Need to sort task depending upon the parent task

Question

Need to sort task such that, It should come in sequence once parent of the task is listed down. It can be approached as a Directed Acyclic Graph(DAG)

In this particular example, The main issue is: "13 ifhj 5,4,9,15" should always come after 5,4,9 and 15.But 13 is coming before 15.

I am providing sample input and expected output.

I/p:

task_no    t_name     Parent_Task_Numbers
    1       task1       0
    2       task2       0
    3       task3       0
    4       task4       1,2
    5       task5       0
    6       task6       2
    7       task7       4
    8       task8       6,1,4
    9       task9       10
    10      task10      1
    11      task11      2
    12      task12      3
    13      task13      5,4,9,15
    15      task15      7,8,2,10

Expected o/p:(This can be different than this, but the main idea should be served)

task_no    t_name     Parent_Task_Numbers
    1       task1       0
    2       task2       0
    3       task3       0
    5       task5       0
    10      task10      1
    4       task4       1,2
    9       task9       10
    6       task6       2
    11      task11      2
    12      task12      3
    7       task7       4
    15      task15      7,8,2,10
    13      task13      5,4,9,15
    8       task8       6,1,4

def sort_parent_tasks1(fileName):
    df1 = pd.read_csv(fileName)
    print("df1.Parent_Task_Numbers.str.split(',')", type(df1.Parent_Task_Numbers.str.split(',')))
    df1.Parent_Task_Numbers.str.split(',').apply(sorted, reverse = True).str.join(',').str.strip(',')
    df3 = df1.sort_values(['Parent_Task_Numbers'])
    df3.to_csv("/fileData/task_files/output/test_generated1.csv")


inputFile = "/fileData/task_files/input/test.csv"
sort_parent_tasks1(inputFile)

What I am getting is: which is WRONG.

tasknumber    taskname    Parent_Task_Numbers
1       task1       0
2       task2       0
3       task3       0
5       task5       0
10      task10      1
4       task4       1,2
9       task9       10
6       task6       2
11      task11      2
12      task12      3
7       task7       4
13      task13      5,4,9,15
8       task8       6,1,4
15      task15      7,8,2,1

Please review [ask], and create a [mcve], that means no images of code! Also reviewing [how to create good, reproducible pandas examples](https://stackoverflow.com/questions/20109391/how-to-make-good-reproducible-pandas-examples) would be helpful here. — user3483203, Jun 18 '19 at 15:23

score 0 · Accepted Answer · answered Jun 24 '19 at 09:16

taking input from https://ipython.org/ipython-doc/dev/parallel/dag_dependencies.html

import networkx as nx
import matplotlib.pyplot as plt
import pandas as pd



def create_DAG(abs_file = "input.csv"):
    G2 = nx.DiGraph()
    df1 = pd.read_csv(abs_file)
    print(df1)

    for index, row in df1.iterrows():
        print("row['tasknumber']" , row['tasknumber'], "row['Parent_Task_Numbers']", row['Parent_Task_Numbers'])
        parents = row['Parent_Task_Numbers'].split(",")
        for p in parents:
            G2.add_edge(int(p), int(row['tasknumber']))

    nodes = df1["tasknumber"].tolist()
    map(G2.add_node, nodes)

    for item in nodes:
            print("successors of {} is {}".format(item, list(G2.successors(item))))

    # now draw the graph:
    nx.draw(G2, with_labels=True, edge_color='r')
    plt.show()
    print("graph printed ")

    return G2


G2 = create_DAG()

order = nx.topological_sort(G2)
print("order :", order)
abs_file = "input.csv"
df2 = pd.read_csv(abs_file)
df3 = pd.DataFrame(columns=df2.columns)

for task in order:
    row1 = df2.loc[df2['tasknumber'] == task]
    df3 = df3.append(row1, ignore_index=True)


print("rearranged data frame is : ", df3)

Need to sort task depending upon the parent task

1 Answers1