HACK Assembler

gistfile1.py

# we need our imports
import string, sys

# global hash table
hashTable = {}
RAMstart = 16 # variables are stored in RAM[16] to... RAM[255]??

# parse the assembly
def Assembler(filename):
	# define the variables we need to
	lineCount = 0

	# open the file
	print "Opening File... "
	f = open(filename, 'r')
	
	print "Looking for symbols"
	
	# first read through
	# scan through once for symbols... into our hash table
	for line in f:
		# skip new lines
		if line[:2] == "//" or line == '\n':
			continue
	
		# parse line
		line = parseLine(line)
		if not line:	
			continue
			
		commandType = cType(line)	
	
		if commandType == 'L_COMMAND': 
			hashTable[line[1:-1]] = lineCount
		else:
			lineCount += 1
			
	# seek file back to start
	f.seek(0);
	
	# second read through - parse
	# loop through it, print out non empty and new-line lines.
	print "Reading, Assembling and Writing Files..."
	
	# open file to write to
	nFile = filename[0:filename.find('.')] + ".hack"
	newFile = open(nFile, "w")
	
	for line in f:
		if line[:2] == "//" or line == '\n':
			continue
	
		# parse the line
		line = parseLine(line)
		
		# check if it's a blank line
		if not line:
			continue
			
		# get the cType 
		commandType = cType(line)
		
		# if we dunno what it is, break system
		if commandType == 'Error':
			print "Assembler Error on line : " + line
				
		# print machine code
		if commandType == 'A_COMMAND':
			machineCode = aCommand(line)
			newFile.write(machineCode + "\n")
		elif commandType == 'C_COMMAND':
			machineCode = cCommand(line)
			newFile.write(machineCode + "\n")
			
	# close the file
	f.close()
	newFile.close()
	
	print "Assembled Successfully"
	
# returns the c-instruction code
def cCommand(input):
	return '111' + comp(input) + dest(input) + jump(input)
	
# get the comp value
def comp(input) :
	if input.find('=') > -1:
		input = input[input.find('=')+1 :]
	elif input.find(';') > -1:
		input = input[0:input.find(';')]
	else:
		return '0000000'
		
	comp = {'0': '0101010', '1': '0111111', '-1': '0111010', 'D': '0001100',
			'A': '0110000', '!D': '0001101', '!A': '0110001', '-D': '0001111',
			'-A': '0110011', 'D+1': '0011111', 'A+1': '0110111', 'D-1': '0001110',
			'A-1': '0110010', 'D+A': '0000010', 'D-A': '0010011', 'A-D': '0000111',
			'D&A': '0000000', 'D|A': '0010101',
			
			'M': '1110000', '!M': '1110001', '-M': '1110011', 'M+1': '1110111',
			'M-1': '1110010', 'D+M': '1000010', 'D-M': '1010011', 'M-D': '1000111',
			'D&M': '1000000', 'D|M': '1010101'}

	return comp[input]

# get the dest value
def dest(input) :
	dest = {'M': '001', 'D': '010', 'MD': '011', 'A': '100', 
			'AM': '101', 'AD': '110', 'AMD': '111'}

	if input.find('=') == -1:
		return '000'
	else:
		destSgmt = input[0: input.find('=')]
		return dest[destSgmt]

	
# get the jump value
def jump(input) :
	jmp = {'JGT': '001', 'JEQ': '010', 'JGE': '011', 'JLT': '100', 
			'JNE': '101', 'JLE': '110', 'JMP': '111'}

	if input.find(';') == -1:
		return '000'
	else:
		jmpSgmt = input[input.find(';')+1 :]
		return jmp[jmpSgmt]
			
# returns a-instruction code			
def aCommand(input):
	#map the special addresses
	symbols = {'SP': '0', 'LCL': '1', 'ARG': '2', 'THIS': '3', 'THAT': '4',
				'R0': '0', 'R1': '1', 'R2': '2', 'R3': '3', 'R4': '4', 'R5': '5',
				'R6': '6', 'R7': '7', 'R8': '8', 'R9': '9', 'R10': '10', 'R11': '11',
				'R12': '12', 'R13': '13', 'R14': '14', 'R15': '15', 'SCREEN': '16384',
				'KBD': '24576' }
	
	global RAMstart
	
	input = input[1:]
	if input in symbols.keys():
		input = symbols[input]
	elif input.isdigit():
		input = input
	elif input in hashTable.keys():
		input = hashTable[input]
	else:
		# it's a variable, store it, return RAM address
		hashTable[input] = RAMstart
		input = RAMstart
		RAMstart += 1
		
	num = bin(int(input))[2:].zfill(15)
	return '0' + str(num)
			
# Get the command type of this line
def cType(c):
	# grab the variables we need
	command = ['A_COMMAND', 'C_COMMAND', 'L_COMMAND', 'ERROR']
	
	# check if it is an A_COMMAND @likethis
	if c[0] == '@':
		return command[0]	# is an A-instruction
	elif c[0] == '(' and c[len(c)-1] == ')':
		return command[2]	# is a label
	elif c.find(';') > -1 or c.find('=') > -1:
		return command[1]	# is C-instruction
	else:
		return command[3] 	# dunno what it is - break assembler error

# parse a line, remove white space and \n
def parseLine(input):
	# parse out white space
	input = string.join(input.split(), "")
	# and new lines
	input = input.rstrip('\n')	
	## check and remove comments
	if input.find('//') > -1:
		input = input[0:input.find('//')]
	return input
	
# get file name
file = raw_input("Enter file name: ")
Assembler(file)