Fix a crash when pickling a BeautifulSoup object that has no

  tree builder. [bug=1934003]

4 files changed, 35 insertions, 1 deletions

diff --git a/CHANGELOG b/CHANGELOG
index af99990..3c807d5 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -33,6 +33,9 @@ Python 2 was revision 605.
   version. "text" still works, but will give a DeprecationWarning.
   [bug=1947038]
 
+* Fix a crash when pickling a BeautifulSoup object that has no
+  tree builder. [bug=1934003]
+
 = 4.10.0 (20210907)
 
 * This is the first release of Beautiful Soup to only support Python
diff --git a/bs4/__init__.py b/bs4/__init__.py
index ddf1a86..75c801c 100644
--- a/bs4/__init__.py
+++ b/bs4/__init__.py
@@ -396,7 +396,7 @@ class BeautifulSoup(Tag):
     def __getstate__(self):
         # Frequently a tree builder can't be pickled.
         d = dict(self.__dict__)
-        if 'builder' in d and not self.builder.picklable:
+        if 'builder' in d and d['builder'] is not None and not self.builder.picklable:
             d['builder'] = None
         return d
     
diff --git a/bs4/tests/test_lxml.py b/bs4/tests/test_lxml.py
index 41319d1..396ca0e 100644
--- a/bs4/tests/test_lxml.py
+++ b/bs4/tests/test_lxml.py
@@ -1,5 +1,6 @@
 """Tests to ensure that the lxml tree builder generates good trees."""
 
+import pickle
 import re
 import warnings
 
@@ -185,3 +186,14 @@ class TestLXMLXMLTreeBuilder(SoupTest, XMLTreeBuilderSmokeTest):
         assert soup.find('prefix:tag2').name == 'tag2'
         assert soup.find('prefix:tag3').name == 'tag3'
         assert soup.subtag.find('prefix:tag3').name == 'tag3'
+
+    def test_pickle_removes_builder(self):
+        # The lxml TreeBuilder is not picklable, so it won't be
+        # preserved in a pickle/unpickle operation.
+
+        soup = self.soup("<a>some markup</a>")
+        assert isinstance(soup.builder, self.default_builder)
+        pickled = pickle.dumps(soup)
+        unpickled = pickle.loads(pickled)
+        assert "some markup" == unpickled.a.string
+        assert unpickled.builder is None
diff --git a/bs4/tests/test_soup.py b/bs4/tests/test_soup.py
index 67845cc..2f53a30 100644
--- a/bs4/tests/test_soup.py
+++ b/bs4/tests/test_soup.py
@@ -4,6 +4,7 @@
 from pdb import set_trace
 import logging
 import os
+import pickle
 import pytest
 import sys
 import tempfile
@@ -384,6 +385,24 @@ class TestNewString(SoupTest):
         assert isinstance(s, Comment)
 
 
+class TestPickle(SoupTest):
+   # Test our ability to pickle the BeautifulSoup object itself.
+
+    def test_normal_pickle(self):
+        soup = self.soup("<a>some markup</a>")
+        pickled = pickle.dumps(soup)
+        unpickled = pickle.loads(pickled)
+        assert "some markup" == unpickled.a.string
+        
+    def test_pickle_with_no_builder(self):
+        # We had a bug that prevented pickling from working if
+        # the builder wasn't set.
+        soup = self.soup("some markup")
+        soup.builder = None
+        pickled = pickle.dumps(soup)
+        unpickled = pickle.loads(pickled)
+        assert "some markup" == unpickled.string
+
 class TestEncodingConversion(SoupTest):
     # Test Beautiful Soup's ability to decode and encode from various
     # encodings.