nojhan/weboob-devel
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[meteofrance] fix #1431, call the url that retrieve all the search results

Browse source
This commit is contained in:
Bezleputh 2014-07-22 10:55:45 +02:00
commit 8ca5c619c5
2 changed files with 21 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

28
modules/meteofrance/browser.py
View file

@ -21,11 +21,7 @@
import urllib import urllib
from weboob.tools.browser import BaseBrowser from weboob.tools.browser import BaseBrowser
from weboob.tools.json import json as simplejson from .pages.meteo import WeatherPage, SearchCitiesPage
from weboob.capabilities.weather import City
from .pages.meteo import WeatherPage
import re
__all__ = ['MeteofranceBrowser'] __all__ = ['MeteofranceBrowser']
@ -34,29 +30,23 @@ class MeteofranceBrowser(BaseBrowser):
DOMAIN = 'www.meteofrance.com' DOMAIN = 'www.meteofrance.com'
PROTOCOL = 'http' PROTOCOL = 'http'
ENCODING = 'utf-8' ENCODING = 'utf-8'
USER_AGENT = BaseBrowser.USER_AGENTS['wget']
WEATHER_URL = '{0}://{1}/previsions-meteo-france/{{city_name}}/{{city_id}}'.format(PROTOCOL, DOMAIN) WEATHER_URL = '{0}://{1}/previsions-meteo-france/{{city_name}}/{{city_id}}'.format(PROTOCOL, DOMAIN)
CITY_SEARCH_URL = '{0}://{1}/mf3-rpc-portlet/rest/lieu/facet/previsions/search/{{city_pattern}}'\ CITY_SEARCH_URL = 'http://www.meteofrance.com/recherche/resultats'
.format(PROTOCOL, DOMAIN)
PAGES = { PAGES = {
WEATHER_URL.format(city_id=".*", city_name=".*"): WeatherPage, WEATHER_URL.format(city_id=".*", city_name=".*"): WeatherPage,
CITY_SEARCH_URL: SearchCitiesPage,
} }
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
BaseBrowser.__init__(self, *args, **kwargs) BaseBrowser.__init__(self, *args, **kwargs)
def iter_city_search(self, pattern): def iter_city_search(self, pattern):
searchurl = self.CITY_SEARCH_URL.format(city_pattern=urllib.quote_plus(pattern.encode('utf-8'))) datas = {'facet': 'previsions',
response = self.openurl(searchurl) 'search-type': 'previsions',
return self.parse_cities_result(response) 'query': pattern}
self.location(self.CITY_SEARCH_URL, data=urllib.urlencode(datas))
def parse_cities_result(self, datas): assert self.is_on_page(SearchCitiesPage)
cities = simplejson.loads(datas.read(), self.ENCODING) return self.page.iter_cities()
re_id = re.compile('\d{5}', re.DOTALL)
for city in cities:
if re_id.match(city['codePostal']):
mcity = City(int(city['codePostal']), u'%s' % city['slug'])
yield mcity
def iter_forecast(self, city_id): def iter_forecast(self, city_id):
mcity = self.iter_city_search(city_id).next() mcity = self.iter_city_search(city_id).next()

14
modules/meteofrance/pages/meteo.py
View file

@ -20,14 +20,24 @@
from weboob.tools.browser import BasePage from weboob.tools.browser import BasePage
from weboob.capabilities.weather import Forecast, Current from weboob.capabilities.weather import Forecast, Current, City
import datetime import datetime
import re
__all__ = ['WeatherPage'] __all__ = ['WeatherPage']
class SearchCitiesPage(BasePage):
def iter_cities(self):
list = self.document.getroot().xpath('//ul[@class="list-style-1"]/li/a')
for a in list:
m = re.search('/.*/(.*)/(\d{5})', a.attrib.get('href'))
if m:
mcity = City(int(m.group(2)), u'%s' % m.group(1))
yield mcity
class WeatherPage(BasePage): class WeatherPage(BasePage):
def get_temp_without_unit(self, temp_str): def get_temp_without_unit(self, temp_str):
# It seems that the mechanize module give us some old style # It seems that the mechanize module give us some old style