Pull in _NodeBase from actorgraph; include short-hash in repr

ffmpeg/nodes.py

@@ -1,25 +1,72 @@
 from __future__ import unicode_literals
 
 from builtins import object
+import copy
 import hashlib
-import json
 
 
-class Node(object):
+def _recursive_repr(item):
-    """Node base"""
+    """Hack around python `repr` to deterministically represent dictionaries.
-    def __init__(self, parents, name, *args, **kwargs):
+
+    This is able to represent more things than json.dumps, since it does not require things to be JSON serializable
+    (e.g. datetimes).
+    """
+    if isinstance(item, basestring):
+        result = str(item)
+    elif isinstance(item, list):
+        result = '[{}]'.format(', '.join([_recursive_repr(x) for x in item]))
+    elif isinstance(item, dict):
+        kv_pairs = ['{}: {}'.format(_recursive_repr(k), _recursive_repr(item[k])) for k in sorted(item)]
+        result = '{' + ', '.join(kv_pairs) + '}'
+    else:
+        result = repr(item)
+    return result
+
+
+def _create_hash(item):
+    hasher = hashlib.sha224()
+    repr_ = _recursive_repr(item)
+    hasher.update(repr_.encode('utf-8'))
+    return hasher.hexdigest()
+
+
+class _NodeBase(object):
+    @property
+    def hash(self):
+        if self._hash is None:
+            self._update_hash()
+        return self._hash
+
+    def __init__(self, parents, name):
         parent_hashes = [hash(parent) for parent in parents]
         assert len(parent_hashes) == len(set(parent_hashes)), 'Same node cannot be included as parent multiple times'
         self._parents = parents
         self._hash = None
         self._name = name
-        self._args = args
+
-        self._kwargs = kwargs
+    def _transplant(self, new_parents):
+        other = copy.copy(self)
+        other._parents = copy.copy(new_parents)
+        return other
+
+    @property
+    def _repr_args(self):
+        raise NotImplementedError()
+
+    @property
+    def _repr_kwargs(self):
+        raise NotImplementedError()
+
+    @property
+    def _short_hash(self):
+        return '{:x}'.format(abs(hash(self)))[:12]
 
     def __repr__(self):
-        formatted_props = ['{}'.format(arg) for arg in self._args]
+        args = self._repr_args
-        formatted_props += ['{}={!r}'.format(key, self._kwargs[key]) for key in sorted(self._kwargs)]
+        kwargs = self._repr_kwargs
-        return '{}({})'.format(self._name, ','.join(formatted_props))
+        formatted_props = ['{!r}'.format(arg) for arg in args]
+        formatted_props += ['{}={!r}'.format(key, kwargs[key]) for key in sorted(kwargs)]
+        return '{}({}) <{}>'.format(self._name, ', '.join(formatted_props), self._short_hash)
 
     def __hash__(self):
         if self._hash is None:
@@ -30,9 +77,8 @@ class Node(object):
         return hash(self) == hash(other)
 
     def _update_hash(self):
-        props = {'args': self._args, 'kwargs': self._kwargs}
+        props = {'args': self._repr_args, 'kwargs': self._repr_kwargs}
-        props_str = json.dumps(props, sort_keys=True).encode('utf-8')
+        my_hash = _create_hash(props)
-        my_hash = hashlib.md5(props_str).hexdigest()
         parent_hashes = [str(hash(parent)) for parent in self._parents]
         hashes = parent_hashes + [my_hash]
         hashes_str = ','.join(hashes).encode('utf-8')
@@ -40,6 +86,22 @@ class Node(object):
         self._hash = int(hash_str, base=16)
 
 
+class Node(_NodeBase):
+    """Node base"""
+    def __init__(self, parents, name, *args, **kwargs):
+        super(Node, self).__init__(parents, name)
+        self._args = args
+        self._kwargs = kwargs
+
+    @property
+    def _repr_args(self):
+        return self._args
+
+    @property
+    def _repr_kwargs(self):
+        return self._kwargs
+
+
 class InputNode(Node):
     """InputNode type"""
     def __init__(self, name, *args, **kwargs):

ffmpeg/tests/test_ffmpeg.py

@@ -73,12 +73,12 @@ def test_repr():
     trim3 = ffmpeg.trim(in_file, start_frame=50, end_frame=60)
     concatted = ffmpeg.concat(trim1, trim2, trim3)
     output = ffmpeg.output(concatted, 'dummy2.mp4')
-    assert repr(in_file) == "input(filename={!r})".format('dummy.mp4')
+    assert repr(in_file) == "input(filename={!r}) <{}>".format('dummy.mp4', in_file._short_hash)
-    assert repr(trim1) == "trim(end_frame=20,start_frame=10)"
+    assert repr(trim1) == "trim(end_frame=20, start_frame=10) <{}>".format(trim1._short_hash)
-    assert repr(trim2) == "trim(end_frame=40,start_frame=30)"
+    assert repr(trim2) == "trim(end_frame=40, start_frame=30) <{}>".format(trim2._short_hash)
-    assert repr(trim3) == "trim(end_frame=60,start_frame=50)"
+    assert repr(trim3) == "trim(end_frame=60, start_frame=50) <{}>".format(trim3._short_hash)
-    assert repr(concatted) == "concat(n=3)"
+    assert repr(concatted) == "concat(n=3) <{}>".format(concatted._short_hash)
-    assert repr(output) == "output(filename={!r})".format('dummy2.mp4')
+    assert repr(output) == "output(filename={!r}) <{}>".format('dummy2.mp4', output._short_hash)
 
 
 def test_get_args_simple():