rudimentary support for --stats in --dbt --json mode

data_diff/__main__.py

@@ -317,6 +317,7 @@ def main(conf, run, **kw):
                 json_output=kw["json_output"],
                 state=state,
                 where_flag=kw["where"],
+                stats_flag=kw["stats"],
                 columns_flag=kw["columns"],
             )
         else:

data_diff/dbt.py

@@ -62,6 +62,7 @@ class TDiffVars(pydantic.BaseModel):
     include_columns: List[str]
     exclude_columns: List[str]
     dbt_model: Optional[str] = None
+    stats_flag: bool = False
 
 
 def dbt_diff(
@@ -73,6 +74,7 @@ def dbt_diff(
     state: Optional[str] = None,
     log_status_handler: Optional[LogStatusHandler] = None,
     where_flag: Optional[str] = None,
+    stats_flag: bool = False,
     columns_flag: Optional[Tuple[str]] = None,
 ) -> None:
     print_version_info()
@@ -113,7 +115,7 @@ def dbt_diff(
             if log_status_handler:
                 log_status_handler.set_prefix(f"Diffing {model.alias} \n")
 
-            diff_vars = _get_diff_vars(dbt_parser, config, model, where_flag, columns_flag)
+            diff_vars = _get_diff_vars(dbt_parser, config, model, where_flag, stats_flag, columns_flag)
 
             # we won't always have a prod path when using state
             # when the model DNE in prod manifest, skip the model diff
@@ -165,6 +167,7 @@ def _get_diff_vars(
     config: TDatadiffConfig,
     model,
     where_flag: Optional[str] = None,
+    stats_flag: bool = False,
     columns_flag: Optional[Tuple[str]] = None,
 ) -> TDiffVars:
     cli_columns = list(columns_flag) if columns_flag else []
@@ -200,6 +203,7 @@ def _get_diff_vars(
         where_filter=where_flag or datadiff_model_config.where_filter,
         include_columns=cli_columns or datadiff_model_config.include_columns,
         exclude_columns=[] if cli_columns else datadiff_model_config.exclude_columns,
+        stats_flag=stats_flag,
     )
 
 
@@ -338,6 +342,7 @@ def _local_diff(diff_vars: TDiffVars, json_output: bool = False) -> None:
                         "removed": columns_removed,
                         "changed": columns_type_changed,
                     },
+                    stats_only=diff_vars.stats_flag,
                 )
             ),
             flush=True,

data_diff/format.py

@@ -40,6 +40,7 @@ def jsonify(
     dataset2_columns: Columns,
     columns_diff: Dict[str, List[str]],
     with_summary: bool = False,
+    stats_only: bool = False,
 ) -> "JsonDiff":
     """
     Converts the diff result into a JSON-serializable format.
@@ -53,21 +54,31 @@ def jsonify(
     t1_exclusive_rows = []
     t2_exclusive_rows = []
     diff_rows = []
+    rows = None
     schema = [field for field, _ in diff_info.diff_schema]
 
     t1_exclusive_rows, t2_exclusive_rows, diff_rows = _group_rows(diff_info, schema)
 
-    diff_rows_jsonified = []
-    for row in diff_rows:
-        diff_rows_jsonified.append(_jsonify_diff(row, key_columns))
+    if not stats_only:
+        diff_rows_jsonified = []
+        for row in diff_rows:
+            diff_rows_jsonified.append(_jsonify_diff(row, key_columns))
 
-    t1_exclusive_rows_jsonified = []
-    for row in t1_exclusive_rows:
-        t1_exclusive_rows_jsonified.append(_jsonify_exclusive(row, key_columns))
+        t1_exclusive_rows_jsonified = []
+        for row in t1_exclusive_rows:
+            t1_exclusive_rows_jsonified.append(_jsonify_exclusive(row, key_columns))
 
-    t2_exclusive_rows_jsonified = []
-    for row in t2_exclusive_rows:
-        t2_exclusive_rows_jsonified.append(_jsonify_exclusive(row, key_columns))
+        t2_exclusive_rows_jsonified = []
+        for row in t2_exclusive_rows:
+            t2_exclusive_rows_jsonified.append(_jsonify_exclusive(row, key_columns))
+
+        rows = RowsDiff(
+            exclusive=ExclusiveDiff(
+                dataset1=t1_exclusive_rows_jsonified,
+                dataset2=t2_exclusive_rows_jsonified
+            ),
+            diff=diff_rows_jsonified,
+        )
 
     summary = None
     if with_summary:
@@ -87,10 +98,7 @@ def jsonify(
         model=dbt_model,
         dataset1=list(table1.table_path),
         dataset2=list(table2.table_path),
-        rows=RowsDiff(
-            exclusive=ExclusiveDiff(dataset1=t1_exclusive_rows_jsonified, dataset2=t2_exclusive_rows_jsonified),
-            diff=diff_rows_jsonified,
-        ),
+        rows=rows,
         summary=summary,
         columns=columns,
     ).json()
@@ -228,7 +236,7 @@ class JsonDiff:
     model: str
     dataset1: List[str]
     dataset2: List[str]
-    rows: RowsDiff
+    rows: Optional[RowsDiff]
     summary: Optional[JsonDiffSummary]
     columns: Optional[JsonColumnsSummary]