add error handling, mypy not clean

Author: shuoweil

bigframes/blob/_functions.py

@@ -262,7 +262,7 @@ def image_blur_to_bytes_func(
         )
         return json.dumps(result_dict)
     else:
-        return result_dict["content"]
+        return base64.b64encode(result_dict["content"]).decode("utf-8")
 
 
 image_blur_to_bytes_def = FunctionDef(
@@ -384,7 +384,7 @@ def image_resize_to_bytes_func(
         )
         return json.dumps(result_dict)
     else:
-        return result_dict["content"]
+        return base64.b64encode(result_dict["content"]).decode("utf-8")
 
 
 image_resize_to_bytes_def = FunctionDef(
@@ -530,7 +530,7 @@ def image_normalize_to_bytes_func(
 
 
 # Extracts all text from a PDF url
-def pdf_extract_func(src_obj_ref_rt: str) -> str:
+def pdf_extract_func(src_obj_ref_rt: str, verbose: bool) -> str:
     try:
         import io
         import json
@@ -563,8 +563,10 @@ def pdf_extract_func(src_obj_ref_rt: str) -> str:
     except Exception as e:
         result_dict = {"status": str(e), "content": ""}
 
-    result_json = json.dumps(result_dict)
-    return result_json
+    if verbose:
+        return json.dumps(result_dict)
+    else:
+        return result_dict["content"]
 
 
 pdf_extract_def = FunctionDef(
@@ -573,7 +575,9 @@ def pdf_extract_func(src_obj_ref_rt: str) -> str:
 
 
 # Extracts text from a PDF url and chunks it simultaneously
-def pdf_chunk_func(src_obj_ref_rt: str, chunk_size: int, overlap_size: int) -> str:
+def pdf_chunk_func(
+    src_obj_ref_rt: str, chunk_size: int, overlap_size: int, verbose: bool
+) -> str:
     try:
         import io
         import json
@@ -619,8 +623,10 @@ def pdf_chunk_func(src_obj_ref_rt: str, chunk_size: int, overlap_size: int) -> s
     except Exception as e:
         result_dict = {"status": str(e), "content": []}
 
-    result_json = json.dumps(result_dict)
-    return result_json
+    if verbose:
+        return json.dumps(result_dict)
+    else:
+        return json.dumps(result_dict["content"])
 
 
 pdf_chunk_def = FunctionDef(

bigframes/operations/blob.py

@@ -422,6 +422,12 @@ def image_blur(
             ).rename("blurred_content")
 
             if verbose:
+                blurred_content_b64_series = res._apply_unary_op(
+                    ops.JSONValue(json_path="$.content")
+                )
+                blurred_content_series = bbq.sql_scalar(
+                    "FROM_BASE64({0})", columns=[blurred_content_b64_series]
+                )
                 blurred_status_series = res._apply_unary_op(
                     ops.JSONValue(json_path="$.status")
                 )
@@ -431,7 +437,10 @@ def image_blur(
                 results_struct = bbq.struct(results_df).rename("blurred_results")
                 return results_struct
             else:
-                return blurred_content_series
+                blurred_bytes = bbq.sql_scalar(
+                    "FROM_BASE64({0})", columns=[res]
+                ).rename("blurred_bytes")
+                return blurred_bytes
 
         if isinstance(dst, str):
             dst = os.path.join(dst, "")
@@ -547,16 +556,26 @@ def image_resize(
             ).rename("resized_content")
 
             if verbose:
+                resized_content_b64_series = res._apply_unary_op(
+                    ops.JSONValue(json_path="$.content")
+                )
+                resized_content_series = bbq.sql_scalar(
+                    "FROM_BASE64({0})", columns=[resized_content_b64_series]
+                )
+
                 resized_status_series = res._apply_unary_op(
                     ops.JSONValue(json_path="$.status")
                 )
                 results_df = bpd.DataFrame(
                     {"status": resized_status_series, "content": resized_content_series}
                 )
-                results_struct = bbq.strcut(results_df).rename("resized_results")
+                results_struct = bbq.struct(results_df).rename("resized_results")
                 return results_struct
             else:
-                return resized_content_series
+                resized_bytes = bbq.sql_scalar(
+                    "FROM_BASE64({0})", columns=[res]
+                ).rename("resized_bytes")
+                return resized_bytes
 
         if isinstance(dst, str):
             dst = os.path.join(dst, "")

tests/system/large/blob/test_function.py

@@ -52,39 +52,50 @@ def images_output_uris(images_output_folder: str) -> list[str]:
 
 
 @pytest.mark.parametrize(
-    "verbose, expected_type",
+    "verbose",
     [
-        (True, "struct"),
-        (False, "json"),
+        (True),
+        (False),
     ],
 )
 def test_blob_exif(
     bq_connection: str,
     session: bigframes.Session,
     verbose: bool,
-    expected_type: str,
 ):
     exif_image_df = session.from_glob_path(
         "gs://bigframes_blob_test/images_exif/*",
         name="blob_col",
         connection=bq_connection,
-        verbose=verbose,
     )
 
     actual = exif_image_df["blob_col"].blob.exif(
-        engine="pillow", connection=bq_connection
-    )
-    expected = bpd.Series(
-        ['{"ExifOffset": 47, "Make": "MyCamera"}'],
-        session=session,
-        dtype=dtypes.JSON_DTYPE,
-    )
-    pd.testing.assert_series_equal(
-        actual.to_pandas(),
-        expected.to_pandas(),
-        check_dtype=False,
-        check_index_type=False,
+        engine="pillow", connection=bq_connection, verbose=verbose
     )
+    if verbose:
+        assert hasattr(actual, "struct")
+        actual_exploded = actual.struct.explode()
+        assert "status" in actual_exploded.columns
+        assert "content" in actual_exploded.columns
+
+        status_series = actual_exploded["status"]
+        assert status_series.dtype == dtypes.STRING_DTYPE
+
+        content_series = actual_exploded["content"]
+        assert content_series.dtype == dtypes.JSON_DTYPE
+
+    else:
+        expected = bpd.Series(
+            ['{"ExifOffset": 47, "Make": "MyCamera"}'],
+            session=session,
+            dtype=dtypes.JSON_DTYPE,
+        )
+        pd.testing.assert_series_equal(
+            actual.to_pandas(),
+            expected.to_pandas(),
+            check_dtype=False,
+            check_index_type=False,
+        )
 
 
 @pytest.mark.parametrize("verbose", [True, False])