Test price (#163)

* Fix price after merge * Bump coverage
scrapinghub · Sep 5, 2019 · ced8822 · ced8822
1 parent a47e551
commit ced8822
Show file tree

Hide file tree

Showing 3 changed files with 48 additions and 16 deletions.
diff --git a/src/arche/rules/price.py b/src/arche/rules/price.py
@@ -35,7 +35,7 @@ def compare_was_now(df: pd.DataFrame, tagged_fields: TaggedFields):
         result.add_error(
             f"{price_less_percent} ({len(df_prices_less)}) of "
             f"items with {price_was_field} < {price_field}",
-            errors={error: list(df_prices_less.index)},
+            errors={error: set(df_prices_less.index)},
         )
 
     df_prices_equals = pd.DataFrame(
@@ -52,15 +52,14 @@ def compare_was_now(df: pd.DataFrame, tagged_fields: TaggedFields):
             ),
             errors=(
                 {
-                    f"Prices equal for {len(df_prices_equals)} items": list(
+                    f"Prices equal for {len(df_prices_equals)} items": set(
                         df_prices_equals.index
                     )
                 }
             ),
         )
 
     result.items_count = len(df.index)
-
     return result
 
 
@@ -98,7 +97,7 @@ def compare_prices_for_same_urls(
 
     errors = {}
     for url, group in missing_urls.groupby(missing_urls):
-        errors[f"Missing {url}"] = pd.Series(group.index)
+        errors[f"Missing {url}"] = set(group.index)
 
     if not missing_urls.empty:
         result.add_info(
@@ -207,12 +206,6 @@ def compare_prices_for_same_names(
         return result
 
     name_field = name_field[0]
-
-    product_url_field = tagged_fields.get("product_url_field")
-    if not product_url_field:
-        result.add_info("product_url_field tag is not set")
-    else:
-        product_url_field = product_url_field[0]
     source_df = source_df[source_df[name_field].notnull()]
     target_df = target_df[target_df[name_field].notnull()]
 
@@ -228,13 +221,13 @@ def compare_prices_for_same_names(
 
     errors = {}
     for name, group in missing_names.groupby(missing_names):
-        errors[f"Missing {name}"] = pd.Series(group.index)
+        errors[f"Missing {name}"] = set(group.index)
 
-    if missing_names:
+    if not missing_names.empty:
         result.add_info(
             f"{len(missing_names)} names missing from the tested job", errors=errors
         )
-    if new_names:
+    if not new_names.empty:
         result.add_info(f"{len(new_names)} new names in the tested job")
     result.add_info(f"{len(same_names)} same names in both jobs")
 

diff --git a/tests/rules/test_price.py b/tests/rules/test_price.py
@@ -21,14 +21,14 @@
                 (
                     "22.22% (2) of items with original_price < sale_price",
                     None,
-                    {"Past price is less than current for 2 items": [0, 8]},
+                    {"Past price is less than current for 2 items": {0, 8}},
                 )
             ],
             Level.WARNING: [
                 (
                     "11.11% (1) of items with original_price = sale_price",
                     None,
-                    {"Prices equal for 1 items": [1]},
+                    {"Prices equal for 1 items": {1}},
                 )
             ],
         },
@@ -124,3 +124,42 @@ def test_compare_names_for_same_urls(
         pd.DataFrame(source_data), pd.DataFrame(target_data), tagged_fields
     )
     assert result == create_result("Compare Names Per Url", expected_messages)
+
+
+@pytest.mark.parametrize(
+    "source_data, target_data, tagged_fields, expected_messages",
+    [
+        (
+            {"name": ["Coffee", "Tea", "Juice"], "price": [3.0, 5.0, 2.0]},
+            {"name": ["Coffee", "Tea", "Wine"], "price": [4.0, 4.8, 20.0]},
+            {"name_field": ["name"], "product_price_field": ["price"]},
+            {
+                Level.INFO: [
+                    (
+                        "1 names missing from the tested job",
+                        None,
+                        {"Missing Wine": {2}},
+                    ),
+                    ("1 new names in the tested job",),
+                    ("2 same names in both jobs",),
+                ],
+                Level.ERROR: [
+                    (
+                        "2 checked, 1 errors",
+                        (
+                            "different price for Coffee\nsource price is 3.0 for 0\n"
+                            "target price is 4.0 for 0"
+                        ),
+                    )
+                ],
+            },
+        )
+    ],
+)
+def test_compare_prices_for_same_names(
+    source_data, target_data, tagged_fields, expected_messages
+):
+    result = p.compare_prices_for_same_names(
+        pd.DataFrame(source_data), pd.DataFrame(target_data), tagged_fields
+    )
+    assert result == create_result("Compare Prices For Same Names", expected_messages)
diff --git a/tox.ini b/tox.ini
@@ -53,4 +53,4 @@ mock_use_standalone_module = true
 branch = true
 
 [coverage:report]
-fail_under = 78
+fail_under = 81