Nabble - MySQL - C++

Training & jobs

2008-11-05T01:30:27Z

“Join us to make future”
FOR OPT/F-1 STUDENTS
FREE TRAINING
FOOD ACCOMODATION
H1B PROCESSING
FOR L1/ L2/ H1/ H4/ EAD/ GC
EXCELLENT OPPRTUNITIES
TRAINING
PLACEMENT
Highly Competitive offers for New H1b Transfers

About Us:
V2 technology inc is serving NJ since 2005, for us our employees are
of utmost importance. Our highly skilled and dedicated instructors
train you explicitly in market-related technologies for today and
tomorrow. We will work with you in developing marketing strategies
and finding the assignments of your choice.
We not only help you get a job but we build your long lasting
careers!!!
What are we Looking for:
Excellent communication skills.
Valid F1/ OPT/ CPT/ H1/ H4/ L1 or valid work status
(EAD, GC).
Bachelors’ degree in CS/ IT or previous IT experience.
Preferably (not mandatory) Master's Degree in Computers/
Electronics or previous IT experience.
Willing to relocate anywhere in USA.

What’s The Deal:
H1-B sponsorship to F1/ OPT/ H4/ L1/ L2/ E3/ EAD
Transfer of H1 & L1 visas.
Green Card sponsorship through PERM
Job focused professional training in
Java/ J2EE
.NET
Documentum
Share point
Oracle
Data Modeling
Relocation assistance- airfare, hotel accommodation, car rental
etc
Guarantee lowest bench period.
Employee referral program.
Effective Resume writing help, Marketing and Placement.
Technical mock interviews. Preparation of In Person Interviews

100% Guaranteed placement
Get in Touch:
Contact immediately to know more about the opportunities with us
at:609-843-0312
or
write mail at jobs@v2techinc.com
Contact person : John

Training & jobs

2008-09-22T23:18:15Z

“Join us to make future”
FOR OPT/F-1 STUDENTS
FREE TRAINING
FOOD ACCOMODATION
H1B PROCESSING FOR L1/ L2/ H1/ H4/ EAD/ GC
EXCELLENT OPPRTUNITIES
TRAINING
PLACEMENT
Highly Competitive offers for New H1b Transfers
About Us:
V2 technology inc is serving NJ since 2005, for us our employees are of utmost importance. Our highly skilled and dedicated instructors train you explicitly in market-related technologies for today and tomorrow. We will work with you in developing marketing strategies and finding the assignments of your choice.
We not only help you get a job but we build your long lasting careers!!!
What are we Looking for:
Excellent communication skills.
Valid F1/ OPT/ CPT/ H1/ H4/ L1 or valid work status (EAD, GC).
Bachelors’ degree in CS/ IT or previous IT experience.
Preferably (not mandatory) Master's Degree in Computers/ Electronics or previous IT experience.
Willing to relocate anywhere in USA.
What’s The Deal:
H1-B sponsorship to F1/ OPT/ H4/ L1/ L2/ E3/ EAD
Transfer of H1 & L1 visas.
Green Card sponsorship through PERM
Job focused professional training in
Java/ J2EE
.NET
Documentum
Share point
Oracle
Data Modeling
Relocation assistance- airfare, hotel accommodation, car rental etc
Guarantee lowest bench period.
Employee referral program.
Effective Resume writing help, Marketing and Placement.
Technical mock interviews. Preparation of In Person Interviews
100% Guaranteed placement
Get in Touch:
Contact immediately to know more about the opportunities with us at:609-843-0312
or
write mail at jobs@v2techinc.com
Contact person : John

Re: dbdriver sigsegv errors.

2008-06-18T22:10:40Z

Hello,

answering my own question here...

So I spent about 5 hours figuring out that I was using an old version of
mysqlpp.dll in my system32 folder. When upgrading to mysql6 i decided
to rebuild the mysql++ dll against the mysql6 headers, but i forgot to
update the system32 folder! No wonder it compiled but didn't run well...

So that is my, "don't be stupid like me" post.

Daniel Kolbo wrote:

> Hello,
>
> I am using mysql6. mysql++ seems to be 'working' and 'installed'
> correctly; however, at the end of the execution of my programs i am
> receiving SIGSEGV errors.
> When I installed mysql++ simple1 was built. This version of simple1
> does NOT produce the error. However, when I copy the code and try to
> build it myself I receive the error at the end of the file when
> running the application. I am able to compile, but when I run the
> program it has to abort, and when i debug I get the SIGSEGV error at
> end of the file.
>
> Here is the commands codeblocks used to compile my project titled
> 'temp' that contains the exact same code as simple1.cpp.
>
> mingw32-g++.exe -Wall -fexceptions -g
> -IE:\mysql++-3.0.3\include_mysql++ -IE:\mysql++-3.0.3\include_mysql6
> -c E:\C++\codeblocks\Practice\temp\main.cpp -o obj\Debug\main.o
> mingw32-g++.exe -Wall -fexceptions -g
> -IE:\mysql++-3.0.3\include_mysql++ -IE:\mysql++-3.0.3\include_mysql6
> -c E:\C++\codeblocks\Practice\temp\printdata.cpp -o obj\Debug\printdata.o
> mingw32-g++.exe -o bin\Debug\temp.exe obj\Debug\main.o
> obj\Debug\printdata.o obj\Debug\cmdline.o
> E:\mysql++-3.0.3\include_lib_a_def\libmysqlpp.a
> Output size is 1.92 MB
> Process terminated with status 0 (0 minutes, 9 seconds)
> 0 errors, 0 warnings
>
> "E:\mysql++-3.0.3\include_mysql6" contains the exact contents of
> C:\Program Files\MySQL\MySQL Server 6.0\include
> "E:\mysql++-3.0.3\include_mysql++" contains all the mysql++ header
> files (the same as the contents of the "include" folder when running
> the default install.bat)
>
> In codeblocks' debugger I get a SIGSEGV. This occurs at the end of
> the file, right around the return 0 in my int main() c++ console
> application. I think the error is occuring when the connection
> destructor is being called. Here is the backtrace.
> #0 6B3C5834 mysqlpp::DBDriver::~DBDriver(this=0x3e3f80)
> (lib/dbdriver.cpp:67)
> #1 6B3C21B7 mysqlpp::Connection::~Connection(this=0x22ff30)
> (lib/connection.cpp:67)
> #2 0040185A main(argc=1, argv=0x3e3f60)
> (E:/C++/codeblocks/Practice/temp/main.cpp:68)
>
> Here is the code for the destructor for dbdriver
> DBDriver::~DBDriver()
> {
> if (connected()) {
> disconnect();
> }
>
> OptionList::const_iterator it;
> for (it = applied_options_.begin(); it != applied_options_.end();
> ++it) {
> delete *it;//THIS IS LINE 67 where the SIGSEGV occurs
> }
> }
>
> I don't think this has to do with mysql6, b/c the default build of
> simple1 does not create the error whereas mine does create the error.
> Hopefully, this means there is a solution.
> What is going on here? Did I link the project correctly?
> Thanks,
>
>

dbdriver sigsegv errors.

