Re : RE: 'long long' not supported

#1

Just for info, I had the same KIND OF problem on windows platform event w=
ith 2.0b2 while trying to integrate MYSQL server sources (4.11/4.12a/5.02=
b), MYSQL++(1.7/2.0) and JASPER (free JPEG2000 Library).
I saw then the =
following type redefinition problems :
- MYSQL sources defines typedef l=
ong long my_longlong =3D> OK
- MYSQL++ defines typedef __int64 longlong=

- JASPER defines typedef long long longlong

When you try to remove th=
e definition on MYSQL++ side, you can go on compiling further but you rea=
ch SQLString::longlong types.

MYSQL++ team, do you think it could be i=
nteresting and not destructing regarding MYSQL++ API to redfines 64 bits =
types to a name specific to MYSQL++(ex mysqlpp_longlong) as it is done in=
MYSQL sources C API to avoid naming conflicts ?

NP.

Quote:

------=
---- Message d'origine -----------
Exp=E9diteur : "Purcell Dan Cliffor=
d \(UMR-Student\)" <dcpty9 (AT) umr (DOT) edu
A : "MySQL++ Mailing List" <pluspl=
us (AT) lists (DOT) mysql.com
CC :
Date : Wed, 20 Jul 2005 23:45:28 -0500
=
Objet : RE: 'long long' not supported
Warren Young said:

If y=
ou don't need this feature, you can #define NO_LONG_LONGS, but this
wi=
ll only affect MySQL++, not the C API. Also, there were problems with
=
this mode until very recently in the MySQL++ v2.0 development; I'm not

=

Quote:

even sure if the fixes are in beta2. You may have to use the repositor=
y
version.

Using the #define NO_LONG_LONGS got rid of that pro=
blem, but how can I get around the problem of all of these redefinitions =

which result from this change.

Quote:

g++ -g -Wall -W -I/usr/local/inc=
lude -I/usr/include/mysql -c -o test.o test.cpp
In file included fro=
m /usr/local/include/query.h:49,
from /usr/local/incl=
ude/mysql++.h:45,
from test.cpp:6:
/usr/local/incl=
ude/coldata.h:262: redefinition of `template <class Str
longlong mysq=
lpp:

perator+ (mysqlpp::ColData_Tmpl<Str>, long int)'
/usr/local/incl=
ude/coldata.h:255: `template <class Str> long int
mysqlpp:

perator+ (=
mysqlpp::ColData_Tmpl<Str>, long int)' previously
declared here
/us=
r/local/include/coldata.h:262: redefinition of `template <class Str
l=
onglong mysqlpp:

perator+ (long int, mysqlpp::ColData_Tmpl<Str>)'
/us=
r/local/include/coldata.h:255: `template <class Str> long int
mysqlpp:=

perator+ (long int, mysqlpp::ColData_Tmpl<Str>)' previously
declared=
here
/usr/local/include/coldata.h:262: redefinition of `template <cla=
ss Str
longlong mysqlpp:

perator- (mysqlpp::ColData_Tmpl<Str>, long =
int)'
/usr/local/include/coldata.h:255: `template <class Str> long int=

mysqlpp:

perator- (mysqlpp::ColData_Tmpl<Str>, long int)' previously=

declared here
/usr/local/include/coldata.h:262: redefinition of `t=
emplate <class Str
longlong mysqlpp:

perator- (long int, mysqlpp::Co=
lData_Tmpl<Str>)'
/usr/local/include/coldata.h:255: `template <class S=
tr> long int
mysqlpp:

perator- (long int, mysqlpp::ColData_Tmpl<Str>)=
' previously
declared here

... the same error is given for the=
other operators * , / , % , & , ^ , | , << ,

... then these e=
rrors

In file included from /usr/local/include/manip.h:47,
=
from /usr/local/include/vallist.h:31,
f=
rom /usr/local/include/row.h:34,
from /usr/local/incl=
ude/result.h:38,
from /usr/local/include/connection.h=
:39,
from /usr/local/include/query.h:50,
=
from /usr/local/include/mysql++.h:45,
from te=
st.cpp:6:
/usr/local/include/coldata.h: In instantiation of `mysqlpp::=
ColData_Tmpl<mysqlpp::const_string>':
/usr/local/include/datetime.h:14=
7: instantiated from here
/usr/local/include/coldata.h:196: `mysqlpp=
::ColData_Tmpl<Str>:

perator
longlong () const [with Str =3D mysqlpp:=
:const_string]' has already been
declared in `mysqlpp::ColData_Tmpl<my=
sqlpp::const_string>'
/usr/local/include/coldata.h:200: `mysqlpp::ColD=
ata_Tmpl<Str>:

perator
ulonglong () const [with Str =3D mysqlpp::cons=
t_string]' has already been
declared in `mysqlpp::ColData_Tmpl<mysqlpp=
::const_string>'
In file included from /usr/local/include/vallist.h:31=
,
from /usr/local/include/row.h:34,
=
from /usr/local/include/result.h:38,
from /usr/loc=
al/include/connection.h:39,
from /usr/local/include/q=
uery.h:50,
from /usr/local/include/mysql++.h:45,
=
from test.cpp:6:
/usr/local/include/coldata.h: In inst=
antiation of `mysqlpp::ColData_Tmpl<string>':
/usr/local/include/manip=
..h:171: instantiated from here
/usr/local/include/coldata.h:196: `my=
sqlpp::ColData_Tmpl<Str>:

perator
longlong () const [with Str =3D str=
ing]' has already been declared in
`mysqlpp::ColData_Tmpl<string>'
=
/usr/local/include/coldata.h:200: `mysqlpp::ColData_Tmpl<Str>:

perator

=

Quote:

ulonglong () const [with Str =3D string]' has already been declared in=

`mysqlpp::ColData_Tmpl<string>'
make: *** [test.o] Error 1

t=
hanks for your consideration,
-Dan

--
MySQL++ Mailing List=

For list archives: http://lists.mysql.com/plusplus
To unsubscribe: =
http://lists.mysql.com/plusplus?unsu...) sfr (DOT) fr

Nicolas P=
ERRIN

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsu...ie.nctu.edu.tw

#2

np (AT) sfr (DOT) fr wrote:

Quote:

Just for info, I had the same KIND OF problem on windows platform event with 2.0b2 while trying to integrate MYSQL server sources (4.11/4.12a/5.02b), MYSQL++(1.7/2.0) and JASPER (free JPEG2000 Library).
I saw then the following type redefinition problems :
- MYSQL sources defines typedef long long my_longlong => OK
- MYSQL++ defines typedef __int64 longlong
- JASPER defines typedef long long longlong

That should only cause a problem when you say "using namespace mysqlpp".
If you are having this problem, you shouldn't hoist these definitions
up into the global namespace; fully qualify symbols instead:

mysqlpp::longlong x;

Quote:

MYSQL++ team, do you think it could be interesting and not
destructing regarding MYSQL++ API to redfines 64 bits types to a name
specific to MYSQL++(ex mysqlpp_longlong) as it is done in MYSQL
sources C API to avoid naming conflicts ?

It's not necessary, and I'm sure there's a lot of code out there that
depends on this typedef. We've broken a lot of things in MySQL++ 2.0,
but only where necessary. This isn't such a case.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsu...ie.nctu.edu.tw

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Re : RE: 'long long' not supported

mailing.database.mysql-plusplus mailing.database.mysql-plusplus

Re : RE: 'long long' not supported - 07-21-2005 , 04:15 AM

Re: Re : RE: 'long long' not supported - 07-21-2005 , 07:53 PM