from scrapy import Selector

# Create a selector object from a secret website
sel = Selector(  text = html)

# Select all hyperlinks of div elements belonging to class "course-block"
course_as = sel.css( 'div.course-block > a' )

# Selecting all href attributes chaining with css
hrefs_from_css = course_as.css( '::attr(href)' )

# Selecting all href attributes chaining with xpath
hrefs_from_xpath = course_as.xpath( './@href')