Basically a fork from https://github.com/blochberger/sokman but with the intention of adding a visual interface as well
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
68 lines
2.3 KiB
68 lines
2.3 KiB
from typing import Set
|
|
import pickle
|
|
from django.db.models import Count
|
|
|
|
from pathlib import Path
|
|
|
|
from django.core.management.base import BaseCommand
|
|
from django.db.models import Count, Q
|
|
|
|
import sok.management.commands.dblpimport as dblp
|
|
|
|
from sok.models import Publication, SearchTerm
|
|
|
|
|
|
class Command(BaseCommand):
|
|
|
|
def echo(self, msg: str, bold=True):
|
|
if bold:
|
|
msg = self.style.HTTP_INFO(msg)
|
|
self.stdout.write(msg)
|
|
|
|
# BaseCommand
|
|
|
|
def handle(self, *args, **options):
|
|
publications_found: Set[str] = set()
|
|
publications_peer_reviewed: Set[str] = set()
|
|
publications_relevant: Set[str] = set()
|
|
|
|
self.echo("Loading DBLP dump...")
|
|
all_cite_keys = dblp.get_all_cite_keys(dblp.DUMP_PATH)
|
|
|
|
for search_term in SearchTerm.objects.all():
|
|
# DBLP search result
|
|
self.echo(f"Searching DBLP for '{search_term}'")
|
|
query, results, total = dblp.PublicationResult.from_search(search_term.name, 1000)
|
|
for result in results:
|
|
if result.cite_key not in all_cite_keys:
|
|
continue
|
|
publications_found.add(result.cite_key)
|
|
if result.is_peer_reviewed:
|
|
publications_peer_reviewed.add(result.cite_key)
|
|
|
|
# Relevant publications
|
|
for publication in Publication.objects.filter(
|
|
publicationsource__search_term=search_term,
|
|
exclusion_criteria__isnull=True,
|
|
).distinct():
|
|
publications_relevant.add(publication.cite_key)
|
|
|
|
self.cache_path = Path('.choices.semanticscholar.pickle')
|
|
if self.cache_path.exists():
|
|
self.echo("Loading choices from snowballing...")
|
|
with self.cache_path.open('rb') as f:
|
|
self.cache = pickle.load(f)
|
|
self.echo(f"{len(self.cache)} publications excluded during snowballing")
|
|
self.echo(f"{len(Publication.objects.filter(variant_of__isnull=True))} publications included in total so far")
|
|
|
|
# Output
|
|
self.echo(f"Total publications: {len(publications_found):4d}", bold=True)
|
|
self.echo(f"- peer reviewed: {len(publications_peer_reviewed):4d}", bold=True)
|
|
self.echo(f"- relevant: {len(publications_relevant):4d}", bold=True)
|
|
|
|
pubs = Publication.objects.values('title').annotate(count=Count('referenced_by')).values('title', 'year', 'count').order_by('count').reverse()
|
|
for p in pubs[:15]: print(f"{p['title']} ({p['year']}): {p['count']} citations")
|
|
|
|
years = Publication.objects.values('year').annotate(count=Count('year')).values('year', 'count')
|
|
for y in years: print(f"{y['year']}, {y['count']}")
|
|
|
|
|