python-ox/ox/web/siteparser.py

72 lines
2.6 KiB
Python
Raw Normal View History

2010-07-07 23:25:57 +00:00
# -*- coding: utf-8 -*-
# vi:si:et:sw=4:sts=4:ts=4
import re
from ..cache import readUrlUnicode
from .. import stripTags, decodeHtml
from ..utils import datetime
2010-07-07 23:25:57 +00:00
def cleanup(key, data, data_type):
if data:
if isinstance(data[0], basestring):
#FIXME: some types need stripTags
#data = [stripTags(decodeHtml(p)).strip() for p in data]
data = [decodeHtml(p).strip() for p in data]
elif isinstance(data[0], list) or isinstance(data[0], tuple):
data = [cleanup(key, p, data_type) for p in data]
2010-07-10 08:24:56 +00:00
while len(data) == 1 and not isinstance(data, basestring):
2010-07-07 23:25:57 +00:00
data = data[0]
if data_type == 'list' and isinstance(data, basestring):
data = [data, ]
elif data_type != 'list':
data = ''
return data
class SiteParser(dict):
baseUrl = ''
regex = {}
def getUrl(self, page):
return "%s%s" % (self.baseUrl, page)
2010-10-08 16:07:39 +00:00
def readUrlUnicode(self, url, timeout):
return readUrlUnicode(url, timeout=timeout)
2010-07-12 08:52:26 +00:00
def __init__(self, timeout=-1):
2010-07-07 23:25:57 +00:00
for key in self.regex:
url = self.getUrl(self.regex[key]['page'])
2010-10-08 16:07:39 +00:00
data = self.readUrlUnicode(url, timeout)
2010-07-07 23:25:57 +00:00
if isinstance(self.regex[key]['re'], basestring):
data = re.compile(self.regex[key]['re'], re.DOTALL).findall(data)
data = cleanup(key, data, self.regex[key]['type'])
2010-07-10 08:24:56 +00:00
elif callable(self.regex[key]['re']):
data = self.regex[key]['re'](data)
2010-07-07 23:25:57 +00:00
else:
for r in self.regex[key]['re']:
2010-07-10 08:24:56 +00:00
if callable(r):
f = r
else:
f = re.compile(r, re.DOTALL).findall
2010-07-07 23:25:57 +00:00
if isinstance(data, basestring):
2010-07-10 08:24:56 +00:00
data = f(data)
2010-07-07 23:25:57 +00:00
else:
2010-07-10 08:24:56 +00:00
data = [f(d) for d in data]
2010-07-07 23:25:57 +00:00
data = cleanup(key, data, self.regex[key]['type'])
def apply_f(f, data):
if data and isinstance(data[0], list):
data = [f(d) for d in data]
else:
data = f(data)
return data
2010-07-10 08:24:56 +00:00
if self.regex[key]['type'] == 'float' and data:
2010-07-07 23:25:57 +00:00
data = apply_f(float, data)
2010-07-12 08:52:26 +00:00
elif self.regex[key]['type'] == 'int' and data:
2010-07-07 23:25:57 +00:00
data = apply_f(int, data)
elif self.regex[key]['type'] == 'date':
parse_date = lambda d: d and datetime.strptime('-'.join(d), '%m-%d-%Y').strftime('%Y-%m-%d')
2010-07-07 23:25:57 +00:00
data = apply_f(parse_date, data)
2010-07-10 08:24:56 +00:00
if data:
self[key] = data
2010-07-07 23:25:57 +00:00