moving stuff

1 files changed, 64 insertions, 0 deletions

diff --git a/scraper/s2-papers.py b/scraper/s2-papers.py
new file mode 100644
index 00000000..4fe1f025
--- /dev/null
+++ b/scraper/s2-papers.py
@@ -0,0 +1,64 @@
+import os
+import sys
+import csv
+import subprocess
+import time
+import random
+import re
+import json
+import click
+from s2 import SemanticScholarAPI
+from util import *
+
+'''
+s2 search API format:
+results
+matchedAuthors
+matchedPresentations
+query
+querySuggestions
+results
+stats
+totalPages
+totalResults
+'''
+
+s2 = SemanticScholarAPI()
+
+@click.command()
+@click.option('--index', '-n', default=0, help='Index of CSV (query,)')
+@click.option('--depth', '-d', default=1, help='Depth to recurse (not implemented).')
+def fetch_papers(index, depth):
+  keys, lines = read_citation_list(index)
+  for line in lines:
+    label = line[0]
+    title = re.sub(r'[^-0-9a-zA-Z ]+', '', line[1])
+    entry_fn = './datasets/s2/entries/{}.json'.format(title)
+    if not os.path.exists(entry_fn):
+      print('not found: {}'.format(entry_fn))
+      continue
+    result = read_json(entry_fn)
+    paper_id = result['id']
+    paper = fetch_paper(paper_id)
+    # get all of the paper's citations
+
+def fetch_paper(paper_id):
+  os.makedirs('./datasets/s2/papers/{}/{}'.format(paper_id[0:2], paper_id), exist_ok=True)
+  paper_fn = './datasets/s2/papers/{}/{}/paper.json'.format(paper_id[0:2], paper_id)
+  if os.path.exists(paper_fn):
+    return read_json(paper_fn)
+  print(paper_id)
+  paper = s2.paper(paper_id)
+  if paper is None:
+    print("Got none paper??")
+    time.sleep(random.randint(20, 30))
+    paper = s2.paper(paper_id)
+    if paper is None:
+      print("Paper not found")
+      return None  
+  write_json(paper_fn, paper)
+  time.sleep(random.randint(5, 10))
+  return paper
+
+if __name__ == '__main__':
+  fetch_papers()