2008-06-18T19:19:20Z

Hello,

I am using mysql6. mysql++ seems to be 'working' and 'installed'
correctly; however, at the end of the execution of my programs i am
receiving SIGSEGV errors.

When I installed mysql++ simple1 was built. This version of simple1
does NOT produce the error. However, when I copy the code and try to
build it myself I receive the error at the end of the file when running
the application. I am able to compile, but when I run the program it
has to abort, and when i debug I get the SIGSEGV error at end of the file.

Here is the commands codeblocks used to compile my project titled 'temp'
that contains the exact same code as simple1.cpp.

mingw32-g++.exe -Wall -fexceptions -g
-IE:\mysql++-3.0.3\include_mysql++ -IE:\mysql++-3.0.3\include_mysql6 -c
E:\C++\codeblocks\Practice\temp\main.cpp -o obj\Debug\main.o
mingw32-g++.exe -Wall -fexceptions -g
-IE:\mysql++-3.0.3\include_mysql++ -IE:\mysql++-3.0.3\include_mysql6 -c
E:\C++\codeblocks\Practice\temp\printdata.cpp -o obj\Debug\printdata.o
mingw32-g++.exe -o bin\Debug\temp.exe obj\Debug\main.o
obj\Debug\printdata.o obj\Debug\cmdline.o
E:\mysql++-3.0.3\include_lib_a_def\libmysqlpp.a
Output size is 1.92 MB
Process terminated with status 0 (0 minutes, 9 seconds)
0 errors, 0 warnings

"E:\mysql++-3.0.3\include_mysql6" contains the exact contents of
C:\Program Files\MySQL\MySQL Server 6.0\include
"E:\mysql++-3.0.3\include_mysql++" contains all the mysql++ header files
(the same as the contents of the "include" folder when running the
default install.bat)

In codeblocks' debugger I get a SIGSEGV. This occurs at the end of the
file, right around the return 0 in my int main() c++ console
application. I think the error is occuring when the connection
destructor is being called. Here is the backtrace.
#0 6B3C5834 mysqlpp::DBDriver::~DBDriver(this=0x3e3f80)
(lib/dbdriver.cpp:67)
#1 6B3C21B7 mysqlpp::Connection::~Connection(this=0x22ff30)
(lib/connection.cpp:67)
#2 0040185A main(argc=1, argv=0x3e3f60)
(E:/C++/codeblocks/Practice/temp/main.cpp:68)

Here is the code for the destructor for dbdriver
DBDriver::~DBDriver()
{
if (connected()) {
disconnect();
}

OptionList::const_iterator it;
for (it = applied_options_.begin(); it != applied_options_.end();
++it) {
delete *it;//THIS IS LINE 67 where the SIGSEGV occurs
}
}

I don't think this has to do with mysql6, b/c the default build of
simple1 does not create the error whereas mine does create the error.
Hopefully, this means there is a solution.

What is going on here? Did I link the project correctly?

Thanks,

Re: sql_create_ macro

2008-06-17T19:05:55Z

Actually can see the difference between what works and what doesn't.

If I do this it doesn't work:

bool database::AddRow (string A,
string B,
string C)
{

RowTyoe newRow (mysqlpp::String (A),
mysqlpp::String (B),
mysqlpp::String (C));

}

If I do this it does

bool database::AddRow (string A,
string B,
string C)
{
mysqlpp::String A2 (A);

RowTyoe newRow (A2,
mysqlpp::String (B),
mysqlpp::String (C));

}

G

2008/6/13 Warren Young <mysqlpp@...>:

> On Jun 12, 2008, at 9:36 PM, Soul Boy wrote:
>
>> I can see no difference between my macro declarations other than the
>> number of parameters.
>
> Please show your SSQLS declarations.
>
>
> --
> MySQL++ Mailing List
> For list archives: http://lists.mysql.com/plusplus
> To unsubscribe:
> http://lists.mysql.com/plusplus?unsub=redgreenandpotatoe@...
>
>

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

RE: mysql++ load local infile

2008-06-17T14:05:40Z

Just wanted to say thanks. My problem had to do with using backslashes instead of forward slashes. I was using \\ instead of /
_________________________________________________________________
The other season of giving begins 6/24/08. Check out the i’m Talkathon.
http://www.imtalkathon.com?source=TXT_EML_WLH_SeasonOfGiving

Re: mysql++ load local infile

2008-06-17T10:10:30Z

What exactly is failing with your program?

I do this:

try {
...
string statement = "load data infile '" + fullFileName;
statement += "' into table daytypes ";
statement += "fields terminated by ',' enclosed by '' ";
statement += "lines terminated by '\\r\\n'";
executeSQLStatement(statement);
}
catch (const mysqlpp::Exception& er) {
...stuff
}

executeSQLStatement() simply invokes:

mysqlpp::Query query = mySQLConn->query();
query << SQLStatement;
query.execute();

I'd have to double check, but I believe that it is important that the
slashes in the file name are forward slashes.

Dennis M. wrote:

> I am using mysql++ to access mySQL from a C++ program.
>
> I am able execute various queries except LOAD DATA LOCAL INFILE.
>
> How do I go about doing this? I have not seen an implementation in any of the examples provided in the mysql++ download.
>
> This is what I am doing right now:
>
> #define HOST "localhost"
> #define DB "databse_name"
> #define USER "anonymous"
> #define PASS ""
>
> .
> .
> .
>
> mysqlpp::Connection con(DB, HOST, USER, PASS);
> mysqlpp::NoExceptions ne(con);
> mysqlpp::Query query = con.query();
>
> CString command;
> command = "LOAD DATA LOCAL INFILE \"file_location\" into TABLE name_of_table;";
> query = con.query(command);
> query.execute();
>
>
>
> Is there anything else that needs to be done for this to work, or do I have my query set up incorrectly?
>
>
> Thanks!
>
> _________________________________________________________________
> Need to know now? Get instant answers with Windows Live Messenger.
> http://www.windowslive.com/messenger/connect_your_way.html?ocid=TXT_TAGLM_WL_Refresh_messenger_062008
>

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

mysql++ load local infile

2008-06-17T09:53:36Z

I am using mysql++ to access mySQL from a C++ program.

I am able execute various queries except LOAD DATA LOCAL INFILE.

