use the lxml parser (elementtidy is now deprecated)

weboob/backends/bnporc/browser.py

@@ -21,17 +21,8 @@ Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 from cStringIO import StringIO
 
 from weboob.tools.browser import BaseBrowser, BrowserIncorrectPassword
-from weboob.tools.parsers.elementtidyparser import ElementTidyParser
 from weboob.backends.bnporc import pages
 
-# Parser
-class BNParser(ElementTidyParser):
-    def parse(self, data, encoding):
-        s = data.read()
-        s = s.replace('<?Pub Caret>', '')
-        data = StringIO(s)
-        return ElementTidyParser.parse(self, data, encoding)
-
 # Browser
 class BNPorc(BaseBrowser):
     DOMAIN = 'www.secure.bnpparibas.net'
@@ -49,7 +40,6 @@ class BNPorc(BaseBrowser):
     is_logging = False
 
     def __init__(self, *args, **kwargs):
-        kwargs['parser'] = BNParser()
         kwargs['history'] = None # need history
         BaseBrowser.__init__(self, *args, **kwargs)
 

weboob/backends/bnporc/pages/account_coming.py

@@ -28,7 +28,6 @@ class AccountComing(BasePage):
 
         for tr in self.document.getiterator('tr'):
             if tr.attrib.get('class', '') == 'hdoc1' or tr.attrib.get('class', '') == 'hdotc1':
-                operation = Operation()
                 tds = tr.findall('td')
                 if len(tds) != 3:
                     continue
@@ -38,10 +37,11 @@ class AccountComing(BasePage):
                 for child in tds[1].getchildren():
                     if child.text: label += child.text
                     if child.tail: label += child.tail
-                label += tds[1].tail
+                if tds[1].tail: label += tds[1].tail
                 label = label.strip()
                 amount = tds[2].text.replace('.','').replace(',','.')
 
+                operation = Operation()
                 operation.setDate(date)
                 operation.setLabel(label)
                 operation.setAmount(float(amount))