from struct import pack,unpack
from random import randint

# convert 16 dna chars into signed 32 bit integer
def bincode(D):
   if D=='A': return 0
   if D=='C': return 1
   if D=='G': return 2
   if D=='T': return 3
#bincode
# better to use hashtable/dictionary
   
def pack16(S): # S is a string of 16 DNA values
   if len(S)!=16: raise Exception("string not 16 chars long")
   V = 0  # value
   first = bincode(S[0])
   if first>=2: first = -4 + first  # G, T are negatives if first
   V = first  # sets last (rightmost) bits of V:
   i = 1
   while i<16:
      V = V * 4  # shift left 2 bits
      V = V + bincode(S[i])
      i += 1
   # while
   return V
#pack16

v = pack16("TAACGTTACATACATC")
print v
v2 = pack16("CAACGTTACATACATC")

DNA = "ACGT"   
def unpack16(V):  # take V as 32bit signed int, return 16-char DNA String
  D = ""
  i = 0
  while i<16:
      c = DNA[V%4]
      D = c + D
      V = V / 4  # sign-extended shift right
      i += 1
  #while
  return D
# unpack16

print unpack16(v)
print unpack16(v2)

def writednabin(D,filename):
    fd = open(filename,"wb")  # open(filename,"rb")
    lend = pack(">i",len(D))
    fd.write(lend)   # z = fd.read(4), then unpack z, y, = unpack(">i",z)
    i = 0
    while i<=len(D)-16:
       v = pack16(D[i:i+16])
       vp = pack(">i",v)
       fd.write(vp)
       i += 16
    # while
    rem = len(D) % 16
    if rem>0:
        r = D[len(D)-rem:] + ('A'*(16-rem))  # pad with dummies
        v = pack16(r)
        vp = pack(">i",v)
        fd.write(vp)
    # if remainder exists
    fd.close()
#writednabin




# s[3:5] gives substring from s[3] to s[4],, 5-3 is length of substring


DNA = "ACGT"

# generate and return random DNA string of length n
def gendna(n):
    s = ""
    while n>0:
        s = s + DNA[ randint(0,3) ]
        n -= 1
    #while
    return s
#gendna

D = gendna(130)
writednabin(D,"test.dna")
print D