Added support for JSON inputs, closes #12

README.md

@@ -5,7 +5,7 @@
 [![Tests](https://github.com/simonw/csv-diff/workflows/Test/badge.svg)](https://github.com/simonw/csv-diff/actions?query=workflow%3ATest)
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/simonw/csv-diff/blob/main/LICENSE)
 
-Tool for viewing the difference between two CSV files. See [Generating a commit log for San Francisco’s official list of trees](https://simonwillison.net/2019/Mar/13/tree-history/) (and the [sf-tree-history repo commit log](https://github.com/simonw/sf-tree-history/commits)) for background information on this project.
+Tool for viewing the difference between two CSV, TSV or JSON files. See [Generating a commit log for San Francisco’s official list of trees](https://simonwillison.net/2019/Mar/13/tree-history/) (and the [sf-tree-history repo commit log](https://github.com/simonw/sf-tree-history/commits)) for background information on this project.
 
 ## Installation
 
@@ -53,6 +53,8 @@ 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`.
 
+You can also feed it JSON files, provided they are a JSON array of objects where each object has the same keys. Use `--format=json` if your input files are JSON.
+
 Use `--show-unchanged` to include full details of the unchanged values for rows with at least one change in the diff output:
 
     % csv-diff one.csv two.csv --key=id --show-unchanged

csv_diff/__init__.py

@@ -26,6 +26,26 @@ def load_csv(fp, key=None, dialect=None):
     return {keyfn(r): r for r in rows}
 
 
+def load_json(fp, key=None):
+    raw_list = json.load(fp)
+    assert isinstance(raw_list, list)
+    if key:
+        keyfn = lambda r: r[key]
+    else:
+        keyfn = lambda r: hashlib.sha1(
+            json.dumps(r, sort_keys=True).encode("utf8")
+        ).hexdigest()
+    return {keyfn(r): _simplify_json_row(r) for r in raw_list}
+
+
+def _simplify_json_row(r):
+    # Convert list/dict values into JSON serialized strings
+    for key, value in r.items():
+        if isinstance(value, (dict, tuple, list)):
+            r[key] = json.dumps(value)
+    return r
+
+
 def compare(previous, current, show_unchanged=False):
     result = {
         "added": [],

csv_diff/cli.py

@@ -1,6 +1,6 @@
 import click
 import json as std_json
-from . import load_csv, compare, human_text
+from . import load_csv, load_json, compare, human_text
 
 
 @click.command()
@@ -18,9 +18,9 @@
 )
 @click.option(
     "--format",
-    type=click.Choice(["csv", "tsv"]),
+    type=click.Choice(["csv", "tsv", "json"]),
     default=None,
-    help="Explicitly specify input format (csv, tsv) instead of auto-detecting",
+    help="Explicitly specify input format (csv, tsv, json) instead of auto-detecting",
 )
 @click.option(
     "--json", type=bool, default=False, help="Output changes as JSON", is_flag=True
@@ -43,16 +43,19 @@
     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"
+    "Diff two CSV or JSON files"
     dialect = {
         "csv": "excel",
         "tsv": "excel-tab",
     }
 
     def load(filename):
-        return load_csv(
-            open(filename, newline=""), key=key, dialect=dialect.get(format)
-        )
+        if format == "json":
+            return load_json(open(filename), key=key)
+        else:
+            return load_csv(
+                open(filename, newline=""), key=key, dialect=dialect.get(format)
+            )
 
     diff = compare(load(previous), load(current), show_unchanged)
     if json:

setup.py

@@ -15,7 +15,7 @@ def get_long_description():
 
 setup(
     name="csv-diff",
-    description="Python CLI tool and library for diffing CSV files",
+    description="Python CLI tool and library for diffing CSV and JSON files",
     long_description=get_long_description(),
     long_description_content_type="text/markdown",
     author="Simon Willison",

tests/test_cli.py

@@ -1,7 +1,9 @@
 from click.testing import CliRunner
-from csv_diff import cli
+from csv_diff import cli, load_csv
+import csv
 import pytest
 from .test_csv_diff import ONE, ONE_TSV, TWO, TWO_TSV, THREE, FIVE
+import io
 import json
 from textwrap import dedent
 
@@ -15,6 +17,29 @@ def tsv_files(tmpdir):
     return str(one), str(two)
 
 
+@pytest.fixture
+def json_files(tmpdir):
+    one = tmpdir / "one.json"
+    one.write(
+        json.dumps(
+            [
+                {"id": 1, "name": "Cleo", "nested": {"foo": 3}},
+                {"id": 2, "name": "Pancakes", "nested": {"foo": 3}},
+            ]
+        )
+    )
+    two = tmpdir / "two.json"
+    two.write(
+        json.dumps(
+            [
+                {"id": 1, "name": "Cleo", "nested": {"foo": 3, "bar": 5}},
+                {"id": 2, "name": "Pancakes!", "nested": {"foo": 3}},
+            ]
+        )
+    )
+    return str(one), str(two)
+
+
 def test_human_cli(tmpdir):
     one = tmpdir / "one.csv"
     one.write(ONE)
@@ -101,6 +126,26 @@ def test_tsv_files(tsv_files):
     } == json.loads(result.output.strip())
 
 
+def test_json_files(json_files):
+    one, two = json_files
+    result = CliRunner().invoke(
+        cli.cli,
+        [one, two, "--key", "id", "--json", "--format", "json"],
+        catch_exceptions=False,
+    )
+    assert 0 == result.exit_code
+    assert {
+        "added": [],
+        "removed": [],
+        "changed": [
+            {"key": 1, "changes": {"nested": ['{"foo": 3}', '{"foo": 3, "bar": 5}']}},
+            {"key": 2, "changes": {"name": ["Pancakes", "Pancakes!"]}},
+        ],
+        "columns_added": [],
+        "columns_removed": [],
+    } == json.loads(result.output.strip())
+
+
 def test_sniff_format(tsv_files):
     one, two = tsv_files
     result = CliRunner().invoke(cli.cli, [one, two, "--key", "id", "--json"])