Upload New File

__author__ = 'debjit'

import glob

import os

import subprocess

import sys

import argparse

import re

import csv

import gzip

import igraph

import ast

from igraph import *

from matplotlib import *

import matplotlib.pyplot as plt

import time

import codecs

output_path = r'D:\SWP\Multi-Hop-Knowledge-Paths-Human-Needs-master\scr\graph_model\output'

def read_file(data,purpose,output_path,g):

   f = codecs.open(data,encoding='utf-8',mode='r')

   id_knowledge={}

   id_done=[]

   for line in f:

      #line = line.decode('utf-8')

      id = line.split('\t')[0].strip()

      print(id)

      if id not in id_done:

             concepts = line.split('\t')[-1]

             knowledge = make_sub_graph(id,purpose,concepts,g)

             id_knowledge[id]=knowledge

             id_done.append(id)

   return id_knowledge

def make_sub_graph(id,purpose,concepts,g):

      vertex = ['status', 'approval', 'safety', 'competition', 'health', 'family', 'love','romance','food', 'eating', 'independent', 'power', 'order', 'curiosity', 'calm','peace' ,'honor', 'belonging', 'social', 'money','save_money', 'idealism', 'rest']

      concepts = concepts.strip()

      if concepts !='No Knowledge for this sentence':

         concepts = ast.literal_eval(concepts)

      else:

         concepts = [] 

      concept_know_human=[]

      knowledge=[]

      nodes_1=[]

      nodes_2=[]

      edge_relation = []

      edges=[]

      if concepts!=[]:

       for i in range(len(concepts)):

        for j in range(i+1,len(concepts)):

            if concepts[i]!=concepts[j]:

                try:

                    s = g.get_all_shortest_paths(concepts[i],concepts[j],mode='OUT')

                    k = g.get_all_shortest_paths(concepts[j],concepts[i],mode='OUT')

                    s = [list(x) for x in set(tuple(x) for x in s)]

                    k = [list(x) for x in set(tuple(x) for x in k)]

                    s.extend(k)

                    if len(s)==1:

                        if len(s[0])==1:

                            continue

                        else: 

                            y, w = extract_paths_between_concepts(s,g)

                            nodes_1.append(y)

                            nodes_2.append(w)

                    elif s==[]: 

                            continue

                    else:                    

                        y, w = extract_paths_between_concepts(s,g)

                        nodes_1.append(y)

                        nodes_2.append(w)

                except ValueError:

                            continue

       for i in concepts:

         for j in vertex:

            if i!=j:

                try:

                    s = g.get_all_shortest_paths(i,j,mode='OUT')

                    k = g.get_all_shortest_paths(j,i,mode='OUT')

                    s = [list(x) for x in set(tuple(x) for x in s)]

                    k = [list(x) for x in set(tuple(x) for x in k)]

                    s.extend(k)

                    if len(s)==1:

                        if len(s[0])==1:

                            continue #"No Knowledge for "+concepts[i]+" and "+concepts[j]

                        else: 

                            y, w = extract_paths_between_concepts(s,g)

                            nodes_1.append(y)

                            nodes_2.append(w)

                            #edge_relation.append(v)

                    elif s==[]: 

                            continue

                    else:                    

                        y, w = extract_paths_between_concepts(s,g)

                        nodes_1.append(y)

                        nodes_2.append(w)

                        #edge_relation.append(v)

                except ValueError:

                            continue

       g_1=Graph(directed=True)

       flat_list_1=[]

       flat_list_2=[]

       nodes_1.extend(nodes_2)

       for sublist in nodes_1:

         for item in sublist:

             flat_list_1.append(item)

       unique_nodes = list(set(flat_list_1).union(set(flat_list_2)))

       for i in concepts:

        try:

         neighboring_nodes=g.neighbors(i)

         for j in range(len(neighboring_nodes)):

              unique_nodes.append(g.vs[neighboring_nodes[j]]["name"])

        except ValueError:

                            continue

       for i in vertex:

        try:

         neighboring_nodes=g.neighbors(i)

         for j in range(len(neighboring_nodes)):

            unique_nodes.append(g.vs[neighboring_nodes[j]]["name"])

        except ValueError:

                            continue  

       unique_nodes=list(set(unique_nodes))

       g_1 = g.subgraph(unique_nodes)   

       g_1.write_pickle(output_path+"_"+purpose+'/'+id)

      return knowledge

def extract_paths_between_concepts(s,g):  

    edge_relation=[]

    nodes_1=[]

    nodes_2=[]

    edge=[]

    edge_relation=[]

    for i in range(len(s)):

      temp = ''

      for j in range(len(s[i])-1):

            nodes_1.append(g.vs[s[i][j]]["name"])

            nodes_2.append(g.vs[s[i][j+1]]["name"])

    return(nodes_1,nodes_2)

def main():

    parser = argparse.ArgumentParser()

    parser.add_argument("inputfile", help=".txt file containing the input text", nargs='?')

    parser.add_argument("graphpath", help=".txt file containing the input text", nargs='?')

    parser.add_argument("outputpath", help=".txt file containing the input text", nargs='?')

    parser.add_argument("--purpose", help="train or dev or test", nargs='?')

    args = parser.parse_args()

    g = Graph(directed=True)

    g = read(args.graphpath, format="pickle")

    id_knowledge = read_file(args.inputfile,args.purpose,args.outputpath,g)

    #id_knowledge = read_file(r'C:\Users\PC\Desktop\Coli neu\SWP\Multi-Hop-Knowledge-Paths-Human-Needs-master\training.txt',"train", r'C:\Users\PC\Desktop\Coli neu\SWP\Multi-Hop-Knowledge-Paths-Human-Needs-master\output',g) 

if __name__ == '__main__':

    main()