def search_diffbot_cache(term):
response = requests.get('http://api.diffbot.com/v3/search',
params={
'token': config.credentials.diffbot,
'query':
requests.utils.quote('"{}"'.format(term)),
'col': 'GLOBAL-INDEX'
}).json()
if not response.get('objects'):
if response.get('error'):
print("Response Error '{}' (code: {})".format(
response['error'], response['errorCode']))
else:
print("NO RESULTS")
results = []
for object in response.get('objects', []):
if object.get('text'):
pr = PageRequest(object.get('pageUrl'), term, run=False)
pr.extract_sentences(object.get('text'))
result = {
"title": object.get('title'),
"url": object.get('pageUrl'),
'search_provider': 'diffbot',
"author": object.get('author'),
"date": parse_date(object.get('date', '')).isoformat(),
"doc": object.get('text'),
"sentences": pr.sentences,
"variants": list(pr.variants)
}
results.append(result)
return results
def search_diffbot_cache(term):
response = requests.get('http://api.diffbot.com/v3/search', params={
'token': config.credentials.diffbot,
'query': requests.utils.quote('"{}"'.format(term)),
'col': 'GLOBAL-INDEX'
}).json()
if not response.get('objects'):
if response.get('error'):
print("Response Error '{}' (code: {})".format(response['error'], response['errorCode']))
else:
print("NO RESULTS")
results = []
for object in response.get('objects', []):
if object.get('text'):
pr = PageRequest(object.get('pageUrl'), term, run=False)
pr.extract_sentences(object.get('text'))
result = {
"title": object.get('title'),
"url": object.get('pageUrl'),
'search_provider': 'diffbot',
"author": object.get('author'),
"date": parse_date(object.get('date', '')).isoformat(),
"doc": object.get('text'),
"sentences": pr.sentences,
"variants": list(pr.variants)
}
results.append(result)
return results
def test_extract_html_features():
from serapis.extract import PageRequest
test_request = PageRequest("http://thescene.whro.org/hear-cool-stuff", 'defenestration', run=False)
test_html = "<div><p><em><strong>de-fen-es-tra-tion</strong></em> (dee-fen-uh-STRAY-shun) | n. the act of throwing someone or something out of a window</p></div><div>"
test_request.get_html_features(test_html)
assert test_request.features['highlighted']
def test_extract_html_features():
from serapis.extract import PageRequest
test_request = PageRequest("http://thescene.whro.org/hear-cool-stuff",
'defenestration',
run=False)
test_html = "<div><p><em><strong>de-fen-es-tra-tion</strong></em> (dee-fen-uh-STRAY-shun) | n. the act of throwing someone or something out of a window</p></div><div>"
test_request.get_html_features(test_html)
assert test_request.features['highlighted']
def extract_wrapper(url_object, term):
try:
result = PageRequest(url_object['url'], term).structured
except Exception:
import traceback
log.error("Failed to get page {} -- {}".format(url_object['url'],
traceback.format_exc()))
return url_object
return merge_dict(url_object, result)
def test_page_structure():
from serapis.extract import PageRequest
p = PageRequest(
'http://nytimes.com/2015/10/04/technology/scouring-the-web-to-make-new-words-lookupable.html',
"lookupable")
assert p.structured['title'] == test_response['title']
assert p.structured['url'] == test_response['url']
assert p.structured['author'] == test_response['author']
assert len(p.structured['doc']) > 0
def search_duckduckgo(term):
result = []
try:
req = requests.get(
'http://api.duckduckgo.com/?q={}&format=json'.format(term)).json()
except:
return result
if req['AbstractSource'] not in config.duckduckgo_sources:
return result
if req.get('Abstract'):
pr = PageRequest(req['AbstractURL'], term, run=False)
pr.extract_sentences(req['Abstract'])
result.append({
'title': req['Heading'],
'url': req['AbstractURL'],
'search_provider': 'duckduckgo',
'author': None,
'date': None,
'source': req['AbstractSource'],
'doc': req['Abstract'],
"sentences": pr.sentences,
"variants": list(pr.variants)
})
if req.get('Definition'):
pr = PageRequest(req['DefinitionURL'], term, run=False)
pr.extract_sentences(req['Definition'])
result.append({
'title': req['Heading'],
'url': req['DefinitionURL'],
'source': req['DefinitionSource'],
'search_provider': 'duckduckgo',
'author': None,
'date': None,
'doc': req['Definition'],
"sentences": pr.sentences,
"variants": list(pr.variants)
})
log.info("Searching DuckDuckGo for '{}' returned {} results".format(
term, len(result)))
return result
def search_duckduckgo(term):
result = []
try:
req = requests.get('http://api.duckduckgo.com/?q={}&format=json'.format(term)).json()
except:
return result
if req['AbstractSource'] not in config.duckduckgo_sources:
return result
if req.get('Abstract'):
pr = PageRequest(req['AbstractURL'], term, run=False)
pr.extract_sentences(req['Abstract'])
result.append({
'title': req['Heading'],
'url': req['AbstractURL'],
'search_provider': 'duckduckgo',
'author': None,
'date': None,
'source': req['AbstractSource'],
'doc': req['Abstract'],
"sentences": pr.sentences,
"variants": list(pr.variants)
})
if req.get('Definition'):
pr = PageRequest(req['DefinitionURL'], term, run=False)
pr.extract_sentences(req['Definition'])
result.append({
'title': req['Heading'],
'url': req['DefinitionURL'],
'source': req['DefinitionSource'],
'search_provider': 'duckduckgo',
'author': None,
'date': None,
'doc': req['Definition'],
"sentences": pr.sentences,
"variants": list(pr.variants)
})
log.info("Searching DuckDuckGo for '{}' returned {} results".format(term, len(result)))
return result
def test_page_request():
from serapis.extract import PageRequest
p = PageRequest(
'http://nytimes.com/2015/10/04/technology/scouring-the-web-to-make-new-words-lookupable.html',
"lookupable")
assert p.response
