I'd like to generate completely random piece of html source, possibly from a grammar. I want to do this in python but I'm not sure how to proceed -- is there a library that takes a grammar and just randomly follows its rules, printing the path?
Ideas?
import urllib
html = urllib.urlopen('http://random.yahoo.com/bin/ryl').read()
I think that pulling a random page is much easier to implement and will be far more random than anything you could program yourself. Any program designed to produce random pages will still have to adhere to whatever rules defining the structure of html. Since humans are much better and breaking rules than machines, a random page from the web is more likely to contain structures you won't get from a randomizer.
You don't have to use yahoo, there are probably other random link generators, or you could build your own.
It's quite easy to roll your own random html generator that looks very much like a top-down parser. Here's a base!
def RandomHtml():
yield '<html><body>'
yield '<body>'
yield RandomBody()
yield '</body></html>'
def RandomBody():
yield RandomSection()
if random.randrange(2) == 0:
yield RandomBody()
def RandomSection():
yield '<h1>'
yield RandomSentence()
yield '</h1>'
sentences = random.randrange(5, 20)
for _ in xrange(sentences):
yield RandomSentence()
def RandomSentence():
words = random.randrange(5, 15)
yield (' '.join(RandomWord() for _ in xrange(words)) + '.').capitalize()
def RandomWord():
chars = random.randrange(2, 10)
return ''.join(random.choice(string.ascii_lowercase) for _ in xrange(chars))
def Output(generator):
if isinstance(generator, str):
print generator
else:
for g in generator: Output(g)
Output(RandomHtml())