How do I go about doing this? I have not seen an implementation in any of the examples provided in the mysql++ download.

This is what I am doing right now:

#define HOST "localhost"
#define DB "databse_name"
#define USER "anonymous"
#define PASS ""

.
.
.

mysqlpp::Connection con(DB, HOST, USER, PASS);
mysqlpp::NoExceptions ne(con);
mysqlpp::Query query = con.query();

CString command;
command = "LOAD DATA LOCAL INFILE \"file_location\" into TABLE name_of_table;";
query = con.query(command);
query.execute();

Is there anything else that needs to be done for this to work, or do I have my query set up incorrectly?

Thanks!

_________________________________________________________________
Need to know now? Get instant answers with Windows Live Messenger.
http://www.windowslive.com/messenger/connect_your_way.html?ocid=TXT_TAGLM_WL_Refresh_messenger_062008

Re: sql_create_ macro

2008-06-13T05:05:49Z

On Jun 12, 2008, at 9:36 PM, Soul Boy wrote:

> I can see no difference between my macro declarations other than the
> number of parameters.

Please show your SSQLS declarations.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

sql_create_ macro

2008-06-12T20:36:00Z

Hi I am trying to use the sql_create_4 and sql_create_7 macros to
create a classes which I can pass to
query.insert on. Actually I have created several classes with this
macro and they all work. I just added
a new one with 4 values and suddenly I get the error:

/usr/local/include/mysql++/query.h: In member function
`mysqlpp::Query& mysqlpp::Query::insert(const T&) [with T = AcctMncTb
()(mysqlpp::String, mysqlpp::String, mysqlpp::String,
mysqlpp::String)]':
database.cpp:347: instantiated from here
/usr/local/include/mysql++/query.h:958: error: request for member
`value_list' in `v', which is of non-class type `AcctMncTb
()(mysqlpp::String, mysqlpp::String, mysqlpp::String,
mysqlpp::String)'
/usr/local/include/mysql++/query.h:958: error: request for member
`field_list' in `v', which is of non-class type `AcctMncTb
()(mysqlpp::String, mysqlpp::String, mysqlpp::String,
mysqlpp::String)'
/usr/local/include/mysql++/query.h:958: error: request for member
`table' in `v', which is of non-class type `AcctMncTb
()(mysqlpp::String, mysqlpp::String, mysqlpp::String,
mysqlpp::String)'
*** Error code 1
make: Fatal error: Command failed for target `database.o'

I can see no difference between my macro declarations other than the
number of parameters.

I could see one post on the mailing list referencing these errors but
I couldn't see an answer.

I am using g++ on solaris mysql++ version 3.0.3 mysql version 5.0.51

Any ideas ?

Thanks

G

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: mysqlpp::Query Issue

2008-06-02T22:54:31Z

Hi Warren,

I was saying that if I declare mysqlpp::Query sqlquery(); instead of mysqlpp::Query sqlquery = m_pCommon->commonDBCon.query(); then I was not able to use 'sqlquery' object for another transaction. I tried to do it again to see what happens but it is not compiling because of stream "<<" operation.
Anyway, It seems all is fine now and I also removed CriticalSection calls and now using another Mutex object.
SysMutex is predefined object which I have been using other places and destructor of this class takes care of releasing.

Here is final code...

SysMutex_var mutex = m_pCommon->m_mutexDBAccess;
if (!mutex.waitForMutex(INFINITE))
{
LOG_WARN(logger,"Tel_app::Load:Error getting mutex");
return -1;
}
try
{
// Load the app
mysqlpp::Query sqlquery = m_pCommon->commonDBCon.query();
mysqlpp::StoreQueryResult sqlresults;

// Form query first
sqlquery << "select * from Applications where AppDN=" << AppDN;
sqlresults = sqlquery.store();
if(sqlresults) {
//for (size_t i = 0; i < res.num_rows(); ++i) { // one app only
m_nAppDN = sqlresults[0]["AppDN"];
m_sAppName = sqlresults[0]["AppName"];
m_sAppDesc = sqlresults[0]["AppDescription"];
//}
}
else {
LOG_ERROR(logger,"Failed to get AppInfo");
return -1;
}
// Now, Load Application Steps
sqlquery << "select * from ApplicationSteps where AppDN=" << AppDN;
sqlresults = sqlquery.store();

AppSteps_Module temp2;
if (sqlresults) {
for (size_t i = 0; i < sqlresults.num_rows(); ++i) {
temp2.StepNumber = sqlresults[i]["StepNumber"];
int StepType = sqlresults[i]["StepType"];
temp2.StepType = (Step_Type)StepType;
temp2.PromptID= sqlresults[i]["PromptID"];
temp2.PromptFileName = sqlresults[i]["PromptFileName"];
temp2.PromptText = sqlresults[i]["PromptText"];
temp2.MaxDigits = sqlresults[i]["MaxDigits"];
temp2.MaxSeconds = sqlresults[i]["MaxSeconds"];
temp2.TermTones = sqlresults[i]["TermTones"];
temp2.ValidDigits = sqlresults[i]["ValidDigits"];
AppSteps.push_back(temp2);
}

}
else {
LOG_ERROR(logger,"Failed to get AppSteps");
return -1;
}

// Now, Load Application Step Items
sqlquery << "select * from ApplicationStepItems where AppDN=" << AppDN;
sqlresults = sqlquery.store();
AppStepItems_Module temp;
if (sqlresults) {
for (size_t i = 0; i < sqlresults.num_rows(); ++i) {
//AppStepItems.push_back(sqlresults[i]);
temp.StepNumber = sqlresults[i]["StepNumber"];
temp.AppKeypadId = sqlresults[i]["AppKeypadId"];
temp.NextStepNumber = sqlresults[i]["NextStepNumber"];
AppStepItems.push_back(temp);
}
}
else {
LOG_ERROR(logger,"Failed to get AppStepItems");
return -1;
}
} // try
catch (const mysqlpp::BadQuery& er) {
// Handle any query errors
LOG_ERROR(logger,"TelApp:Query error: " << er.what() );
return -1;
}
catch (const mysqlpp::BadConversion& er) {
// Handle bad conversions; e.g. type mismatch populating 'stock'
LOG_ERROR(logger,"TelApp:Conversion error: " << er.what() << " tretrieved data size: " << er.retrieved <<", actual size: " << er.actual_size);
return -1;
}
catch (const mysqlpp::Exception& er) {
// Catch-all for any other MySQL++ exceptions
LOG_ERROR(logger,"TelApp:Exception: " << er.what() );
return -1;
}
catch(...){
LOG_ERROR(logger,"TelApp:Unknown Exception");
return -1;
}

----- Original Message ----
From: Warren Young <mysqlpp@...>
To: MySQL++ Mailing List <plusplus@...>
Sent: Monday, June 2, 2008 2:00:44 AM
Subject: Re: mysqlpp::Query Issue

onlyreply-sql@... wrote:
> I think the problem was with
> default constructor on query object not properly being initialized.
> Once I initialized sqlquery object like this ... mysqlpp::Query
> sqlquery = commonDBCon.query(); , the problem went away

Are you saying that it doesn't work when you pass the query string to
Connection::query()? The examples do this in several places
successfully, and this is an intended feature, so if you can show an
instance where it doesn't work, I'd be interested in seeing compilable
code that shows it.

