def get_instruction_info(self, data, addr):
(instrTxt, instrLen) = skwrapper.disasm(data, addr)
if instrLen == 0:
return None
result = InstructionInfo()
result.length = instrLen
return result
def get_instruction_info(self, data, addr): (instrTxt, instrLen) = skwrapper.disasm(data, addr) if instrLen == 0: return None result = InstructionInfo() result.length = instrLen rccs = r'(?:C|NC|Z|NZ|M|P|PE|PO)' regexes = [ \ r'^(?:JP|JR) '+rccs+r',\$(.*)$', # 0: conditional jump eg: JP PE,#DEAD r'^(?:JP|JR) \$(.*)$', # 1: unconditional jump eg: JP #DEAD r'^(?:JP|JR) \((?:HL|IX|IY)\)$', # 2: unconditional indirect eg: JP (IX) r'^DJNZ \$(.*)$', # 3: dec, jump if not zero eg: DJNZ #DEAD r'^CALL '+rccs+r',\$(.*)$', # 4: conditional call eg: CALL PE,#DEAD r'^CALL \$(.*)$', # 5: unconditional call eg: CALL #DEAD r'^RET '+rccs+'$', # 6: conditional return r'^(?:RET|RETN|RETI)$', # 7: return, return (nmi), return (interrupt) ] m = None for (i,regex) in enumerate(regexes): m = re.match(regex, instrTxt) if not m: continue if i==0 or i==3: dest = int(m.group(1), 16) result.add_branch(BranchType.TrueBranch, dest) result.add_branch(BranchType.FalseBranch, addr + instrLen) pass elif i==1: dest = int(m.group(1), 16) result.add_branch(BranchType.UnconditionalBranch, dest) pass elif i==2: result.add_branch(BranchType.IndirectBranch) pass elif i==4 or i==5: dest = int(m.group(1), 16) result.add_branch(BranchType.CallDestination, dest) pass elif i==6: pass # conditional returns don't end block elif i==7: result.add_branch(BranchType.FunctionReturn) break return result
def get_instruction_text(self, data, addr):
(instrTxt, instrLen) = skwrapper.disasm(data, addr)
if instrLen == 0:
return None
result = []
atoms = [t for t in re.split(r'([, ()\+])', instrTxt) if t] # delimeters kept if in capture group
result.append(InstructionTextToken(InstructionTextTokenType.InstructionToken, atoms[0]))
if atoms[1:]:
result.append(InstructionTextToken(InstructionTextTokenType.TextToken, ' '))
#
for atom in atoms[1:]:
if not atom or atom == ' ':
continue
# PROBLEM: cond 'C' conflicts with register C
# eg: "RET C" is it "RET <reg>" or "REG <cc>" ?
# eg: "CALL C" is it "CALL <reg>" or "CALL C,$0000" ?
elif atom == 'C' and atoms[0] in ['CALL','RET']:
# flag, condition code
result.append(InstructionTextToken(InstructionTextTokenType.TextToken, atom))
elif atom in self.reg16_strs or atom in self.reg8_strs:
result.append(InstructionTextToken(InstructionTextTokenType.RegisterToken, atom))
elif atom in self.cond_strs:
result.append(InstructionTextToken(InstructionTextTokenType.TextToken, atom))
elif atom[0] == '#':
result.append(InstructionTextToken(InstructionTextTokenType.IntegerToken, atom, int(atom[1:],16)))
elif atom[0] == '$':
if len(atom)==5:
result.append(InstructionTextToken(InstructionTextTokenType.PossibleAddressToken, atom, int(atom[1:],16)))
else:
result.append(InstructionTextToken(InstructionTextTokenType.IntegerToken, atom, int(atom[1:],16)))
elif atom.isdigit():
result.append(InstructionTextToken(InstructionTextTokenType.IntegerToken, atom, int(atom)))
elif atom == '(':
result.append(InstructionTextToken(InstructionTextTokenType.BeginMemoryOperandToken, atom))
elif atom == ')':
result.append(InstructionTextToken(InstructionTextTokenType.EndMemoryOperandToken, atom))
elif atom == '+':
result.append(InstructionTextToken(InstructionTextTokenType.TextToken, atom))
elif atom == ',':
result.append(InstructionTextToken(InstructionTextTokenType.OperandSeparatorToken, atom))
else:
raise Exception('unfamiliar token: %s from instruction %s' % (tok, instrTxt))
return result, instrLen
def get_instruction_text(self, data, addr): (instrTxt, instrLen) = skwrapper.disasm(data, addr) tokens = [InstructionTextToken(InstructionTextTokenType.TextToken, instrTxt)] return tokens, instrLen