def __main__():
def get_content():
try:
return open(FILE)
except FileNotFoundError as e:
resp = requests.get(URL)
f = open(FILE, 'wb')
f.write(resp.content)
print(resp.content)
f.close()
return resp.content
return None
c = get_content()
soup = bs4.BeautifulSoup(get_content(), "lxml")
lis = soup.select('ul#container > li')
events = [li_to_event(li) for li in lis]
c = UnscrollClient()
# W3C Web Standards
TITLE = 'W3C Web Standards'
favthumb = c.cache_thumbnail(
'https://2.bp.blogspot.com/-70GFD8HsG3I/VMKLC7IoiBI/AAAAAAAAIck/GCu0LIY3PCU/s1600/Logo%2BW3C.png'
)
c.delete_scroll_with_title(TITLE)
c.__batch__(scroll_title=TITLE, thumbnail=favthumb['url'], events=events)
print(len(events))
def create(newsgroup, dir, maxyear):
_title = '{}'.format(newsgroup)
api = UnscrollClient()
api.delete_scroll_with_title(_title)
favthumb = api.cache_thumbnail(THUMBNAIL_URL)
with_thumbnail = favthumb.get('url')
scroll = api.create_or_retrieve_scroll(
_title,
description='Usenet message board archives',
link='https://archive.org/details/usenethistorical',
with_thumbnail=favthumb['url'],
subtitle='Collection via Usenet Historical Collection',
)
newsgroup_to_events(newsgroup, scroll, api, dir, maxyear)
def load_data(begin=None,
end=None,
title=None,
slug=None,
thumbnail_url=None,
delete=False):
# Get the file listing
_link = 'https://archive.org/details/{}'.format(slug)
_r = requests.get('https://archive.org/metadata/{}'.format(slug))
_data = _r.json()
# Get metadata
_md = _data.get('metadata')
_title = title
if (title is None):
_title = _md.get('title')
_description = _md.get('description')
_events = extract_events(_data, 'MP3', _link, begin, end)
_thumbnail_url = thumbnail_url
if (thumbnail_url is None):
_thumbnail_url = extract_thumbnail(_data)
api = UnscrollClient()
_thumb = api.cache_thumbnail(_thumbnail_url)
_with_thumbnail = _thumb.get('url')
if delete is True:
api.delete_scroll_with_title(_title)
print('XXXXXXX{}'.format(_title))
scroll = api.create_or_retrieve_scroll(
_title,
subtitle='via Archive.org',
public=True,
description=_description,
link=_link,
citation='',
with_thumbnail=_with_thumbnail,
)
for event in _events:
pprint.pprint(event)
j = api.create_event(event, scroll)
pprint.pprint(j.json())
def __main__():
c = UnscrollClient()
c.login()
c.delete_scroll_with_title('Amazon PR')
thumbnail = 'http://media.corporate-ir.net/media_files/IROL/17/176060/img/logos/amazon_logo_RGB.jpg'
favthumb = c.cache_thumbnail(thumbnail)
scroll = c.create_or_retrieve_scroll('Amazon PR',
description='A set of press releases from the Amazon Press Room.',
link='http://phx.corporate-ir.net/phoenix.zhtml?c=176060&p=irol-news&nyo=0',
citation='Amazon Press Room',
with_thumbnail=favthumb.get('url'))
print(scroll)
get_releases(c, scroll)
def __main__():
events = []
title = 'IETF RFCs'
c = UnscrollClient()
c.delete_scroll_with_title('IETF RFCs')
favthumb = c.cache_thumbnail(
'https://ietf.org/media/images/ietf-logo.original.png')
# Load RFCs
read = ''
with open('cache/rfc/rfc-index.xml', 'r') as f:
read = f.read()
parsed = xmltodict.parse(read)
docs = parsed['rfc-index']['rfc-entry']
events = [rfc_to_event(x) for x in docs]
# Do it
scroll = c.__batch__(scroll_title=title,
thumbnail=favthumb['url'],