> LeaveCriticalSection(&m_csDBAccess);

Not MySQL++ related, but this is highly error-prone code. You should be
using an RAII pattern here to make sure that all code paths that exit
the function release the critical section mutex. It appears fine now,
but future maintenance could add a code path that accidentally doesn't
release the mutex. See the code for MySQL++'s Transaction class -- it's
short -- for the general idea.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=onlyreply-sql@...

Re: loading shared library error

2008-06-02T22:38:08Z

On Tue, Jun 3, 2008 at 3:35 AM, Warren Young <mysqlpp@...> wrote:
> yh wrote:
>>
>> What was I missing?
>
> You will find a deeper study of README-examples.txt rewarding.

Oops, my apology.

Thank you.

Kind Regards,

Jim

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: loading shared library error

2008-06-02T10:35:13Z

yh wrote:
> What was I missing?

You will find a deeper study of README-examples.txt rewarding.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

RE: loading shared library error

2008-06-02T05:47:47Z

Maybe you have to add a path to your LD_LIBRARY_PATH environment variable, pointing to the path under /home/my/Install that contains the libmysqlpp.so that you just built and installed.

Will

-----Original Message-----
From: yh [mailto:jupiter.hce@...]
Sent: Mon 6/2/2008 6:32 PM
To: plusplus@...
Subject: loading shared library error

Hi,

I installed MySQL++ to a local directory, but could not run the
example code. What was I missing?

[mysql++-3.0.3]$ ./configure --prefix=/home/my/Install
[mysql++-3.0.3]$ make
[mysql++-3.0.3]$ make install

[mysql++-3.0.3]$ ./simple1
./simple1: error while loading shared libraries: libmysqlpp.so.3:
cannot open shared object file: No such file or directory

Thank you.

Jim

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=william.dowling@...

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

loading shared library error

2008-06-02T05:28:34Z

Hi,

I installed MySQL++ to a local directory, but could not run the
example code. What was I missing?

[mysql++-3.0.3]$ ./configure --prefix=/home/my/Install
[mysql++-3.0.3]$ make
[mysql++-3.0.3]$ make install

[mysql++-3.0.3]$ ./simple1
./simple1: error while loading shared libraries: libmysqlpp.so.3:
cannot open shared object file: No such file or directory

Thank you.

Jim

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: mysqlpp::Query Issue

2008-06-02T05:13:31Z

Hi:

Try mutex instead critical section, mutexes have the advantage that can be
accessed by the threads of different processes. Critical section is often used
to protect a shared resource in order to synchronize the access order and (one
of the most important restriction for me) you cannot have a wait with a timeout
like mutex or semaphore,

Best regards,

Alex

----- Original Message -----
From: "Warren Young" <mysqlpp@...>
To: "MySQL++ Mailing List" <plusplus@...>
Sent: Monday, June 02, 2008 11:00 AM
Subject: Re: mysqlpp::Query Issue

> onlyreply-sql@... wrote:
>> I think the problem was with
>> default constructor on query object not properly being initialized.
>> Once I initialized sqlquery object like this ... mysqlpp::Query
>> sqlquery = commonDBCon.query(); , the problem went away
>
> Are you saying that it doesn't work when you pass the query string to
> Connection::query()? The examples do this in several places successfully, and
> this is an intended feature, so if you can show an instance where it doesn't
> work, I'd be interested in seeing compilable code that shows it.
>
>> LeaveCriticalSection(&m_csDBAccess);
>
> Not MySQL++ related, but this is highly error-prone code. You should be using
> an RAII pattern here to make sure that all code paths that exit the function
> release the critical section mutex. It appears fine now, but future
> maintenance could add a code path that accidentally doesn't release the mutex.
> See the code for MySQL++'s Transaction class -- it's short -- for the general
> idea.
>
> --
> MySQL++ Mailing List
> For list archives: http://lists.mysql.com/plusplus
> To unsubscribe:
> http://lists.mysql.com/plusplus?unsub=alexis.rodriguez@...
>
>

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: mysqlpp::Query Issue

2008-06-02T01:59:06Z

onlyreply-sql@... wrote:
> I think the problem was with
> default constructor on query object not properly being initialized.
> Once I initialized sqlquery object like this ... mysqlpp::Query
> sqlquery = commonDBCon.query(); , the problem went away

Are you saying that it doesn't work when you pass the query string to
Connection::query()? The examples do this in several places
successfully, and this is an intended feature, so if you can show an
instance where it doesn't work, I'd be interested in seeing compilable
code that shows it.

> LeaveCriticalSection(&m_csDBAccess);

Not MySQL++ related, but this is highly error-prone code. You should be
using an RAII pattern here to make sure that all code paths that exit
the function release the critical section mutex. It appears fine now,
but future maintenance could add a code path that accidentally doesn't
release the mutex. See the code for MySQL++'s Transaction class -- it's
short -- for the general idea.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: mysqlpp::Query Issue

2008-06-01T21:49:04Z

Thanks Warren for detailed reply !
Sorry for garbled code and after reading your reply & user manual , finally I got everything working. I think the problem was with default constructor on query object not properly being initialized. Once I initialized sqlquery object like this ... mysqlpp::Query sqlquery = commonDBCon.query(); , the problem went away and I'm able to use this object with multiple queries/tables. I also removed 'reset' & 'clear' function calls. Here is the final code and please take a look when you get chance and let me know if it's not right way of doing it. I can store result directly in my container also but issue is I use some enum types & also want to restrict some values.

