Created
August 26, 2012 03:43
-
-
Save zokis/3473678 to your computer and use it in GitHub Desktop.
URL Parser
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
""" | |
http://www.gmail.com/g/mail?fr=3&ui=2 | |
http://gmail.com | |
ssh://marcelo@git.com | |
ftp://ime.usp.com.br/~denis/?model=SistemasDeArquivo | |
""" | |
import unittest | |
import string | |
def analisador_url(url): | |
url_parseada = {} | |
querystring = None | |
if '?' in url: | |
url, querystring = url.split('?') | |
parametros = {} | |
# if querystring: | |
# for params in querystring.split('&'): | |
# chave, valor = params.split('=') | |
# parametros[chave] = valor | |
if querystring: | |
parametros = dict([tuple(chave_valor.split('=')) for chave_valor in [item for item in querystring.split('&')]]) | |
url_parseada['querystring'] = parametros | |
url_parseada['protocolo'], host_dominio_path = url.split('://') | |
url_parseada['user'] = None | |
if '@' in url: | |
url_parseada['user'], host_dominio_path = host_dominio_path.split('@') | |
url_parseada['path'] = None | |
host_dominio = host_dominio_path | |
if '/' in host_dominio_path: | |
host_dominio, url_parseada['path'] = string.split(host_dominio_path, '/', 1) | |
url_parseada['path'] = '/' + url_parseada['path'] | |
host_dominio_itens = host_dominio.split('.') | |
url_parseada['host'] = '' | |
url_parseada['dominio'] = host_dominio_itens[0] | |
if len(host_dominio_itens) > 2: | |
url_parseada['host'], url_parseada['dominio'] = host_dominio_itens[0], '.'.join(host_dominio_itens[1:]) | |
return url_parseada | |
class DojoUrls(unittest.TestCase): | |
def test_deve_retornar_protocolo(self): | |
self.assertEquals('http', | |
analisador_url('http://')['protocolo']) | |
self.assertEquals('ssh', | |
analisador_url('ssh://marcelo@git.com')['protocolo']) | |
self.assertEquals('ftp', | |
analisador_url('ftp://ime.usp.com.br/~denis/?model=SistemasDeArquivo')['protocolo']) | |
def test_deve_retornar_host(self): | |
self.assertEquals('www', | |
analisador_url('http://www.gmail.com')['host']) | |
self.assertEquals('ime', | |
analisador_url('http://ime.usp.com.br')['host']) | |
def test_deve_retornar_dominio(self): | |
self.assertEquals('gmail.com', | |
analisador_url('http://www.gmail.com')['dominio']) | |
self.assertEquals('usp.com.br', | |
analisador_url('http://ime.usp.com.br/~denis/?model=SistemasDeArquivo')['dominio']) | |
def test_deve_retornar_path(self): | |
self.assertEquals('/g/mail', | |
analisador_url('http://www.gmail.com/g/mail?fr=3&ui=2')['path']) | |
self.assertEquals('/~denis/', | |
analisador_url('ftp://ime.usp.com.br/~denis/?model=SistemasDeArquivo')['path']) | |
def test_deve_retornar_querystring(self): | |
self.assertEquals({'model': 'SistemasDeArquivo'}, | |
analisador_url('ftp://ime.usp.com.br/~denis/?model=SistemasDeArquivo')['querystring']) | |
self.assertEquals({'fr': '3', 'ui': '2'}, | |
analisador_url('http://www.gmail.com/g/mail?fr=3&ui=2')['querystring']) | |
def test_deve_retornar_user(self): | |
self.assertEquals('marcelo', | |
analisador_url('ssh://marcelo@git.com')['user']) | |
self.assertEquals('weverton', | |
analisador_url('ssh://weverton@git.com')['user']) | |
self.assertEquals(None, | |
analisador_url('ssh://git.com')['user']) | |
if __name__ == '__main__': | |
unittest.main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment