letitplay/cryptowizards-smartcontract

CryptoWizards smart-contract

Cryptowizards are digital characters. Different and colorful they can be bought, collected, and auctioned. But ultimately they exist to compete in tournaments for real and virtual prizes. To read more please visit https://wizards.one

Cryptowizards are built on top of EOS blackchain. Current version of smart-contract deplyed at wizardstoken is available here.

Building instruction

eosiocpp -o build/wizards.wast wizContract.cpp
eosiocpp -g build/wizards.abi wizContract.cpp

Your wast file may differ based on parameters set in eosiocpp.

cryptowizards-smartcontract's Issues

difference from textual battle rules

CryptoWizards-SmartContract/pvp/pvp.cpp

Lines 501 to 504 in 84e9a06

	round.max_attack = getWizOffenceAt(fighter1, genoIndex) * powerCoef;
	round.cur_attack = next_rand(round.max_attack + 1);
	round.max_defense_opp = getWizDeffenceAt(fighter2, genoIndex);
	round.cur_defense_opp = next_rand(round.max_defense_opp + 1);

Here, the actual attack and defense can be 0.
But the DNA battles description said they are both at least 1.
So, which one is the intended behavior, the code or the document?

outdated contract published?

CryptoWizards-SmartContract/pvp/pvp.cpp

Line 576 in 84e9a06

runFightRoundGenotype(fighter1, fighter2, damageGot1, damageGot2, attrIndex1, attrIndex2, curLog);

This does not seem to be the latest version of the contract.
In the published version, the default value for the powerCoef of runFightRoundGenotype is commented out, so this call would not compile.

market should disallow 0.0000 EOS sales

CryptoWizards-SmartContract/marketplace/marketplace.cpp

Line 48 in 777cca2

eosio_assert( price.amount >= 0, "must bet positive quantity" );

That >= should be > instead.

When trying to buy, a deal created for 0.0000 EOS fails with the following message:

Error Buying Wizard

Error details:
eosio_assert_message_exception
eosio_assert_message assertion failure
assertion failure with message: must transfer positive quantity

So it would be better to disallow to create such deals in the first place.

buffer size seems wrong

CryptoWizards-SmartContract/wizContract.cpp

Line 295 in ce31e5f

sha256((char *)mixedChar, sizeof(mixedChar), &result);

Why is the size equal to sizeof(mixedChar)?
To me, it looks like the size argument should be the size of uint32_t ar where we point to.
By coincidence, both seem to be 4 bytes, but if I'm right, it's still a dormant bug if something has to change.

random number generation: bug in improbable case

CryptoWizards-SmartContract/wizContract.cpp

Line 188 in 928e611

uint8_t randInd = 0;

This is a bug in handling the improbable case (still a bit probable for very small lengths though).
The line declares a new variable instead of assigning to the existing one.

Note that some compiler options (like -Wshadow shadowing warnings for GCC) may prevent such bugs.

shop items generation: array index out of bounds

CryptoWizards-SmartContract/wizContract.cpp

Line 126 in 928e611

uint32_t ind = 1 + shopArray[itemType] % ragvar.sum;

This is a bug.
shopArray[itemType] should in fact be shopArray[i].
The item type can be 0..9, but shopArray is currently of length 2.

How does this not crash, at least sometimes?