Bug#417475: FTBFS with GCC 4.3: missing #includes

[Jan Wagemakers]

Jaakko Hyvätti schreef:

[warning: suggest parentheses around && within ||]
> Enough ramblimng.  I have the parentheses added in my code now anyway, and 
> I suggest you can leave them as they are there in that version,

Ok.

Today, I have upgraded gcc-snapshot to version 20070613-1 and tested picprog
with that. I have not found any new problem. 

I'll wait a while before I bother my sponsor to upload a new picprog.deb,
and will test picprog when new versions of gcc-snapshot are released.


-- 
Met vriendelijke groetjes - Jan Wagemakers -

  Linux 2.6.21.1 i686 GNU/Linux


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Jaakko Hyvätti]

On Sat, 9 Jun 2007, Jan Wagemakers wrote:

picport.cc:181: warning: suggest parentheses around && within ||


etc..


These warnings can probably be solved by placing parentheses, but I'm not
sure about where exactly to place them. Anyone suggestions?


Well these warnings are kind of useful for newbie programmers, but when 
you know the precedence rules of && and ||, the extra parentheses only 
confuse, you start looking at it and trying to figure out why they are 
there.  Should probably have used warnings options that turn those off. 
Enough ramblimng.  I have the parentheses added in my code now anyway, and 
I suggest you can leave them as they are there in that version, the code 
is correct and if parentheses get added to wrong place it will break.  If 
you can live with the warnings.


Thanks for letting me know of the warnings and the include file thing.

Jaakko


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Jan Wagemakers]

Martin Michlmayr schreef:

>> Thanks for this report/patch. At the moment, gcc-snapshot >
>> 20070326-1 seems not to be available for i386¹ so at the moment I
>> can not test it myself. 
> I uploaded a new version which finally builds on x86, so this will
> show up in the archive within the next few days.

Indeed :-)

> Note that some other header files got cleaned up since I filed my bug
> so the patch I sent might need an update.

It seems that now "#include " is also needed to compile. 

*** /tmp/picprog-1.8.3.orig/program.cc  2007-06-09 12:08:32.570077447 +0200
--- program.cc  2007-06-09 11:58:07.546671127 +0200
***
*** 27,35 
--- 27,37 
  
  */
  
+ #include 
  #include 
  #include 
  
+ #include 
  #include 
  #include 
  #include 


I still get some warnings, but it seems to compile fine. 

 pts/3  jan /tmp/picprog-1.8.3$ make clean
rm -f core main.o picport.o hexfile.o program.o picprog
 pts/3  jan /tmp/picprog-1.8.3$ make dep
g++ -M -O2 -Wall -W -Wwrite-strings *.cc > .depend
 pts/3  jan /tmp/picprog-1.8.3$ make
g++ -O2 -Wall -W -Wwrite-strings   -c -o main.o main.cc
g++ -O2 -Wall -W -Wwrite-strings   -c -o picport.o picport.cc
picport.cc: In constructor 'picport::picport(const char*, bool, bool)':
picport.cc:181: warning: suggest parentheses around && within ||
picport.cc: In static member function 'static void picport::delay(long int)':
picport.cc:263: warning: suggest parentheses around && within ||
picport.cc:283: warning: suggest parentheses around && within ||
g++ -O2 -Wall -W -Wwrite-strings   -c -o hexfile.o hexfile.cc
hexfile.cc: In member function 'int hexfile::load(const char*)':
hexfile.cc:415: warning: suggest parentheses around + or - in operand of &
hexfile.cc:544: warning: suggest parentheses around + or - in operand of &
hexfile.cc: In member function 'int hexfile::program(const char*, bool, bool, 
bool, bool)':
hexfile.cc:1108: warning: suggest parentheses around arithmetic in operand of |
g++ -O2 -Wall -W -Wwrite-strings   -c -o program.o program.cc
g++ -s main.o picport.o hexfile.o program.o -o picprog
 pts/3  jan /tmp/picprog-1.8.3$ 


These warnings can probably be solved by placing parentheses, but I'm not
sure about where exactly to place them. Anyone suggestions?


