add media type

Author: Matt Sokoloff

labelbox/data/serialization/labelbox_v1/label.py

@@ -212,7 +212,7 @@ def _infer_media_type(self) -> str:
                 #  in each label json payload before converting.
                 #  We will eventually provide the media type in the export.
                 raise TypeError(
-                    "Can't infer data type from row data. row_data: {self.row_data[:200]}"
+                    f"Can't infer data type from row data. row_data: {self.row_data[:200]}"
                 )
 
     def _has_object_annotations(self):

tests/data/assets/labelbox_v1/unkown_media_type_export.json

@@ -0,0 +1 @@
+[{"ID": "ckw3ce1mc78b50zc30dqf0qhj", "DataRow ID": "ckw3cctc41uqg0zo5023e59hn", "Labeled Data": "https://storage.labelbox.com/ckk4q1vgapsau07324awnsjq2%2F8821d3e2-9059-b616-9d4a-9723da3ea073-im1?Expires=1638367029433&KeyName=labelbox-assets-key-3&Signature=FPOQz-alx3gHMK30ib1iPqJj0W0", "Label": {"objects": [{"featureId": "ckw3ce58u00003e66w9rh0onm", "schemaId": "ckw3cdy207b6t0zbn3sh52xoh", "color": "#1CE6FF", "title": "obj", "value": "obj", "polygon": [{"x": 99.405, "y": 56.15}, {"x": 111.421, "y": 99.129}, {"x": 146.082, "y": 80.413}, {"x": 118.815, "y": 47.369}], "instanceURI": "https://api.labelbox.com/masks/feature/ckw3ce58u00003e66w9rh0onm?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2s0cTF2Z3djMHZwMDcwNHhoeDdtNHZrIiwib3JnYW5pemF0aW9uSWQiOiJja2s0cTF2Z2Fwc2F1MDczMjRhd25zanEyIiwiaWF0IjoxNjM3MTU3NDI5LCJleHAiOjE2Mzk3NDk0Mjl9.L4PvjcpSIWV_9R5_M7c_24sj79wtserE_2hkx3ZeCMU"}], "classifications": [], "relationships": []}, "Created By": "msokoloff@labelbox.com", "Project Name": "test", "Created At": "2021-11-17T09:48:56.000Z", "Updated At": "2021-11-17T09:48:56.305Z", "Seconds to Label": 2.239, "External ID": "im1", "Agreement": -1, "Benchmark Agreement": -1, "Benchmark ID": null, "Dataset Name": "no-name", "Reviews": [], "View Label": "https://editor.labelbox.com?project=ckw3cd90b38br0zd4dh4n1xou&label=ckw3ce1mc78b50zc30dqf0qhj", "Has Open Issues": 0, "Skipped": false}, {"ID": "ckw3cea3f7b9t0zbn2tgp2y83", "DataRow ID": "ckw3cctc41uqo0zo5gpma1mr2", "Labeled Data": "https://storage.labelbox.com/ckk4q1vgapsau07324awnsjq2%2F1bc65970-9880-78b4-d298-7a7ef7f8f3fc-im3?Expires=1638367029433&KeyName=labelbox-assets-key-3&Signature=GZUsyQqYYlQPWBYv7GApFYlHXAc", "Label": {"objects": [], "classifications": [{"featureId": "ckw3ced5e00023e66236meh70", "schemaId": "ckw3cdy207b6v0zbn11gp0zz4", "title": "classification", "value": "classification", "answer": {"featureId": "ckw3ced5e00013e6652355ejd", "schemaId": "ckw3cdy207b6w0zbn2hgp3321", "title": "op1", "value": "op_1"}}], "relationships": []}, "Created By": "msokoloff@labelbox.com", "Project Name": "test", "Created At": "2021-11-17T09:49:02.000Z", "Updated At": "2021-11-17T09:49:02.220Z", "Seconds to Label": 5.373, "External ID": "im3", "Agreement": -1, "Benchmark Agreement": -1, "Benchmark ID": null, "Dataset Name": "no-name", "Reviews": [], "View Label": "https://editor.labelbox.com?project=ckw3cd90b38br0zd4dh4n1xou&label=ckw3cea3f7b9t0zbn2tgp2y83", "Has Open Issues": 0, "Skipped": false}, {"ID": "ckw3cec4v78ex0zc3aodwdekw", "DataRow ID": "ckw3cctc41uqs0zo52cy6eus1", "Labeled Data": "https://storage.labelbox.com/ckk4q1vgapsau07324awnsjq2%2Fdb42c0e8-e005-3305-ed35-b021f109b6a7-im4?Expires=1638367029433&KeyName=labelbox-assets-key-3&Signature=Hms9mqcIyWNDzoJUdvMa6_hRKY4", "Label": {"objects": [{"featureId": "ckw3cefl900033e66k41q6zpc", "schemaId": "ckw3cdy207b6t0zbn3sh52xoh", "color": "#1CE6FF", "title": "obj", "value": "obj", "polygon": [{"x": 69.58, "y": 42.292}, {"x": 64.932, "y": 74.128}, {"x": 91.888, "y": 64.601}, {"x": 86.775, "y": 41.828}], "instanceURI": "https://api.labelbox.com/masks/feature/ckw3cefl900033e66k41q6zpc?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiJja2s0cTF2Z3djMHZwMDcwNHhoeDdtNHZrIiwib3JnYW5pemF0aW9uSWQiOiJja2s0cTF2Z2Fwc2F1MDczMjRhd25zanEyIiwiaWF0IjoxNjM3MTU3NDI5LCJleHAiOjE2Mzk3NDk0Mjl9.L4PvjcpSIWV_9R5_M7c_24sj79wtserE_2hkx3ZeCMU"}], "classifications": [{"featureId": "ckw3ceijf00053e669zaplftd", "schemaId": "ckw3cdy207b6v0zbn11gp0zz4", "title": "classification", "value": "classification", "answer": {"featureId": "ckw3ceijf00043e665ex22xkp", "schemaId": "ckw3cdy207b6y0zbn77201rux", "title": "op2", "value": "op_2"}}], "relationships": []}, "Created By": "msokoloff@labelbox.com", "Project Name": "test", "Created At": "2021-11-17T09:49:15.000Z", "Updated At": "2021-11-17T09:49:15.785Z", "Seconds to Label": 5, "External ID": "im4", "Agreement": -1, "Benchmark Agreement": -1, "Benchmark ID": null, "Dataset Name": "no-name", "Reviews": [], "View Label": "https://editor.labelbox.com?project=ckw3cd90b38br0zd4dh4n1xou&label=ckw3cec4v78ex0zc3aodwdekw", "Has Open Issues": 0, "Skipped": false}, {"ID": "ckw3ce1s34c1i0zbp32067q4v", "DataRow ID": "ckw3cctc41uqk0zo52n31egs1", "Labeled Data": "https://storage.labelbox.com/ckk4q1vgapsau07324awnsjq2%2F402cbd62-9127-5b50-57d6-d77aaf89f643-im2?Expires=1638367029433&KeyName=labelbox-assets-key-3&Signature=QIwHFUXN1mjBn8K4ZLWVQGQekmE", "Label": {}, "Created By": "msokoloff@labelbox.com", "Project Name": "test", "Created At": "2021-11-17T09:48:59.000Z", "Updated At": "2021-11-17T09:49:02.000Z", "Seconds to Label": 3.524, "External ID": "im2", "Agreement": -1, "Benchmark Agreement": -1, "Benchmark ID": null, "Dataset Name": "no-name", "Reviews": [], "View Label": "https://editor.labelbox.com?project=ckw3cd90b38br0zd4dh4n1xou&label=ckw3ce1s34c1i0zbp32067q4v", "Has Open Issues": 0, "Skipped": true}]

