From 14d88ae54bdd2a824b4172308366a3e513b29de9 Mon Sep 17 00:00:00 2001
From: Michael Abed <michaelabed@gmail.com>
Date: Fri, 27 Jan 2012 01:22:46 -0500
Subject: initial commit

---
 asmblr.py | 114 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 114 insertions(+)
 create mode 100755 asmblr.py

(limited to 'asmblr.py')

diff --git a/asmblr.py b/asmblr.py
new file mode 100755
index 0000000..52431bf
--- /dev/null
+++ b/asmblr.py
@@ -0,0 +1,114 @@
+#!/usr/bin/env python2
+
+import sys
+
+
+def tokenize(i):
+    if i[0] == ';': return None
+    x = i.strip().split(' ')
+    if len(x) == 3 and ',' in x[1]:
+        x[1] = x[1].split(',')[0]
+    return x
+
+def genaddresses(tokens):
+    offset = 31
+    lblcnt = 0
+    table = {}
+    for t in tokens:
+        if ':' not in t[0]:
+            continue
+        label = t[0].split(':')[0]
+        table[label] = offset + lblcnt
+        lblcnt += 1
+    return table
+
+def gencode(tokens, addresses, ops):
+    out = []
+    regs = {'R1': 0, 'R2': 1, 'R3': 3, 'R4': 4, 'R5' : 5, 'R6' : 6}
+    for t in tokens:
+        word = t[0]
+        if word == "halt":
+            out.append(opcodes[word])
+        elif word in ["inc", "je", "jne", "jmp"]:
+            out.append(handle1arg(t, ops, regs, addresses))
+        elif word in ["add", "sub", "xor", "cmp"]:
+            out.append(handle2arg(t, ops, regs))
+        elif word == "mov":
+            out.append(handlemov(t, ops, regs))
+    return out
+
+def handle1arg(l, ops, regs, ads):
+    op = ops[l[0]] << 12 
+    if l[0] == 'inc':
+        op += regs[l[1]]
+    else:
+        try:
+            op += int(l[1], 16)
+        except ValueError:
+            op += ads[l[1]]
+    return op
+
+def handle2arg(l, ops, regs):
+    op = ops[l[0]] << 12
+    op += regs[l[1]] << 6
+    op += regs[l[2]]
+    return op
+
+def handlemov(l, ops, regs):
+    if "[" in l[1]:
+        op = ops['cpt'] << 12
+        op += regs[l[1][1:3]] << 6
+        op += regs[l[2]]
+    elif "[" in l[2]:
+        op = ops['cpf'] << 12
+        op += regs[l[1]] << 6
+        op += regs[l[2][1:3]]
+    elif "R" in l[1] and "R" in l[2]:
+        op = ops['cpy'] << 12
+        op += regs[l[1]] << 6
+        op += regs[l[2]]
+    else:
+        op = ops['set'] << 12
+        op += regs[l[1]] << 6
+        op += int(l[2], 16)
+    return op
+
+def formatcode(code):
+    lines = []
+    for i in code:
+        lines.append("{:0>4X}".format(i))
+    return ''.join(lines)
+
+f = sys.argv[1]
+
+opcodes = {
+        'halt': 0x0,
+        'inc' : 0x1,
+        'jmp' : 0x2,
+        'je'  : 0x3,
+        'jne' : 0x4,
+        'add' : 0x5,
+        'sub' : 0x6,
+        'xor' : 0x7,
+        'cmp' : 0x8,
+        'set' : 0x9,
+        'cpy' : 0xA,
+        'cpt' : 0xB,
+        'cpf' : 0xC }
+
+prgm = open(f).readlines()
+tokens = []
+
+for i in prgm:
+    tokens.append(tokenize(i))
+
+tokens = filter(None, tokens)
+
+addresses = genaddresses(tokens)
+
+code = gencode(tokens, addresses, opcodes)
+print code
+
+binary = formatcode(code)
+
+print binary
-- 
cgit v1.2.3