-- 
Met vriendelijke groetjes - Jan Wagemakers -

... Linux inside


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Martin Michlmayr]

* Jan Wagemakers <[EMAIL PROTECTED]> [2007-06-05 12:03]:
> Thanks for this report/patch. At the moment, gcc-snapshot >
> 20070326-1 seems not to be available for i386¹ so at the moment I
> can not test it myself. 

I uploaded a new version which finally builds on x86, so this will
show up in the archive within the next few days.

Note that some other header files got cleaned up since I filed my bug
so the patch I sent might need an update.
-- 
Martin Michlmayr
http://www.cyrius.com/

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Jan Wagemakers]

Hello Jaakko,

Just to inform you, a while back I have received this
 bugreport about
picprog and the upcomming GCC 4.3.


-- 
Met vriendelijke groetjes - Jan Wagemakers -

 - Debian GNU/Linux lenny/sid - Up : 37 days 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Jan Wagemakers]

Martin Michlmayr schreef:

> Your package fails to build with GCC 4.3.  Version 4.3 has not been
> released yet but I'm building with a snapshot in order to find errors
> and give people an advance warning.  In GCC 4.3, the C++ header
> dependencies have been cleaned up.  The advantage of this is that
> programs will compile faster.  The downside is that you actually
> need to directly #include everything you use (but you really should
> do this anyway, otherwise your program won't work with any compiler
> other than GCC).  Some background of this can be found at
> http://gcc.gnu.org/PR28080
> 
> You can reproduce this problem with gcc-snapshot (20070326-1 or higher)
> from unstable.

Thanks for this report/patch. At the moment, gcc-snapshot > 20070326-1 seems not
to be available for i386¹ so at the moment I can not test it myself. 

 - I will forward your report to upstream (Jaakko Hyvätti), so he is also aware
   of this problem.

 - I will wait a while before I create a new package that fix this bug, so I
   can test it myself.


[¹] 
-- 
Met vriendelijke groetjes - Jan Wagemakers -

  Linux 2.6.21.1 i686 GNU/Linux


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#417475: FTBFS with GCC 4.3: missing #includes

[Martin Michlmayr]

Package: picprog
Version: 1.8.3-1
Usertags: ftbfs-gcc-4.3
Tags: patch

Your package fails to build with GCC 4.3.  Version 4.3 has not been
released yet but I'm building with a snapshot in order to find errors
and give people an advance warning.  In GCC 4.3, the C++ header
dependencies have been cleaned up.  The advantage of this is that
programs will compile faster.  The downside is that you actually
need to directly #include everything you use (but you really should
do this anyway, otherwise your program won't work with any compiler
other than GCC).  Some background of this can be found at
http://gcc.gnu.org/PR28080

You can reproduce this problem with gcc-snapshot (20070326-1 or higher)
from unstable.

> Automatic build of picprog_1.8.3-1 on coconut0 by sbuild/ia64 0.49
...
> hexfile.cc: In member function 'int hexfile::program(const char*, bool, bool, 
> bool, bool)':
> hexfile.cc:1108: warning: suggest parentheses around arithmetic in operand of 
> |
> g++ -O2 -Wall -W -Wwrite-strings   -c -o program.o program.cc
> program.cc: In member function 'void program::usage(option*, char*)':
> program.cc:457: error: 'exit' was not declared in this scope
> make[1]: *** [program.o] Error 1
> make[1]: Leaving directory `/build/tbm/picprog-1.8.3'

--- program.cc~ 2007-04-02 20:59:09.0 +
+++ program.cc  2007-04-02 20:59:17.0 +
@@ -27,6 +27,7 @@
 
 */
 
+#include 
 #include 
 #include 
 

-- 
Martin Michlmayr
http://www.cyrius.com/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]