diff options
Diffstat (limited to 'bs4')
| -rw-r--r-- | bs4/element.py | 20 | ||||
| -rw-r--r-- | bs4/tests/test_tree.py | 17 |
2 files changed, 34 insertions, 3 deletions
diff --git a/bs4/element.py b/bs4/element.py index e0da4d2..11bf8c3 100644 --- a/bs4/element.py +++ b/bs4/element.py @@ -802,6 +802,14 @@ class PageElement(object): yield i i = i.parent + @property + def decomposed(self): + """Check whether a PageElement has been decomposed. + + :rtype: bool + """ + return getattr(self, '_decomposed', False) or False + # Old non-property versions of the generators, for backwards # compatibility with BS3. def nextGenerator(self): @@ -1211,15 +1219,21 @@ class Tag(PageElement): This element will be removed from the tree and wiped out; so will everything beneath it. + + The behavior of a decomposed PageElement is undefined and you + should never use one for anything, but if you need to _check_ + whether an element has been decomposed, you can use the + `decomposed` property. """ self.extract() i = self while i is not None: - next = i.next_element + n = i.next_element i.__dict__.clear() i.contents = [] - i = next - + i._decomposed = True + i = n + def clear(self, decompose=False): """Wipe out all children of this PageElement by calling extract() on them. diff --git a/bs4/tests/test_tree.py b/bs4/tests/test_tree.py index 3251e0e..80aaaff 100644 --- a/bs4/tests/test_tree.py +++ b/bs4/tests/test_tree.py @@ -1275,6 +1275,23 @@ class TestTreeModification(SoupTest): a.clear(decompose=True) self.assertEqual(0, len(em.contents)) + + def test_decompose(self): + # Test PageElement.decompose() and PageElement.decomposed + soup = self.soup("<p><a>String <em>Italicized</em></a></p><p>Another para</p>") + p1, p2 = soup.find_all('p') + a = p1.a + text = p1.em.string + for i in [p1, p2, a, text]: + self.assertEqual(False, i.decomposed) + + # This sets p1 and everything beneath it to decomposed. + p1.decompose() + for i in [p1, a, text]: + self.assertEqual(True, i.decomposed) + # p2 is unaffected. + self.assertEqual(False, p2.decomposed) + def test_string_set(self): """Tag.string = 'string'""" soup = self.soup("<a></a> <b><c></c></b>") |