From 3492874d5e8b1922fa836e279ff10c6a6c732d5b Mon Sep 17 00:00:00 2001
From: Clemens Wolff <clewolff@microsoft.com>
Date: Mon, 13 Jan 2020 09:07:50 -0500
Subject: [PATCH] Enable import of CSVs with arbitrary column order

---
 .../tests/data/example_out_of_order_columns.csv   |  4 ++++
 app/api/tests/test_api.py                         | 15 +++++++++++++++
 app/api/utils.py                                  |  5 +++--
 3 files changed, 22 insertions(+), 2 deletions(-)
 create mode 100644 app/api/tests/data/example_out_of_order_columns.csv

diff --git a/app/api/tests/data/example_out_of_order_columns.csv b/app/api/tests/data/example_out_of_order_columns.csv
new file mode 100644
index 00000000..4bba91c8
--- /dev/null
+++ b/app/api/tests/data/example_out_of_order_columns.csv
@@ -0,0 +1,4 @@
+label,foo,text,bar,baz
+Positive,foo1,AAA,barA,baz
+Positive,foo2,BBB,barB,bazz
+Negative,foo3,CCC,barC,bazzz
\ No newline at end of file
diff --git a/app/api/tests/test_api.py b/app/api/tests/test_api.py
index 0a77bb7c..c753e79f 100644
--- a/app/api/tests/test_api.py
+++ b/app/api/tests/test_api.py
@@ -1054,6 +1054,21 @@ class TestUploader(APITestCase):
                                 file_format='csv',
                                 expected_status=status.HTTP_201_CREATED)
 
+    def test_can_upload_classification_csv_with_out_of_order_columns(self):
+        self.upload_test_helper(project_id=self.classification_project.id,
+                                filename='example_out_of_order_columns.csv',
+                                file_format='csv',
+                                expected_status=status.HTTP_201_CREATED)
+
+        self.label_test_helper(
+            project_id=self.classification_project.id,
+            expected_labels=[
+                {'text': 'Positive'},
+                {'text': 'Negative'},
+            ],
+            expected_label_keys=[],
+        )
+
     def test_can_upload_csv_with_non_utf8_encoding(self):
         self.upload_test_helper(project_id=self.classification_project.id,
                                 filename='example.utf16.csv',
diff --git a/app/api/utils.py b/app/api/utils.py
index 396a66b4..f40a0137 100644
--- a/app/api/utils.py
+++ b/app/api/utils.py
@@ -356,8 +356,9 @@ class ExcelParser(FileParser):
                 data.append({'text': row[0]})
             # Text, labels and metadata columns
             elif len(row) == len(columns) and len(row) >= 2:
-                text, label = row[:2]
-                meta = json.dumps(dict(zip(columns[2:], row[2:])))
+                datum = dict(zip(columns, row))
+                text, label = datum.pop('text'), datum.pop('label')
+                meta = json.dumps(datum)
                 j = {'text': text, 'labels': [label], 'meta': meta}
                 data.append(j)
             else: