v10.3.1 release notes

docs/release_notes.rst

@@ -16,6 +16,12 @@ Note that it is licensed under GPLv3, so scripts that
 ``import ocrmypdf`` and are released publicly should probably also be
 licensed under GPLv3.
 
+v10.3.1
+=======
+
+-  Fixed a number of test suite failures with pdfminer.six older than veresion 20200420.
+-  Enabled support for pdfminer.six 20200720.
+
 v10.3.0
 =======
 

setup.py

@@ -83,7 +83,7 @@ setup(
         'cffi >= 1.9.1',  # must be a setup and install requirement
         'coloredlogs >= 14.0',  # strictly optional
         'img2pdf >= 0.3.0, < 0.4',  # pure Python, so track HEAD closely
-        'pdfminer.six >= 20191110, <= 20200517',
+        'pdfminer.six >= 20191110, <= 20200720',
         'pikepdf >= 1.14.0, < 2',
         'Pillow >= 7.0.0',
         'pluggy >= 0.13.0',

src/ocrmypdf/pdfinfo/layout.py

@@ -220,8 +220,16 @@ class TextPositionTracker(PDFLayoutAnalyzer):
 
 def get_page_analysis(infile, pageno, pscript5_mode):
     rman = pdfminer.pdfinterp.PDFResourceManager(caching=True)
+    if pdfminer.__version__ < '20200402':
+        # Workaround for https://github.com/pdfminer/pdfminer.six/issues/395
+        disable_boxes_flow = 2
+    else:
+        disable_boxes_flow = None
     dev = TextPositionTracker(
-        rman, laparams=LAParams(all_texts=True, detect_vertical=True, boxes_flow=None)
+        rman,
+        laparams=LAParams(
+            all_texts=True, detect_vertical=True, boxes_flow=disable_boxes_flow
+        ),
     )
     interp = pdfminer.pdfinterp.PDFPageInterpreter(rman, dev)