events=events)
print(len(events))
def __main__():
scroll_thumb = "https://upload.wikimedia.org/wikipedia/commons/0/0b/Studs_Terkel_-_1979-1.jpg"
api = UnscrollClient()
title = "Studs Terkel Interviews"
favthumb = api.cache_thumbnail(scroll_thumb)
with_thumbnail = favthumb.get('url')
api.delete_scroll_with_title(title)
scroll = api.create_or_retrieve_scroll(
title,
description='<b>Via the Studs Terkel Radio Archive at WFMT</b>: '
'In his 45 years on WFMT radio, Studs Terkel talked to the 20th '
'century’s most interesting people.',
link='https://studsterkel.wfmt.com/',
with_thumbnail=with_thumbnail,
subtitle='Collection via WFMT',
)
post_shows(api, scroll)
from bs4 import BeautifulSoup
import requests
from pprint import pprint
from unscroll import UnscrollClient
import datefinder
from random import random
ADOBE_URL = "http://news.adobe.com/views/ajax?js=1&page={}&view_name=bw_press_release&view_display_id=panel_pane_7&view_args=all%2Fall&view_path=news&view_base_path=null&view_dom_id=1&pager_element=0"
c = UnscrollClient(api='http://127.0.0.1',
username='******',
password='******')
c.login()
favicon_url = c.fetch_favicon_url('https://www.adobe.com')
favthumb = c.cache_thumbnail(favicon_url['url'])
c.create_or_retrieve_scroll('Adobe PR', thumbnail=favthumb['url'])
for i in range(1, 92):
pr_url = ADOBE_URL.format(i, )
r = requests.get(pr_url)
r_as_data = r.json()
r_html = r_as_data['display']
parsed = BeautifulSoup(r_html, 'html.parser')
els = parsed.find_all('div', class_='view-inner-wrapper')
events = []
for el in els:
date_source = el.find('div', class_='views-field-created')
date_source_txt = date_source.text
from unscroll import UnscrollClient
from dateparser import parse
import datefinder
from random import random
import re
APPLE_URL = 'https://www.apple.com'
APPLE_PR_URL = 'https://www.apple.com/pr/library'
c = UnscrollClient(api='http://127.0.0.1',
username='******',
password='******')
c.login()
favicon_url = c.fetch_favicon_url(APPLE_URL)
favthumb = c.cache_thumbnail(favicon_url['url'])
print(favthumb)
c.create_or_retrieve_scroll('Apple Press Releases, 2000-2017',
thumbnail=favthumb['url'])
for i in range(1, 66):
pr_url = 'https://www.apple.com/newsroom/archive/?page={}'.format(i, )
print(pr_url)
r = requests.get(pr_url)
parsed = BeautifulSoup(r.content, 'html.parser')
dts = parsed.find_all('a', class_='result__item')
events = []
for dt in dts:
title = dt.find('h3').text
text = dt.find('p').text
class WikipediaText():
year = None
events = []
subject = None
parsed = None
unscroll_client = None
scroll = None
def __init__(self, year=None, subject=None):
self.year = year
self.subject = subject
self.wiki_url = 'https://en.wikipedia.org/wiki/{}'.format(year)
if subject is not None:
self.wiki_url = 'https://en.wikipedia.org/wiki/{}_in_{}'.format(
year, subject)
r = requests.get(self.wiki_url)
self.parsed = BeautifulSoup(r.content, 'html.parser')
self.unscroll_client = UnscrollClient()
self.unscroll_client.login()
favthumb = self.unscroll_client.cache_thumbnail(THUMBNAIL_URL)
subject_title = subject
if subject is None:
subject_title = 'Review'
self.scroll = self.unscroll_client.create_or_retrieve_scroll(
'Wiki Years in {}'.format(subject_title),
