Fix regression in page layout that sometimes returned text lines out of order

CHANGELOG.md

@@ -10,6 +10,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 - `PermissionError` when creating temporary filepaths on windows when running tests ([#469](https://github.com/pdfminer/pdfminer.six/issues/469))
 - Detecting trailer correctly when surrounded with needless whitespace ([#535](https://github.com/pdfminer/pdfminer.six/pull/535))
 - Fix `.paint_path` logic for handling single line segments and extracting point-on-curve positions of Beziér path commands ([#530](https://github.com/pdfminer/pdfminer.six/pull/530))
+- Fix regression in `LTLayoutContainer.group_textboxes` (introduced in 20191107) that sometimes returned text lines out of order ([#659](https://github.com/pdfminer/pdfminer.six/pull/659))
 
 ## Removed
 - Support for Python 3.4 and 3.5 ([#522](https://github.com/pdfminer/pdfminer.six/pull/522))

pdfminer/layout.py

@@ -765,7 +765,7 @@ def isany(obj1, obj2):
             (skip_isany, d, id1, id2, obj1, obj2) = heapq.heappop(dists)
             # Skip objects that are already merged
             if (id1 not in done) and (id2 not in done):
-                if skip_isany and isany(obj1, obj2):
+                if not skip_isany and isany(obj1, obj2):
                     heapq.heappush(dists, (True, d, id1, id2, obj1, obj2))
                     continue
                 if isinstance(obj1, (LTTextBoxVertical, LTTextGroupTBRL)) or \

tests/test_highlevel_extracttext.py

@@ -30,7 +30,11 @@ def run_with_file(sample_path):
     "simple2.pdf": "\f",
     "simple3.pdf": "Hello\n\nHello\nあ\nい\nう\nえ\nお\nあ\nい\nう\nえ\nお\n"
                    "World\n\nWorld\n\n\f",
-    "simple4.pdf": "Text1\nText2\nText3\n\n\f"
+    "simple4.pdf": "Text1\nText2\nText3\n\n\f",
+    "simple5.pdf": "Heading\n\n"
+                   "Link to heading that is working with vim-pandoc.\n\n"
+                   "Link to heading “that is” not working with vim-pandoc.\n\n"
+                   "Subheading\n\nSome “more text”\n\n1\n\n\f"
 }
 
 
@@ -60,6 +64,11 @@ def test_simple4_with_string(self):
         s = run_with_string(test_file)
         self.assertEqual(s, test_strings[test_file])
 
+    def test_simple5_with_string(self):
+        test_file = "simple5.pdf"
+        s = run_with_string(test_file)
+        self.assertEqual(s, test_strings[test_file])
+
     def test_simple1_with_file(self):
         test_file = "simple1.pdf"
         s = run_with_file(test_file)
@@ -80,6 +89,11 @@ def test_simple4_with_file(self):
         s = run_with_file(test_file)
         self.assertEqual(s, test_strings[test_file])
 
+    def test_simple5_with_file(self):
+        test_file = "simple5.pdf"
+        s = run_with_file(test_file)
+        self.assertEqual(s, test_strings[test_file])
+
 
 class TestExtractPages(unittest.TestCase):
     def _get_test_file_path(self):