#! /usr/bin/env python
# from __future__ import division # / operator performs float rather than int division

import os, sys, string, cp

def getSeqRange(theList, rangeString):
	"""rangeString in format 'a-b,c-d,...' 
	returns a string"""
	
	length = len(theList)
	if not type(rangeString) == type(''):
		rangeString='%s-%s' % (rangeString, rangeString)
	coordList = rangeString.split(',')
	
	listOut = [] 
	for coords in coordList:
		if coords[-1] == '-': coords = coords[:-1]
		try:
			c1,c2 = coords.split('-')
			c1,c2 = eval(c1)-1,eval(c2) #convert to 0-index
		except ValueError:
			c1 = eval(coords.split('-')[0])
			c2 = None
		
		if c1 < 0: c1 = 0
		
		if not c2 or c2 > length:
			listOut += theList[c1:]
		else:
			listOut += theList[c1:c2]
					
	return listOut
	
def spaceDelimTable(dataArray, gutter=2, columnGuide=0):
	"""Writes a table with left-justified columns separated by
	space characters. Each column is (gutter) spaces wider than the widest 
	element in the column. dataArray is a list of lists : dataArray[line][element].
	Returns a string."""
		
	lineCount = len(dataArray)
	###figure out number of columns
	colCount = 0
	for line in dataArray:
		colCount = max([colCount, len(line)])
	
	### determine column widths
	colWidths = [0]*colCount
	for j in range(colCount):
		for i in range(lineCount):
			try:
				thisLen = len(dataArray[i][j])
			except IndexError:
				dataArray[i].append('')
				thisLen = 0
			except TypeError:
				thisLen = len(`dataArray[i][j]`)
			
			colWidths[j] = max([colWidths[j], thisLen + gutter])
	
	fstr = '%%-%is'*colCount % tuple(colWidths) + '\n'
	
	outstr = ''
	
	if columnGuide:
		guideList = []
		for w in colWidths:
			guideList.append(columnGuide*(w-gutter))
		
		outstr += fstr % tuple(guideList)
	
	for line in dataArray:
		outstr += fstr % tuple(line)
		outstr = outstr.strip() + '\n'
		
	return outstr

def main():
	"""
%(filename)s

Get a specified column from input file.

{{c         column number (1-based, inclusive), can also specify a range like a-b,c-d,...
	1-}}
{{out		name of outfile. Prints to stdout by default.}}
{{d         delim char
	w         any whitespace character
	t		  tab
	s         space}}
{{h			Prints documentation.}}
{{v			verbosity 
	1}}
{{version   print version info and exit}}

%(version)s
"""
	debug = 0

	# copy from here...
	docstringdict = {'filename':os.path.split(sys.argv[0])[-1],
	'version':'$Id: col.py,v 1.1 2004/10/10 02:53:09 nghoffma Exp $'}
	
	optlist, formattedDocString, formattedSummary = cp.optStringParser(main.__doc__ % docstringdict,
		optWidth=15, lineWidth=60, valOffset = 4, putVals=1)	
	dict = cp.commandparser(options=optlist, usage='Options:\n' + formattedSummary, 
		debug = debug, exitWithUsage=1)
		
	if dict.status('version'):
		sys.exit( 'Version: %(version)s\n' % docstringdict )
	#...to here

	if dict.status('h'):
		print formattedDocString
		sys.exit( 0 )
	v = dict.value('v')
	
	rangeString = dict.value('c')
	
	delimchar = dict.value('d')
	if delimchar == 't':
		splitstr = r"line.strip().split('\t')"
	elif delimchar == 's':
		splitstr = r"line.strip().split(' ')"
	elif delimchar == 'w':
		splitstr = r"line.strip().split()"
	
	#read from a list of fasta files
	infile_list = dict.value('infile_list', 'in','file')

	#write to stdout by default
	writeToFile = 0
	if dict.status('out'):
		outfile = dict.value('out','out','file')
		writeToFile = 1
	else:
		outfile = sys.stdout
		v = 0
	
	listOut = []
	for f in infile_list:
		for line in f.readlines():
			lis = eval(splitstr)
			thisLineList = getSeqRange(lis, rangeString)
			listOut.append(thisLineList)
	
	outfile.write(spaceDelimTable(listOut))
	
	
	if writeToFile:
		outfile.close()
	
	
if __name__ == '__main__':
	main()