def class_tr_to_str(tr: bs4.element.Tag) -> str:
for a in tr.find_all("a"):
a["href"] = BASE_API_URL + a["href"]
data = (tr.find("td", class_="header"), tr.find("td",
class_="description"))
nameSpan = data[0].find("span", class_="element-name")
if data[0].find("span", class_="attribute-type") is not None:
accessType = "param"
type_ = data[0].find("span", class_="param-type").text.strip()
else:
accessType = "func"
if accessType == "param":
attributeMode = data[0].find("span", class_="attribute-mode").text
header = f"`{nameSpan.text} :: {type_}` {attributeMode}"
else:
header = f"`{nameSpan.text}`"
contents = [item for item in data[1].contents if item != " "]
if len(contents) > 0:
if len(contents) > 1 and "\n" not in contents[0]:
description = tomd.convert(
f"<p>{''.join([str(item) for item in contents[:-1]])}</p>"
).strip()
else:
description = contents[0].strip()
return f"{header} - {description}"
else:
return header
def getNode(contentSoup):
node = ArticleNode()
node.depth = contentSoup['depth']
node.id = contentSoup['id']
node.name = contentSoup.find('a', class_='x-wiki-index-item').text.replace(
'/', ' ')
node.url = 'https://www.liaoxuefeng.com' + contentSoup.find(
'a', class_='x-wiki-index-item')['href']
print(node.toString())
content = getUrlContent(node.url)
if content != None:
soup = BeautifulSoup(content, 'lxml')
node.articleHTML = str(
soup.find('div', class_="x-wiki-content x-main-content"))
node.articleMD = tomd.convert(node.articleHTML)
with open('output/' + node.name + '.md', 'w', encoding='utf-8') as fs:
fs.write(node.articleMD)
fs.flush()
for item in contentSoup.find_all('div', depth=str(int(node.depth) + 1)):
node.children.append(getNode(item))
return node
def _get_new_data(self,page_url,soup):
'''
抽取有效数据
:param page_url: 下载页面的 URL
:param soup:
:return: 返回有效数据
'''
data = {}
dbStore = DataOutput()
Htmldownloader = HtmlDownloader()
data['article_url'] = page_url
html = Htmldownloader.download(page_url)
ind_soup = BeautifulSoup(html, 'html.parser')
title = ind_soup.find_all("h1",class_="csdn_top")
data['title']=title[0].string
updated = ind_soup.find_all("span",class_="time")
data['update_date'] = updated[0].string
pageview = ind_soup.find_all("span",class_="txt")
data['pageviewcnt']= pageview[0].string
articlecontent = ind_soup.find_all("div",class_="article_content")
md = tomd.convert(articlecontent[0].prettify())
# the page_content is too long to be saved in mysql blob data type , make that be saved on local folder as a file
data['page_content']=pickle.dumps(md.encode("utf-8"))
dbStore.store_data(data)
print(data)
return data
def get_url_content(url):
str_html = requests.get(url)
html = BeautifulSoup(str_html.text, 'html.parser')
title = html.find('h1').text
date = html.find('span', {'class': 'time'}).text
html.find('div', {'class': 'toc'}).decompose()
html.find('svg').decompose()
html.find('br').decompose()
clean_tags(html)
replace_h_tags(html)
get_image(title, html)
content = ''.join(map(str, html.find('div', {'id': 'content_views'}).contents))
print(content)
exit()
md_content = tomd.convert(content)
path = './content'
file = title + '.md'
file_path = path + '/' + file
if not os.path.exists(path):
os.makedirs(path)
f = open(file_path, 'w', encoding="utf-8")
f.write(str(md_content))
f.close()
def _generate_ics(self):
data = self.data
if data != None:
c = Calendar()
days = list(data.keys())
days.sort()
for xday in days:
day_data = data[xday]
title = day_data[0].text_content()
week, day = title.split(", ")
week = w2n.word_to_num(week.lstrip("Week "))
day = w2n.word_to_num(day.lstrip("Day "))
offset = (week - 1) * 7 + (day)
event_day = self.start_date + timedelta(days=offset)
event_day = event_day.replace(hour=0,
minute=0,
second=0,
microsecond=0)
description = "".join(
[str(html.tostring(el)) for el in day_data])
description = tomd.convert(description)
e = Event(name="Magoosh {}".format(title),
begin=event_day,
end=event_day,
description=description)
e.make_all_day()
c.events.add(e)
with open(self.out_file, 'w') as f:
f.writelines(c)
print("File written to {}".format(self.out_file))
def parse_content(self):
""" parse data from html to markdown """
self.markdown = []
try:
for item in self.data:
if item:
for i in item:
t = str(i)
test = re.search("^!", t)
if test:
self.markdown.append(t)
md = tomd.convert(t)
self.markdown.append(md)
filtered = list(
filter(lambda x: not re.match(r'^\s*$', x), self.markdown))
