Created
October 18, 2022 12:38
-
-
Save GuyAglionby/93c1d6af00e45a802ef3c49f40decfaa to your computer and use it in GitHub Desktop.
Remove duplicate video entries from ACL Anthology data
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import glob | |
import itertools | |
from lxml import etree | |
from networkx.utils import UnionFind | |
from tqdm import tqdm | |
def elems_same(elem1, elem2): | |
return elem1.attrib == elem2.attrib | |
def handle_file(filename): | |
with open(filename, 'r') as f: | |
root = etree.parse(f) | |
for paper in root.xpath('//paper'): | |
videos = paper.xpath('./video') | |
uf = UnionFind(videos) | |
for v1, v2 in itertools.combinations(videos, 2): | |
if elems_same(v1, v2): | |
uf.union(v1, v2) | |
for dupes in uf.to_sets(): | |
if len(dupes) > 1: | |
# remove the last video elems, presumably they were most recently added + for consistency | |
dupes_w_idx = sorted([(dupe, videos.index(dupe)) for dupe in dupes], key=lambda x: x[1]) | |
for elem, _ in dupes_w_idx[1:]: | |
paper.remove(elem) | |
string_rep = etree.tostring(root, pretty_print=True, xml_declaration=True, encoding='UTF-8').decode('utf-8') | |
# hacky | |
bad_paper1 = "\n </paper>\n <paper id=" | |
good_paper1 = "\n </paper>\n <paper id=" | |
string_rep = string_rep.replace(bad_paper1, good_paper1) | |
bad_paper2 = "\n </paper>\n </volume>" | |
good_paper2 = "\n </paper>\n </volume>" | |
string_rep = string_rep.replace(bad_paper2, good_paper2) | |
with open(filename, 'w') as f: | |
f.write(string_rep) | |
def main(): | |
for file in tqdm(list(glob.glob('acl-anthology/data/xml/*.xml'))): | |
handle_file(file) | |
if __name__ == '__main__': | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment