several fixes in the new ICapTranslate capability, module and application

This commit is contained in:
Romain Bignon 2012-04-01 19:32:05 +02:00
commit a8166a4719
5 changed files with 96 additions and 78 deletions

View file

@ -18,21 +18,32 @@
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
"backend for http://translate.google.com"
from weboob.capabilities.translate import ICapTranslate
from weboob.tools.backend import BaseBackend, BackendConfig
from weboob.capabilities.translate import ICapTranslate, Translation, TranslationFail
from weboob.tools.backend import BaseBackend
from .browser import GoogleTranslateBrowser
__all__ = ['GoogleTranslateBackend']
class GoogleTranslateBackend(BaseBackend, ICapTranslate):
MAINTAINER = 'Lucien Loiseau'
EMAIL = 'loiseau.lucien@gmail.com'
VERSION = '0.c'
LICENSE = 'AGPLv3+'
STORAGE = {'seen': {}}
NAME = 'googletranslate'
DESCRIPTION = u'Google translation web service'
BROWSER = GoogleTranslateBrowser
def translate(self, lan_from, lan_to, text):
return self.browser.translate(lan_from, lan_to, text)
translation = Translation(0)
translation.lang_src = lan_from
translation.lang_dst = lan_to
translation.text = self.browser.translate(lan_from, lan_to, text)
if translation.text is None:
raise TranslationFail()
return translation

View file

@ -18,16 +18,16 @@
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
from weboob.tools.browser import BaseBrowser, BrowserHTTPNotFound, BrowserIncorrectPassword
from weboob.tools.browser.decorators import id2url, check_url
from weboob.capabilities.translate import TranslationFail
import urllib
from weboob.tools.browser import BaseBrowser
from .pages import TranslatePage
import urllib
__all__ = ['GoogleTranslateBrowser']
class GoogleTranslateBrowser(BaseBrowser):
DOMAIN = 'translate.google.com'
ENCODING = 'UTF-8'
@ -43,22 +43,17 @@ class GoogleTranslateBrowser(BaseBrowser):
"""
translate 'text' from 'source' language to 'to' language
"""
try:
d = {
'sl': source,
'tl': to,
'js': 'n',
'prev': '_t',
'hl': 'en',
'ie': 'UTF-8',
'layout': '2',
'eotf': '1',
'text': text,
}
self.location('http://'+self.DOMAIN, urllib.urlencode(d))
translation = self.page.get_translation()
return translation
except TranslationFail:
return "no translation available"
d = {
'sl': source.encode('utf-8'),
'tl': to.encode('utf-8'),
'js': 'n',
'prev': '_t',
'hl': 'en',
'ie': 'UTF-8',
'layout': '2',
'eotf': '1',
'text': text.encode('utf-8'),
}
self.location('http://'+self.DOMAIN, urllib.urlencode(d))
translation = self.page.get_translation()
return translation

View file

@ -17,15 +17,17 @@
# You should have received a copy of the GNU Affero General Public License
# along with weboob. If not, see <http://www.gnu.org/licenses/>.
from datetime import date
from weboob.tools.browser import BasePage
__all__ = ['TranslatePage']
class TranslatePage(BasePage):
def get_translation(self):
for span in self.document.getiterator('span'):
if (span.attrib.get('id', '') == 'result_box'):
for children in span.getchildren():
return children.text
boxes = self.parser.select(self.document.getroot(), 'span#result_box', 1).findall('span')
if len(boxes) == 0:
return None
return u'\n'.join([unicode(box.text) for box in boxes])