Usando proxies em Python com httpx

Usando proxies em Python com httpx

httpx e proxies

O httpx é uma biblioteca HTTP moderna para Python que suporta os modos síncrono e assíncrono de operação. Uma vantagem essencial é o suporte nativo a HTTP/2 e a proxies SOCKS sem bibliotecas adicionais.

Instalação

pip install httpx
# Para suporte a SOCKS:
pip install httpx[socks]
# Para HTTP/2:
pip install httpx[http2]

Modo síncrono

Proxy HTTP

import httpx

proxy = "http://user:pass@proxy_ip:port"
response = httpx.get("https://httpbin.org/ip", proxy=proxy)
print(response.json())

Proxies diferentes para HTTP e HTTPS

proxies = {
    "http://": "http://proxy1:port",
    "https://": "http://proxy2:port",
}

with httpx.Client(proxy=proxies) as client:
    response = client.get("https://httpbin.org/ip")
    print(response.json())

Proxy SOCKS5

import httpx

proxy = "socks5://user:pass@proxy_ip:port"
with httpx.Client(proxy=proxy) as client:
    response = client.get("https://httpbin.org/ip")
    print(response.json())

Modo assíncrono

Exemplo básico

import httpx
import asyncio

async def fetch():
    proxy = "http://user:pass@proxy_ip:port"
    async with httpx.AsyncClient(proxy=proxy) as client:
        response = await client.get("https://httpbin.org/ip")
        print(response.json())

asyncio.run(fetch())

Requisições paralelas

import httpx
import asyncio

async def fetch_url(client, url):
    try:
        response = await client.get(url, timeout=10)
        return response.text
    except Exception as e:
        return None

async def main():
    proxy = "http://user:pass@proxy_ip:port"
    urls = [f"https://example.com/page/{i}" for i in range(50)]

    async with httpx.AsyncClient(proxy=proxy) as client:
        tasks = [fetch_url(client, url) for url in urls]
        results = await asyncio.gather(*tasks)
        success = sum(1 for r in results if r)
        print(f"Success: {success}/{len(urls)}")

asyncio.run(main())

HTTP/2 via proxy

O httpx é uma das poucas bibliotecas Python com suporte a HTTP/2:

import httpx

proxy = "http://user:pass@proxy_ip:port"
with httpx.Client(proxy=proxy, http2=True) as client:
    response = client.get("https://httpbin.org/ip")
    print(f"HTTP version: {response.http_version}")
    print(response.json())

O HTTP/2 via proxy funciona pelo método CONNECT — o proxy cria um túnel TCP através do qual é estabelecida a conexão HTTP/2 com o servidor.

Rotação de proxies

import httpx
import random

PROXIES = [
    "http://user:pass@proxy1:port",
    "http://user:pass@proxy2:port",
    "http://user:pass@proxy3:port",
]

def get_random_proxy():
    return random.choice(PROXIES)

# Rotação síncrona
for i in range(10):
    proxy = get_random_proxy()
    with httpx.Client(proxy=proxy) as client:
        resp = client.get("https://httpbin.org/ip")
        print(resp.json())

Rotação assíncrona

import httpx
import asyncio
import random

PROXIES = [
    "http://user:pass@proxy1:port",
    "http://user:pass@proxy2:port",
    "http://user:pass@proxy3:port",
]

async def fetch_with_rotation(url):
    proxy = random.choice(PROXIES)
    async with httpx.AsyncClient(proxy=proxy) as client:
        try:
            resp = await client.get(url, timeout=10)
            return resp.text
        except Exception:
            return None

async def main():
    urls = ["https://example.com"] * 20
    tasks = [fetch_with_rotation(url) for url in urls]
    results = await asyncio.gather(*tasks)
    print(f"Success: {sum(1 for r in results if r)}")

asyncio.run(main())

Configuração do cliente

Timeout

timeout = httpx.Timeout(
    connect=5.0,    # conexão com o proxy e o servidor
    read=10.0,      # leitura da resposta
    write=5.0,      # envio da requisição
    pool=5.0        # espera no pool de conexões
)

client = httpx.Client(proxy=proxy, timeout=timeout)

Cabeçalhos e cookies

headers = {
    "User-Agent": "Mozilla/5.0 ...",
    "Accept-Language": "en-US,en;q=0.9",
}

client = httpx.Client(
    proxy=proxy,
    headers=headers,
    follow_redirects=True,
    max_redirects=10
)

Pool de conexões

limits = httpx.Limits(
    max_connections=100,          # total de conexões
    max_keepalive_connections=20  # conexões keep-alive
)

client = httpx.AsyncClient(proxy=proxy, limits=limits)

Tratamento de erros

import httpx

async def safe_fetch(client, url):
    try:
        response = await client.get(url)
        response.raise_for_status()
        return response.text
    except httpx.ProxyError as e:
        print(f"Proxy error: {e}")
    except httpx.ConnectTimeout:
        print("Connection timeout")
    except httpx.ReadTimeout:
        print("Read timeout")
    except httpx.HTTPStatusError as e:
        print(f"HTTP {e.response.status_code}")
    except httpx.RequestError as e:
        print(f"Request error: {e}")
    return None

httpx vs requests vs aiohttp

Migração a partir do requests

O httpx foi projetado como substituto direto do requests. A migração é mínima:

# requests
import requests
resp = requests.get(url, proxies={"https": proxy})

# httpx
import httpx
resp = httpx.get(url, proxy=proxy)

A principal diferença: no httpx usa-se o parâmetro proxy (no singular) em vez de proxies.

Conclusão

O httpx é a melhor escolha para novos projetos Python que precisam de proxies. O suporte nativo a HTTP/2, SOCKS5 e aos modos síncrono e assíncrono faz dele uma ferramenta versátil. A API compatível com requests simplifica a migração de projetos já existentes.