#!/usr/bin/python
#
#  x2x version 2.  Support seqcluster and sortx
#  April 20, 2009 CER
#      version 4   Write GROUP and ENV files
#  April 30, 2009 CER
#      version 5   write EXP files instead of ENV files
#
Second_Column_Env = 'one_experiment'
import sys,re,os
#process the input parameters
params = sys.argv
if len(params) == 2:
 output_file_name = params[1]
else:
 print("x2x error:  Usage: x2x <output-files-name-root>")
 sys.exit()

first = 1                           #flag
dict  = {}                          #the table of data
i     = 0                           #count the attributes
attr  = []
sortxrun = 1			    #Assume we will find a sortx run
filenames = os.listdir(os.curdir)   #get all of the file names
files_grp = []
for files in filenames:
 test = re.findall(r'.+grp$',files) #pick out the *.grp files
 if test:
  files_grp.append(files)
if not files_grp:			#if we did not find any .grp files from sortx, look for  .csv files from seqcluster
 for files in filenames:
  test = re.findall(r'.+csv$',files) #pick out the *.csv files
  if test:
   files_grp.append(files)
   sortxrun = 0
if not files_grp:			#if we did not find any .grp or .csv files, then abandon all hope...
 print
 print("x2x error:  Did not find .grp (sortx) or .csv (SeqclusterX) files in the local directory.")
 print
 sys.exit()
#
#
#
# build annotation file to move sortx data to XplorSeq
#
#
#build the output file name
group_file_name = output_file_name + '_x2xGROUP.csv'
env_file_name = output_file_name + '_x2xEXP.csv'
output = file(group_file_name,"w")   #open the GROUP file which will get everything
outenv = file(env_file_name,"w")      #open the EXP file which will only get the column headings and sequence names

for allfiles in files_grp:
 input = file(allfiles,'r')           #point to current input file

                                      #Decode cluster type from file name
 if sortxrun:			      				#Parse sortx .grp file names here
  clustype = re.findall(r'_(M\d+|F\d+|N\d+).grp$',allfiles)
 else:								#Parse seqcluster .csv file names here
  clustype = re.findall(r'_(MAX\w+|MIN\w+|AVG\w+).csv$',allfiles)

 myattr = 'clus_' + clustype[0]       #Pre-pend clus_ to cluster type
 attr.append(myattr)                  #save the attribute name

 line = input.readline()              #read the 2nd line of the file

#now start the loops
 while line:
  tokens = re.split(r'( |\t)',line)       #get the key and the attribute value
  key    = re.sub(' ','',tokens[0])   #strip off any spaces
  attrval= re.sub(' ','',tokens[len(tokens)-1])   #strip off any spaces
  attrval= re.sub('\r','',attrval)    #strip end of line chars
  attrval= re.sub('\n','',attrval)    #strip end of line chars


  if first: 
   dict[key]= {}                       #make a dictonary for each key value 

#  print key + " " + attrval
  dict[key][attr[i]] = attrval         #put the attribute value in the table
  line = input.readline()              #get the next line of the file
 i = i + 1                             #count the attribute
 input.close()
 first = 0
#Now write out what was written to debug this
print "-------------------------------------"
print "   "
print "  "
print "  "

attr.sort()
temp = "Key" 

outenv.write(temp + '\t' + 'Experiment0' + '\t\n')	#write the column header to the EXP file

for allattr in attr:
 temp = temp + '\t' + allattr
print temp
temp = temp + '\n'
output.write(temp)		#Write all of the column headers to the GROUP file

key_list = dict.keys()

key_list.sort()

for allkeys in key_list:
 temp = allkeys
 outenv.write(temp + '\t'+ Second_Column_Env + '\t\n')	#write the sequence name to the EXP file
 for allattr in attr:
  temp = temp + '\t' + dict[allkeys][allattr]
 print temp
 temp = temp + '\n'
 output.write(temp)		#Write all of the stat values for this sequence name

output.close()                       #close the GROUP file
outenv.close()			     #close the EXP file