I'm trying to determine the proper way to use prepared statements with QSqlQuery. The docs are not very specific on this subject.
void select(const QSqlDatabase &database) {
QSqlQuery query(database);
query.prepare("SELECT * FROM theUniverse WHERE planet = :planet");
query.bindValue(":planet", "earth");
query.exec();
}
So will this snippet create a permanent prepared statement in the connection database? Will this prepared statement persist between calls to select(), i.e. will it be saved when the function returns and QSqlQuery query is disposed?
Or should I create QSqlQuery on the heap and use the same instance over and over again?
Ok, the idea is that you have to create QSqlQuery on heap, prepare the query and do the following with it:
QSqlQuery::bindValue(s)
QSqlQuery::exec
read data with QSqlQuery::[next|first|last|...]
QSqlQuery::finish
rinse and repeat
the following snipped is useful to create, prepare and retrieve queries on heap:
QSqlDatabase database;
QMap<QString, QSqlQuery *> queries; //dont forget to delete them later!
QSqlQuery *prepareQuery(const QString &query)
{
QSqlQuery *ret = 0;
if (!queries.contains(query)) {
QSqlQuery *q = new QSqlQuery(database);
q->prepare(query);
queries[query] = ret = q;
} else {
ret = queries[query];
}
}
return ret;
}
Related
I am willing to store objects in a database. The purpose is to be able to read / write these objects with the program. The requirements are the following:
Objects can be complex using Qt classes such as QList, QString ... or even can contain other objects that use QObjects
The database should not be readable or modified by human (no text file, and if I use sqlite database, it has to be encrypted in a way)
I should be able to remove, read an object by its name and count the number of objects in the database, without loading everything in the memory
I asked a question here, to do this with a QDataStream with a minimalist example. But it seems it is not the best way to proceed.
Would you have some suggestions regarding the solutions that exist for this purpose?
I have tried the following but it does not fulfill the requirements:
Storing text in sqlite with QtSQL: but the data is accessible by using sqlitemanager for example, can be modified or removed by humans. Moreover, I have no idea regarding the way to store QList for example or other objects that I created and contain QObject (for example, 2 QList)
Storing binary data using QDataStream: in this case, I cannot count the number of objects in my file, neither read or remove a specific object without loading the entire file in memory.
I would be grateful if you could give me some suggestions or provide example, even if the example is minimalist.
I finally found a solution, especially thanks to Igor Tandetnik and thanks to the topic here
I haven't quite finalized, there is a small imperfection because I have to define an object of my user class that I don't use in order to call the readFromDB function to generate my object from the db.
On the other hand, I get this error message "QSqlDatabasePrivate::addDatabase: duplicate connection name 'qt_sql_default_connection', old connection removed" each time I call my database.
Anyway, it's a bit late now, and I think it might help some people so I post this minimalist imperfect code below. I'll post an update in the next few days.
Thanks again.
#include "QString"
#include "QFile"
#include "QDataStream"
#include "qdebug.h"
#include "QtSql"
#include "QSqlDatabase"
#include "qmessagebox.h"
class User
{
protected:
QString name;
QList<QString> childrens;
public:
QString getName(){ return name;}
QList<QString> getChildrens(){ return childrens;}
void setName(QString x) {name = x;}
void setChildrens(QList<QString> x) {childrens = x;}
friend QDataStream &operator<<(QDataStream &out, const User &t)
{
out << t.name << t.childrens;
return out;
}
friend QDataStream &operator>>(QDataStream &in, User &t)
{
QString inname;
QList<QString> inchildrens;
in >> inname >> inchildrens;
t.name = inname;
t.childrens = inchildrens;
return in;
}
QByteArray object2blob( const User& user )
{
QByteArray result;
QDataStream bWrite( &result, QIODevice::WriteOnly );
bWrite << user;
return result;
}
User blob2object( const QByteArray& buffer )
{
User result;
QDataStream bRead( buffer );
bRead >> result;
return result;
}
int saveToDB( const User& user )
{
QSqlDatabase myDB = QSqlDatabase::addDatabase("QSQLITE");
myDB.setDatabaseName( "file.db");
if (!myDB.open())
{
qDebug()<<"Failed to open SQL database of registered users";
}
else
{
qDebug()<<"Successfully opening SQL database of registered users";
QSqlQuery query;
query.prepare( "CREATE TABLE IF NOT EXISTS users (name TEXT, childrens BLOB)" );
if( !query.exec() )
{
qDebug() << query.lastError();
}
else
{
qDebug() << "Table created!";
query.prepare( "INSERT INTO users (name,childrens) VALUES (:name,:childrens)" );
query.bindValue(":name", name);
query.bindValue( ":childrens", object2blob(user) );
query.exec();
}
query.clear();
myDB.close();
}
QSqlDatabase::removeDatabase("UserConnection");