EnterCriticalSection(&m_csDBAccess);
try
{
// Load the app
mysqlpp::Query sqlquery = commonDBCon.query();
mysqlpp::StoreQueryResult sqlresults;

// Form query first
sqlquery << "select * from Applications where AppDN=" << AppDN;
sqlresults = sqlquery.store();
if(sqlresults) {
//for (size_t i = 0; i < res.num_rows(); ++i) { // one app only
m_nAppDN = sqlresults[0]["AppDN"];
m_sAppName = sqlresults[0]["AppName"];
m_sAppDesc = sqlresults[0]["AppDescription"];
//}
}
else {
LOG_ERROR(logger,"Failed to get AppInfo");
LeaveCriticalSection(&m_csDBAccess);
return -1;
}
// Now, Load Application Steps
sqlquery << "select * from ApplicationSteps where AppDN=" << AppDN;
sqlresults = sqlquery.store();

AppSteps_Module temp2;
if (sqlresults) {
for (size_t i = 0; i < sqlresults.num_rows(); ++i) {
temp2.StepNumber = sqlresults[i]["StepNumber"];
int StepType = sqlresults[i]["StepType"];
temp2.StepType = (Step_Type)StepType;
temp2.PromptID= sqlresults[i]["PromptID"];
temp2.PromptFileName = sqlresults[i]["PromptFileName"];
temp2.PromptText = sqlresults[i]["PromptText"];
temp2.MaxDigits = sqlresults[i]["MaxDigits"];
temp2.MaxSeconds = sqlresults[i]["MaxSeconds"];
temp2.TermTones = sqlresults[i]["TermTones"];
temp2.ValidDigits = sqlresults[i]["ValidDigits"];
AppSteps.push_back(temp2);
}
}
else {
LOG_ERROR(logger,"Failed to get AppSteps");
LeaveCriticalSection(&m_csDBAccess);
return -1;
}

// Now, Load Application Step Items
sqlquery << "select * from ApplicationStepItems where AppDN=" << AppDN;
sqlresults = sqlquery.store();
AppStepItems_Module temp;
if (sqlresults) {
for (size_t i = 0; i < sqlresults.num_rows(); ++i) {
//AppStepItems.push_back(sqlresults[i]);
temp.StepNumber = sqlresults[i]["StepNumber"];
temp.AppKeypadId = sqlresults[i]["AppKeypadId"];
temp.NextStepNumber = sqlresults[i]["NextStepNumber"];
AppStepItems.push_back(temp);
}
}
else {
LOG_ERROR(logger,"Failed to get AppStepItems");
LeaveCriticalSection(&m_csDBAccess);
return -1;
}
} // try
catch (const mysqlpp::BadQuery& er) {
// Handle any query errors
LOG_ERROR(logger,"TelApp:Query error: " << er.what() );
LeaveCriticalSection(&m_csDBAccess);
return -1;
}
catch (const mysqlpp::BadConversion& er) {
// Handle bad conversions; e.g. type mismatch populating 'stock'
LOG_ERROR(logger,"TelApp:Conversion error: " << er.what() << " tretrieved data size: " << er.retrieved <<", actual size: " << er.actual_size);
LeaveCriticalSection(&m_csDBAccess);
return -1;
}
catch (const mysqlpp::Exception& er) {
// Catch-all for any other MySQL++ exceptions
LOG_ERROR(logger,"TelApp:Exception: " << er.what() );
LeaveCriticalSection(&m_csDBAccess);
return -1;
}
catch(...){
LOG_ERROR(logger,"TelApp:Unknown Exception");
LeaveCriticalSection(&m_csDBAccess);
return -1;
}

LeaveCriticalSection(&m_csDBAccess);

----- Original Message ----
From: Warren Young <mysqlpp@...>
To: plusplus@...
Sent: Sunday, June 1, 2008 5:27:57 PM
Subject: Re: mysqlpp::Query Issue

(Something in the way you posted this message is messed up: all these
s make reading it a lot harder than it ought to be. Please find
and fix this problem.)

On Jun 1, 2008, at 3:47 AM, onlyreply-sql@... wrote:

> sqlresults.clear();

It's not necessary to clear() the query result object.

> sqlquery.reset();

This reset() call is either misplaced, or unnecessary.

MySQL++ v3 auto-resets the query object after a successful execution
unless you're using template queries.

It doesn't auto-reset with template queries because that would throw
away the template. This doesn't apply here because you're not using
that feature.

It also doesn't auto-reset after an error executing the query because
some people like to send the failed SQL out to the debug log when
reporting query errors. Thus, if I'm wrong and reset() is somehow
needed in your program, the bug may be that you have the reset() call
inside the if { } statement: if the query fails, you need to reset()
the query object before reusing it.

> sprintf(strquery,"select * from ApplicationStepItems where AppDN=
> %d",nAppDN);

It's not relevant to your problem, but I wanted to point out that
there's no need to build the query string separately here. MySQL++
has two built-in mechanisms for building query strings like this
already: template queries, and a C++ stream interface.

Template queries are very much like sprintf(), but I think
inappropriate in this particular instance. They make the most sense
when you re-use the same basic query format multiple times. In your
case, you're giving a different type of query each time.

I'd use the stream interface:

Query q = conn.query("select * from ApplicationStepItems where
AppDN=");
q << nAppDN;

This gives you several advantages. You avoid the need for a temporary
buffer, you avoid the possibility of a buffer overrun, you can use
data types that sprintf() doesn't know how to cope with, and you get
features like automatic quoting and escaping.

> mysqlpp::Query sqlquery2 = pSystem->m_pCommon-
> >commonDBCon.query(strquery);

Here we see that you're not reusing the Query object, so reset()
cannot help. Resetting the first Query object can't affect this one,
even though they use the same Connection object.

> Can anyone please suggest me the solution ?

You haven't given enough information. You just say it doesn't work,
but don't give the error message. I may have accidentally hit on the
solution above, but if you make all these changes and you still get
the same problem, I wouldn't be surprised. Post the error message
you're getting.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=onlyreply-sql@...

Re: mysqlpp::Query Issue

2008-06-01T17:27:57Z

(Something in the way you posted this message is messed up: all these
s make reading it a lot harder than it ought to be. Please find
and fix this problem.)

On Jun 1, 2008, at 3:47 AM, onlyreply-sql@... wrote:

> sqlresults.clear();

It's not necessary to clear() the query result object.

> sqlquery.reset();

This reset() call is either misplaced, or unnecessary.

MySQL++ v3 auto-resets the query object after a successful execution
unless you're using template queries.

It doesn't auto-reset with template queries because that would throw
away the template. This doesn't apply here because you're not using
that feature.

It also doesn't auto-reset after an error executing the query because
some people like to send the failed SQL out to the debug log when
reporting query errors. Thus, if I'm wrong and reset() is somehow
needed in your program, the bug may be that you have the reset() call
inside the if { } statement: if the query fails, you need to reset()
the query object before reusing it.

> sprintf(strquery,"select * from ApplicationStepItems where AppDN=
> %d",nAppDN);

It's not relevant to your problem, but I wanted to point out that
there's no need to build the query string separately here. MySQL++
has two built-in mechanisms for building query strings like this
already: template queries, and a C++ stream interface.