tests/data/serialization/labelbox_v1/test_unknown_media.py

@@ -0,0 +1,44 @@
+import json
+
+import pytest
+
+from labelbox.data.serialization.labelbox_v1.converter import LBV1Converter
+
+
+def test_image():
+    file_path = 'tests/data/assets/labelbox_v1/unkown_media_type_export.json'
+    with open(file_path, 'r') as file:
+        payload = json.load(file)
+
+    collection = list(LBV1Converter.deserialize(payload))
+    # One of the data rows is broken.
+    assert len(collection) != len(payload)
+
+    for row in payload:
+        row['media_type'] = 'image'
+
+    collection = LBV1Converter.deserialize(payload)
+    for idx, serialized in enumerate(LBV1Converter.serialize(collection)):
+        assert serialized.keys() == payload[idx].keys()
+        for key in serialized:
+            if key != 'Label':
+                assert serialized[key] == payload[idx][key]
+            elif key == 'Label':
+                for annotation_a, annotation_b in zip(
+                        serialized[key]['objects'],
+                        payload[idx][key]['objects']):
+                    if not len(annotation_a['classifications']):
+                        # We don't add a classification key to the payload if there is no classifications.
+                        annotation_a.pop('classifications')
+
+                    if isinstance(annotation_b.get('classifications'),
+                                  list) and len(
+                                      annotation_b['classifications']):
+                        if isinstance(annotation_b['classifications'][0], list):
+                            annotation_b['classifications'] = annotation_b[
+                                'classifications'][0]
+
+                    assert annotation_a == annotation_b
+
+
+# After check the nd serializer on this shit.. It should work for almost everything (except the other horse shit..)