Gather discussions with Graph QL API (Need access token and gh cli)

gather_discussion.py

"""Gather Discussions in Github Repo with GraphQL

Returns:
    CSV output file
"""

from ast import parse
import subprocess
import json

LABEL_FILTER = ["sensing", "perception"]


def get_discussions_graphql():
    command = "gh api graphql -f query="

    query_body = """'
    query{
        repository(owner:"autowarefoundation", name:"autoware") {
            discussions(first:100){
                edges{
                    node{
                        title
                        publishedAt
                        lastEditedAt
                        updatedAt
                        url
                        answer{
                            author{
                                login
                            }
                            publishedAt
                        }
                        author {
                            login
                        }
                        authorAssociation
                        labels(first:5) {
                            edges {
                                node {
                                name
                                }
                            }
                        }
                    }
                }
            }
        }
    }
    '
    """


    cp = subprocess.run(command + query_body, shell=True, capture_output=True)
    text = cp.stdout.decode()
    #print(text)
    return json.loads(text)


def get_discussion_list(data):
    datalist = data['data']['repository']['discussions']['edges']
    print("Data len is:", len(datalist))
    return datalist

def parse_discussion(data):

    node = data['node']

    milestone=""
    title = node['title']
    answered = True if node['answer'] else False
    state = "closed" if answered else "open"

    created_date = node['publishedAt']
    last_edited_date = node['lastEditedAt']
    updated_date = node['updatedAt']
    url = node['url']
    author = node['author']['login']
    labels = ""
    authorAssociation = node['authorAssociation']

    if 'labels' in node.keys():
        label_list = node['labels']['edges']
        for ll in label_list:
            labels = labels + "," + ll['node']['name']

    if filter_via_label(labels, LABEL_FILTER):
        return [milestone, title, None , state, labels, last_edited_date, created_date, updated_date, url, author, authorAssociation]
    else:
        return []



def filter_via_label(labels, filter_list):
    for filter in filter_list:
        if filter in labels:
            return True
    return False


def main():

    discussions = get_discussions_graphql()

    d_list = get_discussion_list(discussions)


    csv_list = []
    column_title = ["Milestone", "Title", "Assignee", "Status", "Labels", "Last edited", "opened_at", "updated_at", "URL", "user name", "user association"]

    csv_list.append(column_title)

    for d in d_list:
        parsed = parse_discussion(d)
        if parsed:
            csv_list.append(parsed)


    import csv

    with open("discussions.csv", "w") as f:
        writer = csv.writer(f)
        writer.writerows(csv_list)



if __name__=='__main__':

    main()