mirrors/bookwyrm
1
0
Fork 1
mirror of https://github.com/bookwyrm-social/bookwyrm.git synced 2024-09-23 03:50:09 +00:00
Code Issues Projects Releases Wiki Activity

prefer English language sources for ISNI titles

Browse source
This commit is contained in:
Hugh Rundle 2021-11-22 12:15:06 +11:00
parent fd3ed7fb8c
commit 09c5a3861b
No known key found for this signature in database
GPG key ID: CD23D6039184286B
2 changed files with 22 additions and 20 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
bookwyrm/templates/book/edit/edit_book.html
View file

@ -66,7 +66,7 @@
{% if book_title %}
<a href="{{ match.local_path }}" target="_blank">{% trans "Author of " %}<em>{{ book_title }}</em></a>
{% else %}
<a href="{{ match.id }}" target="_blank">{% trans "Author of " %}<em>{{ alt_title }}</em></a>
<a href="{{ match.id }}" target="_blank">{% if alt_title %}{% trans "Author of " %}<em>{{ alt_title }}</em>{% else %} {% trans "Find more information at isni.org" %}{% endif %}</a>
{% endif %}
{% endwith %}
</p>

40
bookwyrm/utils/isni.py
View file

@ -92,32 +92,34 @@ def find_authors_by_name(name_string, description=False):
# build list of possible authors
possible_authors = []
for element in root.iter("responseRecord"):
personal_name = element.find(".//forename/..")
bio = element.find(".//nameTitle")
if not personal_name:
continue
author = get_author_from_isni(element.find(".//isniUnformatted").text)
if bool(description):
titles = element.findall(".//title")
if titles:
# some of the "titles" in ISNI are a little ...iffy
title_element = [e for e in titles if not e.text.replace('@', '').isnumeric()][0]
title = (
title_element.text.replace('@', '')
if titles is not None
and title_element is not None
and len(title_element.text) > 4
else None
)
author.bio = (
title if title is not None
else bio.text if bio is not None
else "More information at isni.org"
)
titles = []
# prefer title records from LoC+ coop, Australia, Ireland, or Singapore
# in that order
for source in ["LCNACO", "NLA", "N6I", "NLB"]:
for parent in element.findall(f'.//titleOfWork/[@source="{source}"]'):
titles.append(parent.find(".//title"))
for parent in element.findall(f'.//titleOfWork[@subsource="{source}"]'):
titles.append(parent.find(".//title"))
# otherwise just grab the first title listing
titles.append(element.find(".//title"))
if titles is not None:
# some of the "titles" in ISNI are a little ...iffy
# '@' is used by ISNI/OCLC to index the starting point ignoring stop words
# (e.g. "The @Government of no one")
title_elements = [e for e in titles if not e.text.replace('@', '').isnumeric()]
if len(title_elements):
author.bio = title_elements[0].text.replace('@', '')
else:
author.bio = None
possible_authors.append(author)