find_all_predecessors.py

import csv
import pydot

def predecessors(pred, targets):
    d = {}
    graph = pydot.Dot(graph_type='digraph', graph_name='Predecessors')
    for target_activity in targets:
        target_list = []
        d[target_activity] = target_list
        predecessors_impl(graph, pred, None, target_activity, target_list)
    return d, graph

def predecessors_impl(graph, pred, parent_activity, activity, lst):

    # If we have already gotten all predecessors of this activity, we are done.
    if activity in lst:
        return

    # Append the activity to the target list
    lst.append(activity)

    if parent_activity is not None:
        graph.add_edge(pydot.Edge(parent_activity, activity))

    # If the activity has predecessors, visit them recursively
    if activity in pred:
        for p in pred[activity]:
            predecessors_impl(graph, pred, activity, p, lst)

def main():
    with open('Predecessors.csv', newline='') as pred_csv:
        pred = {}
        pred_reader = csv.reader(pred_csv)
        # Skip header row
        next(pred_reader)
        for row in pred_reader:
            assert len(row) == 2
            # New activities are introduced with a dash predecessor, just skip
            # it.
            if row[0] in pred and row[1] != '-':
                pred[row[0]].append(row[1])
            elif row[1] != '-':
                pred[row[0]] = [row[1]]

    with open('Target activites.csv', newline='') as target_csv:
        targets = []
        target_reader = csv.reader(target_csv)
        # Skip header row
        next(target_reader)
        for row in target_reader:
            assert len(row) == 2
            if row[0] not in targets:
                targets.append(row[0])

    # targets = ['CHSC660PAVA010', 'ISA14.1-FORECAST20']

    result_dict, graph = predecessors(pred, targets)
    print("Got result:")
    print(result_dict)
    print("Writing to output.csv and output.dot")
    with open('output.csv', 'w', newline='') as output_csv:
        writer = csv.writer(output_csv)
        for target, target_preds in result_dict.items():
            writer.writerow([target] + target_preds)
    
    graph.write_raw('output.dot')

if __name__ == "__main__":
    main()