def scrape_data(self):
# Create boilerplate recognizer
skindex = Scraper(self.crawl_config)
skindex.load_local_pages()
skindex.add_template_elements()
# Process all by removing boilerplate and extracting information
return skindex.process_all(exclude_data=['cleaned', 'author'])
def scrape_data(self):
# Create boilerplate recognizer
skindex = Scraper(self.crawl_config)
skindex.load_local_pages()
skindex.add_template_elements()
# Process all by removing boilerplate and extracting information
return skindex.process_all(exclude_data=['cleaned', 'author'])
class NewsCrawler(Crawler):
def __init__(self, config, cache=None):
super(NewsCrawler, self).__init__(config, cache)
self.scraper = Scraper(config)
self.template_complete = False
self.data = {}
self.templates_done = 0
async def save_response(self, html_code, url, headers, crawl_date):
# f*****g mess
try:
# just let the indexer save the files as normal and also create a Template
url = url
tree = makeTree(html_code, self.scraper.domain)
if self.templates_done < self.scraper.config["max_templates"]:
self.templates_done += 1
self.scraper.domain_nodes_dict.add_template_elements(tree)
self.scraper.url_to_headers_mapping[url] = headers
self.data[url] = self.scraper.process(url, tree, False,
["cleaned"])
self.data[url]["crawl_date"] = crawl_date
scrape_date = time.strftime("%Y-%m-%dT%H:%M:%S",
time.localtime(time.time()))
self.data[url]["scrape_date"] = scrape_date
except Exception as e:
LOGGER.error(
"CRITICAL ERROR IN SCRAPER for url %r: %r, stack %r",
url,
str(e),
traceback.format_exc(),
)
return
def save_data(self, data):
raise NotImplementedError("save_data has to be implemented")
def save_bulk_data(self, data):
raise NotImplementedError("save_bulk_data has to be implemented")
def finish_leftovers(self):
LOGGER.info("finish leftovers")
if self.data:
image_set = get_image_set(self.data)
LOGGER.info("saving number of documents: %r", len(self.data))
LOGGER.info("found num unique images: %r", len(image_set))
LOGGER.info("saving status code: %r",
self.save_bulk_data(self.data))
return dict(self.scraper.domain_nodes_dict)
class NewsCrawler(Crawler):
def __init__(self, config, cache=None):
super(NewsCrawler, self).__init__(config, cache)
self.scraper = Scraper(config)
self.template_complete = False
self.data = {}
self.templates_done = 0
@asyncio.coroutine
def save_response(self, html_code, url, headers, crawl_date):
# f*****g mess
try:
# just let the indexer save the files as normal and also create a Template
url = url
tree = makeTree(html_code, self.scraper.domain)
if self.templates_done < self.scraper.config['max_templates']:
self.templates_done += 1
self.scraper.domain_nodes_dict.add_template_elements(tree)
self.scraper.url_to_headers_mapping[url] = headers
self.data[url] = self.scraper.process(url, tree, False, ['cleaned'])
self.data[url]['crawl_date'] = crawl_date
scrape_date = time.strftime('%Y-%m-%dT%H:%M:%S', time.localtime(time.time()))
self.data[url]['scrape_date'] = scrape_date
except Exception as e:
LOGGER.error("CRITICAL ERROR IN SCRAPER for url %r: %r, stack %r",
url, str(e), traceback.format_exc())
return
def save_data(self, data):
raise NotImplementedError('save_data has to be implemented')
def save_bulk_data(self, data):
raise NotImplementedError('save_bulk_data has to be implemented')
def finish_leftovers(self):
LOGGER.info('finish leftovers')
if self.data:
image_set = get_image_set(self.data)
LOGGER.info('saving number of documents: %r', len(self.data))
LOGGER.info('found num unique images: %r', len(image_set))
LOGGER.info('saving status code: %r', self.save_bulk_data(self.data))