description='Events spidered from the English Wikipedia pages.',
link='https://en.wikipedia.org/wiki/List_of_years',
with_thumbnail=favthumb.get('url'))
def tidy(self, txt=None):
return re.sub('\[edit\]\s*', '', txt)
def realday(self, monthname=None, day=None):
month = MONTHS_HASH[monthname]
day = int(day)
return date(self.year, month, day)
def wikihtml_to_event(self, date=None, wikihtml=None, kind=None):
sup = wikihtml.find('sup')
if sup is not None:
_ = sup.extract()
contents = [str(x) for x in wikihtml.children]
joined = "".join(contents)
linked = re.sub(r'/wiki/', 'http://en.wikipedia.org/wiki/', joined)
targeted = re.sub(r'href=', 'target="_blank" href=', linked)
bleached = bleach.clean(targeted,
tags=['b', 'i', 'strong', 'em'],
strip=True)
pass1 = re.sub(MONTHS_PREFIX, '', bleached)
pass2 = re.sub(MONTHS_PREFIX, '', pass1)
lastpass = re.sub('^\s*\d+\s*[-–—]\s*', '', pass2)
titles = [
x['title'] for x in wikihtml.find_all('a') if x.has_attr('title')
]
filtered = [x for x in titles if not MONTH_REGEX.match(x)]
title = None
subject = None
if len(filtered) == 0:
title = " ".join(bleached.split(" ")[0:4]) + '...'
else:
title = filtered[0]
subject = title
thumbnail = None
if subject is not None:
image_d = self.unscroll_client.fetch_wiki_thumbnail_data(
title=subject)
image_url = image_d.get('url') if image_d is not None else None
if image_url is not None:
thumbnail_local = self.unscroll_client.cache_local(image_url)
thumbnail_d = self.unscroll_client.post_thumbnail(
thumbnail_local)
if thumbnail_d is not None:
thumbnail = thumbnail_d['url']
if kind == 'birth':
lastpass = '******'.format(lastpass)
elif kind == 'death':
lastpass = '******'.format(lastpass)
ranking = 0
if kind == 'world event':
ranking = 0.9
if kind == 'birth':
ranking = 0.1
if kind == 'death':
ranking = 0.5
dt = datetime.combine(date, datetime.max.time()).isoformat(' ')
wiki_subject = None
if subject is not None:
subject = re.sub(r'\s', '_', subject)
wiki_subject = 'https://en.wikipedia.org/wiki/{}'.format(subject, )
event = {
'title': lastpass,
'text': None,
'resolution': 10,
'ranking': ranking,
'when_happened': dt,
'when_original': None,
'with_thumbnail': thumbnail,
'content_url': wiki_subject,
'source_url': self.wiki_url,
'source_name': 'Wikipedia Event Pages',
'content_type': kind
}
e = self.unscroll_client.create_event(event, self.scroll)
pprint.pprint(e.json())
return event
def descend(self, ul=None, kind=None):
last_date = None
events = []
for d in ul:
if d.name == 'ul':
pass
elif d.name == 'li':
t = re.findall(MONTHS_DAYS, d.text)
if len(t) > 0:
last_date = t[0]
if not (d.find('ul')):
date = self.realday(monthname=last_date[0],
day=last_date[1])
e = self.wikihtml_to_event(date=date,
wikihtml=d,
kind=kind)
# print("A: {}\n".format(e.get('title')))
events.append(e)
elif last_date is not None:
date = self.realday(last_date[0], last_date[1])
e = self.wikihtml_to_event(date=date,
wikihtml=d,
kind=kind)
# print("B: {}\n".format(e.get('title')))
events.append(e)
if len(events) > 0:
return events
def get_events(self):
event_types = {
'#Events': 'world event',
'#Births': 'birth',
'#Deaths': 'death'
}
events = []
for keytype in event_types:
try:
events_h2 = self.parsed.select(keytype)[0].parent
for event in events_h2.next_siblings:
