def find_press_of_journal():
presses = session.query(Press)
for press in presses:
press_id = press.id
press_url = press.url
req = requests.get(press_url, headers=HEADER)
txt = req.text
soup = BeautifulSoup(txt, features="lxml")
press_soup = soup.find("div", "clear-both")
press_soup = press_soup.find_next_sibling()
press_soup = press_soup.find_next_sibling()
p_infos = press_soup.find_all("li")
while p_infos:
for p_info in p_infos:
journal_url = p_info.a["href"]
if const.DBLP_JOURNAL_PREVIX in journal_url:
journal_url = journal_url[:-10]
j_q = session.query(Journal)\
.filter(Journal.dblp_address == journal_url).first()
if j_q:
j_q.press = press_id
session.commit()
press_soup = press_soup.find_next_sibling()
p_infos = press_soup.find_all("li")
session.close()
def update_is_updated_of_volumes():
paper_volume_ids = session.query(distinct(Paper.volume_id))
volume_ids = session.query(Volume.id).filter(
Volume.id.notin_(paper_volume_ids)).all()
for volume_id in volume_ids:
session.query(Volume).filter(Volume.id == volume_id[0]).update(
{Volume.is_updated: False})
session.commit()
session.close()
def update_dblp_id_of_papers():
volume_ids = get_volumes_of_None()
for volume_id in volume_ids:
volume = session.query(Volume).filter(Volume.id == volume_id).first()
if not volume:
continue
req = requests.get(volume.url, headers=HEADER)
txt = req.text
soup = BeautifulSoup(txt, features="lxml")
sibling = soup.body.find("ul", class_="publ-list")
if sibling is None:
return info_of_papers
# sibling = body_main.find_previous_sibling()
while sibling:
# item_name = "entry informal"
# item_name = "entry book"
article_entries = sibling.find_all(
"li", {"class": ["entry data", "entry article"]})
if not len(article_entries):
sibling = sibling.find_next_sibling()
continue
for article_entry in article_entries:
paper_id = article_entry["id"]
paper_doi = get_paper_doi(article_entry)
#paper_title = get_paper_title(article_entry)
#if paper_title is None:
# paper_title = ""
#if len(paper_title) > 255:
# paper_title = paper_title[:255]
#if not paper_title:
# continue
paper_info = session.query(Paper)\
.filter(Paper.doi == paper_doi).first()
if not paper_info:
continue
paper_dblp_id = paper_info.dblp_id
if not paper_dblp_id:
paper_info.dblp_id = paper_id
session.commit()
sibling = sibling.find_next_sibling()
def check_journal_press(journal_url):
journal_query = session.query(Journal)\
.filter(Journal.dblp_address == journal_url).first()
if journal_query:
return True
else:
return False
def get_volumes_of_None():
volume_ids = session.query(distinct(Paper.volume_id))\
.filter(Paper.dblp_id == None)
v_ids = list()
for volume_id in volume_ids:
v_ids.append(volume_id[0])
return v_ids
def check_dblp_address(self):
dblp_addresses = session.query(Journal.dblp_address)
num = 0
for dblp_address in dblp_addresses:
if const.DBLP_JOURNAL_PREVIX in dblp_address[0]:
print(dblp_address[0])
num = num + 1
print(num)
def update_year_of_volumes():
volumes = session.query(Volume).filter(Volume.id >= 34745)
for volume in volumes:
volume_id = volume.id
paper_info = session.query(Paper)\
.filter(Paper.volume_id == volume_id).first()
if not paper_info:
continue
paper_dblp_id = paper_info.dblp_id
paper_bibtex_url = \
const.DBLP_JOURNAL_BIBTEX_PREVIX + paper_dblp_id + \
const.DBLP_JOURNAL_BIBTEX_SUFFIX
per_bibtex = get_paper_bibtex(paper_bibtex_url)
volume_year = per_bibtex.get("year")
volume_info = per_bibtex.get("volume")
volume.year = volume_year
volume.volume = volume_info
session.commit()
def find_volume_by_url(volume_url):
return session.query(Volume).filter(Volume.url == volume_url).first()
def get_all_journals_from_db():
return session.query(Journal).all()
def set_updated_status_of_volumes(volume_id):
volume = session.query(Volume).filter(Volume.id == volume_id).first()
volume.is_updated = True
session.commit()
def query_journal_is_in_volumes(journal_issn):
result = session.query(Volume).filter(Volume.issn == journal_issn).first()
if result:
return True
else:
return False
def query_volumes():
volumes = session.query(Volume.issn).distinct().all()
print(len(volumes))
def query_volumes_by_filter(conditions, limit_num=1000):
query = session.query(Volume).filter(conditions).limit(limit_num)
return query
def collect_journal_papers():
arx_journal_issn = "2331-8422"
# while True:
# conditions = (Volume.is_updated == False)
volume_id = 46770
while volume_id <= 46770:
conditions = (Volume.id == volume_id)
new_volumes = analyze_volumes.query_volumes_by_filter(
conditions) #, limit_num=5)
new_volume_infos = list()
for new_volume in new_volumes:
new_volume_infos.append({
const.VOLUME_ID: new_volume.id,
const.JOURNAL_ISSN: new_volume.issn,
const.VOLUME_URL: new_volume.url
})
