def scrap_genre(self, channel, genre, url):
html = urllib.urlopen(url).read()
soup = BeautifulSoup(html, from_encoding='utf-8')
for show in soup.article('li'):
try:
url = urllib.basejoin(self.BASE_URL, show.a.get('href'))
print url
self.get_serie(channel, genre, url)
except Exception,e:
print "ERROR: %s" % str(e)
def scrape_news():
try:
news_url = "https://mars.nasa.gov/news/"
browser.visit(news_url)
html = browser.html
soup = BeautifulSoup(html, "html.parser")
#find text for title and teaser
article = soup.find_all("div", class_="list_text")
#print(len(article))
#print(article)
#find title and teaser
news_title = soup.article("div", class_="content_title")
news_p = soup.article("div", class_="article_teaser_body")
# Dictionary entry from MARS NEWS
mars_info['news_title'] = news_title
mars_info['news_paragraph'] = news_p
return mars_info
finally:
browser.quit()
def recuperer_infos_conseiller(url):
res = requests.get(f"{AFE_DOMAIN}{url}")
res.raise_for_status()
soup = BeautifulSoup(res.content, "lxml")
header = soup.article.header
titre, *_ = header.h1.contents[0].strip().split(" ", 1)
# on écarte le premier qui est traité à part (date et lieu de naissance)
fields = [
li for li in soup.article("li") if len(li.contents) == 2 and li.strong
][1:]
infos = {
correspondance[li.contents[0].text.strip(": ")]: li.contents[1]
for li in fields
}
bloc_naissance = header.ul("li")[0]
if len(bloc_naissance.contents) > 1:
infos["date_naissance"], *lieu = bloc_naissance.contents[1].split(
" à ")
if lieu:
infos["lieu_naissance"] = lieu[0]
if "email_principal" in infos:
infos["email_principal"] = infos["email_principal"].text.replace(
" chez ", "@")
if "email_autre" in infos:
infos["email_autre"] = infos["email_autre"].text.replace(" chez ", "@")
return {
"titre": titre,
**{
k: v.strip() if isinstance(v, str) else v.text.strip()
for k, v in infos.items()
},
}
def download_chapter(
url=None, filename=None,
main_title=None, title_strip=None, title_re=None, scraper=None
):
# Download a given chapter
from lxml import html, etree
from bs4 import BeautifulSoup, UnicodeDammit
import requests, re
#~ import pdb
#~ pdb.set_trace()
if (url is None or filename is None):
return False
if (scraper is not None):
page = scraper.get(url)
else:
page = requests.get(url)
if page.status_code == 404:
return False
page.encoding="utf-8"
#tree = BeautifulSoup(UnicodeDammit.detwingle(page.text), "html5lib")
tree = BeautifulSoup(page.text, "html5lib")
# Trim down to just the article content
title = ""
btree = tree.article
if not btree:
return False
# remove Next/Previous
for i in btree("a", string=re.compile("(Next|Prev(ious)?|Index)( ?Chapter)?")):
i.decompose()
#~ for i in btree("a", string="Previous Chapter"):
#~ i.decompose()
for i in btree("hr"):
i.unwrap()
for i in btree("span", style=re.compile("float: ?right")):
i.decompose()
for i in btree("span", style=re.compile("(font-family|color|text-align)")):
i.unwrap()
for i in btree("div", class_=re.compile("wpcnt|sharedaddy")):
i.decompose()
for i in btree("p"):
if 'style' in i:
del i['style']
if "Previous Chapter" in btree.p.text:
btree.p.decompose()
# TODO: remove all empty tags
# Want to rewrite chapter links
# pull images from glossary page and embed?
if (main_title is not None and main_title != ""):
title = main_title
else:
doc_title = btree.find("h1", class_="entry-title")
temp_string = ""
if doc_title and isinstance(doc_title, list):
doc_title = doc_title[0]
if doc_title:
for x in doc_title.stripped_strings:
temp_string = "{} {}".format(temp_string, x)
doc_title = temp_string
if ("glossary" in doc_title.lower() or
"index" in doc_title.lower()
):
title = doc_title
else:
t_div = btree("div", class_="entry-content")[0]
if (t_div.u):
t_div.u.unwrap()
if t_div.span:
t_div.span.unwrap()
if t_div.br:
t_div.br.decompose()
if (t_div.b):
st = tree.new_tag("strong")
temp_string = ""
for x in t_div.b.stripped_strings:
temp_string = temp_string + " {}".format(x)
st.string = temp_string
tree.article.div.b.replace_with(st)
titles = t_div.strong
#~ print("titles:{}".format(titles))
if (titles):
if titles.br:
titles.br.decompose()
if not titles.string:
title = ""
for x in titles.stripped_strings:
title = title + " {}".format(x)
else:
title = titles.string
#~ print("strtitle:{}".format(title))
if (re.match('^\s+$', title)):
title = ""
if (title == "" and t_div.h3):
title = t_div.h3.string
#~ print("h3title:{}".format(title))
if (title == ""):
title = doc_title
#~ print("dtitle:{}".format(title))
if not title:
title = tree.title.string
#~ print("title:{}".format(title))
title = title.strip()
title = re.sub(re.compile('\n| |\r|\t| '), ' ', title)
title = title.replace(' ', ' ')
if (title_strip is not None):
#~ print("strip:'{}' title:'{}'".format(title_strip, title))
#title = title.replace(title_strip, '').strip()
title = re.sub(title_strip, '', title).strip()
#~ print("stripped:'{}'".format(title))
if (title_re is not None):
title_re = title_re.strip()
title_re = title_re.rstrip('"')
title_re = title_re.lstrip('"')
title_re = title_re.rstrip("'")
title_re = title_re.lstrip("'")
t_regex = title_re.split('||')
title = re.sub(t_regex[0], t_regex[1], title)
#~ print("re:'{}' title:'{}'".format(title_re,title))
nt = tree.new_tag("section")
nt["epub:type"] = "chapter"
tmp = tree.article("div", class_="entry-meta")
if tmp:
tmp[0].decompose()
tree.article.div.wrap(nt)
tree.article.div.unwrap()
nt = tree.new_tag("body")
tree.article.section.wrap(nt)
nt = tree.new_tag("html")
tree.article.section.wrap(nt)
nt = tree.new_tag("head")
tree.article.section.insert_before(nt)
nt = tree.new_tag("title")
nt.string = title.strip()
tree.article.head.append(nt)
nt = tree.new_tag("link", rel="stylesheet", href="style/main.css")
nt["type"] = "text/css"
tree.article.head.append(nt)
#~ tree = html.fromstring(big_html)
#~ html.html_to_xhtml(tree)
tree = BeautifulSoup(tree.article.html.prettify(formatter="html"), "html5lib")
with open(filename, 'w') as f:
f.write(tree.prettify(formatter="html"))