if event.name == "h2":
break
else:
if event.name == "h3":
pass
if event.name == 'ul':
es = self.descend(ul=event.descendants,
kind=event_types[keytype])
if es is not None:
events += es
except IndexError:
print('No {}'.format(keytype, ))
return events
'https://en.wikipedia.org/wiki/Timeline_of_World_War_II_(1945)',
'https://en.wikipedia.org/wiki/Timeline_of_World_War_II_(1945%E2%80%931991)',
'https://en.wikipedia.org/wiki/Timeline_of_the_Manhattan_Project'
]
api = UnscrollClient()
scroll = api.create_or_retrieve_scroll('WWII audio')
for page in pages:
items = extract_list('', page)
for item in items:
if item is not None:
thumb_url = None
wiki_thumb = api.fetch_wiki_thumbnail_data(item.get('item'))
if wiki_thumb is not None:
thumb = api.cache_thumbnail(wiki_thumb.get('url'))
if thumb is not None:
thumb_url = thumb.get('url')
content_url = item.get('content_url')
if content_url is None:
content_url = page
event = {
'title': item.get('title'),
'text': item.get('text'),
'content_url': content_url,
'source_url': page,
'with_thumbnail': thumb_url,
'when_happened': item.get('when_happened'),
'when_original': item.get('when_original'),
from unscroll import UnscrollClient
c = UnscrollClient()
p = c.cache_thumbnail(
'https://upload.wikimedia.org/wikipedia/commons/b/b2/Donnchadh_mac_Gille-Brighdhe_Seal.jpg'
)
print(p)
p2 = c.fetch_wiki_thumbnail('George_Orwell')
print(p2)
def __main__():
c = UnscrollClient()
c.login()
favthumb = c.cache_thumbnail(THUMBNAIL_IMAGE)
scroll = c.create_or_retrieve_scroll(
'Cooper-Hewitt',
description='Items from the Cooper Hewitt',
link='https://github.com/cooperhewitt/collection',
citation='Cooper-Hewitt Museum Collection',
with_thumbnail=favthumb.get('url'))
conn = sqlite3.connect('/home/unscroll/cache/cooper/objects.db')
conn.row_factory = sqlite3.Row
sqlc = conn.cursor()
i = 0
sqlc.execute("SELECT * FROM objects LIMIT -1 OFFSET {}".format(i))
for row in sqlc.fetchall():
if row['primary_image'] is not None and row['date'] is not None:
# switch to the 300x300 thumbnail
sq = re.sub('z\.jpg', 'sq.jpg', row['primary_image'])
local_sq = re.sub(r'https?://', '', sq)
local = '/home/unscroll/cache/cooper/{}'.format(local_sq, )
i = i + 1
found = False
try:
f = open(local, 'r')
f.close()
found = True
except FileNotFoundError as e:
try:
r = requests.get(sq)
p = pathlib.Path(local)
p.parent.mkdir(parents=True, exist_ok=True)
f = open(local, 'wb')
f.write(r.content)
f.close()
found = True
except ConnectionError as e:
print('[cooperhewitt2.py] ConnectionError: {}'.format(e, ))
print('{}: {}/{}'.format(i, local, found))
ud = UnscrollDate(row['date'], begin=-4000, end=2018)
if ud.is_okay():
with_thumbnail = None
if found:
thumb = c.post_thumbnail(local)
if thumb is not None:
with_thumbnail = thumb.get('url')
d = {
'title':
row['title'],
'text':
row['description'],
'resolution':
ud.resolution,
'ranking':
0,
'content_url':
'https://collection.cooperhewitt.org/objects/{}/'.format(
row['id'], ),
'with_thumbnail':
with_thumbnail,
'source_name':
'Collection Data for Cooper Hewitt, Smithsonian Design Museum',
'source_url':
'https://github.com/cooperhewitt/collection',
'when_happened':
ud.when_happened,
'when_original':
ud.when_original
}
e = c.create_event(d, scroll)