Draw a graph using braille characters

Question

I hacked together a way to make a small graph using Braille unicode codepoints by abusing unicodedata:

def plot(timeline):
  ret = []
  for r1, r2 in zip(timeline[::2], timeline[1::2]):
    character_name = "BRAILLE PATTERN DOTS-"
    dots  = [x for x, y in zip("7321", range(r1))]
    dots += [x for x, y in zip("8654", range(r2))]
    character_name += "".join(sorted(dots))
    ret.append(unicodedata.lookup(character_name) if dots else " ")
  return "".join(ret)


>>> print plot([random.randrange(0,4) for _ in range(100)])
 ⣆⣆⣶⣤⢠⡀⣦⡆⢀⡄⣰⣠⣦⣆⣰⢰⢠⢰⣄⡄ ⣀⣦⡆⣄⣶⣤⣠⡆⣠⣦⣆  ⣰⣴⡄⣤⣀ ⡀⣄⡀⣦⣶⣰⣶⣄⣴         

Barring low-hanging incremental improvements such as memoizing the call to unicodedata.lookup, what would be a more proper way to perform this lookup?

Answer 1

Why zip?

You have:

dots  = [x for x, y in zip("7321", range(r1))]
dots += [x for x, y in zip("8654", range(r2))]

You're never using the second part of that zip. So you're basically making lists for the two ranges, making lists for the two zips, making lists for the outcomes of those things, all to drop most of the information.

You can simplify that to just:

dots = '7321'[:r1] + '8654'[:r2]

character name

Try to avoid building up strings with +=. Instead just use format:

character_name = "BRAILLE PATTERN DOTS-{}".format(join(sorted(dots)))

Answer 2

I'm a bit confused by your list comprehensions:

dots  = [x for x, y in zip("7321", range(r1))]

What's the significance of "7231"? It would benefit from being a named constant, clarifying what it is.

Likewise, using x means I can't inferr anything from that name. r1 and r2 are also meaningless. If you named these differently, then I might be able to follow what's happening even without knowing that this is for creating braile characters. Even if their significance is semi-random, you should note that!

Also, you don't seem to use y. If that's the case, it's more pythonic to use an underscore instead of a y, to make it clear that you don't intend to use that value for anything.

If I'm reading this right, you're just using range(r1) with zip to affect how many characters out of "7231", which is clever and effective but terribly unclear. A comment would clear up confusion here immensely and make it a lot easier to follow.