get maximum history

modules/boursorama/browser.py

@@ -22,9 +22,6 @@
 from weboob.tools.browser import BaseBrowser, BrowserIncorrectPassword
 from .pages import LoginPage, AccountsList, AccountHistory, UpdateInfoPage
 
-from datetime import date
-from dateutil.relativedelta import relativedelta
-
 
 __all__ = ['Boursorama']
 
@@ -83,21 +80,17 @@ class Boursorama(BaseBrowser):
         return None
 
     def get_history(self, account):
-        self.location(account._link_id)
+        link = account._link_id
+
+        while link is not None:
+            self.location(link)
             if not self.is_on_page(AccountHistory):
                 raise NotImplementedError()
 
-        operations = self.page.get_operations()
+            for tr in self.page.get_operations():
-        # load last month as well
+                yield tr
-        target = date.today() - relativedelta(months=1)
-        self.location(account._link_id + ("&month=%d&year=%d" % (target.month, target.year)))
-        operations += self.page.get_operations()
-        # and the month before, just in case you're greedy
-        target = date.today() - relativedelta(months=2)
-        self.location(account._link_id + ("&month=%d&year=%d" % (target.month, target.year)))
-        operations += self.page.get_operations()
 
-        return operations
+            link = self.page.get_next_url()
 
     def transfer(self, from_id, to_id, amount, reason=None):
         raise NotImplementedError()

modules/boursorama/pages/account_history.py

@@ -19,6 +19,7 @@
 # along with weboob. If not, see <http://www.gnu.org/licenses/>.
 
 
+from urlparse import urlparse
 from datetime import date
 import re
 
@@ -79,3 +80,18 @@ class AccountHistory(BasePage):
 
     def get_operations(self):
         return self.operations
+
+    def get_next_url(self):
+        items = self.document.getroot().cssselect('ul.mainmenu li')
+
+        current = False
+        for li in reversed(items):
+            if li.attrib.get('class', '') == 'selected':
+                current = True
+            elif current:
+                a = li.find('a')
+                if 'year' in a.attrib.get('href', ''):
+                    url = urlparse(self.url)
+                    return url.path + a.attrib['href']
+                else:
+                    return None