Template queries are very much like sprintf(), but I think
inappropriate in this particular instance. They make the most sense
when you re-use the same basic query format multiple times. In your
case, you're giving a different type of query each time.

I'd use the stream interface:

Query q = conn.query("select * from ApplicationStepItems where
AppDN=");
q << nAppDN;

This gives you several advantages. You avoid the need for a temporary
buffer, you avoid the possibility of a buffer overrun, you can use
data types that sprintf() doesn't know how to cope with, and you get
features like automatic quoting and escaping.

> mysqlpp::Query sqlquery2 = pSystem->m_pCommon-
> >commonDBCon.query(strquery);

Here we see that you're not reusing the Query object, so reset()
cannot help. Resetting the first Query object can't affect this one,
even though they use the same Connection object.

> Can anyone please suggest me the solution ?

You haven't given enough information. You just say it doesn't work,
but don't give the error message. I may have accidentally hit on the
solution above, but if you make all these changes and you still get
the same problem, I wouldn't be surprised. Post the error message
you're getting.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

mysqlpp::Query Issue

2008-06-01T02:47:16Z

Hi All,

I'm not able to reuse mysqlpp::Query object and program throws exception. Although,I'm calling    reset() function on it. Here si the code snippet...

    mysqlpp::Query sqlquery = commonDBCon.query(strquery);
    if (sqlresults = sqlquery.store()) {
            AppDN   = sqlresults[0]["AppDN"];
            AppName = sqlresults[0]["AppName"];
            AppDesc = sqlresults[0]["AppDescription"];
        sqlresults.clear();
        sqlquery.reset();
    }

    // Now, Load Application Step Items
    sprintf(strquery,"select * from ApplicationStepItems where AppDN=%d",nAppDN);
    mysqlpp::Query sqlquery2 = pSystem->m_pCommon->commonDBCon.query(strquery);
    //sqlquery.storein(AppStepItems);
    AppStepItems_Module temp;
    if (sqlresults = sqlquery2.store()) {
        for (size_t i = 0; i < sqlresults.num_rows(); ++i) {
            //AppStepItems.push_back(sqlresults[i]);
            temp.StepNumber = sqlresults[i]["StepNumber"];
            temp.AppKeypadId = sqlresults[i]["AppKeypadId"];
            temp.NextStepNumber = sqlresults[i]["NextStepNumber"];
            AppStepItems.push_back(temp);
      }
}

Can anyone please suggest me the solution ?
My requirements is just to access multiple tables and store records in appropriate structures/templates.

thanks

Re: Deadlock exception

2008-05-30T05:15:23Z

Warren Young wrote:
> use a separate Connection from each thread?

Sorry, wooly thinking... Of course you can get deadlocks with multiple
connections....that's the only way, in fact. Need more caffeine....

Anyway, this is almost certainly not a bug in MySQL++. Not that MySQL++
never has bugs, but that there's no code in MySQL++ for generating this
error code. Unless MySQL++ is somehow corrupting your query strings, I
don't see how the problem could be in MySQL++.

It could be a bug in the server, but that's unlikely if you're running
the current GA release. If you're running something else, at least try
your program against the current GA release on a test server.

If the problem continues when running against the current GA release of
the server, the server is reporting a legitimate problem to your
program. Your program is somehow mistreating the database, and it is
not happy about it.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: Deadlock exception

2008-05-30T04:46:35Z

António Santos wrote:
>
> This happened when I was using transactions but then I removed the use
> of transactions and I get the same result.
> Do anyone know why this is happening? I have a mutex that assures that
> two threads do not affect the same rows at the same time.

First, I want to point out that this error comes from the MySQL server.
MySQL++ is only reporting it.

A Google search for ER_LOCK_DEADLOCK (the underlying MySQL error code)
says there are many more causes for this error than trying to modify
rows at the same time.

Rather than try to fight all of these cases, why not follow the advice
in the MySQL++ user manual, and use a separate Connection from each
thread? You cannot cause a MySQL server deadlock in this situation.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Deadlock exception

2008-05-30T01:30:27Z

Hi there,

I've been using mysql++ 3.0 lately in a multithreaded program and
sometimes, when I issue an INSERT query, I get an exception:
Deadlock found when trying to get lock; try restarting transaction
This happened when I was using transactions but then I removed the use
of transactions and I get the same result.
Do anyone know why this is happening? I have a mutex that assures that
two threads do not affect the same rows at the same time.

Thanks!

Antonio Santos
Portugal

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

MySQL example code

2008-05-29T08:31:00Z

Hi,

Are there any good reference / example code to show how to connect MySQL
database, what is the whole procedule to get query response (data in
buffer) using multithreads and cache?

Thank you.

Kind Regards,

Jim

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: MySQL example code

2008-05-29T07:48:12Z

jupiter.hce@... wrote:
>
> Are there any good reference / example code to show how to connect MySQL
> database, what is the whole procedule to get query response (data in
> buffer) using multithreads and cache?

Read the user manual:

http://tangentsoft.net/mysql++/doc/html/userman/

At the very least, read chapters 1 through 3, plus 7.

Then, read README-examples.txt in the MySQL++ distribution, along with
the README for your platform and/or tool chain. Then play with the
examples.

This should answer most of your questions.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: Miscellaneous questions - compile flags and StoreQueryResult

2008-05-26T20:29:19Z

On May 24, 2008, at 11:24 PM, alpha@... wrote:

> Is there any way can I get rid of need for adding these include and
> library flags every time I want to compile?

These headers are segregated for good reason.

Perhaps you're typing the compilation command by hand each time? If
so, you need to start learning any or all of: Bourne shell scripting,
Make, and Automake. Once you have your build process automated, the
build commands can be as long as you like.

> Also I read that sometimes I might need the flags "-L/usr/lib/mysql
> -L/usr/local/lib/mysql++", though my test script didn't need them.
> In what
> situations would I need these flags?

You'd need them if the libraries in question were stored in the named
directories instead of one searched automatically by the compiler.

> how can I find the number of fields

ResultBase::num_fields():

http://tangentsoft.net/mysql++/doc/html/refman/classmysqlpp_1_1ResultBase.html

Or, get the Fields object from the result object, and walk through
that list.

> *Since StoreQueryResult seems to read the result all at once, is it
> really
> bad/slow/unadvisable for really huge results - say hundreds of
> thousands
> or more rows?

It's covered in the manual: http://tangentsoft.net/mysql++/doc/html/userman/tutorial.html#querytypes

> *How can I uninstall MySQL++?

If you built from source:

$ cd /wherever/you/unpacked/it
# make uninstall

If you installed a binary package, you uninstall it the same way you
uninstall any other package on your system.

