Let's say you have a result object.

For example, your program may have done:

pqxx::result r = w.exec("SELECT * FROM mytable");

Now how do you access the data inside r?

The simplest way is array indexing. A result acts as an array of tuples, and a tuple acts as an array of fields.

for (int rownum=0; rownum < r.size(); ++rownum)
{
  const result::tuple row = r[rownum];
  for (int colnum=0; colnum < row.size(); ++colnum)
  {
    const result::field = row[colnum];
    std::cout << field.c_str() << '\t';
  }
  std::cout << std::endl;
}

But results and rows also define const_iterator types:

for (pqxx::result::const_iterator row = r.begin();
     row != r.end();
     ++row)
 {
   for (pqxx::result::tuple::const_iterator field = row->begin();
        field != row->end();
        ++field)
     std::cout << field->c_str() << '\t';
   std::cout << std::endl;
 }

They also have const_reverse_iterator types, which iterate backwards from rbegin() to rend() exclusive.

All these iterator types provide one extra bit of convenience that you won't normally find in C++ iterators: referential transparency. You don't need to dereference them to get to the row or field they refer to. That is, instead of row->end() you can also choose to say row.end(). Similarly, you may prefer field.c_str() over field->c_str().

This becomes really helpful with the array-indexing operator. With regular C++ iterators you would need ugly expressions like (*row)[0] or row->operator[](0). With the iterator types defined by the result and tuple classes you can simply say row[0].