filtered = [x.replace('\n', '') for x in filtered]
filtered = [x.replace('\t', '') for x in filtered]
self.markdown = filtered
except RuntimeError as err:
print(Fore.RED, "[!] Recursion {0}".format(err))
pass
self.markdown.append('\n' + self.url)
def anno_edit(request, pk):
'''公告编辑'''
anno = Announcement.objects.get(pk=pk)
# 当为post请求时，修改数据
if request.method == "POST":
form = AnnoForm(request.POST, instance=anno)
if form.is_valid():
form.save()
registerinfo = {
'title': '修改成功',
'subtitle': '数据更新成功',
'status': 'success', }
context = {
'registerinfo': registerinfo,
'anno': Announcement.objects.all(),
}
return render(request, 'backend/annolist.html', context=context)
else:
registerinfo = {
'title': '错误',
'subtitle': '数据填写错误',
'status': 'error', }
context = {
'form': form,
'registerinfo': registerinfo,
'anno': anno,
}
return render(request, 'backend/annoedit.html', context=context)
# 当请求不是post时，渲染form
else:
anno.body = tomd.convert(anno.body)
context = {
'anno': anno,
}
return render(request, 'backend/annoedit.html', context=context)
def anno_edit(request, pk):
'''公告编辑'''
anno = Announcement.objects.get(pk=pk)
# 当为post请求时，修改数据
if request.method == "POST":
form = AnnoForm(request.POST, instance=anno)
if form.is_valid():
form.save()
registerinfo = {
'title': '修改成功',
'subtitle': '数据更新成功',
'status': 'success', }
context = {
'registerinfo': registerinfo,
'anno': Announcement.objects.all(),
}
return render(request, 'backend/annolist.html', context=context)
else:
registerinfo = {
'title': '错误',
'subtitle': '数据填写错误',
'status': 'error', }
context = {
'form': form,
'registerinfo': registerinfo,
'anno': anno,
}
return render(request, 'backend/annoedit.html', context=context)
# 当请求不是post时，渲染form
else:
anno.body = tomd.convert(anno.body)
context = {
'anno': anno,
}
return render(request, 'backend/annoedit.html', context=context)
def get(self, request, pk):
anno = Announcement.objects.get(pk=pk)
anno.body = tomd.convert(anno.body)
context = {"anno": anno}
return render(request,
"my_admin/announcement_detail.html",
context=context)
def run():
print('Fetching content...', end='')
title, author, description, cta, img_url = get_meta_data()
article = tomd.convert(get_article(cta))
print('Done')
date = datetime.now().strftime('%Y%m%d')
commitMessage = f'{title} by {author}'
fileName = os.path.join('blinks', f'{date[:4]}', f'{date}-{title}-{author}.md')
print('Building output...', end='')
# Convert to markdown, add source
output = f'\n# {title}\n*{author}*\n\n>{description}\n\n{article}\n\nSource: [{commitMessage}](https://www.blinkist.com{cta})'
print('Done')
print(f'Committing file {fileName}...', end='')
g = Github(repoToken)
repo = g.get_repo(repoName)
try:
repo.create_file(fileName, commitMessage, output)
print('Done')
return 'OK'
except GithubException:
print('already exists')
return 'File Exists'
async def wiki_updated(name, feed_data, feed, feeds):
time_latest_entry = feed_data['time_latest_entry']
for i, entry in enumerate(feed.entries):
if get_formatted_time(entry) > time_latest_entry:
info_log('Found new wiki entry made on ' + entry.updated)
summary = ''
if re.search('<p.*?>.*?<\/p>', entry.summary):
summary = html.unescape(
tomd.convert(
re.search('<p.*?>.*?<\/p>', entry.summary).group()))
summary = re.sub(r'\((\/\S*)\)',
r'(https://wiki.factorio.com\1)', summary)
summary = re.sub('<bdi>|<\/bdi>', '', summary)
embed = discord.Embed(
title=f'{entry.author} changed {entry.title}',
color=14103594,
timestamp=datetime.datetime(*entry.updated_parsed[0:6]),
url=entry.link,
description=summary)
channel = client.get_channel(feed_data['channel'])
await channel.send(embed=embed)
else:
break
feeds[name]['time_latest_entry'] = get_formatted_time(feed.entries[0])
with open('feeds.json', 'w') as f:
json.dump(feeds, f)
def getMarkdown(entry, attr, selector, *args, **kwargs):
if args or kwargs:
if args and kwargs:
snippet = soup.find_all(
entry, attrs={attr, selector})[args[0]].find(
kwargs['opt_entry'],
attrs={kwargs['opt_attr'], kwargs['opt_selector']})
elif kwargs and len(args) == 0:
snippet = soup.find(entry, attrs={attr, selector}).find(
kwargs['opt_entry'],
attrs={kwargs['opt_attr'], kwargs['opt_selector']})
uls = snippet.find('ul')
if uls != None:
snippet.ul.append(soup.new_tag('p'))
snippet = tomd.convert(str(snippet)).replace('- ', '\n- ').strip()
return snippet
return tomd.convert(str(soup.find(entry, attrs={attr, selector})))
def parse_article_content(bsObj, directory, title):