It would appear you need to get a book on Linux software development.
Of the questions not already answered in the MySQL++ manual, the rest
relate to topics you're expected to know as a Linux programmer. It is
inappropriate to post such general questions to a special-purpose
mailing list like this one.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: Failing to compile MySQL++

2008-05-26T20:13:26Z

On May 24, 2008, at 1:21 PM, FreezeBurnX wrote:

> if(!connect_to_db(argc,argv,con)){

This function is part of the examples in MySQL++ v2.x, not part of the
MySQL++ library proper. It exists only for the examples; it's not
something to be reused. Instead, create your connection with the
Connection constructor, or with Connection::connect().

Or, better, switch to MySQL++ 3.x. Among its many improvements are
examples that don't hide this aspect of MySQL++ use within unreusable
functions. Look through the first parts of this section of the v3
user manual to see the difference:

http://tangentsoft.net/mysql++/doc/html/userman/tutorial.html

> I've been searching about this for
> more than 4 hours.

A mailing list search for "connect_to_db" pulls up 48 posts, some of
which certainly contain this answer because it's been asked and
answered several times before. I believe it's covered in the Tutorial
chapter of the v2 user manual, too.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Miscellaneous questions - compile flags and StoreQueryResult

2008-05-24T22:24:08Z

I got MySQL++ working, but when compiling I need to add include and
library flags, e.g.:

g++ test.cpp -I/usr/include/mysql -I/usr/local/include/mysql++ -lmysqlpp

Is there any way can I get rid of need for adding these include and
library flags every time I want to compile?

Also I read that sometimes I might need the flags "-L/usr/lib/mysql
-L/usr/local/lib/mysql++", though my test script didn't need them. In what
situations would I need these flags?

And a few more questions:

*In a StoreQueryResult object, how can I find the number of fields, so I
can iterate over both rows and fields in nested for() loops?

*Since StoreQueryResult seems to read the result all at once, is it really
bad/slow/unadvisable for really huge results - say hundreds of thousands
or more rows?

*How can I uninstall MySQL++?

Thanks.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Failing to compile MySQL++

2008-05-24T12:21:57Z

Hello!
#include "include/mysql++.h"

#include <iostream>
#include <iomanip>
#include <string>

using namespace std;

int main(int argc, char *argv[])
{
mysqlpp::Connection con(false);
if(!connect_to_db(argc,argv,con)){
return 1;
}
return 0;
}

Gives me the error
C:\mysql++\main.cpp|12|error: `connect_to_db' was not declared in this
scope|

I have successfully compiled the dll and the examples are working, but I
don't understand what I am doing wrong. I've been searching about this for
more than 4 hours. MySQL version 5.0 is installed, I'm using gcc 3.4.5 and
mingw. IDE is Code::Blocks
Thank you!
--
-Bernardo

--
-Bernardo

Re: SSQLS Limitation (again)

2008-05-23T18:16:20Z

Warren Young wrote:

On May 22, 2008, at 6:16 AM, Brett M wrote:

one of the tables I need to read (not
write to) has 31 columns, I only need the first seven and the very last
column.

Then you don't need 31 column support.

[...]

You only need as many columns declared in the SSQLS as you actually pull
from the DB. (Technically not even that: if you pull more cols from the DB
than is in the SSQLS, the extra just gets sliced off, in v3.)

Excellent, thanks Watten, that works. I did try something very similar to
this but it didn't work, seems so obvious now.

this is an open
source application and I hope many people will want to use it, I can't
expect them all to do the same thing on their systems.

Anyone not interested in changing their MySQL++ code is likely to want
binaries from you, not source. So, if you ever do need more than 25 of
those columns at once, it's probably still not Game Over.

Its an open source application for Linux (and being a gentoo fanboy) I will
be distributing it as source.

[...]

Sure, but technically that then requires that you use LGPL for your own
code, or something compatible.

It also means you're one step behind whenever we change the definition of
the SSQLSes.

Third strike: it isn't necessary, as described above.

I hope to release my application as LGPL anyway but as you point out in
your first answer my problem is solved regardless.

Thanks for the quick reply.
Cheers,
Brett

Re: Searching in MySQL++ mailing list archive

2008-05-22T16:35:39Z

On Mon, May 19, 2008 at 07:10:28AM -0600, Warren Young wrote:
> Yes, it's broken, and seems to have been for quite some time. Since
> this also affects all the other MySQL mail lists, I expected it would
> be fixed without any of us needing to add to the list of complaints.
> I was clearly wrong...

Try MARC in the meantime:

http://marc.info/?l=mysql-plusplus

- Chris

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: license question

2008-05-22T16:30:41Z

Warren Young wrote:

> On May 22, 2008, at 4:25 PM, support wrote:
>
>> mysql++ isn't much use without the mysql C api...
>
> That's true.
>
> Are you just wishing it were not so, or were you about to propose some
> change that *makes* it not so?
>
>> I want to create a closed source app. I want to use mysql++. I don't
>> link to mysql C api or touch it, but mysql++ does.
>> I want to include the mysql++ dll and the mysql C api dll with my app
>> when I
>> distribute it so it 'just works'. I don't mind making the source for
>> those available if I have to. I will not allow any of my code to become
>> GPL or LGPL.
>
> I'm sure you can find a lawyer who is willing to argue that case for
> you. For a fee, of course.
>

SO that is a no.
>> I do not want to spend money.
>
> Something for nothing, eh? :)
>
No sir. Don't assume too much. This is not politics or emotion for me.
The project MUST be closed source. Thats the way it is. Out of my hands.
If money is to be spent, it will be spent on an existing MS SQL
infrastructure. It seems mysql++ can't fit my requirements then. Unless
I can find some info on mysql pricing and work it in the budget.

>>I'm sure you can find a better place to ask that question.

Ok.
Well mysql++ looks like nice work there anyway. :) cheers

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: MySQL++ and Gentoo

2008-05-22T16:21:26Z

On Wed, Oct 24, 2007 at 10:49:38PM +0200, Michael Hanselmann wrote:
> On Wed, Oct 24, 2007 at 04:49:48AM -0600, Warren Young wrote:
> > Michael Hanselmann wrote:
> >> Actually, I'm a Gentoo dev, too, and just bumped the package last week.
> >
> > Thanks! The current .ebuild file is still fine, then?
>
> I have troubles understanding your question. The version in the portage
> tree works, yes.

I was reading through old mailing list archives and saw my name mentioned
in this old thread.

I took a quick look at the difference between gentoo's latest ebuild
and mysql++'s. I've attached the diff for reference.

The official gentoo ebuild is still on version 2.3.2, and they apply
a patch for gcc-4.3. I've attached their patch as well, for reference.

