Add trade history levelDB, trade_MP call, clean up MetaDex populateRPCTransactionObject

[zathras-crypto]

1) gettransaction_MP now works for any metadex order not just open ones 2) populateRPCTransactionObject now parses off the wire for metadex just like all other types 3) new levelDB MP_tradelist 4) every completed trade added to tradelistDB with txids, addresses, properties, and amounts 5) new gettrade_MP call added 6) gettrade_MP will iter metadex map to determine if offer still active 7) new getMatchedTrades function added using tradelistDB to return all trades matched with a supplied txid 8) gettrade_MP will call getMatchedTrades and supply to user over RPC 9) mscrpc 7 function to dump tradelistDB

example http://pastie.org/pastes/9698921/text?key=a5n5cqglbd80c7tfytwgxw

[dexX7]

That's pretty awesome. Will start some tests in a few minutes.

Sidenote: git rebase -i HEAD~26 can be used to combine (stash) some commits for easier code review (e.g. grouped into persistence, rpc, ...). Overwriting already pushed stuff is possible with -f in git push remotename -f branchname. That's of course no demand to do so, but might be useful from time to time. :)

[ghost]

this looks good to me ... @m21 to expedite ? so we can have usable functionality for the rest of the RPCs

[dexX7]

Tested a few transactions, works great! I posted some comments inline.

The only thing I'm really not sure about is mastercore_rpc.cpp#L2125-L2133:

for (md_PricesMap::iterator it = prices.begin(); it != prices.end(); ++it)
{
    md_Set & indexes = (it->second);
    for (md_Set::iterator it = indexes.begin(); it != indexes.end(); ++it)
    {
         CMPMetaDEx obj = *it;
         if( obj.getHash().GetHex() == hash.GetHex() ) orderOpen = true;
    }
}

Declaring variables multiple times seems to be legal, but how do you get away with it?

The iterator it is declared two times here and both loops modify it.

[m21]

They are local to the for-loops Dexx: http://stackoverflow.com/questions/7959573/declaring-variables-inside-loops-good-practice-or-bad-practice-2-parter

[dexX7]

@m21: the link doesn't really nail it, but it makes sense and works. Interesting realization for me.. :)

See this example: http://ideone.com/pc8h37