return dict(self.scraper.domain_nodes_dict)
'index_filter_regexps': [
],
'index_required_regexps': [
'2015', '2014'
],
'max_saved_responses': 100,
'max_workers': 10,
})
crawl.start(CRAWL_CONFIG)
# Indexing
SCRAPE_CONFIG = CRAWL_CONFIG.copy()
SCRAPE_CONFIG.update({
'template_proportion': 0.09,
'max_templates': 1000
})
skindex = Scraper(SCRAPE_CONFIG)
skindex.load_local_pages()
skindex.add_template_elements()
res = skindex.process_all(remove_visuals=True)
# Crawling
CRAWL_CONFIG = DEFAULT_CRAWL_CONFIG
CRAWL_CONFIG.update({
'seed_urls': ['http://www.techcrunch.com/'],
'collections_path': '/Users/pascal/egoroot/sky_collections',
'collection_name': 'techie',
# Optional
'crawl_filter_regexps': [],
'crawl_required_regexps': ['2015', '2014'],
'index_filter_regexps': [],
'index_required_regexps': ['2015', '2014'],
'max_saved_responses': 100,
'max_workers': 10,
})
crawl.start(CRAWL_CONFIG)
# Indexing
SCRAPE_CONFIG = CRAWL_CONFIG.copy()
SCRAPE_CONFIG.update({'template_proportion': 0.09, 'max_templates': 1000})
skindex = Scraper(SCRAPE_CONFIG)
skindex.load_local_pages()
skindex.add_template_elements()
res = skindex.process_all(remove_visuals=True)
def post(self):
CRAWL_CONFIG = DEFAULT_CRAWL_CONFIG
CRAWL_CONFIG.update({
'collections_path': os.path.join(os.path.expanduser('~'), 'sky_collections/'),
# 'max_workers': 10,
})
args = self.request.arguments
print(args)
for arg in args:
value = args[arg][0].decode('utf8')
if value and arg != 'url' and arg != 'checkboxcache':
print('pre', arg, CRAWL_CONFIG[arg])
if isinstance(CRAWL_CONFIG[arg], list):
CRAWL_CONFIG[arg] = [int(value)] if is_numeric(value) else [value]
else:
CRAWL_CONFIG[arg] = int(value) if is_numeric(value) else value
print('post', arg, CRAWL_CONFIG[arg])
url = self.get_argument('url', '')
use_cache = self.get_argument('checkboxcache', '')
domain = extractDomain(url)
CRAWL_CONFIG['seed_urls'] = [url]
CRAWL_CONFIG['collection_name'] = domain[7:]
if use_cache != 'on':
col_path = os.path.join(CRAWL_CONFIG['collections_path'],
CRAWL_CONFIG['collection_name'])
print(col_path)
if os.path.exists(col_path):
shutil.rmtree(col_path)
crawl.start(CRAWL_CONFIG)
SCRAPE_CONFIG = CRAWL_CONFIG.copy()
SCRAPE_CONFIG.update({
'template_proportion': 0.4,
'max_templates': 100
})
skindex = Scraper(SCRAPE_CONFIG)
skindex.load_local_pages()
skindex.add_template_elements()
res = skindex.process_all(remove_visuals=True,
maxn=CRAWL_CONFIG['max_saved_responses'])
items = []
for num, url in enumerate(res):
if num == CRAWL_CONFIG['max_saved_responses']:
break
dc = res[url]
dc['url'] = url
dc['source_name'] = domain
dc['images'] = [x for x in reversed(dc['images'][:5])]
# dc['blobs'] = [TextBlob(x) for x in dc['body'] if dc['body']]
items.append(dc)
# this is quite out of place like this
print('num unique images', len(get_image_set({x['url']: x for x in items})))
if items and 'money' in items[0]:
items = sorted(items, key=lambda x: len(x['money']), reverse=True)
self.render('page_template.html', items=items, cached=False)
def post(self):
CRAWL_CONFIG = DEFAULT_CRAWL_CONFIG.copy()
CRAWL_CONFIG.update(
{
"collections_path": os.path.join(os.path.expanduser("~"), "sky_view_collections/"),
# 'max_workers': 10,
}
)
args = self.request.arguments
print(args)
for arg in args:
value = args[arg][0].decode("utf8")
if value and arg != "url" and arg != "checkboxcache":
print("pre", arg, CRAWL_CONFIG[arg])
if isinstance(CRAWL_CONFIG[arg], list):
CRAWL_CONFIG[arg] = [int(value)] if is_numeric(value) else value.split(", ")
else:
CRAWL_CONFIG[arg] = int(value) if is_numeric(value) else value.split(", ")[0]
print("post", arg, CRAWL_CONFIG[arg])