I think mysql++ is supported well enough in Gentoo now that you could
probably remove the ebuild from your source tarball and nobody would
notice. :-) The Gentoo world lives off the portage tree anyway.

I'm not an active Gentoo user anymore, so I'm probably not the best
maintainer for the ebuild.

- Chris

--- mysql++.ebuild 2008-05-16 19:35:07.000000000 -0400
+++ /home/cdfrey/incoming/mysql++-2.3.2.ebuild 2008-05-22 18:48:43.000000000 -0400
@@ -1,43 +1,58 @@
-# Copyright 1999-2004 Gentoo Foundation
+# Copyright 1999-2008 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
+# $Header: /var/cvsroot/gentoo-x86/dev-db/mysql++/mysql++-2.3.2.ebuild,v 1.3 2008/04/21 03:00:49 dirtyepic Exp $

-inherit gcc eutils gnuconfig
+inherit eutils

DESCRIPTION="C++ API interface to the MySQL database"
-# This is the download page but includes links to other places
HOMEPAGE="http://tangentsoft.net/mysql++/"
-SRC_URI_BASE="http://tangentsoft.net/mysql++/releases"
-SRC_URI="${SRC_URI_BASE}/${P}.tar.gz"
+SRC_URI="http://www.tangentsoft.net/mysql++/releases/${P}.tar.gz"

LICENSE="LGPL-2"
SLOT="0"
-KEYWORDS="x86 ~alpha ~hppa ~mips ~sparc ~ppc ~amd64"
+KEYWORDS="~alpha ~amd64 ~hppa ~mips ~ppc ~sparc ~x86"
IUSE=""

-DEPEND=">=dev-db/mysql-3.23.49"
+DEPEND=">=sys-devel/gcc-3"
+RDEPEND="${DEPEND}
+ >=virtual/mysql-4.0"

src_unpack() {
- unpack ${P}.tar.gz
+ unpack ${A}
+ cd "${S}"
+
+ epatch "${FILESDIR}"/${P}-gcc-4.3.patch
+
+ for i in "${S}"/lib/*.h ; do
+ sed -i \
+ -e '/#include </s,mysql.h,mysql/mysql.h,g' \
+ -e '/#include </s,mysql_version.h,mysql/mysql_version.h,g' \
+ "${i}" || die "Failed to sed ${i} for fixing MySQL includes"
+ done
}

src_compile() {
- gnuconfig_update
- # not including the directives to where MySQL is because it seems to find it
- # just fine without
+ local myconf
+ # we want C++ exceptions turned on
+ myconf="--enable-exceptions"
+ # give threads a try
+ myconf="${myconf} --enable-thread-check"
+ # not including the directives to where MySQL is because it seems to
+ # find it just fine without
+
# force the cflags into place otherwise they get totally ignored by
# configure
- CFLAGS="${CFLAGS}" CXXFLAGS="${CFLAGS} ${CXXFLAGS}" econf || \
- die "econf failed"
+ CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS}" \
+ econf ${myconf} || die "econf failed"

emake || die "unable to make"
}

src_install() {
- make DESTDIR=${D} install || die
+ emake DESTDIR="${D}" install || die
# install the docs and HTML pages
- dodoc README* LICENSE.txt
+ dodoc README* CREDITS ChangeLog HACKERS Wishlist
dodoc doc/*
- dohtml doc/html/*
+ cp -ra doc/html "${D}"/usr/share/doc/${PF}/html
prepalldocs
}
-

diff -Naur mysql++-2.3.2-orig/lib/const_string.h mysql++-2.3.2/lib/const_string.h
--- mysql++-2.3.2-orig/lib/const_string.h 2008-03-22 16:16:39.000000000 -0600
+++ mysql++-2.3.2/lib/const_string.h 2008-03-22 16:18:47.000000000 -0600
@@ -35,6 +35,7 @@
#include <iostream>
#include <stdexcept>
#include <string>
+#include <cstring>

namespace mysqlpp {

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: license question

2008-05-22T15:43:32Z

On May 22, 2008, at 4:25 PM, support wrote:

> mysql++ isn't much use without the mysql C api...

That's true.

Are you just wishing it were not so, or were you about to propose some
change that *makes* it not so?

> I want to create a closed source app. I want to use mysql++. I don't
> link to mysql C api or touch it, but mysql++ does.
> I want to include the mysql++ dll and the mysql C api dll with my
> app when I
> distribute it so it 'just works'. I don't mind making the source for
> those available if I have to. I will not allow any of my code to
> become
> GPL or LGPL.

I'm sure you can find a lawyer who is willing to argue that case for
you. For a fee, of course.

> I do not want to spend money.

Something for nothing, eh? :)

Hard cash isn't the only token that'll make this sociolegal carousel
go 'round. If you read the MySQL++ Wishlist, you'll find a plan for
making MySQL++ talk to other databases. If you're willing to work
with us to develop that, it solves your problem.

Just offering it as an option to consider...

> what about the mysql ODBC connector?

I'm sure you can find a better place to ask that question.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...

Re: license question

2008-05-22T15:22:20Z

Warren Young wrote:

> > On May 22, 2008, at 1:52 PM, support wrote:
> >
>
>> >> I'm evaluating mysql++ and I would like to create a closed source app
>> >> using it. Likely it would stay in house but we need to be able to
>> >> distribute it in the future if we should decide to do so. As long as we
>> >> link dynamically, this is possible, correct?
>>
> >
> > IANAL, and I don't speak for any other entity, not even the company I
> > work for. My informed opinion is, "yes". :)
> >
>
>> >> Can we distribute the mysql++ .dll/.so binary with our application?
>>
> >
> > Yes, as long as you're willing to provide the source code that
> > produced that DLL to anyone who asks. The exact terms of this
> > requirement are spelled out in the LGPL.
> >
>
>> >> Same question for the mysql C api .dll/.so..?
>>
> >
> > The MySQL C API is under the GPL, not the LGPL, so it doesn't matter
> > how you link: your app falls under the GPL unless you buy a commercial
> > redistribution license from MySQL, Inc.
> >
>

oops, sorry I reponded to you and not list..

Now i'm confused. mysql++ isn't much use without the mysql C api... So
it really IS NOT possible to do what I want, unless I buy a license from
MySQL? I cannot open-source this app.

Wait, maybe you misunderstood my second question. ok:
I want to create a closed source app. I want to use mysql++. I don't link to mysql C api or touch it, but mysql++ does.
I want to include the mysql++ dll and the mysql C api dll with my app when I
distribute it so it 'just works'. I don't mind making the source for
those available if I have to. I will not allow any of my code to become
GPL or LGPL. I do not want to spend money. Is it possible?

If not, what about the mysql ODBC connector?

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsub=lists@...