def __init__(self, page_url):
response = None
self.url = None
try:
response = self.get_response_object(page_url)
except Exception:
try:
# time.sleep(1)
response = self.get_response_object(page_url)
except Exception:
try:
# time.sleep(5)
response = self.get_response_object(page_url)
except Exception:
print "\nTried 3 times. Cannot access url: %s. \nHence, cannot make HTML_page_Obj\n" % page_url
if response != None:
self.url = page_url
try:
self.short_url = response.headers.getparam('Link')
except Exception:
self.short_url = ""
self.charset = response.headers.getparam('charset')
self.headers = {'charset': self.charset}
for i in response.headers:
self.headers[i] = response.headers[i].split(';')[0]
self.html = response.read() #the actual html
self.html = text_manip.ensure_UTF8(self.html)
self.html_soup = None
self.link_dict = None
self.all_style = None
def __init__(self, wikipedia_image_url=None, direct_url=None):
self.direct_image_link = None
self.saved_path = None
self.saved_filename = None
if direct_url is not None:
self.direct_image_link = direct_url
elif wikipedia_image_url is not None:
img_page = HTML_page_Obj(wikipedia_image_url)
img_soup = img_page.make_soup()
# print img_soup.find('div', _class="mw-filepage-resolutioninfo")
self.main_img_link = img_soup.find(
'div', class_="fullMedia").find(
'a', class_="internal").get('href')
if self.main_img_link[:9] == "//upload.":
self.main_img_link = "https:" + self.main_img_link
img_links_soup = img_soup.find(
"div", class_="mw-filepage-resolutioninfo")
self.img_px_to_links_map = {
} #maps the product of height and width to tuples of pixel dimensions and their links
for img_link in img_links_soup.find_all('a'):
actual_link = img_link.get('href')
if actual_link[:9] == "//upload.":
actual_link = "https:" + actual_link
img_link_text = text_manip.ensure_UTF8(
re.sub(",", "", img_link.get_text()))
# print text_manip.ensure_ASCII("\n\nimg_link_text: %s"%img_link_text)
img_dimensions = re.findall('\d+', img_link_text)
width = int(img_dimensions[0])
height = int(img_dimensions[1])
if len(img_dimensions) == 2:
self.img_px_to_links_map[width * height] = (
actual_link, width, height) #does mapping
else:
print "Cannot use image %s, which is at %s with dimensions %s" % (
wikipedia_image_url, img_link.get('href'),
img_dimensions)
def html_soup_and_prettify(self):
if self.html_soup == None:
self.html_soup = self.make_soup()
self.html = text_manip.ensure_UTF8(self.html_soup.prettify())
return self.html_soup
def __init__(self, wikipedia_url, just_heading=False):
# inherit items from the HTML_page_Obj:
self._page = HTML_page_Obj(wikipedia_url)
self.url = self._page.url
self.wiki_domain = text_manip.extract_website(
self.url) #gets en.wikipedia.org, etc. in different languages
# make the soup of the HTML and get the HTML, prettified:
soup = self._page.html_soup_and_prettify()
self.html = self._page.html #prettified, in UTF-8 format
# Get the heading of the Wikipedia page:
self.heading = text_manip.ensure_UTF8(
BeautifulSoup('''%s''' % soup.find(
"h1", id="firstHeading",
class_="firstHeading")).get_text().strip())
if just_heading == False: # prevents a lot of unecessary processing if True
# Make a soup of the article html, set the article html:
article_soup = soup.find(
"div", id="mw-content-text") #might come in handy later
self.article_html = text_manip.ensure_UTF8(article_soup.prettify())
temp_art_html = self.article_html
temp_art_html = text_manip.remove_HTML_perfect(
html=temp_art_html, tag="div", class_list=["reflist"])
temp_art_html = text_manip.remove_after(
regex='<span.*?id="References"', text=temp_art_html)
temp_art_html = text_manip.remove_after(
regex='<span.*?id="See_also"', text=temp_art_html)
temp_art_html = text_manip.remove_after(
regex='<span.*?id="Further_reading"', text=temp_art_html)
temp_art_html = text_manip.remove_after(
regex='<span.*?id="External_links"', text=temp_art_html)
article_soup = BeautifulSoup(temp_art_html)
# Get the links from the article HTML:
self.article_links = [] # real links, from article
self.img_links = [] # real direct links, from article
self.direct_img_to_img_map = {}
self.media_links = [] # real links, from article
not_allowed_in_url_from_article = [
"/wiki/Special:", "/wiki/Template:", "/wiki/Portal:",
"/wiki/Wikipedia:", "/wiki/Help:", "redlink=1"
]
for link_tag in article_soup.find_all('a'):
link = link_tag.get('href')
if link is not None and link[:6] == "/wiki/": #only extracts links to wikipedia
allowed_flag = True
for allowed_test in not_allowed_in_url_from_article:
if allowed_test in link:
allowed_flag = False
break
if allowed_flag:
link = text_manip.ensure_UTF8(link)
mapped_link = "https://" + self.wiki_domain + link
if link[:11] != "/wiki/File:":
self.article_links.append(mapped_link)
else:
if link[-4:] == '.ogg': #video or audio file
self.media_links.append(mapped_link)
else: # image file
img_tag = link_tag.img
# print "\n\n"
# print "link_tag :",link_tag
# print "\timg_tag :",img_tag
if img_tag is not None:
direct_img_link = img_tag.get('src')
# print "\t\tdirect_img_link :",direct_img_link
self.img_links.append(direct_img_link)
self.direct_img_to_img_map[
direct_img_link] = mapped_link
self.article_links = list(set(self.article_links))
self.img_links = list(set(self.img_links))
self.media_links = list(set(self.media_links))
# replace the /wiki/ links for actual links. Presetve the old html, and use replaced_html from now on
self.replaced_html = self.html
for link_tag in soup.find_all('a'):
link = link_tag.get('href')
if link is not None:
if link[:6] == "/wiki/":
mapped_link = "https://" + self.wiki_domain + link
self.replaced_html = self.replaced_html.replace(
'href="%s' % link, 'href="%s' % mapped_link)
