From 63a5503d4612d76d54121fbee51eeaa8ad594da7 Mon Sep 17 00:00:00 2001
From: Holly Becker <becker.holly@gmail.com>
Date: Fri, 29 Jan 2016 17:00:45 -0800
Subject: [PATCH] METS: Handle empty rows in metadata.csv

Skip empty rows when parsing a metadata.csv
---
 .../archivematicaCreateMETSMetadataCSV.py     |  3 +++
 src/MCPClient/tests/test_create_aip_mets.py   | 19 +++++++++++++++++++
 2 files changed, 22 insertions(+)

diff --git a/src/MCPClient/lib/clientScripts/archivematicaCreateMETSMetadataCSV.py b/src/MCPClient/lib/clientScripts/archivematicaCreateMETSMetadataCSV.py
index 105fa2c6b2..73aeaa9ee8 100755
--- a/src/MCPClient/lib/clientScripts/archivematicaCreateMETSMetadataCSV.py
+++ b/src/MCPClient/lib/clientScripts/archivematicaCreateMETSMetadataCSV.py
@@ -57,6 +57,7 @@ def parseMetadata(SIPPath):
             print >>sys.stderr, "error parsing: ", metadataCSVFilePath
             traceback.print_exc(file=sys.stderr)
             sharedVariablesAcrossModules.globalErrorCount += 1
+            continue
         # Provide warning if this file already has differing metadata
         # Not using all_metadata.update(csv_metadata) because of that
         for entry, values in csv_metadata.iteritems():
@@ -99,6 +100,8 @@ def parseMetadataCSV(metadataCSVFilePath):
         header = [h.strip() for h in header[1:]]
         # Parse data
         for row in reader:
+            if not row:
+                continue
             entry_name = row[0]
             if entry_name.endswith("/"):
                 entry_name = entry_name[:-1]
diff --git a/src/MCPClient/tests/test_create_aip_mets.py b/src/MCPClient/tests/test_create_aip_mets.py
index cfa1ad7605..89bf6def94 100644
--- a/src/MCPClient/tests/test_create_aip_mets.py
+++ b/src/MCPClient/tests/test_create_aip_mets.py
@@ -348,3 +348,22 @@ def test_parse_metadata_csv_non_ascii(self):
         assert 'objects/foo.jpg' in dc
         assert 'dc.title' in dc['objects/foo.jpg']
         assert dc['objects/foo.jpg']['dc.title'] == [u'元気です'.encode('utf8')]
+
+    def test_parse_metadata_csv_blank_rows(self):
+        # Create metadata.csv
+        data = [
+            ['Filename', 'dc.title', 'dc.type', 'dc.type', 'dc.type'],
+            ['objects/foo.jpg', 'Foo', 'Photograph', 'Still image', 'Picture'],
+            [],
+        ]
+        with open('metadata.csv', 'wb') as f:
+            writer = csv.writer(f)
+            for row in data:
+                writer.writerow(row)
+
+        # Run test
+        dc = archivematicaCreateMETSMetadataCSV.parseMetadataCSV('metadata.csv')
+        # Verify
+        assert dc
+        assert len(dc) == 1
+        assert 'objects/foo.jpg' in dc