return 0;
}
User readFromDB( QString name )
{
User result;
QSqlDatabase myDB = QSqlDatabase::addDatabase("QSQLITE");
myDB.setDatabaseName( "file.db");
if (!myDB.open())
{
qDebug()<<"Failed to open SQL database of registered users";
}
else
{
QSqlQuery query;
query.prepare( "SELECT * FROM users WHERE name ='"+ name +"'" );
//query.bindValue( 0, name );
if ( query.exec() && query.next() ) {
result = blob2object( query.value( 1 ).toByteArray() );
}
query.clear();
myDB.close();
}
QSqlDatabase::removeDatabase("UserConnection");
qDebug()<<result.getChildrens();
return result;
}
};
////////////////////////////////////////////////////////////////
int main()
{
User u;
u.setName("Georges");
u.setChildrens(QList<QString>()<<"Jeanne"<<"Jean");
u.saveToDB(u);
User v;
v.setName("Alex");
v.setChildrens(QList<QString>()<<"Matthew");
v.saveToDB(v);
User w;
w.setName("Mario");
w.saveToDB(w);
User to_read; //to improve here
User a = to_read.readFromDB("Georges");
qDebug()<<a.getChildrens();
return 0;
}
I'm new to Pro-C and Oracle (but not databases) so I probably need some fundamental help on what I'm doing wrong. I wrote the following Pro*C program to query for all the tables owned by a specific user:
#include <stdio.h>
EXEC SQL INCLUDE SQLCA;
int main() {
int i;
char oraCreds[10] = "user/pass";
char tables[50][50], parts[50][50];
exec sql connect :oraCreds;
if (sqlca.sqlcode != 0) {
printf("ERROR abbreviated\n");
return(0);
}
EXEC SQL SELECT TABLE_NAME
INTO :tables
FROM USER_TABLES;
if (sqlca.sqlcode != 0) {
printf("ERROR: Could not query database for user tables\n");
printf("\tError Code: %d\n", sqlca.sqlcode);
printf("\t%.*s\n", sqlca.sqlerrm.sqlerrml, sqlca.sqlerrm.sqlerrmc);
} else {
for (i = 0; i < 50; i++) {
printf("\nTable %d: %s\n", i+1, tables[i]);
}
}
// Code below inserted here
return(0);
}
This produces the following error:
ERROR: Could not query database for user tables
Error Code: 1403
ORA-01403: no data found
The problem is that there are records in user_tables and in sqlplus I can do a typical SELECT TABLE_NAME FROM USER_TABLES; in order to verify that.
In an attempt to verify some of the basic functionality of above program, I inserted the following query at the comment.
EXEC SQL SELECT NAME, PART_NO
INTO :tables, :parts
FROM ITEMS;
if (sqlca.sqlcode != 0) {
printf("ERROR: Could not query database for user tables\n");
printf("\tError Code: %d\n", sqlca.sqlcode);
printf("\t%.*s\n", sqlca.sqlerrm.sqlerrml, sqlca.sqlerrm.sqlerrmc);
} else {
for (i = 0; i < 50; i++) {
printf("\nTable %d: %s\n", i+1, tables[i]);
}
}
This second query gives the expected results and prints out one line for each item's name.
If I understand correctly, the table, user_tables, is a standard part of Oracle, so is there something special about it that would be throwing off my query? I've started reading some alternate ways to query for the user's tables, but I would love to know why the above method does not work.
I need to implement a Linux Kernel Driver, that (in the first step) only forwards all file operations to another file (in later steps, this should be managed and manipulated, but I don't want to discuss this here).
My idea is the following, but when reading, the kernel crashes:
static struct {
struct file *file;
char *file_name;
int open;
} file_out_data = {
.file_name = "/any_file",
.open = 0,
};
int memory_open(struct inode *inode, struct file *filp) {
PRINTK("<1>open memory module\n");
/*
* We don't want to talk to two processes at the same time
*/
if (file_out_data.open)
return -EBUSY;
/*
* Initialize the message
*/
Message_Ptr = Message;
try_module_get(THIS_MODULE);
file_out_data.file = filp_open(file_out_data.file_name, filp->f_flags, filp->f_mode); //here should be another return handling in case of fail
file_out_data.open++;
/* Success */
return 0;
}
int memory_release(struct inode *inode, struct file *filp) {
PRINTK("<1>release memory module\n");
/*
* We're now ready for our next caller
*/
file_out_data.open--;
filp_close(file_out_data.file,NULL);
module_put(THIS_MODULE);
/* Success */
return 0;
}
ssize_t memory_read(struct file *filp, char *buf,
size_t count, loff_t *f_pos) {
PRINTK("<1>read memory module \n");
ret=file_out_data.file->f_op->read(file_out_data.file,buf,count,f_pos); //corrected one, false one is to find in the history
return ret;
}
So, can anyone please tell me why?
Don't use set_fs() as there is no reason to do it.
Use file->f_fop->read() instead of the vfs_read. Take a look at the file and file_operations structures.
Why are you incrementing file_out_data.open twice and decrementing it once? This could cause you to use file_out_data.file after it has been closed.
You want to write memory in your file ou read?
Because you are reading and not writing...
possible i'm wrong
I have inherited a SQLite database that is poorly commented and I need to understand what it is querying and receiving. Is there a way to see the value a sqlite3_stmt pointer is pointing to? Is it something that can be printed to the console in a printf() or NSLog()?
Thanks
The best way to do it is to trace the compiled statement. I've found a solution with the help of this question:
The sqlite3_trace command triggers a callback, which you need to code yourself. Once the trace is turned on it will execute for the rest of the program.
I've got all my database access routines in a single class, and I've used this as the callback function (it's a C function, not a method).
void traceCallback( void* udp, const char* sql )
{
printf("{SQL} [%s]\n", sql);
}
I turned the trace on in an init database method: it's turned on once the database is opened.
-(void)initializeDatabase
{
NSString *path = [self createEditableCopyOfDatabaseIfNeeded];
// open the db
if (sqlite3_open([path UTF8String], &db) == SQLITE_OK)
sqlite3_trace(db, traceCallback, NULL);
else {
// error - cleanup
sqlite3_close(db);
NSLog(#"Error opening db");
NSLog(#"Path: %#",path);
}
}
This will turn a SQL statement with bound variables into a string.
From this:
const char *sql2 = "select a.key, b.key from words a left outer join known_words b on a.key = b.word_id where a.word_foreign = :word_foreign";
To this:
{SQL} [select a.key, b.key from words a left outer join known_words b on a.key = b.word_id where a.word_foreign = 'bak.']
I'm using Visual C++ 2005 and would like to know the simplest way to connect to a MS SQL Server and execute a query.
I'm looking for something as simple as ADO.NET's SqlCommand class with it's ExecuteNonQuery(), ExecuteScalar() and ExecuteReader().
Sigh offered an answer using CDatabase and ODBC.
Can anybody demonstrate how it would be done using ATL consumer templates for OleDb?
Also what about returning a scalar value from the query?
With MFC use CDatabase and ExecuteSQL if going via a ODBC connection.
CDatabase db(ODBCConnectionString);
db.Open();
db.ExecuteSQL(blah);
db.Close();
You should be able to use OTL for this. It's pretty much:
#define OTL_ODBC_MSSQL_2008 // Compile OTL 4/ODBC, MS SQL 2008
//#define OTL_ODBC // Compile OTL 4/ODBC. Uncomment this when used with MS SQL 7.0/ 2000
#include <otlv4.h> // include the OTL 4.0 header file
#include <stdio>
int main()
{
otl_connect db; // connect object
otl_connect::otl_initialize(); // initialize ODBC environment
try
{
int myint;
db.rlogon("scott/tiger#mssql2008"); // connect to the database
otl_stream select(10, "select someint from test_tab", db);
while (!select.eof())
{
select >> myint;
std::cout<<"myint = " << myint << std::endl;
}
}
catch(otl_exception& p)
{
std::cerr << p.code << std::endl; // print out error code
std::cerr << p.sqlstate << std::endl; // print out error SQLSTATE
std::cerr << p.msg << std::endl; // print out error message
std::cerr << p.stm_text << std::endl; // print out SQL that caused the error
std::cerr << p.var_info << std::endl; // print out the variable that caused the error
}
db.logoff(); // disconnect from the database
return 0;
}
The nice thing about OTL, IMO, is that it's very fast, portable (I've used it on numerous platforms), and connects to a great many different databases.
I used this recently:
#include <ole2.h>
#import "msado15.dll" no_namespace rename("EOF", "EndOfFile")
#include <oledb.h>
void CMyDlg::OnBnClickedButton1()
{
if ( FAILED(::CoInitialize(NULL)) )
return;
_RecordsetPtr pRs = NULL;
//use your connection string here
_bstr_t strCnn(_T("Provider=SQLNCLI;Server=.\\SQLExpress;AttachDBFilename=C:\\Program Files\\Microsoft SQL Server\\MSSQL.1\\MSSQL\\Data\\db\\db.mdf;Database=mydb;Trusted_Connection=Yes;MARS Connection=true"));
_bstr_t a_Select(_T("select * from Table"));
try {
pRs.CreateInstance(__uuidof(Recordset));
pRs->Open(a_Select.AllocSysString(), strCnn.AllocSysString(), adOpenStatic, adLockReadOnly, adCmdText);
//obtain entire restult as comma separated text:
CString text((LPCWSTR)pRs->GetString(adClipString, -1, _T(","), _T(""), _T("NULL")));
//iterate thru recordset:
long count = pRs->GetRecordCount();
COleVariant var;
CString strColumn1;
CString column1(_T("column1_name"));
for(int i = 1; i <= count; i++)
{
var = pRs->GetFields()->GetItem(column1.AllocSysString())->GetValue();
strColumn1 = (LPCTSTR)_bstr_t(var);
}
}
catch(_com_error& e) {
CString err((LPCTSTR)(e.Description()));
MessageBox(err, _T("error"), MB_OK);
_asm nop; //
}
// Clean up objects before exit.
if (pRs)
if (pRs->State == adStateOpen)
pRs->Close();
::CoUninitialize();
}
Try the Microsoft Enterprise Library. A version should be available here for C++. The SQlHelper class impliments the methods you are looking for from the old ADO days. If you can get your hands on version 2 you can even use the same syntax.