1 files changed, 302 insertions, 0 deletions
diff --git a/scraper/util.py b/scraper/util.py
new file mode 100644
index 00000000..47e5a4aa
--- /dev/null
+++ b/scraper/util.py
@@ -0,0 +1,302 @@
+import re
+import os
+import csv
+import string
+import codecs
+import gspread
+import simplejson as json
+from oauth2client.service_account import ServiceAccountCredentials
+
+def read_citation_list(index=0):
+  filename = './datasets/citations.csv'
+  if index > 0:
+    fn, ext = os.path.splitext(filename)
+    filename = fn + '-' + str(index) + ext
+  with open(filename, 'r') as f:
+    reader = csv.reader(f)
+    lines = list(reader)
+    keys = lines[0]
+    lines = lines[1:]
+    return keys, lines
+
+def unfussy_reader(reader):
+  while True:
+    try:
+      yield next(reader)
+    except StopIteration:
+      return
+    except csv.Error:
+      print(csv.Error)
+      # log the problem or whatever
+      continue
+
+def read_csv(fn, keys=True, create=False):
+  try:
+    with open(fn, 'r', newline='', encoding='utf-8') as f:
+      # reader = csv.reader( (line.replace('\0','') for line in f) )
+      reader = csv.reader(f)
+      lines = list(unfussy_reader(reader))
+      if keys:
+        keys = lines[0]
+        lines = lines[1:]
+        return keys, lines
+      return lines
+  except:
+    if create:
+      return []
+    raise
+
+def csv_writer(fn):
+  with open(fn, 'w', newline='', encoding='utf-8') as f:
+    return csv.writer(f)
+
+def write_csv(fn, keys, rows):
+  with open(fn, 'w', newline='', encoding='utf-8') as f:
+    writer = csv.writer(f)
+    if keys is not None:
+      writer.writerow(keys)
+    for row in rows:
+      writer.writerow(row)
+
+def read_text(fn):
+  with open(fn, 'r') as f:
+    return f.read()
+
+def read_json(fn):
+  with open(fn, 'r') as json_file:
+    return json.load(json_file)
+
+def write_json(fn, data):
+  with open(fn, 'w') as outfile:
+    json.dump(data, outfile)
+
+def write_report(fn, title=None, keys=None, rows=[]):
+  with open(fn, 'w') as f:
+    f.write("<!doctype html>")
+    f.write("<html>")
+    f.write("<head>")
+    f.write("<meta charset='utf-8'>")
+    if title is not None:
+      f.write("<title>{}</title>".format(title))
+    f.write("<link rel='stylesheet' href='reports.css'>")
+    f.write("</head>")
+    f.write("<body>")
+    if title is not None:
+      f.write("<h2>{}</h2>".format(title))
+    count = write_table(f, keys=keys, rows=rows)
+    f.write("</body>")
+    f.write("</html>")
+    print("{} {}".format(fn, count))
+
+def percent(m, n):
+  if n == 0:
+    return 100
+  return round(m / n * 100)
+
+class NameLine(object):
+  def __init__(self, s):
+    self.s = s.strip()
+  def __str__(self):
+    return '<span class="name">' + self.s + '</span>'
+
+class BoldLine(object):
+  def __init__(self, s):
+    self.s = s.strip()
+  def __str__(self):
+    return '<b>' + self.s + '</b>'
+
+class LinkLine(object):
+  def __init__(self, href, txt):
+    self.href = href
+    self.txt = txt.strip()
+  def __str__(self):
+    if self.href:
+      return '<a href="{}">{}</a>'.format(self.href, self.txt)
+    else:
+      return '<span class="gray">{}</a>'.format(self.txt)
+
+def write_table(f, keys, rows):
+  count = 0
+  f.write("<table border='1' cellpadding='3' cellspacing='3'>")
+  if keys is not None:
+    for key in keys:
+      f.write("<th>{}</th>".format(key))
+  for row in rows:
+    if row is None:
+       return
+    count += 1
+    f.write("<tr>")
+    for cell in row:
+      if isinstance(cell, list) or isinstance(cell, tuple):
+        f.write("<td>{}</td>".format('<br/>'.join(str(x) for x in cell)))
+      else:
+        f.write("<td>{}</td>".format(cell))
+    f.write("</tr>")
+  f.write("</table>")
+  return count
+
+def paper_path(key='papers', paper_id=''):
+  return '{}/{}/{}/{}/paper.json'.format('./datasets/s2', key, paper_id[0:2], paper_id)
+
+class DbPaper(object):
+  def __init__(self, paper_id):
+    self.paper_id = paper_id
+    self.data = read_json(paper_path('db_papers', paper_id))
+  @property
+  def title(self):
+    return self.data['title']
+  @property
+  def journal(self):
+    return self.data['journalName']
+  @property
+  def year(self):
+    return self.data['year'] if 'year' in self.data else ''
+  @property
+  def authors(self):
+    return [ (author['ids'][0] if len(author['ids']) else '', author['name']) for author in self.data['authors'] ]
+  @property
+  def pdf_link(self):
+    if self.data['s2PdfUrl']:
+      return self.data['s2PdfUrl']
+    if len(self.data['pdfUrls']):
+      return self.data['pdfUrls'][0]
+    return None
+  def record(self):
+    return [ self.paper_id, self.title, self.journal, self.year ]
+
+class RawPaper(object):
+  def __init__(self, paper_id):
+    self.paper_id = paper_id
+    data = read_json(paper_path('raw_papers', paper_id))
+    # print(data)
+    if 'paper' not in data:
+      print(data)
+      self.data = None
+      return None
+    self.data = data['paper']
+  @property
+  def title(self):
+    return self.data['title']['text']
+  @property
+  def year(self):
+    return self.data['year']['text']
+  @property
+  def journal(self):
+    if 'journal' in self.data and 'name' in self.data['journal']:
+      return self.data['journal']['name']
+    else:
+      return 'Unknown'
+  @property
+  def authors(self):
+    return [ (author[0]['ids'][0] if len(author[0]['ids']) else '', author[0]['name']) for author in self.data['authors'] ]
+  @property
+  def pdf_link(self):
+    if 'primaryPaperLink' in self.data:
+      return self.data['primaryPaperLink']
+    return None
+  def record(self):
+    return [ self.paper_id, self.title, self.journal, self.year ]
+
+def load_paper(paper_id):
+  if os.path.exists(paper_path('db_papers', paper_id)):
+    # print('db paper')
+    return DbPaper(paper_id)
+  if os.path.exists(paper_path('raw_papers', paper_id)):
+    # print('raw paper')
+    return RawPaper(paper_id)
+  print('no paper')
+  return None
+
+def dedupe(a):
+  p = {}
+  for s in a:
+    p[s] = None
+  ss = sorted(p.keys())
+  return ss
+
+def read_headings(fn, paper):
+  headings = []
+  found_abstract = False
+  found_authors = []
+  journal = paper.journal.lower()
+  authors = [ (a[0], a[1], a[1].lower(),) for a in paper.authors ]
+  with open(fn, 'r') as f:
+    for line in f.readlines():
+      line = re.sub(r"\S*@\S*\s?", '', line)
+      l = line.lower().strip()
+      if len(l) < 5:
+        continue
+      if line[0] == 'a' or line[0] == 'b' or line[0] == 'c' or line[0] == '1' or line[0] == '2' or line[0] == '3' or line[0] == '4':
+        line = line[1:]
+      line = line.strip("∗†‡")
+      line = line.replace("ﬂ", "fl").replace('ﬀ', 'ff').replace('ﬃ', 'f‌f‌i').replace('ﬄ', 'f‌f‌l')
+      line = line.strip()
+      if 'abstract' in l:
+        found_abstract = True
+        break
+      if journal and journal in l:
+        continue
+      names = [s.strip() for s in re.split(',| and ', l)]
+      was_found = False
+      for name in names:
+        found = find_authors(authors, name)
+        if found:
+          was_found = True
+          # print("found {}".format(found[1]))
+          if found[0]:
+            found_authors.append(found)
+            continue
+      headings.append(line.strip())
+  return headings, found_abstract
+
+def find_authors(authors, line):
+  for a in authors:
+    if a[2] in line:
+      return a
+  return None
+
+class AddressBook (object):
+  def __init__(self):
+    entities = {}
+    lookup = {}
+    keys, data = fetch_google_sheet('institutions')
+    # keys, data = read_csv('reports/pdf_institutions_deduped.csv', keys=True)
+    for index, line in enumerate(data):
+      if line[0] == line[1] or line[0] not in entities:
+        entities[line[0]] = index
+      lookup[line[1].lower().strip()] = line[0]
+    self.data = data
+    self.lookup = lookup
+    self.entities = entities
+
+  def find(self, address):
+    address = address.lower().strip().strip(string.digits)
+    if address in self.lookup:
+      entity = self.lookup[address]
+      index = self.entities[entity]
+      return self.data[index]
+    for part in address.split(','):
+      part = part.strip().replace('  ', ' ')
+      if part in self.lookup:
+        entity = self.lookup[part]
+        index = self.entities[entity]
+        return self.data[index]
+    return None
+
+def fetch_spreadsheet():
+  scope = ['https://spreadsheets.google.com/feeds','https://www.googleapis.com/auth/drive']
+  credentials = ServiceAccountCredentials.from_json_keyfile_name('./.creds/Megapixels-ef28f91112a9.json', scope)
+  docid = "1denb7TjYsN9igHyvYah7fQ0daABW32Z30lwV7QrDJQc"
+  client = gspread.authorize(credentials)
+  spreadsheet = client.open_by_key(docid)
+  return spreadsheet
+
+def fetch_worksheet(name="institutions"):
+  spreadsheet = fetch_spreadsheet()
+  return spreadsheet.worksheet(name)
+
+def fetch_google_sheet(name="institutions"):
+  rows = fetch_worksheet(name).get_all_values()
+  keys = rows[0]
+  lines = rows[1:]
+  return keys, lines