/internal.h +++ b/src/internal.h @@ -86,10 +86,8 @@ #define STRCASEPREFIX(a, b) (c_strncasecmp(a, b, strlen(b)) == 0) #define STRSKIP(a, b) (STRPREFIX(a, b) ? (a) + strlen(b) : NULL)
-#define STREQ_NULLABLE(a, b) \ - ((a) ? (b) && STREQ((a), (b)) : !(b)) -#define STRNEQ_NULLABLE(a, b) \ - ((a) ? !(b) || STRNEQ((a), (b)) : !!(b)) +#define STREQ_NULLABLE(a, b) (g_strcmp0(a, b) == 0) +#define STRNEQ_NULLABLE(a, b) (g_strcmp0(a, b) != 0) Why not use g_strcmp0 directly? Because I don't really care about replacing everyting.
Yeah, on one hand using the glib functions directly might make it easier for someone familiar with glib to understand the code when just glancing at it. On the other hand, replacing all these macros with direct calls to glib seems like it's undoing a lot of work, and creating code-churn/obscuring true "git blame", and will just need to be done again in 2029 when glib is no longer the flavor of the day, and we want to convert everything to use "hlib" or "ilib" (or maybe by then it will be "postapocalypticlib" - think about it, having all our code sprinkled with postapocalyptic_dystopian_strcmp0() will be *so* future proof!) (The preceding paragraph was only half sarcastic, same for this sentence.)
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
participants (5)
-
Daniel P. Berrangé -
Ján Tomko -
Laine Stump -
Pavel Hrdina -
Peter Krempa