add hash and mem_id (#96)

selectolax/lexbor.pyi

@@ -37,6 +37,8 @@ class LexborCSSSelector:
     def any_matches(self, query: str, node: "LexborNode") -> bool: ...
 
 class LexborNode:
+    @property
+    def mem_id(self) -> int: ...
     @property
     def child(self) -> None | "LexborNode": ...
     @property
@@ -51,6 +53,7 @@ class LexborNode:
     def last_child(self) -> None | "LexborNode": ...
     @property
     def html(self) -> str | None: ...
+    def __hash__(self) -> int: ...
     def text_lexbor(self) -> str: ...
     def text(
         self, deep: bool = True, separator: str = "", strip: bool = False

selectolax/lexbor/node.pxi

@@ -27,6 +27,10 @@ cdef class LexborNode:
         self.node = node
         return self
 
+    @property
+    def mem_id(self):
+        return <size_t> self.node
+
     @property
     def child(self):
         """Alias for the `first_child` property."""
@@ -101,6 +105,9 @@ cdef class LexborNode:
             return html
         return None
 
+    def __hash__(self):
+        return self.mem_id
+
     def text_lexbor(self):
         """Returns the text of the node including text of all its child nodes.
 

selectolax/modest/node.pxi

@@ -214,6 +214,16 @@ cdef class Node:
         cdef _Attributes attributes = _Attributes.create(self.node, self.parser.decode_errors)
         return attributes
 
+    @property
+    def mem_id(self):
+        """Get the mem_id attribute of the node.
+
+        Returns
+        -------
+        text : int
+        """
+        return <size_t> self.node
+
     @property
     def id(self):
         """Get the id attribute of the node.
@@ -229,6 +239,9 @@ cdef class Node:
         attr = myhtml_attribute_by_key(self.node, key, 2)
         return None if attr == NULL else attr.value.data.decode(_ENCODING, self.parser.decode_errors)
 
+    def __hash__(self):
+        return self.mem_id
+
     def text(self, bool deep=True, str separator='', bool strip=False):
         """Returns the text of the node including text of all its child nodes.
 

selectolax/parser.pyi

@@ -72,6 +72,18 @@ class Node:
 
         Returns None if id does not set."""
         ...
+
+    def mem_id(self) -> int:
+        """Get the mem_id of the node.
+
+        Returns 0 if mem_id does not set."""
+        ...
+
+    def __hash__(self) -> int:
+        """ Get the hash of this node
+        :return: int
+        """
+        ...
     def text(self, deep: bool = True, separator: str = "", strip: bool = False) -> str:
         """Returns the text of the node including text of all its child nodes."""
         ...

tests/test_nodes.py

@@ -492,6 +492,13 @@ def test_script_contain(parser):
     """)
     assert html_parser.scripts_contain('super_value')
 
+@pytest.mark.parametrize(*_PARSERS_PARAMETRIZER)
+def test_hash_nodes(parser):
+    tree = parser("""<div><p><strong>J</strong>ohn</p><p>Doe</p></div>""")
+    node = tree.css_first("div")
+    node_dict = {"node": node}
+    assert node.mem_id == hash(node)
+
 
 @pytest.mark.parametrize(*_PARSERS_PARAMETRIZER)
 def test_srcs_contain(parser):
@@ -552,3 +559,4 @@ def test_merge_text_nodes(parser):
     assert node.html == "<div><p>John</p><p>Doe</p></div>"
     text = tree.text(deep=True, separator=" ", strip=True)
     assert text == "John Doe"
+