53 lines
1.4 KiB
Python
53 lines
1.4 KiB
Python
from pyquery import PyQuery as pq
|
|
from feedgen.feed import FeedGenerator
|
|
from collections import namedtuple
|
|
import multiprocessing
|
|
from datetime import datetime
|
|
|
|
Entry = namedtuple('Entry', 'url fe')
|
|
|
|
class NvidiaParser:
|
|
NAME = 'NVidia Research'
|
|
URL = '/nvidia'
|
|
CACHE_TIMEOUT = 3600
|
|
|
|
def parseNvidiaDate(entry):
|
|
dom = pq(entry.url)
|
|
print(entry.url)
|
|
|
|
time = dom('.field--name-field-publication-date').find('time').attr.datetime
|
|
time = datetime.strptime(time, '%Y-%m-%dT%H:%M:%S%z')
|
|
entry.fe.pubDate(time)
|
|
|
|
def getRss(self):
|
|
root_url = 'https://research.nvidia.com'
|
|
d = pq(root_url +'/publications')
|
|
|
|
fg = FeedGenerator()
|
|
fg.id(root_url)
|
|
fg.title('NVidia Research')
|
|
fg.link(href=root_url, rel='alternate')
|
|
fg.logo(root_url + '/themes/custom/nvidia/favicon.ico')
|
|
fg.description('NVidia Research papers')
|
|
|
|
entries = []
|
|
print('RSS GOT')
|
|
|
|
for elem in d('.views-field-title').items():
|
|
link = elem.find('a')
|
|
url = root_url + link.attr.href
|
|
title = link.text()
|
|
|
|
fe = fg.add_entry()
|
|
fe.id(url)
|
|
fe.title(title)
|
|
fe.link(href=url)
|
|
|
|
entries.append(Entry(url, fe))
|
|
|
|
with multiprocessing.Pool(8) as p:
|
|
p.map(NvidiaParser.parseNvidiaDate, entries)
|
|
|
|
return fg.rss_str()
|
|
|