# 1. Find html.
html = bsObj.find('div', {'class': 'article_content'})
md = tomd.convert(html.prettify())
# 2. Write to the file.
with open('%s/%s.md' % (directory, title), 'w', encoding='utf-8') as f:
f.write(md)
def __str__(self) -> str:
html = (
"<h1> {}</h1>\n<pre><code>Exercise ID {}</code></pre>\n<h2> Assignment </h2>{}\n"
.format(self.data.title, self.id, self.data.assignment) +
self.get_pre_exercise_code() + self.get_instructions() +
self.get_sample_code() + self.get_anwsers() + self.get_hints() +
self.get_solution())
return tomd.convert(html)
def get_class_description(soup: bs4.BeautifulSoup, class_: str) -> str:
a = soup.select(f"tr > td.header > a[href=\"{class_}.html\"]")
if len(a) == 1:
descriptionTag = a[0].parent.parent.find("td", class_="description")
descriptionRaw = "".join(
[str(content) for content in descriptionTag.contents])
return tomd.convert(f"<p>{descriptionRaw}</p>").strip()
return None
def download_statement(url):
req = requests.get(url)
filename = _extract_name(url)
dirname = _extract_dirname(url)
if req.status_code == 200:
statement = _extract_statement(req.content)
markdown = tomd.convert(str(statement))
_save_file(markdown, filename, dirname)
print(f'Successful download of {filename}!')
else:
print(f'We cannot get the request of {filename}!')
def extract_text(status: Status) -> str:
as_html = status["content"]
text: str = tomd.convert(as_html)
tags = status["tags"]
mentions = status["mentions"]
text = clean_tags(text, tags)
text = clean_mentions(text, mentions)
text = text.replace("<br />", "\n")
text = clean_links(text)
text = html.unescape(text)
return text
def get_event_description(div: bs4.element.Tag) -> str:
for a in div.find_all("a"):
a["href"] = BASE_API_URL + a["href"]
data = (div.select("div.element-content > p"),
div.find("div", class_="detail-content"))
paragraphs = []
for p in data[0]:
contents = p.contents
if not (len(contents) == 1 and len(contents[0].strip()) == 0):
paragraphs.append(html.unescape(tomd.convert(str(p))))
return "\n".join([p.strip().replace("\n", "") for p in paragraphs])
def getArticle(id):
url = "https://blog-console-api.csdn.net/v1/editor/getArticle?id={}".format(id)
print(url)
response = requests.get(url,headers=headers);
responseJson = json.loads(response.content)
if responseJson['data']['markdowncontent'] == '':
text = tomd.convert(responseJson['data']['content'])
# parser = Parser(responseJson['data']['content'])
responseJson['data']["markdowncontent"] = text
# print(''.join(parser.outputs))
write_txt(json.dumps(responseJson['data'],ensure_ascii=False),pwd+"/articles/"+id+".txt")
time.sleep(1)
def getList(entry, attr, selector, element):
list = soup.find(entry, attrs={attr: selector}).find_all(element)
list_md = ''
counter = 0
for idx, val in enumerate(list):
if element == 'p':
if val.get('class') == [u'counter']:
counter += 1
list_md += str(counter) + '. ' + val.text.strip() + '\n'
else:
list_md += '\n' + tomd.convert(str(val)).strip() + '\n'
else:
list_md += '- ' + val.text.strip() + '\n'
return list_md
def get_meta_data():
container = get_element_from_request('https://www.blinkist.com/nc/daily', 'div', 'daily-book__container')
title = container.find('h3', 'daily-book__headline').string.strip()
author = container.find('div', 'daily-book__author').string.strip()
# description = container.find('div', 'dailyV2__free-book__description').string.strip()
description_html = container.find('div', 'book-tabs__content-inner')
description = tomd.convert(str(description_html).strip())
# cta = container.find('div', 'dailyV2__free-book__cta').a['href']
cta = container.find('a', 'daily-book__cta').get('href')
img_url = container.find('img')['src']
