def add_doi(xml_node, collection_id, volume_id, force=False):
if 'id' in xml_node.attrib:
# normal paper
paper_id = int(xml_node.attrib['id'])
else:
# frontmatter
paper_id = 0
anth_id = build_anthology_id(collection_id, volume_id, paper_id)
new_doi_text = f'{data.DOI_PREFIX}{anth_id}'
doi_url = f'{data.DOI_URL_PREFIX}{data.DOI_PREFIX}{anth_id}'
if not test_url(doi_url):
print(f"-> [{anth_id}] Skipping since DOI {doi_url} doesn't exist")
return False
doi = xml_node.find('doi')
if doi is not None:
print(
f'-> [{anth_id}] Cowardly refusing to overwrite existing DOI {doi.text} (use --force)',
file=sys.stderr)
return False
else:
doi = make_simple_element('doi', text=new_doi_text)
print(f'Adding DOI {new_doi_text}', file=sys.stderr)
xml_node.append(doi)
return True
def get_anth_url(volume_id, paper_id=None, width=4):
return "https://www.aclweb.org/anthology/{volume_id}-{paper_id:0{width}d}".format(
volume_id=volume_id, paper_id=paper_id, width=width
)
if __name__ == "__main__":
filename = sys.argv[1]
outfilename = sys.argv[2]
tree = etree.parse(filename)
volume = tree.getroot()
for paper in volume.findall("paper"):
if "href" in paper.attrib:
if not test_url(paper.attrib["href"]):
sys.stderr.write(
"{}:{} removing href attribute: {}\n".format(
filename, paper.sourceline, paper.attrib["href"]
)
)
del paper.attrib["href"]
href = paper.find("href")
if href is not None:
assert len(href) == 0
if not test_url(href.text):
sys.stderr.write(
"{}:{} removing href element: {}\n".format(
filename, href.sourceline, href.text
)