Add .inner_html property

Author: rushter

CHANGES.md

@@ -6,6 +6,7 @@
 - Fix docstring for `css_first` method
 - Fix memory leak in `merge_text_nodes` for lexbor backend
 - Update lexbor backend
+- Add `.inner_html` property. Allows to get and set inner HTML of a node.
 
 
 ## Version 0.3.34

docs/examples.rst

@@ -173,6 +173,82 @@ Ensure exactly one match exists, otherwise raise an error.
 
     ValueError: Expected 1 match, but found 2 matches
 
+HTML manipulation
+-----------------
+
+Getting HTML data back
+~~~~~~~~~~~~~~~~~~~~~~
+
+You can get HTML data back using `.html` or `.inner_html` properties.
+They can be called on any node.
+
+.. code-block:: python
+
+    from selectolax.lexbor import LexborHTMLParser
+    html = """
+    <div id="main">
+      <div>Hi there</div>
+      <div id="updated">2021-08-15</div>
+     </div>
+    """
+    parser = LexborHTMLParser(html)
+    node = parser.css_first("#main")
+    print("Inner html:\n")
+    print(node.inner_html)
+    print("\nOuter html:\n")
+    print(node.html)
+
+**Output:**
+
+.. code-block:: text
+    Inner html:
+
+      <div>Hi there</div>
+      <div id="updated">2021-08-15</div>
+
+    Outer html:
+
+    <div id="main">
+      <div>Hi there</div>
+      <div id="updated">2021-08-15</div>
+     </div>
+
+
+Changing HTML
+~~~~~~~~~~~~~~
+
+You can also change HTML by setting the `.inner_html` property.
+
+.. code-block:: python
+
+    from selectolax.lexbor import LexborHTMLParser
+    html = """
+    <div id="main">
+      <div>Hi there</div>
+     </div>
+    """
+    parser = LexborHTMLParser(html)
+    node = parser.css_first("#main")
+    print("Old html:\n")
+    print(node.html)
+
+    node.inner_html = "<span>Test</span>"
+    print("\nNew html:\n")
+    print(node.inner)
+
+**Output:**
+
+    Old html:
+
+    <div id="main">
+      <div>Hi there</div>
+     </div>
+
+    New html:
+
+    <div id="main"><span>Test</span></div>
+
+
 DOM Navigation
 --------------
 

selectolax/lexbor.pxd

@@ -215,6 +215,8 @@ cdef extern from "lexbor/html/html.h" nogil:
 
         size_t  ref_count
 
+    ctypedef struct lxb_html_element_t
+
     # Functions
     lxb_html_document_t * lxb_html_document_create()
     lxb_status_t lxb_html_document_parse(lxb_html_document_t *document,  const lxb_char_t *html, size_t size)
@@ -223,6 +225,9 @@ cdef extern from "lexbor/html/html.h" nogil:
     lxb_dom_element_t * lxb_dom_document_element(lxb_dom_document_t *document)
 
     lxb_status_t lxb_html_serialize_tree_str(lxb_dom_node_t *node, lexbor_str_t *str)
+    lxb_status_t lxb_html_serialize_deep_str(lxb_dom_node_t *node, lexbor_str_t *str)
+    lxb_html_element_t* lxb_html_element_inner_html_set(lxb_html_element_t *element,
+                                                        const lxb_char_t *html, size_t size)
 
 cdef class LexborNode:
     cdef:

selectolax/lexbor.pyi

@@ -71,6 +71,34 @@ class LexborSelector:
         """
         ...
 
+    @property
+    def inner_html(self) -> str:
+        """Return HTML representation of the child nodes.
+
+        Works similar to innerHTML in JavaScript.
+        Unlike `.html` propery, does not inlcude current node.
+        Can be used to set HTML as well. See the setter docstring.
+
+        Returns
+        -------
+        text : str
+        """
+        ...
+
+    @inner_html.setter
+    def inner_html(self, html: str):
+        """Set inner HTML to the specified HTML.
+
+        Replaces existing data inside the node.
+        Works similar to innerHTML in JavaScript.
+
+        Parameters
+        ----------
+        html : str
+
+        """
+        ...
+
 class LexborCSSSelector:
     def __init__(self): ...
     def find(self, query: str, node: LexborNode) -> list[LexborNode]: ...
@@ -608,6 +636,34 @@ class LexborNode:
         """
         ...
 
+    @property
+    def inner_html(self) -> str:
+        """Return HTML representation of the child nodes.
+
+        Works similar to innerHTML in JavaScript.
+        Unlike `.html` propery, does not inlcude current node.
+        Can be used to set HTML as well. See the setter docstring.
+
+        Returns
+        -------
+        text : str
+        """
+        ...
+
+    @inner_html.setter
+    def inner_html(self, html: str):
+        """Set inner HTML to the specified HTML.
+
+        Replaces existing data inside the node.
+        Works similar to innerHTML in JavaScript.
+
+        Parameters
+        ----------
+        html : str
+
+        """
+        ...
+
 class LexborHTMLParser:
     """The lexbor HTML parser.
 

selectolax/lexbor.pyx

@@ -386,3 +386,32 @@ cdef class LexborHTMLParser:
         # faster to check if the document is empty which should determine if we have a root
         if self.document != NULL:
             self.root.unwrap_tags(tags, delete_empty=delete_empty)
+
+    @property
+    def inner_html(self):
+        """Return HTML representation of the child nodes.
+
+        Works similar to innerHTML in JavaScript.
+        Unlike `.html` propery, does not inlcude current node.
+        Can be used to set HTML as well. See the setter docstring.
+
+        Returns
+        -------
+        text : str
+        """
+        return self.root.inner_html
+
+    @inner_html.setter
+    def inner_html(self, str html):
+        """Set inner HTML to the specified HTML.
+
+        Replaces existing data inside the node.
+        Works similar to innerHTML in JavaScript.
+
+        Parameters
+        ----------
+        html : str
+
+        """
+        self.root.inner_html = html
+

selectolax/lexbor/node.pxi

@@ -11,6 +11,10 @@ ctypedef fused str_or_LexborNode:
     bytes
     LexborNode
 
+ctypedef fused str_or_bytes:
+    str
+    bytes
+
 cdef inline bytes to_bytes(str_or_LexborNode value):
     cdef bytes bytes_val
     if isinstance(value, unicode):
@@ -889,6 +893,48 @@ cdef class LexborNode:
             container.append(py_text)
             return container.text
 
+    @property
+    def inner_html(self):
+        """Return HTML representation of the child nodes.
+
+        Works similar to innerHTML in JavaScript.
+        Unlike `.html` propery, does not inlcude current node.
+        Can be used to set HTML as well. See the setter docstring.
+
+        Returns
+        -------
+        text : str
+        """
+
+        cdef lexbor_str_t *lxb_str
+        cdef lxb_status_t status
+
+        lxb_str = lexbor_str_create()
+        status = lxb_html_serialize_deep_str(self.node, lxb_str)
+        if status == 0 and lxb_str.data:
+            html = lxb_str.data.decode(_ENCODING).replace('<-undef>', '')
+            lexbor_str_destroy(lxb_str,  self.node.owner_document.text, True)
+        return html
+
+    @inner_html.setter
+    def inner_html(self, str html):
+        """Set inner HTML to the specified HTML.
+
+        Replaces existing data inside the node.
+        Works similar to innerHTML in JavaScript.
+
+        Parameters
+        ----------
+        html : str
+
+        """
+        cdef bytes bytes_val
+        bytes_val = <bytes>html.encode("utf-8")
+        lxb_html_element_inner_html_set(
+                <lxb_html_element_t  *>self.node,
+                <lxb_char_t *> bytes_val, len(bytes_val)
+        )
+
 
 @cython.internal
 @cython.final

tests/test_lexbor.py

@@ -0,0 +1,20 @@
+"""Tests for functionality that is only supported by lexbor backend."""
+
+from selectolax.lexbor import LexborHTMLParser
+
+
+def test_reads_inner_html():
+    html = """<div id="main"><div>Hi</div><div id="updated">2025-09-27</div></div>"""
+    parser = LexborHTMLParser(html)
+    actual = parser.css_first("#main").inner_html
+    expected = """<div>Hi</div><div id="updated">2025-09-27</div>"""
+    assert actual == expected
+
+
+def test_sets_inner_html():
+    html = """<div id="main"><div>Hi</div><div id="updated">2025-09-27</div></div>"""
+    parser = LexborHTMLParser(html)
+    expected = "<span>Test</span>"
+    parser.css_first("#main").inner_html = "<span>Test</span>"
+    actual = parser.css_first("#main").inner_html
+    assert actual == expected