Exclude a list of paths from diff. (#81)

* Exclude a list of paths from diff.

Gives the ability of excluding a list of jsonpaths (e.g. datetime, uuids, etc) from the diff.

---------

Co-authored-by: ppirooznia <payam.pirooznia@viasat.com>

README.md

@@ -25,6 +25,10 @@ Diff JSON and JSON-like structures in Python.
 >>> diff({'a': [0, {'b': 4}, 1]}, {'a': [0, {'b': 5}, 1]})
 {'a': {1: {'b': 5}}}
 
+# You can exclude some jsonpaths from the diff (doesn't work if the value types are different)
+>>> diff({'a': 1, 'b': {'b1': 20, 'b2': 21}, 'c': 3},  {'a': 1, 'b': {'b1': 22, 'b2': 23}, 'c': 30}, exclude_paths=['b.b1', 'c'])
+{'b': {'b2': 23}}
+
 # ...but similarity is taken into account
 >>> diff({'a': [0, {'b': 4}, 1]}, {'a': [0, {'c': 5}, 1]})
 {'a': {insert: [(1, {'c': 5})], delete: [1]}}

jsondiff/__init__.py

@@ -891,7 +891,7 @@ def _set_diff(self, a, b):
         s = s_common / n_tot if n_tot != 0 else 1.0
         return self.options.syntax.emit_set_diff(a, b, s, added, removed), s
 
-    def _dict_diff(self, a, b):
+    def _dict_diff(self, a, b, exclude_paths, path):
         """
         Computes the difference between two dictionaries.
         """
@@ -903,32 +903,42 @@ def _dict_diff(self, a, b):
         added = {}
         changed = {}
         for k, v in a.items():
+            new_path = f'{path}.{k}' if path else k
+            if new_path in exclude_paths:
+                continue
             w = b.get(k, missing)
             if w is missing:
                 nremoved += 1
                 removed[k] = v
             else:
                 nmatched += 1
-                d, s = self._obj_diff(v, w)
+                d, s = self._obj_diff(v, w, exclude_paths, new_path)
                 if s < 1.0:
                     changed[k] = d
                 smatched += 0.5 + 0.5 * s
         for k, v in b.items():
             if k not in a:
+                new_path = f'{path}.{k}' if path else k
+                if new_path in exclude_paths:
+                    continue
                 nadded += 1
                 added[k] = v
         n_tot = nremoved + nmatched + nadded
         s = smatched / n_tot if n_tot != 0 else 1.0
         return self.options.syntax.emit_dict_diff(a, b, s, added, changed, removed), s
 
-    def _obj_diff(self, a, b):
+    def _obj_diff(self, a, b, exclude_paths=None, path=''):
         """
         Computes the difference between any two JSON-compatible objects.
         """
+        if not exclude_paths:
+            exclude_paths = []
+        if path in exclude_paths:
+            return {}, 1.0
         if a is b:
             return self.options.syntax.emit_value_diff(a, b, 1.0), 1.0
         if isinstance(a, dict) and isinstance(b, dict):
-            return self._dict_diff(a, b)
+            return self._dict_diff(a, b, exclude_paths, path)
         elif isinstance(a, tuple) and isinstance(b, tuple):
             return self._list_diff(a, b)
         elif isinstance(a, list) and isinstance(b, list):
@@ -940,15 +950,21 @@ def _obj_diff(self, a, b):
         else:
             return self.options.syntax.emit_value_diff(a, b, 1.0), 1.0
 
-    def diff(self, a, b, fp=None):
+    def diff(self, a, b, fp=None, exclude_paths: list = None) -> dict:
         """
         Computes the difference between two JSON structures.
+        :param a: The original JSON structure.
+        :param b: The modified JSON structure.
+        :param fp: Optional file pointer to dump the diff to.
+        :param exclude_paths: Optional list of string paths to exclude from the diff.
         """
+        if not exclude_paths:
+            exclude_paths = []
         if self.options.load:
             a = self.options.loader(a)
             b = self.options.loader(b)
 
-        d, s = self._obj_diff(a, b)
+        d, s = self._obj_diff(a, b, exclude_paths)
 
         if self.options.marshal or self.options.dump:
             d = self.marshal(d)

tests/test_jsondiff.py

@@ -318,3 +318,16 @@ def test_yaml_dump_string_fp(self):
         buffer = io.StringIO()
         dumper(data, buffer)
         self.assertEqual(expected, buffer.getvalue())
+
+    def test_exclude_paths(self):
+        differ = JsonDiffer()
+
+        a = {'a': 1, 'b': {'b1': 20, 'b2': 21}, 'c': 3}
+        b = {'a': 1, 'b': {'b1': 22, 'b2': 23}, 'c': 30}
+
+        exclude_paths = ['b.b1', 'c']
+
+        d = differ.diff(a, b, exclude_paths=exclude_paths)
+
+        # The diff should only contain changes that are not in the exclude_paths
+        self.assertEqual({'b': {'b2': 23}}, d)