spider_infracciones.py

# -*- coding: utf-8 -*-
# scrapy crawl spider-infracciones -a url=http://www.impo.com.uy/bases/notificaciones-cgm/1-2017/ -o infracciones-1-2017.json
import scrapy

class SpiderInfraccionesSpider(scrapy.Spider):
    name = "spider-infracciones"
    allowed_domains = ["impo.com.uy"]
    start_urls = ['http://www.impo.com.uy/bases/notificaciones-cgm/15-2017/']

    def __init__(self, url, *args, **kwargs):
        super(SpiderInfraccionesSpider, self).__init__(*args, **kwargs)
        self.start_urls = [url]

    def parse(self, response):
        records = response.css('table.tabla_en_texto > tr')
        records.pop(0)

        for record in records:
            tds = record.css('td > pre')
            yield {
                'matricula': tds[0].css('::text').extract_first() or '',
                'fecha_hora': tds[1].css('::text').extract_first() or '',
                'interseccion': tds[2].css('::text').extract_first() or '',
                'intervenido': tds[3].css('::text').extract_first() or '',
                'articulo': tds[4].css('::text').extract_first() or '',
                'valor_ur': tds[5].css('::text').extract_first() or ''
            }