diff options
| author | Jules Laplace <julescarbon@gmail.com> | 2018-11-25 22:19:15 +0100 |
|---|---|---|
| committer | Jules Laplace <julescarbon@gmail.com> | 2018-11-25 22:19:15 +0100 |
| commit | ee3d0d98e19f1d8177d85af1866fd0ee431fe9ea (patch) | |
| tree | 41372528e78d4328bc2a47bbbabac7e809c58894 /s2-search.py | |
| parent | 255b8178af1e25a71fd23703d30c0d1f74911f47 (diff) | |
moving stuff
Diffstat (limited to 's2-search.py')
| -rw-r--r-- | s2-search.py | 46 |
1 files changed, 0 insertions, 46 deletions
diff --git a/s2-search.py b/s2-search.py deleted file mode 100644 index 44bab08b..00000000 --- a/s2-search.py +++ /dev/null @@ -1,46 +0,0 @@ -import os -import sys -import csv -import subprocess -import time -import random -import re -import simplejson as json -import click -from s2 import SemanticScholarAPI -from util import * - -''' -s2 search API format: -results -matchedAuthors -matchedPresentations -query -querySuggestions -results -stats -totalPages -totalResults -''' - -@click.command() -@click.option('--index', '-n', default=0, help='Index of CSV (query,)') -def fetch_entries(index): - keys, lines = read_citation_list(index) - s2 = SemanticScholarAPI() - 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): - results = s2.search(title) - write_json(dump_fn, results) - if len(results['results']) == 0: - print("No results for {}".format(title)) - else: - print(title) - write_json(entry_fn, results['results'][0]) - time.sleep(random.randint(10, 20)) - -if __name__ == '__main__': - fetch_entries() |
