QxOrm  1.2.9
C++ Object Relational Mapping library
eos::portable_iarchive Class Reference

Portable binary input archive using little endian format. More...

#include <portable_iarchive.hpp>

List of all members.

Classes

struct  dummy

Public Member Functions

template<class T >
boost::disable_if
< boost::is_arithmetic< T >
>::type 
load (T &t, dummy< 1 >=0)
 Default fall through for non-arithmetic types (ie. strings)
void load (bool &b)
template<typename T >
boost::enable_if
< boost::is_integral< T >
>::type 
load (T &t, dummy< 2 >=0)
 Load integer types.
template<typename T >
boost::enable_if
< boost::is_floating_point< T >
>::type 
load (T &t, dummy< 3 >=0)
 Load floating point types.
 portable_iarchive (std::istream &is, unsigned flags=0)
 always construct on a stream using ios::binary mode!

Private Member Functions

signed char load_signed_char ()

Detailed Description

Portable binary input archive using little endian format.

This archive addresses integer size, endianness and floating point types so that data can be transferred across different systems. There may still be constraints as to what systems are compatible and the user will have to take care that e.g. a very large int being saved on a 64 bit machine will result in a portable_archive_exception if loaded into an int on a 32 bit system. A possible workaround to this would be to use fixed types like boost::uint64_t in your serialization structures.

Note:
The class is based on the portable binary example by Robert Ramey and uses Beman Dawes endian library plus fp_utilities by Johan Rade.
Robert Ramey changed his example in boost 1.36 and no longer inherits from binary_iarchive_impl but goes down to basic_binary_iprimitive!?

Definition at line 152 of file portable_iarchive.hpp.


Constructor & Destructor Documentation

eos::portable_iarchive::portable_iarchive ( std::istream &  is,
unsigned  flags = 0 
) [inline]

always construct on a stream using ios::binary mode!

Definition at line 278 of file portable_iarchive.hpp.


Member Function Documentation

template<class T >
boost::disable_if<boost::is_arithmetic<T> >::type eos::portable_iarchive::load ( T &  t,
dummy< 1 >  = 0 
) [inline]

Default fall through for non-arithmetic types (ie. strings)

Definition at line 175 of file portable_iarchive.hpp.

void eos::portable_iarchive::load ( bool &  b) [inline]

Special case loading bool type, preserving compatibility to integer types - this is somewhat redundant but simply treating bool as integer type generates lots of warnings

Definition at line 183 of file portable_iarchive.hpp.

template<typename T >
boost::enable_if<boost::is_integral<T> >::type eos::portable_iarchive::load ( T &  t,
dummy< 2 >  = 0 
) [inline]

Load integer types.

First we load the size information ie. the number of bytes that hold the actual data. Then we retrieve the data and transform it to the original value by using load_little_endian.

Definition at line 202 of file portable_iarchive.hpp.

template<typename T >
boost::enable_if<boost::is_floating_point<T> >::type eos::portable_iarchive::load ( T &  t,
dummy< 3 >  = 0 
) [inline]

Load floating point types.

We simply rely on fp_traits to set the bit pattern from the (unsigned) integral type that was stored in the stream.

Note:
by Johan Rade (author of the floating point utilities library): Be warned that the math::detail::fp_traits<T>::type::get_bits() function is *not* guaranteed to give you all bits of the floating point number. It will give you all bits if and only if there is an integer type that has the same size as the floating point you are copying from. It will not give you all bits for double if there is no uint64_t. It will not give you all bits for long double if sizeof(long double) > 8 or there is no uint64_t.

The member fp_traits<T>::type::coverage will tell you whether all bits are copied. This is a typedef for either math::detail::all_bits or math::detail::not_all_bits.

If the function does not copy all bits, then it will copy the most significant bits. So if you serialize and deserialize the way you describe, and fp_traits<T>::type::coverage is math::detail::not_all_bits, then your floating point numbers will be truncated. This will introduce small rounding off errors.

Note:
treat nan values using fpclassify

Definition at line 253 of file portable_iarchive.hpp.

signed char eos::portable_iarchive::load_signed_char ( ) [inline, private]

Definition at line 163 of file portable_iarchive.hpp.


The documentation for this class was generated from the following file: