This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| <!DOCTYPE html> | |
| <html> | |
| <head> | |
| <title></title> | |
| </head> | |
| <body> | |
| <a href="sas.svg">BannerImage</a> | |
| <a href="http://www.google.com">Google</a> | |
| <a href="https://www.dimensaoalfa.com.br">Notícias Dimensão ALfa</a> | |
| <a href="www.wsricardo.blogspot.com">Meu Blog</a> | |
| <a href="#">Banner</a> | |
| <a href="noticias/n1.html">Notícias</a> | |
| <a href="http://youtube.com">Youtube></a> | |
| <a href="https://wikipedia.org">Wikipédia</a> | |
| </body> | |
| </html> |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| """ | |
| Author: WSRicardo | |
| Youtube: https://www.youtube.com/@dimensaoalfa | |
| Blog: https://wsricardo.blogspot.com | |
| Site: www.dimensaoalfa.com.br | |
| Github: www.github.com/wsricardo | |
| """ | |
| from bs4 import BeautifulSoup | |
| def get_links( webcontent ): | |
| """ | |
| get_links | |
| arguments | |
| webcoment - html text | |
| return list[ {'url':'', 'text':'' } ] | |
| Return list links found in html content. | |
| """ | |
| soup = BeautifulSoup(webcontent, 'html.parser' ) | |
| linksall = soup.find_all( 'a' ) | |
| links = [ ] | |
| for link in linksall: | |
| l = link.get( 'href' ) | |
| if l: | |
| if ( 'http' in l or 'https' in l ) and len( link.text ) > 3 : | |
| links.append( { | |
| 'url': l, | |
| 'text': link.text.strip( ) | |
| } ) | |
| return links | |
| if __name__== "__main__": | |
| html = '' | |
| with open( "main.html", "r", encoding='utf-8' ) as fl: | |
| html = fl.read() | |
| print( html ) | |
| print( get_links( html ) ) |
Nenhum comentário:
Postar um comentário