return title, author, description, cta, img_url
def convert(self):
# 处理图片;
index = 1
for e in self.contentQ(self.rule.img_tag):
q = pq(e)
img_src = self.rule.find_img(q)
img_src_cur = self.rule.save_pic(self.url, self.title, index,
img_src)
if q[0].tag != "img":
q.replace_with(pq('<img src="' + img_src_cur + '"/>'))
else:
q.attr(src=img_src_cur)
index += 1
# 转换成markdown;
self.rule.save_md(self.title, tomd.convert(self.contentQ))
def extract_news(link):
page = requests.get(link)
soup = bs4.BeautifulSoup(page.text, 'html.parser')
article_html = soup.select(".article")[0]
article = str(article_html.encode('utf-8'))
md = ''.join(
tomd.convert(article).replace("\\n", "\n").replace(
"\\xc2", "").replace("\\xa0", "").replace("\n\n\n", "\n").replace(
"\n\n", "\n").replace("\\xe2\\x80\\x99", "'").split("|"))
md = md.replace("\\xe2", "").replace("\\x86",
"").replace("\\x92", "").replace(
"\\x80", "").replace("\\93", "")
md = extras.get_links(md)
return (md)
async def embed_fff(number):
"""
Returns a discord.Embed object derived from an fff number
"""
link = f"https://factorio.com/blog/post/fff-{number}"
response = await get_soup(link)
if response[0] == 200:
soup = response[1]
titleList = soup.find_all("h2")
em = discord.Embed(title=titleList[0].string.strip(),
url=link,
colour=discord.Colour.dark_green())
titleList = titleList[1:]
if len(titleList) == 0:
titleList = soup.find_all("h4")
if len(titleList) == 0:
titleList = soup.find_all("h3")
for title in titleList:
# Check for smaller font tag and append it to the title
result = fontEx.search(str(title))
if len([group for group in result.groups()
if group is not None]) == 1:
name = result.group(1)
else:
name = result.group(1) + result.group(3)
content = str(title.next_sibling.next_sibling)
if "<p>" not in content:
continue
if "<ol>" in content:
itemCount = 1
while "<li>" in content:
content = content.replace("<li>", f"{itemCount}. ", 1)
itemCount += 1
if "<ul>" in content:
content = content.replace("<li>", "- ")
for item in ["<ol>", "</ol>", "<ul>", "</ul>", "</li>", "<br/>"]:
content = content.replace(item, "")
# Escape Discord formatting characters
for item in ["*", "_"]:
content = content.replace(item, "\\" + item)
content = content.replace("\n\n", "\n")
em.add_field(name=name.replace("amp;", ""),
value=tomd.convert(content).strip())
else:
em = discord.Embed(title="Error",
description=f"Couldn't find FFF #{number}.",
colour=discord.Colour.red())
return em
def write_md(url, md_count):
html = urlopen(url)
bsObj = BeautifulSoup(html, 'html.parser')
title = bsObj.find('div', {'class': 'article-title-box'})
title_convert = (title.h1).get_text()
# print(title)
# title_convert = tomd.convert(title.prettify())
md = bsObj.find('div', {'class': 'article_content'})
# print(md.prettify())
convert = tomd.convert(md.prettify())
md_name = str(md_count) + '.md'
with open(md_name, 'w') as f:
f.write(title_convert)
f.write(convert)
def getContent(url):
page = []
(year,month) = re.search('http://scienceblogs.com/evolgen/(\d+)/(\d+)',url).groups()
print("processing %s %s"%(month,year),end='\r')
out = open('evolgen.md','wt')
html = requests.get(url = url)
bs = BeautifulSoup(html.content,'html.parser')
header = bs.find('h1',{'class':'title entry-title'})
page.append("#" + header.text)
abbr = bs.find('abbr',{'class':'published'})
page.append("Date: %s"%abbr.text)
content = bs.find('div',{'class':'content entry-content'})
for tag in content:
if tag == "\n":
continue
page.append(tomd.convert(str(tag)))
return page
def define_tr_to_str(tr: bs4.element.Tag) -> str:
for a in tr.find_all("a"):
a["href"] = BASE_API_URL + a["href"]
data = (tr.find("td",
class_="header").string, tr.find("td",
class_="description"))
