allow no tags in XML input

ianalyzer_readers/extract.py

@@ -274,6 +274,8 @@ class XML(Extractor):
         tags:
             Tags to select. Each of these can be a `Tag` object, or a callable that
             takes the document metadata as input and returns a `Tag`.
+
+            If no tags are provided, the extractor will work form the starting tag.
             
             Tags represent a query to select tags from current tag (e.g. the entry tag of
             the document). If you provide multiple, they are chained: each Tag query is
@@ -341,6 +343,8 @@ def _select(self, tags: Iterable[TagSpecification], soup: bs4.PageElement, metad
             tag = resolve_tag_specification(tags[0], metadata)
             for result in tag.find_in_soup(soup):
                 yield result
+        else:
+            yield soup
 
 
     def _apply(self, soup_top, soup_entry, *nargs, **kwargs):

tests/xml/test_xml_extraction.py

@@ -63,7 +63,7 @@ def test_xml_transform(tmpdir):
 
 
 def test_xml_no_tag(tmpdir):
-    extractor = XML(CurrentTag())
+    extractor = XML()
     reader = make_test_reader(extractor, Tag('play'), Tag('character'), basic_doc, tmpdir)
     assert_extractor_output(reader, 'HAMLET')
 
@@ -80,7 +80,7 @@ def test_xml_no_tag(tmpdir):
 
 
 def test_xml_attribute(tmpdir):
-    extractor = XML(CurrentTag(), attribute='character')
+    extractor = XML(attribute='character')
     reader = make_test_reader(extractor, Tag('play'), Tag('lines'), doc_with_attribute, tmpdir)
     assert_extractor_output(reader, 'HAMLET')
 
@@ -100,7 +100,7 @@ def test_xml_attribute(tmpdir):
 '''
 
 def test_xml_flatten(tmpdir):
-    extractor = XML(CurrentTag(), flatten=True)
+    extractor = XML(flatten=True)
     reader = make_test_reader(extractor, Tag('play'), Tag('lines'), doc_multiline, tmpdir)
     expected = 'My hour is almost come, When I to sulph\'rous and tormenting flames Must render up myself.'
     assert_extractor_output(reader, expected)