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

[wordreference] make result cleaner

Browse source
This commit is contained in:
Julien Veyssier 2013-02-10 19:04:45 +01:00 • committed by Romain Bignon
commit 7432cfc465
1 changed files with 6 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

9
modules/wordreference/pages.py
View file

@ -19,9 +19,11 @@
from weboob.tools.browser import BasePage from weboob.tools.browser import BasePage
import re
__all__ = ['TranslatePage'] __all__ = ['TranslatePage']
LAST_THING_IN_PARENTHESIS = re.compile("\([^)]\)$")
class TranslatePage(BasePage): class TranslatePage(BasePage):
@ -31,15 +33,16 @@ class TranslatePage(BasePage):
prev_was_nums1 = False prev_was_nums1 = False
for td in tr.getiterator('td'): for td in tr.getiterator('td'):
if prev_was_nums1: if prev_was_nums1:
return u''+td.text_content().split(';')[0].strip() result = u''+td.text_content().split(';')[0].strip()
result = LAST_THING_IN_PARENTHESIS.sub("",result)
return result
if td.attrib.get('class','') == 'nums1': if td.attrib.get('class','') == 'nums1':
prev_was_nums1 = True prev_was_nums1 = True
# if only one signification is found # if only one signification is found
for div in self.document.getiterator('div'): for div in self.document.getiterator('div'):
if div.attrib.get('class','') == "trans clickable": if div.attrib.get('class','') == "trans clickable":
names = u''+" ".join(div.text_content().split(']')[1].split()[1:]).split(';')[0] names = u''+" ".join(div.text_content().split(']')[1].split()[1:]).split(';')[0]
if ")" in names: names = LAST_THING_IN_PARENTHESIS.sub("",names)
names = names.split(")")[1]
return names.strip() return names.strip()
# another numerotation possibility... # another numerotation possibility...
for table in self.document.getiterator('table'): for table in self.document.getiterator('table'):