add_paper_script.py

# Added a new paper to the list.
# Author: Yinyu Nie
# Updated data: 23 June, 2021

# Usage:
# python add_paper_script.py
# follow the instructions to add details for each paper.

import sys
from libs import existing_topics
import urllib.request
import re


if __name__ == '__main__':
    '''Decide topic name'''
    print('''Topic name list:
    [0] 3D Detection & Segmentation;
    [1] Shape Representation;
    [2] Shape & Scene Completion;
    [3] Shape Reconstruction & Generation;
    [4] 3D Scene Understanding;
    [5] 3D Scene Reconstruction & Generation;
    [6] NeRF;
    [7] About Human Body;
    [8] General Methods;
    [9] Others (inc. Networks in Classification, Matching, Registration, Alignment, Depth, Normal, Pose, Keypoints, etc.);
    [10] Survey, Resources and Tools.
    ''')
    topic_id = input('Please input the topic ID. ([q]/[quit] to exit.):\n')
    if topic_id == 'q' or topic_id == 'quit':
        sys.exit(0)
    topic_name = existing_topics[int(topic_id)]

    '''Decide publication name'''
    pub_name = input('Please input publication. Default for Arxiv. ([q]/[quit] to exit.):\n')
    if not pub_name:
        pub_name = 'Arxiv'
    elif pub_name == 'q' or pub_name == 'quit':
        sys.exit(0)

    '''Decide website link'''
    link_addr = input('Please input paper link. ([q]/[quit] to exit.):\n')
    if link_addr == 'q' or link_addr == 'quit':
        sys.exit(0)

    if 'arxiv.org' in link_addr:
        if link_addr.endswith('.pdf'):
            arxiv_id = '.'.join(link_addr.split('/')[-1].split('.')[:-1])
            link_addr = 'https://arxiv.org/abs/' + str(arxiv_id)
        else:
            arxiv_id = link_addr.split('/')[-1]
        html_text = urllib.request.urlopen(link_addr).read()
        html_text = html_text.decode("utf8")
        pattern = re.compile(r'Title:.*?</span>(.+?)</h1>')
        re_result = re.search(pattern, html_text)
        paper_title = re_result.group(1)
        pub_date = int(arxiv_id.split('.')[0])
    else:
        paper_title = input('Please paper title. ([q]/[quit] to exit.):\n')
        if paper_title == 'q' or paper_title == 'quit':
            sys.exit(0)
        pub_date = None

    other_info = {}
    while True:
        info_item = input(
            'Please give optional info, e.g., Project=https://yinyunie.github.io/Total3D/. ([q]/[quit] to exit. [Enter] to pass):\n')
        if info_item == 'q' or info_item == 'quit':
            sys.exit(0)
        if info_item:
            items = info_item.split('=')
            value = '='.join(items[1:])
            other_info[items[0].strip()] = value.strip()
        else:
            break

    markdown_line = '- [[%s](%s)] %s' % (pub_name, link_addr, paper_title)
    for key, item in other_info.items():
        markdown_line += ' [[%s](%s)]' % (key, item)
    markdown_line += '\n'

    '''Insert to README.md'''
    with open('./README.md', 'r') as file:
        paper_list = file.read()

    # check if already existed
    if paper_title.strip() in paper_list:
        print('This paper is already in library.')
        sys.exit(0)

    # insert to paper list
    topic_search = re.search('## ' + topic_name + '.*?\n', paper_list)
    insert_point = topic_search.end(0)
    if pub_date is not None:
        if pub_date <= 2100:
            date_search = re.search('#### Before 2021.*?\n', paper_list[topic_search.end(0):])
            insert_point = topic_search.end(0) + date_search.end(0)

    new_paper_list = paper_list[:insert_point] + markdown_line + paper_list[insert_point:]

    with open('./README.md', 'w') as file:
        file.write(new_paper_list)