From 276fe498679070b4ca22d213c06c6b859c6c7fdd Mon Sep 17 00:00:00 2001
From: "James R. Barlow" <jim@purplerock.ca>
Date: Fri, 4 Dec 2015 03:07:53 -0800
Subject: [PATCH] Better error messages for input file not found or invalid

Not as good finding a general way to deal with ruffus exceptions, but
better than nil.
---
 ocrmypdf/main.py   | 17 ++++++++++++++++-
 tests/test_main.py | 17 +++++++++++++++++
 2 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/ocrmypdf/main.py b/ocrmypdf/main.py
index 97ca7e0c..d2ce1187 100755
--- a/ocrmypdf/main.py
+++ b/ocrmypdf/main.py
@@ -880,10 +880,16 @@ def available_cpu_count():
     return 1
 
 
+def cleanup_ruffus_error_message(msg):
+    msg = re.sub(r'\s+', r' ', msg, re.MULTILINE)
+    msg = re.sub(r"\((.+?)\)", r'\1', msg, re.MULTILINE)
+    msg = msg.strip()
+    return msg
+
+
 def run_pipeline():
     if not options.jobs or options.jobs == 1:
         options.jobs = available_cpu_count()
-
     try:
         cmdline.run(options)
     except ruffus_exceptions.RethrownJobError as e:
@@ -898,6 +904,15 @@ def run_pipeline():
                 return eval(
                     exc_value,
                     {'ExitCode': ExitCode}, {'exc_value': exc_value})
+            elif exc_name == 'ruffus.ruffus_exceptions.MissingInputFileError':
+                print(cleanup_ruffus_error_message(exc_value))
+                return ExitCode.input_file
+            elif exc_name == 'builtins.TypeError':
+                if task_name == 'split_pages':
+                    print("Input file '{0}' is not a valid PDF".format(
+                        options.input_file))
+                    return ExitCode.input_file
+
         return ExitCode.other_error
 
     if not validate_pdfa(options.output_file, _log):
diff --git a/tests/test_main.py b/tests/test_main.py
index 14dee87c..cfa51ed2 100644
--- a/tests/test_main.py
+++ b/tests/test_main.py
@@ -337,3 +337,20 @@ def test_uppercase_extension():
         os.unlink(_make_input("UPPERCASE.PDF"))
 
 
+def test_input_file_not_found():
+    input_file = "does not exist.pdf"
+    sh, out, err = run_ocrmypdf_sh(
+        _make_input(input_file),
+        _make_output("will not happen.pdf"))
+    assert sh.returncode == ExitCode.input_file
+    assert (input_file in out or input_file in err)
+
+
+def test_input_file_not_a_pdf():
+    input_file = __file__  # Try to OCR this file
+    sh, out, err = run_ocrmypdf_sh(
+        _make_input(input_file),
+        _make_output("will not happen.pdf"))
+    assert sh.returncode == ExitCode.input_file
+    assert (input_file in out or input_file in err)
+