--show-unchanged option, closes #9

Author: simonw

README.md

@@ -52,6 +52,17 @@ The `--key=id` option means that the `id` column should be treated as the unique
 
 The tool will automatically detect if your files are comma- or tab-separated. You can over-ride this automatic detection and force the tool to use a specific format using `--format=tsv` or `--format=csv`.
 
+Use `--show-unchanged` to include full details of the unchanged rows in the diff output:
+
+    % csv-diff one.csv two.csv --key=id --show-unchanged
+    1 row changed
+
+      id: 1
+        age: "4" => "5"
+
+        Unchanged:
+          name: "Cleo"
+
 You can use the `--json` option to get a machine-readable difference:
 
     $ csv-diff one.csv two.csv --key=id --json

csv_diff/__init__.py

@@ -24,7 +24,7 @@ def load_csv(fp, key=None, dialect=None):
     return {keyfn(r): r for r in rows}
 
 
-def compare(previous, current):
+def compare(previous, current, show_unchanged=False):
     result = {
         "added": [],
         "removed": [],
@@ -59,19 +59,24 @@ def compare(previous, current):
         for id in changed:
             d = list(diff(previous[id], current[id], ignore=ignore_columns))
             if d:
-                result["changed"].append(
-                    {
-                        "key": id,
-                        "changes": {
-                            field: [prev_value, current_value]
-                            for _, field, (prev_value, current_value) in d
-                        },
+                changes = {
+                    "key": id,
+                    "changes": {
+                        field: [prev_value, current_value]
+                        for _, field, (prev_value, current_value) in d
+                    },
+                }
+                if show_unchanged:
+                    changes["unchanged"] = {
+                        field: value
+                        for field, value in previous[id].items()
+                        if field not in changes["changes"] and field != "id"
                     }
-                )
+                result["changed"].append(changes)
     return result
 
 
-def human_text(result, key=None, singular=None, plural=None):
+def human_text(result, key=None, singular=None, plural=None, show_unchanged=False):
     singular = singular or "row"
     plural = plural or "rows"
     title = []
@@ -116,6 +121,13 @@ def human_text(result, key=None, singular=None, plural=None):
                 )
             block.append("")
             change_blocks.append("\n".join(block))
+            if details.get("unchanged"):
+                block = []
+                block.append("    Unchanged:")
+                for field, value in details["unchanged"].items():
+                    block.append('      {}: "{}"'.format(field, value))
+                block.append("")
+                change_blocks.append("\n".join(block))
         summary.append("\n".join(change_blocks))
     if result["added"]:
         fragment = "{} {} added".format(

csv_diff/cli.py

@@ -37,7 +37,12 @@
     default=None,
     help="Plural word to use, e.g. 'trees' for '2 trees'",
 )
-def cli(previous, current, key, format, json, singular, plural):
+@click.option(
+    "--show-unchanged",
+    is_flag=True,
+    help="Show unchanged fields for rows with at least one change",
+)
+def cli(previous, current, key, format, json, singular, plural, show_unchanged):
     "Diff two CSV files"
     dialect = {
         "csv": "excel",
@@ -49,7 +54,7 @@ def load(filename):
             open(filename, newline=""), key=key, dialect=dialect.get(format)
         )
 
-    diff = compare(load(previous), load(current))
+    diff = compare(load(previous), load(current), show_unchanged)
     if json:
         print(std_json.dumps(diff, indent=4))
     else:

tests/test_human_text.py

@@ -21,6 +21,28 @@ def test_row_changed():
     )
 
 
+def test_row_changed_show_unchanged():
+    diff = compare(
+        load_csv(io.StringIO(ONE), key="id"),
+        load_csv(io.StringIO(TWO), key="id"),
+        show_unchanged=True,
+    )
+    assert (
+        dedent(
+            """
+    1 row changed
+
+      id: 1
+        age: "4" => "5"
+
+        Unchanged:
+          name: "Cleo"
+    """
+        ).strip()
+        == human_text(diff, "id")
+    )
+
+
 def test_row_added():
     diff = compare(
         load_csv(io.StringIO(THREE), key="id"), load_csv(io.StringIO(TWO), key="id")