Home | Docs | Forums | Lists | Bugs | Planet | Store | GMN | Get Gentoo!

Full Text Bug Listing

In file included from ../dist/../cxx/cxx_app.cpp:18:
../dist/../include/db_cxx.h:52:22: error: iostream.h: No such file or directory
In file included from ../dist/../cxx/cxx_app.cpp:18:
../dist/../include/db_cxx.h:397: error: ‘ostream’ has not been declared
../dist/../include/db_cxx.h:480: error: ISO C++ forbids declaration of
‘ostream’ with no type
../dist/../include/db_cxx.h:480: error: expected ‘;’ before ‘*’ token
../dist/../include/db_cxx.h:545: error: ‘ostream’ has not been declared
../dist/../include/db_cxx.h:549: error: ‘ostream’ has not been declared
../dist/../cxx/cxx_app.cpp:41: error: expected constructor, destructor, or type
conversion before ‘*’ token
../dist/../cxx/cxx_app.cpp: In static member function ‘static void
DbEnv::_destroy_check(const char*, int)’:
../dist/../cxx/cxx_app.cpp:51: error: ‘ostream’ was not declared in this
scope
../dist/../cxx/cxx_app.cpp:51: error: ‘out’ was not declared in this scope
../dist/../cxx/cxx_app.cpp:53: error: ‘error_stream_’ was not declared in
this scope
../dist/../cxx/cxx_app.cpp:55: error: ‘cerr’ was not declared in this scope
../dist/../cxx/cxx_app.cpp: In static member function ‘static void
DbEnv::_stream_error_function(const char*, char*)’:
../dist/../cxx/cxx_app.cpp:284: error: ‘error_stream_’ was not declared in
this scope
../dist/../cxx/cxx_app.cpp: At global scope:
../dist/../cxx/cxx_app.cpp:300: error: variable or field ‘set_error_stream’
declared void
../dist/../cxx/cxx_app.cpp:300: error: ‘ostream’ was not declared in this
scope
../dist/../cxx/cxx_app.cpp:300: error: ‘stream’ was not declared in this
scope
make: *** [cxx_app.o] Error 1


Reproducible: Always

Steps to Reproduce:

It's quite a road we've taken together:
*db-3.2.9 (25 Jul 2002)
  25 Jul 2002; Spider <spider@gentoo.org> :
  Add a fix to remove -fno-exceptions in CXXFLAGS as it doesnt build then

But no other distro maintains db-3.2 and neither should we. Unless someone has
grown seriously attached to it, I will use this as a tracker bug to remove its
reverse dependencies.

[22:30] <loki_val> !meta -v python
[22:30] <jeeves> loki_val: Package: dev-lang/python Herd: python Maintainer:
python
[22:30] <jeeves> loki_val: (python) carlo, dev-zero, hawking, lordvan, lucass,
pythonhead
[22:32] <hawking> loki_val: yes?
[22:32] <hawking> :P
[22:32] <loki_val> dev-lang/python-2.3.6-r4             !build? berkdb?
=sys-libs/db-3*
[22:32] <loki_val> Any reason for that that you know of?
[22:33] <loki_val> -r is >=
[22:34] <loki_val> sry, 2.3.6-r3 is >=db-3.1
[22:34] <loki_val> number got lost in the mail
[22:35] <hawking> hmm no, I don't care about those versions though, will punt
them when bug 217221 is resolved.
[22:35] <jeeves> hawking: https://bugs.gentoo.org/217221 nor, P2, All,
hanno@gentoo.org->python@gentoo.org, NEW, pending, dev-lang/python: Buffer
overflow in zlib extension (CVE-2008-1721)
[22:36] <loki_val> hawking: excellent.

I'm having the same problem on gcc-4.3.1.  This is my only blocker so far.  Can
I modify the python ebuild to depend on >=db4.2?

Created an attachment (id=159147) [details]
fix db-3.2 compile with gcc-4.3.x

Here is some decision. I hope it'll be useful

the patch fixed the prolem for me.

after I added them the lib compiled fine on my amd64 arch.

best thanks

(In reply to comment #4)
> Created an attachment (id=159147) [edit] [details]
> fix db-3.2 compile with gcc-4.3.x
> 
> Here is some decision. I hope it'll be useful
> 

your patch fixed the build on my amd64 system with gcc 4.3.1 

I can confirm that Grigor Lekarov's patch makes db-3.2.9_p2 (added to the tree
today) compile with gcc-4.3.1-r1 on my system.

Without the patch, it didn't compile with same same error as everyone else is
getting.

Added to db-3.2.9_p2.

Your original patch had a bad #elif (missing argument), that I fixed up, along
with a better #if check (safe on 4.2 too).

*** Bug 235003 has been marked as a duplicate of this bug. ***

(In reply to comment #8)
> Added to db-3.2.9_p2.
> 
> Your original patch had a bad #elif (missing argument), that I fixed up, along
> with a better #if check (safe on 4.2 too).
> 

Sorry about #elif (it's my bad habit), and thanks for fix. It'll be pleasure if
I can help in future for my favorite Linux.