--extra feature, closes #38

Author: simonw

README.md

@@ -66,6 +66,8 @@ Use `--show-unchanged` to include full details of the unchanged values for rows
         Unchanged:
           name: "Cleo"
 
+### JSON output
+
 You can use the `--json` option to get a machine-readable difference:
 
     $ csv-diff one.csv two.csv --key=id --json
@@ -99,6 +101,29 @@ You can use the `--json` option to get a machine-readable difference:
         "columns_removed": []
     }
 
+### Adding templated extras
+
+You can specify additional keys to be displayed in the human-readable format using the `--extra` option:
+
+    --extra name "Python format string with {id} for variables"
+
+For example, to output a link to `https://news.ycombinator.com/latest?id={id}` for each item with an ID, you could use this:
+
+```bash
+csv-diff one.csv two.csv --key=id \
+  --extra latest "https://news.ycombinator.com/latest?id={id}"
+```
+These extras display something like this:
+```
+1 row changed
+
+  id: 41459472
+    points: "24" => "25"
+    numComments: "5" => "6"
+  extras:
+    latest: https://news.ycombinator.com/latest?id=41459472
+```
+
 ## As a Python library
 
 You can also import the Python library into your own code like so:

csv_diff/__init__.py

@@ -108,7 +108,7 @@ def compare(previous, current, show_unchanged=False):
     return result
 
 
-def human_text(result, key=None, singular=None, plural=None, show_unchanged=False):
+def human_text(result, key=None, singular=None, plural=None, current=None, extras=None):
     singular = singular or "row"
     plural = plural or "rows"
     title = []
@@ -151,6 +151,9 @@ def human_text(result, key=None, singular=None, plural=None, show_unchanged=Fals
                 block.append(
                     '    {}: "{}" => "{}"'.format(field, prev_value, current_value)
                 )
+            if extras:
+                current_item = current[details["key"]]
+                block.append(human_extras(current_item, extras))
             block.append("")
             change_blocks.append("\n".join(block))
             if details.get("unchanged"):
@@ -170,7 +173,10 @@ def human_text(result, key=None, singular=None, plural=None, show_unchanged=Fals
             summary.append(fragment + "\n")
         rows = []
         for row in result["added"]:
-            rows.append(human_row(row, prefix="  "))
+            to_append = human_row(row, prefix="  ")
+            if extras:
+                to_append += "\n" + human_extras(row, extras)
+            rows.append(to_append)
         summary.append("\n\n".join(rows))
         summary.append("")
     if result["removed"]:
@@ -182,7 +188,10 @@ def human_text(result, key=None, singular=None, plural=None, show_unchanged=Fals
             summary.append(fragment + "\n")
         rows = []
         for row in result["removed"]:
-            rows.append(human_row(row, prefix="  "))
+            to_append = human_row(row, prefix="  ")
+            if extras:
+                to_append += "\n" + human_extras(row, extras)
+            rows.append(to_append)
         summary.append("\n\n".join(rows))
         summary.append("")
     return (", ".join(title) + "\n\n" + ("\n".join(summary))).strip()
@@ -193,3 +202,11 @@ def human_row(row, prefix=""):
     for key, value in row.items():
         bits.append("{}{}: {}".format(prefix, key, value))
     return "\n".join(bits)
+
+
+def human_extras(row, extras):
+    bits = []
+    bits.append("  extras:")
+    for key, fmt in extras:
+        bits.append("    {}: {}".format(key, fmt.format(**row)))
+    return "\n".join(bits)

csv_diff/cli.py

@@ -42,13 +42,26 @@
     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):
+@click.option(
+    "extras",
+    "--extra",
+    type=(str, str),
+    multiple=True,
+    help="key: format string - define extra fields to display",
+)
+def cli(previous, current, key, format, json, singular, plural, show_unchanged, extras):
     "Diff two CSV or JSON files"
     dialect = {
         "csv": "excel",
         "tsv": "excel-tab",
     }
 
+    if extras and json:
+        raise click.UsageError(
+            "Extra fields are not supported in JSON output mode",
+            ctx=click.get_current_context(),
+        )
+
     def load(filename):
         if format == "json":
             return load_json(open(filename), key=key)
@@ -57,8 +70,13 @@ def load(filename):
                 open(filename, newline=""), key=key, dialect=dialect.get(format)
             )
 
-    diff = compare(load(previous), load(current), show_unchanged)
+    previous_data = load(previous)
+    current_data = load(current)
+
+    diff = compare(previous_data, current_data, show_unchanged)
     if json:
         print(std_json.dumps(diff, indent=4))
     else:
-        print(human_text(diff, key, singular, plural))
+        print(
+            human_text(diff, key, singular, plural, current=current_data, extras=extras)
+        )

tests/test_cli.py

@@ -234,3 +234,69 @@ def test_semicolon_delimited(tmpdir):
         "columns_added": [],
         "columns_removed": [],
     } == json.loads(result.output.strip())
+
+
+def test_diff_with_extras(tmpdir):
+    one = tmpdir / "one.json"
+    two = tmpdir / "two.json"
+    one.write(
+        json.dumps(
+            [
+                {"id": 1, "name": "Cleo", "type": "dog"},
+                {"id": 2, "name": "Suna", "type": "chicken"},
+            ]
+        )
+    )
+    two.write(
+        json.dumps(
+            [
+                {"id": 2, "name": "Suna", "type": "pretty chicken"},
+                {"id": 3, "name": "Artie", "type": "bunny"},
+            ]
+        )
+    )
+    result = CliRunner().invoke(
+        cli.cli,
+        [
+            str(one),
+            str(two),
+            "--key",
+            "id",
+            "--format",
+            "json",
+            "--extra",
+            "search",
+            "https://www.google.com/search?q={name}",
+        ],
+        catch_exceptions=False,
+    )
+    assert result.exit_code == 0
+    expected = dedent(
+        """
+    1 row changed, 1 row added, 1 row removed
+
+    1 row changed
+
+      id: 2
+        type: "chicken" => "pretty chicken"
+      extras:
+        search: https://www.google.com/search?q=Suna
+
+    1 row added
+
+      id: 3
+      name: Artie
+      type: bunny
+      extras:
+        search: https://www.google.com/search?q=Artie
+
+    1 row removed
+
+      id: 1
+      name: Cleo
+      type: dog
+      extras:
+        search: https://www.google.com/search?q=Cleo
+    """
+    ).strip()
+    assert result.output.strip() == expected