def save_page(url, title, intro):
try:
w = '/wiki/'
name = url[url.find(w) + len(w):]
file = open(name + '.txt', 'a', encoding='utf-8')
file.write(url + "\n")
file.write(clean(h2t(title)) + ": " + "\n")
file.write(clean(h2t(intro)))
file.close()
except Exception as e:
print(str(e))
return "Error"
return name
def get_card(soup, final, kwargs):
"""Getting cards if present, here started the pain"""
# common card
if card := soup.find("div", class_="g mnr-c g-blk"):
if desc := card.find("span", class_="hgKElc"):
final.append(s(None, "Google Info Card:", h2t(str(desc))))
return
def parse_reference(self, text, full_chap, title, emb_color):
# Remove cross references
for sup in text.find_all("sup", {"class": "crossreference"}):
sup.decompose()
# Change headers to markdown
for h3 in text.find_all("h3"):
h3.name = "b"
for h4 in text.find_all("h4"):
h4.name = "b"
text = h2t(str(text))
pages = []
raw = list(pagify(text, page_length=4000))
size = len(raw)
for i, page in enumerate(raw, 1):
emb = discord.Embed(title=title,
description=page,
colour=emb_color)
emb.url = full_chap
emb.set_footer(
text=f"Powered by Biblegateway.com | Page {i}/{size}")
pages.append(emb)
return pages
def parseContent(self):
if self.html is not None:
data = h2t(self.html.decode('UTF-8'))
self.logger.info("Content Parsed")
return data
else:
return None
def parser_text(self, text, soup=None, cards: bool = True):
"""My bad logic for scraping"""
if not soup:
soup = BeautifulSoup(text, features="html.parser")
final = []
kwargs = {"stats": h2t(str(soup.find("div", id="result-stats")))}
if cards:
get_card(soup, final, kwargs)
for res in soup.findAll("div", class_="g"):
if name := res.find("div", class_="yuRUbf"):
url = name.a["href"]
if title := name.find("h3", "LC20lb DKV0Md"):
title = title.text
else:
title = url
def html2text(s):
s = re.compile('</*en-media[^>]*?>').sub('', s)
return h2t(s)
return query
def get_card(soup, final, kwargs):
"""Getting cards if present, here started the pain"""
# common card
if card := soup.find("div", class_="g mnr-c g-blk"):
if desc := card.find("span", class_="hgKElc"):
final.append(s(None, "Google Info Card:", h2t(str(desc))))
return
# another webpull card: what is the language JetBrains made? TODO fix this, depends on too many classes as of now
if card := soup.find("div", class_="kp-blk c2xzTb"):
if head := card.find("div", class_="Z0LcW XcVN5d AZCkJd"):
if desc := card.find("div", class_="iKJnec"):
final.append(s(None, f"Answer: {head.text}", h2t(str(desc))))
return
# calculator card
if card := soup.find("div", class_="tyYmIf"):
if question := card.find("span", class_="vUGUtc"):
if answer := card.find("span", class_="qv3Wpe"):
tmp = h2t(str(question)).strip("\n")
final.append(
s(None, "Google Calculator:",
f"**{tmp}** {h2t(str(answer))}"))
return
# sidepage card
if card := soup.find("div", class_="liYKde g VjDLd"):
if thumbnail := card.find("g-img", attrs={"data-lpage": True}):
#! python3 from html2text import html2text as h2t import pyperclip pyperclip.copy(h2t(pyperclip.paste()))
def html2text(value):
"""Convert HTML text to plaintext"""
return h2t(value)
def html2text(s):
s = re.compile('</*en-media[^>]*?>').sub('', s)
return h2t(s)
def html2text(text):
return h2t(text).replace('#','').replace('**','').replace('__','')
def raw_text(self, html_field): # makes practically raw text from html
#delete additional signs after h2t function
return sub('[\t\n\r*#]+', ' ', h2t(html_field))
if cards:
get_card(soup, final, kwargs)
for res in soup.findAll("div", class_="g"):
if name := res.find("div", class_="yuRUbf"):
url = name.a["href"]
if title := name.find("h3", "LC20lb DKV0Md"):
title = title.text
else:
title = url
else:
url = None
title = None
if desc := res.find("div", class_="IsZvec"):
if remove := desc.find("span", class_="f"):
remove.decompose()
if final_desc := desc.find_all("span"):
desc = h2t(str(final_desc[-1]))[:500]
else:
desc = "Nothing found"
else:
desc = "Not found"
if title:
final.append(s(url, title, desc.replace("\n", " ")))
return final, kwargs
def parser_image(self, html):
# first 2 are google static logo images
return self.link_regex.findall(html)[2:], {}