contents = data[1].contents
if len(contents) > 0:
if len(contents) > 1 and "\n" not in contents[0]:
description = tomd.convert(
f"<p>{''.join([str(item) for item in contents[:-1]])}</p>"
).strip()
else:
description = contents[0].split('\n')[0].strip()
return f"`{data[0]}` - {description}"
else:
return f"`{data[0]}`"
def parse(package):
package_name = get_package(package)
url = 'https://github.com/%s/releases.atom' % package_name
feed = feedparser.parse(url)
entries = []
for item in feed['entries']:
owner, repo = package_name.split('/')
version = re.search('(?<=Repository/)[0-9]+/(.+)', item['id']).group(1)
authors = item['authors'][0]['name'] if 'authors' in item and item['authors'] and item['authors'][0] and item['authors'][0]['name'] else None,
author = authors[0]
content = ""
for obj in item['content']:
if obj['type'] == 'text/html':
content += obj['value']
content = content.replace("<br />", "\n")
content = tomd.convert(content[:1024])
while content.startswith('\n'):
content = content[1:]
while content.endswith('\n'):
content = content[:-1]
entries.append({
"embeds": [{
"title": "New release: %s" % version,
"description": package_name,
"url": item['link'],
"thumbnail": {
"url": "https://github.com/%s.png" % owner,
},
"author": {
"name": author,
"url": "https://github.com/%s" % author,
"icon_url": "https://github.com/%s.png" % author,
},
"fields": [{
"name": item['title_detail']['value'],
"value": content[:1024],
}],
"footer": {
"text": time.strftime("%a %d %b, %Y at %I:%M %p", item['updated_parsed']),
}
}],
"version": version,
"package_name": package_name,
})
return entries
def get_wiki_description(soup):
"""
Returns the first paragraph of a wiki page BeautifulSoup
"""
if soup.select(".mw-parser-output > p"):
pNum = 0
if headerEx.search(
str(
soup.select(
".mw-body-content > #mw-content-text > .mw-parser-output > p"
)[0])):
pNum = 1
return tomd.convert(
str(
soup.select(
".mw-body-content > #mw-content-text > .mw-parser-output > p"
)[pNum])).strip().replace("<br/>", "\n")
return ""
def anno_edit(request, pk):
"""公告编辑"""
anno = Announcement.objects.get(pk=pk)
# 当为post请求时，修改数据
if request.method == "POST":
form = AnnoForm(request.POST, instance=anno)
if form.is_valid():
form.save()
messages.success(request, "数据更新成功", extra_tags="修改成功")
return HttpResponseRedirect(reverse("sspanel:backend_anno"))
else:
messages.error(request, "数据填写错误", extra_tags="错误")
context = {"form": form, "anno": anno}
return render(request, "backend/annoedit.html", context=context)
# 当请求不是post时，渲染form
else:
anno.body = tomd.convert(anno.body)
context = {"anno": anno}
return render(request, "backend/annoedit.html", context=context)
def convert(self, name="", selector=""):
if not name:
name = self.title
if not selector:
selector = self.xpath
# 提取文章内容;
contentQ = self.rootQ(selector)
# 处理图片;
index = 1
for e in contentQ(self.rule.img_tag):
q = pq(e)
img_src = self.rule.find_img(q)
img_src_cur = self.rule.save_pic(self.url, name, index, img_src)
if q[0].tag != "img":
q.replace_with(pq('<img src="' + img_src_cur + '"/>'))
else:
q.attr(src=img_src_cur)
index += 1
# 转换成markdown;
self.rule.save_md(name, tomd.convert(contentQ))
<i>italic</i> <b><i>bold italic</i></b> <em>em</em> <strong>strong</strong> aa <strong> strong </strong> aa </p> <hr/> <table> <thead> <tr class="1"> <th>th1</th> <th>th2</th> </tr> </thead> <tbody> <tr> <td>td</td> <td>td</td> </tr> <tr> <td>td</td> <td>td</td> </tr></tbody></table> """ print(Tomd(string).markdown) print(tomd.convert(string))
import tomd
f = open('in.txt', mode='r', encoding='UTF-8')
contents = f.read()
f_out = open('out.md', 'w+', encoding='UTF-8')
soup = tomd.convert(contents)
f_out.write(soup)
f.close()
f_out.close()
