Hi,
During compilation of some bluetooth apps I get lots of warings like this:
In file included from /usr/include/sys/socket.h:35,
from connectiondlg.h:31,
from connectiondlg.cpp:24:
/usr/include/bits/socket.h:91: warning: `PF_BLUETOOTH' redefined
/usr/include/bluetooth/bluetooth.h:43: warning: this is the location of the previous definition
/usr/include/bits/socket.h:123: warning: `AF_BLUETOOTH' redefined
/usr/include/bluetooth/bluetooth.h:42: warning: this is the location of the previous definition
In file included from connectiondlg.h:35,
from connectiondlg.cpp:24:
/usr/include/bluetooth/hci.h:482: warning: ANSI C++ forbids zero-size array `condition'
/usr/include/bluetooth/hci.h:909: warning: ANSI C++ forbids zero-size array `data'
/usr/include/bluetooth/hci.h:1046: warning: ANSI C++ forbids zero-size array `dev_req'
/usr/include/bluetooth/hci.h:1052: warning: ANSI C++ forbids zero-size array `conn_info'
/usr/include/bluetooth/hci.h:1058: warning: ANSI C++ forbids zero-size array `conn_info'
Would be nice if bluetooth.h would not override stuff already defined
sys/socket.h:123:#define AF_BLUETOOTH PF_BLUETOOTH
I would suggest that bluetooth.h includes itself sys/socket.h before checking if AF_BLUETOOTH
is defined. This removed the dependency on how headers are ordered in source files.
Works for me ;)
hci.h may be due to old g++2.94.4 used on debian. No idea about a workaround :(
Achim
--
To me vi is Zen. To use vi is to practice zen. Every command is
a koan. Profound to the user, unintelligible to the uninitiated.
You discover truth everytime you use it.
-- [email protected]
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel
Hi Max,
> Actually sys/socket.h didn't have PF_\AF_ BLUETOOTH defined up till now
> I guess. Mine for example (latest Fedora 1) still doesn't. Looks like
> GLIBC folks pulled Bluetooth related things into standard headers. Which
> is cool. We'll just have to #ifdef defines that may clash.
first, the file for this definitions is bits/socket.h which is included
by sys/socket.h of course. I checked my GLIBC from Debian Sid (Unstable)
which is libc6-2.3.2.ds1-11 and indeed it contains the PF_/AF_BLUETOOTH
defines. Now that the GLIBC folks included these defines, you must
include sys/socket.h before you include bluetooth/bluetooth.h. Otherwise
the compiler will complain with redefines.
I checked some more of the socket related include files and it seems
that it ok to include sys/socket.h from other include files. For example
netinet/in.h. Actually I don't like the idea to include too much files
from another header file. What do you think?
Regards
Marcel
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel
On Thu, 2004-01-22 at 05:34, Achim Bohnet wrote:
> > > hci.h may be due to old g++2.94.4 used on debian. No idea about a workaround :(
> >
> > include "sys/socket.h" before you include "bluetooth/bluetooth.h".
>
> Hi Marcel,
>
> That one solution, but why isn't it done by bluetooth.h itself? Makes it header order
> independent, and fixes _once and forever_ annoying warnings like that.
> (that was my last try to convience you, promised ;)
Actually sys/socket.h didn't have PF_\AF_ BLUETOOTH defined up till now
I guess. Mine for example (latest Fedora 1) still doesn't. Looks like
GLIBC folks pulled Bluetooth related things into standard headers. Which
is cool. We'll just have to #ifdef defines that may clash.
> What about condition[0] etc in hci.h. Should this be contition[] etc?
>
> ach@allee(0) ~ $ fgrep '[0]' /usr/include/bluetooth/hci.h
> uint8_t condition[0];
> uint8_t data[0];
> struct hci_dev_req dev_req[0]; /* hci_dev_req structures */
> struct hci_conn_info conn_info[0];
> struct hci_conn_info conn_info[0];
No those should be [0] no []. I'm not sure why your GCC does not like
them.
Max
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel
Hi Achim,
> That one solution, but why isn't it done by bluetooth.h itself? Makes it header order
> independent, and fixes _once and forever_ annoying warnings like that.
> (that was my last try to convience you, promised ;)
it is not a nice programming style to include header files from another
header file. From my view only really needed headers should be included
from bluetooth.h and sys/socket.h is not needed at this point. It is
first needed when you start socket programming and if you do so I think
we can expect that you include sys/socket.h first.
> What about condition[0] etc in hci.h. Should this be contition[] etc?
>
> ach@allee(0) ~ $ fgrep '[0]' /usr/include/bluetooth/hci.h
> uint8_t condition[0];
> uint8_t data[0];
> struct hci_dev_req dev_req[0]; /* hci_dev_req structures */
> struct hci_conn_info conn_info[0];
> struct hci_conn_info conn_info[0];
Maybe this is compiler problem, because this statement is correct.