url = self.get_argument("url", "")
use_cache = self.get_argument("checkboxcache", "")
domain = extractDomain(url)
CRAWL_CONFIG["seed_urls"] = [url]
CRAWL_CONFIG["collection_name"] = domain[7:]
if use_cache != "on":
col_path = os.path.join(CRAWL_CONFIG["collections_path"], CRAWL_CONFIG["collection_name"])
print(col_path)
if os.path.exists(col_path):
shutil.rmtree(col_path)
crawl.start(CRAWL_CONFIG)
SCRAPE_CONFIG = CRAWL_CONFIG.copy()
SCRAPE_CONFIG.update({"template_proportion": 0.4, "max_templates": 100})
skindex = Scraper(SCRAPE_CONFIG)
skindex.load_local_pages()
skindex.add_template_elements()
res = skindex.process_all(remove_visuals=True, maxn=CRAWL_CONFIG["max_saved_responses"])
items = []
for num, url in enumerate(res):
if num == CRAWL_CONFIG["max_saved_responses"]:
break
dc = res[url]
dc["url"] = url
dc["source_name"] = domain
dc["images"] = [x for x in reversed(dc["images"][:5])]
# dc['blobs'] = [TextBlob(x) for x in dc['body'] if dc['body']]
items.append(dc)
# this is quite out of place like this
print("num unique images", len(get_image_set({x["url"]: x for x in items})))
if items and "money" in items[0]:
items = sorted(items, key=lambda x: len(x["money"]), reverse=True)
self.render("page_template.html", items=items, cached=False)
def post(self):
CRAWL_CONFIG = DEFAULT_CRAWL_CONFIG.copy()
CRAWL_CONFIG.update({
'collections_path': os.path.join(os.path.expanduser('~'), 'sky_view_collections/'),
# 'max_workers': 10,
})
args = self.request.arguments
print(args)
for arg in args:
value = args[arg][0].decode('utf8')
if value and arg != 'url' and arg != 'checkboxcache':
print('pre', arg, CRAWL_CONFIG[arg])
if isinstance(CRAWL_CONFIG[arg], list):
CRAWL_CONFIG[arg] = [int(value)] if is_numeric(value) else value.split(', ')
else:
CRAWL_CONFIG[arg] = int(value) if is_numeric(value) else value.split(', ')[0]
print('post', arg, CRAWL_CONFIG[arg])
url = self.get_argument('url', '')
use_cache = self.get_argument('checkboxcache', '')
domain = extractDomain(url)
CRAWL_CONFIG['seed_urls'] = [url]
if domain.startswith("http"):
CRAWL_CONFIG['collection_name'] = domain.split("/")[2]
else:
CRAWL_CONFIG['collection_name'] = domain.split("/")[0]
if use_cache != 'on':
col_path = os.path.join(CRAWL_CONFIG['collections_path'],
CRAWL_CONFIG['collection_name'])
print(col_path)
if os.path.exists(col_path):
shutil.rmtree(col_path)
crawl.start(CRAWL_CONFIG)
SCRAPE_CONFIG = CRAWL_CONFIG.copy()
SCRAPE_CONFIG.update({
'template_proportion': 0.4,
'max_templates': 100,
})
skindex = Scraper(SCRAPE_CONFIG)
skindex.load_local_pages()
skindex.add_template_elements()
res = skindex.process_all(remove_visuals=True,
maxn=CRAWL_CONFIG['max_saved_responses'])
items = []
for num, url in enumerate(res):
if num == CRAWL_CONFIG['max_saved_responses']:
break
dc = res[url]
dc['url'] = url
dc['source_name'] = domain
dc['images'] = [x for x in reversed(dc['images'][:5])]
# dc['blobs'] = [TextBlob(x) for x in dc['body'] if dc['body']]
items.append(dc)
# this is quite out of place like this
print('num unique images', len(get_image_set({x['url']: x for x in items})))
if items and 'money' in items[0]:
items = sorted(items, key=lambda x: len(x['money']), reverse=True)
self.render('page_template.html', items=items, cached=False)
def __init__(self, config, cache=None):
super(NewsCrawler, self).__init__(config, cache)
self.scraper = Scraper(config)
self.template_complete = False
self.data = {}
self.templates_done = 0
def __init__(self, config, cache=None):
super(NewsCrawler, self).__init__(config, cache)
self.scraper = Scraper(config)
self.template_complete = False
self.data = {}
self.templates_done = 0
