fix: tail issue in prune_unwanted_nodes() (#808)

* remove incorrect tail handling

remove the incorrect tail handling in prune_unwanted_nodes() as the correct handling is already done in delete_element()

* Update unit_tests.py

Author: arcombe012

tests/unit_tests.py

@@ -930,6 +930,12 @@ def test_htmlprocessing():
     processed = trafilatura.htmlprocessing.handle_textnode(node, options)
     assert processed.tail == "tail" and processed.text == "text"
 
+    # fix for bug 807
+    node = html.fragment_fromstring("<div><p><span>span</span> span tail</p> p tail </div>")
+    assert node.text_content() == "span span tail p tail "
+    prune = etree.XPath(".//span")
+    processed = trafilatura.htmlprocessing.prune_unwanted_nodes(node, [prune])
+    assert node.text_content() == " span tail p tail "
 
 
 def test_extraction_options():

trafilatura/htmlprocessing.py

@@ -103,13 +103,7 @@ def prune_unwanted_nodes(
     for expression in nodelist:
         for subtree in expression(tree):
             # preserve tail text from deletion
-            if subtree.tail is not None:
-                prev = subtree.getprevious()
-                if prev is None:
-                    prev = subtree.getparent()
-                if prev is not None:
-                    # There is a previous node, append text to its tail
-                    prev.tail = (prev.tail or "") + " " + subtree.tail
+            # tail is by default preserved by delete_element()
             # remove the node
             delete_element(subtree)