h/plugins/wolframalpha.py

58 lines
1.4 KiB
Python
Raw Normal View History

import re
from util import hook, http
2010-03-01 02:32:41 +00:00
@hook.command('wa')
@hook.command
def wolframalpha(inp):
".wa/.wolframalpha <query> -- scrapes Wolfram Alpha's" \
"results for <query>"
url = "http://www.wolframalpha.com/input/?asynchronous=false"
h = http.get_html(url, i=inp)
pods = h.xpath("//div[@class='pod ']")
pod_texts = []
2010-03-01 02:32:41 +00:00
for pod in pods:
2010-03-13 07:04:59 +00:00
heading = pod.find('h2')
if heading is not None:
heading = heading.text_content().strip()
if heading.startswith('Input'):
continue
else:
continue
results = []
for image in pod.xpath('div/div[@class="output"]/img'):
alt = image.attrib['alt'].strip()
alt = alt.replace('\\n', '; ')
alt = re.sub(r'\s+', ' ', alt)
if alt:
results.append(alt)
if results:
pod_texts.append(heading + ' ' + '|'.join(results))
2010-03-01 02:32:41 +00:00
ret = '. '.join(pod_texts)
if not pod_texts:
return 'no results'
2010-02-17 23:24:52 +00:00
ret = re.sub(r'\\(.)', r'\1', ret)
def unicode_sub(match):
return unichr(int(match.group(1), 16))
ret = re.sub(r'\\:([0-9a-z]{4})', unicode_sub, ret)
2010-03-01 02:32:41 +00:00
if len(ret) > 430:
ret = ret[:ret.rfind(' ', 0, 430)]
ret = re.sub(r'\W+$', '', ret) + '...'
if not ret:
return 'no result'
return ret