* translaboob is a weboob application for automatic translation

* new capability is defined, ICapTranslate
* googletranslate is a backend for the translaboob application

modules/googletranslate/__init__.py

+"GoogleTranslateBackend init"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012  Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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 .backend import GoogleTranslateBackend
+
+
+__all__ = ['GoogleTranslateBackend']

modules/googletranslate/backend.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# 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 .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)

modules/googletranslate/browser.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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 weboob.tools.browser import BaseBrowser, BrowserHTTPNotFound, BrowserIncorrectPassword
+from weboob.tools.browser.decorators import id2url, check_url
+from weboob.capabilities.translate import TranslationFail
+
+from .pages import TranslatePage
+
+import urllib
+
+__all__ = ['GoogleTranslateBrowser']
+
+class GoogleTranslateBrowser(BaseBrowser):
+    DOMAIN = 'translate.google.com'
+    ENCODING = 'UTF-8'
+    USER_AGENT = BaseBrowser.USER_AGENTS['desktop_firefox']
+    PAGES = {
+        'https?://translate\.google\.com': TranslatePage
+        }
+
+    def __init__(self, *args, **kwargs):
+        BaseBrowser.__init__(self, *args, **kwargs)
+
+    def translate(self, source, to, text):
+        """
+        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"
+

modules/googletranslate/pages.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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
+
+
+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
+
+

modules/googletranslate/tools.py

+"tools for lefigaro backend"
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2011  Julien Hebert
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+import re
+
+def id2url(_id):
+    "return an url from an id"
+    regexp2 = re.compile("(\w+).([0-9]+).(.*$)")
+    match = regexp2.match(_id)
+    if match:
+        return 'http://www.20minutes.fr/%s/%s/%s' % (match.group(1),
+                                                     match.group(2),
+                                                     match.group(3))
+    else:
+        raise ValueError("id doesn't match")
+
+
+def url2id(url):
+    "return an id from an url"
+    return url
+
+
+def rssid(entry):
+    return url2id(entry.id)

scripts/translaboob

+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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 weboob.applications.translaboob import Translaboob
+
+
+if __name__ == '__main__':
+    Translaboob.run()

weboob/applications/translaboob/__init__.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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 .translaboob import Translaboob
+
+__all__ = ['Translaboob']

weboob/applications/translaboob/translaboob.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012  Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+import sys
+
+import os
+import sys
+import codecs
+import locale
+
+from weboob.capabilities.translate import ICapTranslate
+from weboob.tools.application.repl import ReplApplication
+
+__all__ = ['Translaboob']
+
+class Translaboob(ReplApplication):
+    APPNAME = 'translaboob'
+    VERSION = '0.c'
+    COPYRIGHT = 'Copyright(C) 2012 Lucien Loiseau'
+    DESCRIPTION = 'Console application to translate text from one language to another'
+    CAPS = ICapTranslate
+    
+    def main(self, argv):
+        return ReplApplication.main(self, argv)
+
+    def do_translate(self, line):
+        lan_from, lan_to, text = self.parse_command_args(line, 3, 1)
+        """
+        translate <FROM> <TO> <TEXT>
+        translate from one language to another, 
+        <FROM> : source language
+        <TO>   : destination language
+        <TEXT> : language to translate, standart input if - is given
+        """
+        if not text or text == '-':
+            text = self.acquire_input()
+
+        print "from : "+lan_from+" to : "+lan_to 
+        print "<source>"
+        print text
+        print "</source>"
+
+        for backend, translation  in self.do('translate', lan_from, lan_to, text):
+          print "<translation "+backend.name+">"
+          print translation
+          print "</translation>"
+          
+
+
+
+
+
+
+

weboob/capabilities/translate.py

+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2012 Lucien Loiseau
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# 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, datetime
+
+from .base import CapBaseObject, Field, StringField, DateField, DecimalField, IntField
+from .collection import ICapCollection
+
+
+__all__ = ['TranslationFail', 'ICapTranslate']
+
+
+class TranslationFail(Exception):
+    """
+    Raised when no translation matches the given request
+    """
+
+    def __init__(self, msg='No Translation Available'):
+        Exception.__init__(self, msg)
+
+
+class ICapTranslate(ICapCollection):
+    """
+    Capability of online translation website to translate word or sentence
+    """
+    def translate(self, source_language, destination_language, request):
+        """
+        perfom a translation 
+        
+        :param source_language: language in which the request is written
+        :param destination_language: language to translate the request into
+        :param request: the sentence to be translated
+        """
+        raise TranslationFail()
+
+
+
+
+