paste* backends: "Paste not found" exception

weboob/backends/pastealacon/browser.py

@@ -20,7 +20,9 @@
 from mechanize import RobustFactory
 import re
 
-from weboob.tools.browser import BaseBrowser, BrowserUnavailable
+from weboob.tools.browser import BaseBrowser, BrowserUnavailable, BrowserHTTPNotFound
+
+from weboob.capabilities.paste import PasteNotFound
 
 from .pages import PastePage, CaptchaPage, PostPage
 
@@ -50,7 +52,10 @@ class PastealaconBrowser(BaseBrowser):
         This is the fastest and safest method if you only want the content.
         Returns unicode.
         """
-        return self.readurl('http://%s/pastebin.php?dl=%s' % (self.DOMAIN, _id)).decode(self.ENCODING)
+        try:
+            return self.readurl('http://%s/pastebin.php?dl=%s' % (self.DOMAIN, _id), if_fail='raise').decode(self.ENCODING)
+        except BrowserHTTPNotFound:
+            raise PasteNotFound()
 
     def post_paste(self, paste):
         self.home()

weboob/backends/pastealacon/pages.py

@@ -18,16 +18,23 @@
 # along with weboob. If not, see <http://www.gnu.org/licenses/>.
 
 
-from weboob.tools.browser import BasePage
-
 import re
 
+from weboob.tools.browser import BasePage, BrokenPageError
+
+from weboob.capabilities.paste import PasteNotFound
+
 __all__ = ['PastePage', 'PostPage', 'CaptchaPage']
 
 class PastePage(BasePage):
     def fill_paste(self, paste):
         root = self.document.getroot()
-        header = self.parser.select(root, 'id("content")//h3', 1, 'xpath')
+        try:
+            # there is no 404, try to detect if there really is a content
+            self.parser.select(root, 'id("content")/div[@class="syntax"]//ol', 1, 'xpath')
+        except BrokenPageError:
+            raise PasteNotFound()
+        header = self.parser.select(root, 'id("content")/h3', 1, 'xpath')
         matches = re.match(r'Posted by (?P<author>.+) on (?P<date>.+) \(', header.text)
         paste.title = matches.groupdict().get('author')
         paste.contents = self.parser.select(root, '//textarea[@id="code"]', 1, 'xpath').text

weboob/backends/pastealacon/test.py

@@ -17,9 +17,13 @@
 # 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.test import BackendTest
 from weboob.capabilities.base import NotLoaded
 from weboob.tools.browser import BrowserUnavailable
+
+from weboob.capabilities.paste import PasteNotFound
+
 from .paste import PastealaconPaste
 
 class PastealaconTest(BackendTest):
@@ -53,3 +57,12 @@ class PastealaconTest(BackendTest):
     def test_spam(self):
         p = PastealaconPaste(None, title='viagra', contents='http://example.com/')
         self.assertRaises(BrowserUnavailable, self.backend.post_paste, p)
+
+    def test_notfound(self):
+        # html method
+        p = self.backend.get_paste('424242424242424242424242424242424242')
+        self.assertRaises(PasteNotFound, self.backend.fillobj, p, ['title'])
+
+        # raw method
+        p = self.backend.get_paste('424242424242424242424242424242424242')
+        self.assertRaises(PasteNotFound, self.backend.fillobj, p, ['contents'])

weboob/backends/pastebin/browser.py

@@ -18,7 +18,9 @@
 # along with weboob. If not, see <http://www.gnu.org/licenses/>.
 
 
-from weboob.tools.browser import BaseBrowser
+from weboob.tools.browser import BaseBrowser, BrowserHTTPNotFound
+
+from weboob.capabilities.paste import PasteNotFound
 
 from .pages import PastePage, PostPage
 
@@ -43,8 +45,11 @@ class PastebinBrowser(BaseBrowser):
         """
         Get as much as information possible from the paste page
         """
+        try:
             self.location(paste.page_url)
             return self.page.fill_paste(paste)
+        except BrowserHTTPNotFound:
+            raise PasteNotFound()
 
     def get_contents(self, _id):
         """
@@ -52,7 +57,10 @@ class PastebinBrowser(BaseBrowser):
         This is the fastest and safest method if you only want the content.
         Returns unicode.
         """
-        return self.readurl('http://%s/raw.php?i=%s' % (self.DOMAIN, _id)).decode(self.ENCODING)
+        try:
+            return self.readurl('http://%s/raw.php?i=%s' % (self.DOMAIN, _id), if_fail='raise').decode(self.ENCODING)
+        except BrowserHTTPNotFound:
+            raise PasteNotFound()
 
     def post_paste(self, paste):
         self.home()

weboob/backends/pastebin/test.py

@@ -20,6 +20,7 @@
 from weboob.tools.test import BackendTest
 from .paste import PastebinPaste
 from weboob.capabilities.base import NotLoaded
+from weboob.capabilities.paste import PasteNotFound
 
 class PastebinTest(BackendTest):
     BACKEND = 'pastebin'
@@ -57,3 +58,12 @@ class PastebinTest(BackendTest):
         p2 = self.backend.get_paste(p1.id)
         self.backend.fillobj(p2, ['contents'])
         assert p2.contents == p1.contents
+
+    def test_notfound(self):
+        # html method
+        p = self.backend.get_paste('weboooooooooooooooooooooooooob')
+        self.assertRaises(PasteNotFound, self.backend.fillobj, p, ['title'])
+
+        # raw method
+        p = self.backend.get_paste('weboooooooooooooooooooooooooob')
+        self.assertRaises(PasteNotFound, self.backend.fillobj, p, ['contents'])

weboob/capabilities/paste.py

@@ -21,9 +21,12 @@
 from .base import IBaseCap, CapBaseObject, NotLoaded
 
 
-__all__ = ['BasePaste', 'ICapPaste']
+__all__ = ['PasteNotFound', 'BasePaste', 'ICapPaste']
 
 
+class PasteNotFound(Exception):
+    pass
+
 class BasePaste(CapBaseObject):
     """
     Represents a pasted text.