import re
import os
import glob

# Python 2.7.11 written by Ali Khalighifar on July 31, 2017
# This script generates the summary table for all the outputs of EstimateS
# It selects the last line of each table (the number of tables is 2100)

home = 'Working Directory'
finalDir = home + '\\outputs'
os.chdir(finalDir)

# Making a new table to add the summaries of 2100 tables
table = open('table.txt', 'w')
table.write('Rep\t'+'Abund Error\t'+'Mean Abund\t'+'Days\t'+
            'Individuals (computed)\t'+'S(est)\t'+'S(est) 95% CI Lower Bound\t'
            +'S(est) 95% CI Upper Bound\t'+'S(est) SD\t'+'S Mean (runs)\t'
            +'Singletons Mean\t'+'Singletons SD (runs)\t'+'Doubletons Mean\t'
            +'Doubletons SD (runs)\t'+'Uniques Mean\t'+'Uniques SD (runs)\t'
            +'Duplicates Mean\t'+'Duplicates SD (runs)\t'+'ACE Mean\t'
            +'ACE  SD (runs)\t'+'ICE Mean\t'+'ICE SD (runs)\t'+'Chao 1 Mean\t'
            +'Chao 1 95% CI Lower Bound\t'+'Chao 1 95% CI Upper Bound\t'
            +'Chao 1 SD (analytical)\t'+'Chao 2 Mean\t'+'Chao 2 95% CI Lower Bound\t'
            +'Chao 2 95% CI Upper Bound\t'+'Chao 2 SD (analytical)\t'+'Jack 1 Mean\t'
            +'Jack 1 SD (analytical)\t'+'Jack 2 Mean\t'+'Jack 2 SD (runs)\t'
            +'Bootstrap Mean\t'+'Bootstrap  SD (runs)\t'+'MMRuns Mean\t'
            +'MMMeans (1 run)\t'+'Cole Rarefaction\t'+'Cole SD (analytical)\t'
            +'Alpha Mean\t'+'Alpha SD (analytical)\t'+'Shannon Mean\t'
            +'Shannon SD (runs)\t'+'Shannon Exponential Mean\t'
            +'Shannon Exponential SD (runs)\t'+'Simpson Inv Mean\t'
            +'Simpson Inv SD (runs)\n')
os.chdir(baseDir)
files = [txt_file for txt_file in glob.iglob ('*.txt')]
for txtFile in files:
    abund_type = ', '.join(re.findall(r'(High)', txtFile))
    err_type = ', '.join(re.findall(r'^(nine)', txtFile))
    days = ', '.join(re.findall(r'(\d{1,4}days)', txtFile))
    days = ', '.join(re.findall(r'(\d{1,4})', days))
    rep = ', '.join(re.findall(r'(Rep\d{1,3})', txtFile))
    rep = ', '.join(re.findall(r'(\d{1,3})', rep))
    lines = open(txtFile, 'r').readlines()[int(days)+2]
    lines = re.split(r'^''\d{1,4}\t', lines)
    del lines[0]
    lines = ', '.join(lines)
    table.write(rep + '\t' + err_type + '\t' + abund_type + '\t' + days +
                '\t' + lines)
table.close()