# if not new_volume_infos:
# print("Finish the update!")
# break
for new_volume_info in new_volume_infos:
if new_volume_info[const.JOURNAL_ISSN] == arx_journal_issn:
continue
print("%s" % new_volume_info[const.VOLUME_URL])
print("VOLUME ID: %s" % new_volume_info[const.VOLUME_ID])
# if new_volume_info[const.VOLUME_ID] in [35956, 35957, 44428,
# 44429, 44430]:
# continue
new_paper_infos = analyze_papers.analyze_papers_of_volume(
new_volume_info[const.VOLUME_URL])
for new_paper_info in new_paper_infos:
paper_query = \
session.query(Paper).filter(
Paper.dblp_id == new_paper_info[const.PAPER_DBLP_ID])\
.first()
if paper_query is not None:
print("Database has the record [dblp_id: %s]" %
paper_query.dblp_id)
continue
else:
print("Database has no record [dblp_id: %s]" %
new_paper_info[const.PAPER_DBLP_ID])
new_paper = Paper(
title=new_paper_info[const.PAPER_TITLE],
journal_issn=new_volume_info[const.JOURNAL_ISSN],
volume_id=new_volume_info[const.VOLUME_ID],
volume=new_paper_info[const.PAPER_VOLUME],
number=new_paper_info[const.PAPER_NUMBER],
start_page=new_paper_info[const.PAPER_START_PAGE],
end_page=new_paper_info[const.PAPER_END_PAGE],
year=new_paper_info[const.PAPER_DATE],
url=new_paper_info[const.PAPER_URL],
doi=new_paper_info[const.PAPER_DOI],
dblp_id=new_paper_info[const.PAPER_DBLP_ID])
session.add(new_paper)
session.flush()
session.refresh(new_paper)
new_paper_id = new_paper.id
# add authors
author_infos = new_paper_info[const.PAPER_AUTHOR]
order = 1
for author_info in author_infos:
author_query = \
session.query(Author).filter(
Author.title == author_info["author_title"]).first()
if author_query:
author_id = author_query.id
else:
new_author = Author(
title=author_info["author_title"],
name=author_info["author_name"],
dblp_url=author_info["author_dblp_url"])
session.add(new_author)
session.flush()
session.refresh(new_author)
author_id = new_author.id
session.add(
PaperAuthor(paper_id=new_paper_id,
author_id=author_id,
order=order))
order += 1
session.commit()
volume = \
session.query(Volume).filter(
Volume.id == new_volume_info[const.VOLUME_ID]).first()
volume.is_updated = True
session.commit()
print("Volume - [ID: %s] has been processed!" %
new_volume_info[const.VOLUME_ID])
volume_id = volume_id + 1
def find_journal_by_issn(journal_issn):
return session.query(Journal).filter(Journal.issn == journal_issn).first()
def find_journal_by_title(journal_title):
return session.query(Journal).filter(Journal.name == journal_title).first()
from databases.db_engine import engine, session from information_collection import analyze_journal from information_collection import analyze_journal_urls from information_collection.journal_info import Journal from publish_lib import generate_random print(session.query(Journal).filter().count())