Last active
August 29, 2015 14:04
-
-
Save jy00295005/6caa7472d4be23600e40 to your computer and use it in GitHub Desktop.
use three search keywords in nuclear physics to search in solr index, get 3 sets of data as our clustering's truth ground
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import requests | |
from pony.orm import * | |
##define the mysql database and the table Truth with pony | |
db = Database('mysql', host='localhost', user='root', passwd='xxxxxxxxx', db='clustering') | |
# sql_debug(True) | |
db.generate_mapping(create_tables=True) | |
class Truth(db.Entity): | |
id = PrimaryKey(int, auto=True) | |
systemid = Required(LongStr) | |
RefCode = Required(str) | |
Title = Required(str) | |
Abstract = Optional(LongStr) | |
label = Required(int) | |
## solr query and write results into a tuple | |
solr_base_url = 'http://xx.xx.xx.xxx:8080/xxxxxx/select/' | |
rows = 100 | |
words = ['higgs','Neutrino*','dark matter'] | |
iter_number = 2 ##manually change iter number of seerach key words | |
params=[ | |
('q','(PSShortName_Group:NSF OR PSShortName_Group:FP) AND Title:(%s)' % words[iter_number] ), | |
('start',0), | |
('rows',rows), | |
('wt','python') | |
] | |
r = requests.get(solr_base_url, params=params) | |
rsp = eval( r.text ) | |
# print rsp['response']['numFound'] | |
data = [ | |
(doc['id'],doc['RefCode'],doc['Title'],doc['Abstract'] ) | |
for doc in rsp['response']['docs'] if 'Abstract' in doc | |
] | |
# print len(data) | |
## insert tuple into Truth table | |
with db_session: | |
for (id, RefCode, Title, Abstract) in data: | |
Truth( | |
systemid = id, | |
RefCode = RefCode, | |
Title = Title, | |
Abstract = Abstract, | |
label = iter_number | |
) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment