Re: [Boost-users] [serialization] Wrapper + make_nvp woes
Thank you for the fast reply, Robert.
Another idea is to cast your character array to a ... character array - like static_cast< int [64] &>(my array address) but that would require a constant size which might not work for you.
I'm not certain I quite understand this. Perhaps I didn't get my
thoughts down coherently. I'm seeking to write my serialization like
this (sorry I forgot to include this in my initial msg):
char some_array[5] = "test";
....
ar & make_nvp("some_array", make_char_array(some_array, 5));
And obtain the following output:
output.xml:
...
Finally you can use the blunt instrument approach.
save(.... std::string temp_string; // copy data into temp_string ar << temp_string
load .. do the opposite.
Which is pretty much what my serialization wrapper is doing (this is in fact, what I wrote initially, before trying to make it a serialization wrapper). While I can certainly do the blunt approach, it will require me to split my serialization function :(
Hmmm - I think I understand better what you want.
I'm not sure the easiest way to do it.
Maybe one might specialize nvp
Thank you for the fast reply, Robert.
Another idea is to cast your character array to a ... character array - like static_cast< int [64] &>(my array address) but that would require a constant size which might not work for you.
I'm not certain I quite understand this. Perhaps I didn't get my thoughts down coherently. I'm seeking to write my serialization like this (sorry I forgot to include this in my initial msg):
char some_array[5] = "test"; .... ar & make_nvp("some_array", make_char_array(some_array, 5));
And obtain the following output:
output.xml: ...
test output.txt: .... 4 test .... ---------------------------
My aforementioned make_char_array() works fine for text archives--just not when the result is composed with make_nvp().
Finally you can use the blunt instrument approach.
save(.... std::string temp_string; // copy data into temp_string ar << temp_string
load .. do the opposite.
Which is pretty much what my serialization wrapper is doing (this is in fact, what I wrote initially, before trying to make it a serialization wrapper). While I can certainly do the blunt approach, it will require me to split my serialization function :(
Maybe one might specialize nvp
(...
I hadn't thought of this. I'm just now getting the opportunity to use a modern compiler, so I'm feeling my way through all these new template abilities.
The "standard" way would be to make a derivation of xml_archive
Actually, that was the direction I was just starting to investigate this morning. Thanks for all your ideas. I feel certain I'll be able to get something going now. -Dan
participants (2)
-
Dan Thill
-
Robert Ramey