Fix text coming in reverse order with boxes flow disabled (#399)

Closes #398

CHANGELOG.md

@@ -9,6 +9,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 - Allow boxes_flow LAParam to be passed as None, validate the input, and update documentation ([#395](https://github.com/pdfminer/pdfminer.six/pull/395))
 
 ### Fixed
+- Text no longer comes in reverse order when advanced layout analysis is disabled ([#398](https://github.com/pdfminer/pdfminer.six/pull/398))
 - Updated misleading documentation for `word_margin` and `char_margin` ([#407](https://github.com/pdfminer/pdfminer.six/pull/407))
 - Ignore ValueError when converting font encoding differences ([#389](https://github.com/pdfminer/pdfminer.six/pull/389))
 - Grouping of text lines outside of parent container bounding box ([#386](https://github.com/pdfminer/pdfminer.six/pull/386))

pdfminer/layout.py

@@ -800,9 +800,9 @@ class LTLayoutContainer(LTContainer):
         if laparams.boxes_flow is None:
             def getkey(box):
                 if isinstance(box, LTTextBoxVertical):
-                    return (0, -box.x1, box.y0)
+                    return (0, -box.x1, -box.y0)
                 else:
-                    return (1, box.y0, box.x0)
+                    return (1, -box.y0, box.x0)
             textboxes.sort(key=getkey)
         else:
             self.groups = self.group_textboxes(laparams, textboxes)

tests/test_highlevel_extracttext.py

@@ -2,11 +2,14 @@ import unittest
 
 from helpers import absolute_sample_path
 from pdfminer.high_level import extract_text
+from pdfminer.layout import LAParams
 
 
-def run_with_string(sample_path):
+def run_with_string(sample_path, laparams=None):
+    if laparams is None:
+        laparams = {}
     absolute_path = absolute_sample_path(sample_path)
-    s = extract_text(absolute_path)
+    s = extract_text(absolute_path, laparams=LAParams(**laparams))
     return s
 
 
@@ -21,6 +24,9 @@ test_strings = {
     "simple1.pdf": "Hello \n\nWorld\n\nHello \n\nWorld\n\n"
                    "H e l l o  \n\nW o r l d\n\n"
                    "H e l l o  \n\nW o r l d\n\n\f",
+    "simple1.pdf_no_boxes_flow": "Hello \nWorld\nHello \nWorld\n"
+                                 "H e l l o  \nW o r l d\n"
+                                 "H e l l o  \nW o r l d\n\f",
     "simple2.pdf": "\f",
     "simple3.pdf": "Hello\n\nHello\n\nWorld\n\nWorld\n\n\f",
 }
@@ -32,6 +38,11 @@ class TestExtractText(unittest.TestCase):
         s = run_with_string(test_file)
         self.assertEqual(s, test_strings[test_file])
 
+    def test_simple1_no_boxes_flow(self):
+        test_file = "simple1.pdf"
+        s = run_with_string(test_file, laparams={"boxes_flow": None})
+        self.assertEqual(s, test_strings["simple1.pdf_no_boxes_flow"])
+
     def test_simple2_with_string(self):
         test_file = "simple2.pdf"
         s = run_with_string(test_file)