#!/usr/local/bin/python
# Jan.  4, 2006, Dr. Brian Fristensky, University of Manitoba

# Description: Convert a file of trees to a GDE
# flat file. 

# Synopsis: tree2flat.py infile outfile

# Files: infile      file of trees in Phylip tree format
#        outfile     GDE flat file, containing one or more
#                    trees

import sys
import os
import string
import re

# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
# Read a file containing one or more trees, where a tree
# can be defined by the regular expression 
#
# (.*);
#
# In practice, it is more complicated, because we need to
# strip off leading and trailing whitespace, including
# newline characters. The quickest way seems to be to 
# read in lines and concatenate them into big lines, stripping
# whitespace as we go. When a line contains a semicolon (;),
# we break it into two lines and begin a new tree.
#
def GETTREES(FN) :
    TREES = []
    FILE = open(FN,'r')

    # Read the entire file into a single, very long line
    BIGLINE = ""
    LINE = FILE.readline()
    while LINE != '':
          BIGLINE = BIGLINE + LINE.strip()
	  LINE = FILE.readline()
    FILE.close()

    # Create a list containing all trees in BIGLINE
    #I don't know why these lines don't work
#    p = re.compile('\(.*\)\;') 
#    p = re.compile('\(.*\);')
#    TREES = p.findall(BIGLINE)

    # This loses the ;, so we have to add it back when 
    # writing to the output file
    TREES = BIGLINE.split(';')
    return TREES

# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
# For each tree in TREES, write an entry in the form
#
# "name
# tree
#
# where the name is the input file name, followed by a number,
# and tree is the tree on a single line
def WRITETREES(OUTFILE,IFN,TREES) :

    # Create a base name for the trees, using IFN. 
    # Truncate the file extension, if any, and truncate
    # to 8 characters, to leave some room so that we
    # can add a number to the name for each tree
    I = IFN.rfind('.')
    if I > -1 :
       TREENAME = IFN[0:I]
    else :
       TREENAME = IFN
    if len(TREENAME) > 8 :
       TREENAME = TREENAME[0:8]

    # Write the trees
    J = 1
    for T in TREES:

       # BIGLINE.split(;) can create an extra blank line
       # at the end of a file. It is therefore necessary
       # to filter out empty lines.
       OUTTREE = T.strip()
       if len(OUTTREE) > 0 :
          LINE = '\"' + TREENAME + str(J) + '\n'	   
          OUTFILE.write(LINE)
          OUTFILE.write(OUTTREE + ';\n')
          J = J + 1

    
#======================== MAIN PROCEDURE ==========================

#---------- Set global variables
IFN = sys.argv[1]
OFN = sys.argv[2]
OUTFILE = open(OFN,'w')

# Read in list of tokens
if os.path.exists(IFN) :
   TREES = GETTREES(IFN)

   #Write the list as a single line of comma-separated 
   # values
   WRITETREES(OUTFILE,IFN,TREES)

OUTFILE.close()