Regards
Marcel
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel
On Thursday 22 January 2004 14:04, you wrote:
> Hi Achim,
>
> > During compilation of some bluetooth apps I get lots of warings like this:
> >
> > In file included from /usr/include/sys/socket.h:35,
> > from connectiondlg.h:31,
> > from connectiondlg.cpp:24:
> > /usr/include/bits/socket.h:91: warning: `PF_BLUETOOTH' redefined
> > /usr/include/bluetooth/bluetooth.h:43: warning: this is the location of the previous definition
> > /usr/include/bits/socket.h:123: warning: `AF_BLUETOOTH' redefined
> > /usr/include/bluetooth/bluetooth.h:42: warning: this is the location of the previous definition
> > In file included from connectiondlg.h:35,
> > from connectiondlg.cpp:24:
> > /usr/include/bluetooth/hci.h:482: warning: ANSI C++ forbids zero-size array `condition'
> > /usr/include/bluetooth/hci.h:909: warning: ANSI C++ forbids zero-size array `data'
> > /usr/include/bluetooth/hci.h:1046: warning: ANSI C++ forbids zero-size array `dev_req'
> > /usr/include/bluetooth/hci.h:1052: warning: ANSI C++ forbids zero-size array `conn_info'
> > /usr/include/bluetooth/hci.h:1058: warning: ANSI C++ forbids zero-size array `conn_info'
> >
> > Would be nice if bluetooth.h would not override stuff already defined
> >
> > sys/socket.h:123:#define AF_BLUETOOTH PF_BLUETOOTH
> >
> > I would suggest that bluetooth.h includes itself sys/socket.h before checking if AF_BLUETOOTH
> > is defined. This removed the dependency on how headers are ordered in source files.
> > Works for me ;)
> >
> > hci.h may be due to old g++2.94.4 used on debian. No idea about a workaround :(
>
> include "sys/socket.h" before you include "bluetooth/bluetooth.h".
Hi Marcel,
That one solution, but why isn't it done by bluetooth.h itself? Makes it header order
independent, and fixes _once and forever_ annoying warnings like that.
(that was my last try to convience you, promised ;)
What about condition[0] etc in hci.h. Should this be contition[] etc?
ach@allee(0) ~ $ fgrep '[0]' /usr/include/bluetooth/hci.h
uint8_t condition[0];
uint8_t data[0];
struct hci_dev_req dev_req[0]; /* hci_dev_req structures */
struct hci_conn_info conn_info[0];
struct hci_conn_info conn_info[0];
Achim
>
> Regards
>
> Marcel
>
>
>
>
--
To me vi is Zen. To use vi is to practice zen. Every command is
a koan. Profound to the user, unintelligible to the uninitiated.
You discover truth everytime you use it.
-- [email protected]
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel
Hi Achim,
> During compilation of some bluetooth apps I get lots of warings like this:
>
> In file included from /usr/include/sys/socket.h:35,
> from connectiondlg.h:31,
> from connectiondlg.cpp:24:
> /usr/include/bits/socket.h:91: warning: `PF_BLUETOOTH' redefined
> /usr/include/bluetooth/bluetooth.h:43: warning: this is the location of the previous definition
> /usr/include/bits/socket.h:123: warning: `AF_BLUETOOTH' redefined
> /usr/include/bluetooth/bluetooth.h:42: warning: this is the location of the previous definition
> In file included from connectiondlg.h:35,
> from connectiondlg.cpp:24:
> /usr/include/bluetooth/hci.h:482: warning: ANSI C++ forbids zero-size array `condition'
> /usr/include/bluetooth/hci.h:909: warning: ANSI C++ forbids zero-size array `data'
> /usr/include/bluetooth/hci.h:1046: warning: ANSI C++ forbids zero-size array `dev_req'
> /usr/include/bluetooth/hci.h:1052: warning: ANSI C++ forbids zero-size array `conn_info'
> /usr/include/bluetooth/hci.h:1058: warning: ANSI C++ forbids zero-size array `conn_info'
>
> Would be nice if bluetooth.h would not override stuff already defined
>
> sys/socket.h:123:#define AF_BLUETOOTH PF_BLUETOOTH
>
> I would suggest that bluetooth.h includes itself sys/socket.h before checking if AF_BLUETOOTH
> is defined. This removed the dependency on how headers are ordered in source files.
> Works for me ;)
>
> hci.h may be due to old g++2.94.4 used on debian. No idea about a workaround :(
include "sys/socket.h" before you include "bluetooth/bluetooth.h".
Regards
Marcel
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Bluez-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-devel