2012-06-20 20:41:04 +00:00
|
|
|
# -*- coding: iso-8859-1 -*-
|
|
|
|
# Copyright (C) 2004-2005 Tristan Seligmann and Jonathan Jacobs
|
|
|
|
# Copyright (C) 2012 Bastian Kleineidam
|
2012-10-11 10:03:12 +00:00
|
|
|
from .util import fetchUrl, getQueryParams
|
2012-06-20 19:58:13 +00:00
|
|
|
|
|
|
|
def queryNamer(paramName, usePageUrl=False):
|
2012-09-26 14:47:39 +00:00
|
|
|
"""Get name from URL query part."""
|
2012-06-20 19:58:13 +00:00
|
|
|
@staticmethod
|
|
|
|
def _namer(imageUrl, pageUrl):
|
|
|
|
url = (imageUrl, pageUrl)[usePageUrl]
|
|
|
|
return getQueryParams(url)[paramName][0]
|
|
|
|
return _namer
|
|
|
|
|
|
|
|
|
|
|
|
def regexNamer(regex):
|
2012-09-26 14:47:39 +00:00
|
|
|
"""Get name from regular expression."""
|
2012-06-20 19:58:13 +00:00
|
|
|
@staticmethod
|
|
|
|
def _namer(imageUrl, pageUrl):
|
2012-11-13 18:12:28 +00:00
|
|
|
mo = regex.search(imageUrl)
|
|
|
|
if mo:
|
|
|
|
return mo.group(1)
|
2012-06-20 19:58:13 +00:00
|
|
|
return _namer
|
|
|
|
|
|
|
|
|
|
|
|
def bounceStarter(latestUrl, nextSearch):
|
2012-09-26 14:47:39 +00:00
|
|
|
"""Get start URL by "bouncing" back and forth one time."""
|
2012-06-20 19:58:13 +00:00
|
|
|
@classmethod
|
|
|
|
def _starter(cls):
|
|
|
|
url = fetchUrl(latestUrl, cls.prevSearch)
|
2012-11-21 20:57:26 +00:00
|
|
|
if not url:
|
|
|
|
raise ValueError("could not find prevSearch pattern %r in %s" % (cls.prevSearch.pattern, latestUrl))
|
|
|
|
url = fetchUrl(url, nextSearch)
|
|
|
|
if not url:
|
|
|
|
raise ValueError("could not find nextSearch pattern %r in %s" % (nextSearch.pattern, latestUrl))
|
2012-06-20 19:58:13 +00:00
|
|
|
return url
|
|
|
|
return _starter
|
|
|
|
|
|
|
|
|
|
|
|
def indirectStarter(baseUrl, latestSearch):
|
2012-09-26 14:47:39 +00:00
|
|
|
"""Get start URL by indirection."""
|
2012-06-20 19:58:13 +00:00
|
|
|
@staticmethod
|
|
|
|
def _starter():
|
2012-11-21 20:57:26 +00:00
|
|
|
url = fetchUrl(baseUrl, latestSearch)
|
|
|
|
if not url:
|
|
|
|
raise ValueError("could not find latestSearch pattern %r in %s" % (latestSearch.pattern, baseUrl))
|
|
|
|
return url
|
2012-06-20 19:58:13 +00:00
|
|
|
return _starter
|
|
|
|
|
|
|
|
|
|
|
|
class IndirectLatestMixin(object):
|
|
|
|
'''
|
|
|
|
Mixin for comics that link to the latest comic from a base page of
|
|
|
|
some kind. This also supports comics which don't link to the last comic
|
|
|
|
from the base page, but the beginning of the latest chapter or similiar
|
|
|
|
schemes. It simulates going forward until it can't find a 'next' link as
|
|
|
|
specified by the 'nextSearch' regex.
|
|
|
|
|
|
|
|
@type baseUrl: C{string}
|
|
|
|
@cvar baseUrl: the URL where the link to the latest comic is found.
|
|
|
|
@type latestSearch C{regex}
|
|
|
|
@cvar latestSearch: a compiled regex for finding the 'latest' URL.
|
|
|
|
@type nextSearch C{regex}
|
|
|
|
@cvar nextSearch: a compiled regex for finding the 'next' URL.
|
|
|
|
'''
|
|
|
|
|
|
|
|
__latestUrl = None
|
|
|
|
|
|
|
|
def getLatestUrl(self):
|
2012-09-26 14:47:39 +00:00
|
|
|
"""Get latest comic URL."""
|
2012-06-20 19:58:13 +00:00
|
|
|
if not self.__latestUrl:
|
|
|
|
self.__latestUrl = fetchUrl(self.baseUrl, self.latestSearch)
|
|
|
|
if hasattr(self, "nextSearch"):
|
|
|
|
nextUrl = fetchUrl(self.__latestUrl, self.nextSearch)
|
|
|
|
while nextUrl:
|
|
|
|
self.__latestUrl = nextUrl
|
|
|
|
nextUrl = fetchUrl(self.__latestUrl, self.nextSearch)
|
|
|
|
return self.__latestUrl
|
|
|
|
|
|
|
|
latestUrl = property(getLatestUrl)
|
|
|
|
|