onvif-python 0.0.7__tar.gz → 0.0.8__tar.gz

{onvif_python-0.0.7/onvif_python.egg-info → onvif_python-0.0.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: onvif-python
-Version: 0.0.7
+Version: 0.0.8
 Summary: A comprehensive and developer-friendly Python library for working with ONVIF-compliant devices
 Author-email: Nirsimetri Technologies® <open@nirsimetri.com>
 Maintainer: Kaburagi
@@ -31,8 +31,8 @@ Classifier: Natural Language :: Indonesian
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE.md
-Requires-Dist: zeep>=4.2.1
-Requires-Dist: requests>=2.28.0
+Requires-Dist: zeep>=4.3.0
+Requires-Dist: requests>=2.32.0
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: black; extra == "dev"
@@ -43,14 +43,11 @@ Dynamic: license-file
 
 [![License](https://img.shields.io/badge/License-MIT-blue)](https://github.com/nirsimetri/onvif-python?tab=MIT-1-ov-file)
 [![DeepWiki](https://img.shields.io/badge/DeepWiki-AI%20Wiki-orange)](https://deepwiki.com/nirsimetri/onvif-python)
-[![Release](https://img.shields.io/badge/Release-v0.0.7-red?logo=archive)](https://github.com/nirsimetri/onvif-python/releases)
+[![Release](https://img.shields.io/badge/Release-v0.0.8-red?logo=archive)](https://github.com/nirsimetri/onvif-python/releases)
 <br>
-[![PyPI](https://img.shields.io/badge/PyPI-0.0.7-yellow?logo=archive)](https://pypi.org/project/onvif-python/)
+[![PyPI](https://img.shields.io/badge/PyPI-0.0.8-yellow?logo=archive)](https://pypi.org/project/onvif-python/)
 [![Downloads](https://img.shields.io/pypi/dm/onvif-python?label=PyPI%20Downloads)](https://clickpy.clickhouse.com/dashboard/onvif-python)
 
-Are you having trouble finding a Python ONVIF library that supports your device?  
-Are you confused about the compatibility of the various ONVIF versions, which are updated every six months?
-
 **This project provides a comprehensive and developer-friendly Python library for working with ONVIF-compliant devices.** It is designed to be reliable, easy to integrate, and flexible enough to support a wide range of ONVIF profiles and services.  
 
 **[ONVIF](https://www.onvif.org) (Open Network Video Interface Forum)** is a global standard for the interface of IP-based physical security products, including network cameras, video recorders, and related systems.  
@@ -154,7 +151,9 @@ Explore more advanced usage and service-specific operations in the [`examples/`]
 
 The `ONVIFClient` class provides various configuration options to customize the connection behavior, caching strategy, security settings, and debugging capabilities. Below is a detailed description of all available parameters:
 
-### Basic Parameters
+
+<details>
+<summary><b>Basic Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
@@ -163,7 +162,10 @@ The `ONVIFClient` class provides various configuration options to customize the
 | `username` | `str` | ✅ Yes | - | Username for device authentication (use digest authentication) |
 | `password` | `str` | ✅ Yes | - | Password for device authentication |
 
-### Connection Parameters
+</details>
+
+<details>
+<summary><b>Connection Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
@@ -171,33 +173,45 @@ The `ONVIFClient` class provides various configuration options to customize the
 | `use_https` | `bool` | ❌ No | `False` | Use HTTPS instead of HTTP for secure communication |
 | `verify_ssl` | `bool` | ❌ No | `True` | Verify SSL certificates when using HTTPS (set to `False` for self-signed certificates) |
 
-### Caching Parameters
+</details>
+
+<details>
+<summary><b>Caching Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
 | `cache` | `CacheMode` | ❌ No | `CacheMode.ALL` | WSDL caching strategy (see [Cache Modes](#cache-modes) below) |
 
-### Feature Parameters
+</details>
+
+<details>
+<summary><b>Feature Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
-| `apply_patch` | `bool` | ❌ No | `True` | Enable zeep patching for better xsd:any field parsing and automatic flattening |
+| `apply_patch` | `bool` | ❌ No | `True` | Enable zeep patching for better `xsd:any` field parsing and automatic flattening |
 | `capture_xml` | `bool` | ❌ No | `False` | Enable XML capture plugin for debugging SOAP requests/responses |
 
-### Cache Modes
+</details>
+
+<details>
+<summary><b>Cache Modes</b></summary> 
 
 The library provides four caching strategies via the `CacheMode` enum:
 
 | Mode | Description | Best For | Startup Speed | Disk Usage | Memory Usage |
 |------|-------------|----------|---------------|------------|--------------|
-| `CacheMode.ALL` | In-memory + disk cache (SQLite) | Production servers, multi-device apps | ⚡⚡⚡ Fast | 💾 High | 🧠 High |
-| `CacheMode.DB` | Disk cache only (SQLite) | Batch jobs, CLI tools | ⚡⚡ Medium | 💾 Medium | 🧠 Low |
-| `CacheMode.MEM` | In-memory cache only | Short-lived scripts, demos | ⚡⚡ Medium | 💾 None | 🧠 Medium |
-| `CacheMode.NONE` | No caching | Testing, debugging | ⚡ Slow | 💾 None | 🧠 Low |
+| `CacheMode.ALL` | In-memory + disk cache (SQLite) | Production servers, multi-device apps | Fast | High | High |
+| `CacheMode.DB` | Disk cache only (SQLite) | Batch jobs, CLI tools | Medium | Medium | Low |
+| `CacheMode.MEM` | In-memory cache only | Short-lived scripts, demos | Medium | None | Medium |
+| `CacheMode.NONE` | No caching | Testing, debugging | Slow | None | Low |
 
 **Recommendation:** Use `CacheMode.ALL` (default) for production applications to maximize performance.
 
-### Usage Examples
+</details>
+
+<details>
+<summary><b>Usage Examples</b></summary>
 
 **Basic Connection:**
 ```python
@@ -218,8 +232,7 @@ client = ONVIFClient(
     "admin", 
     "password",
     timeout=30,
-    use_https=True,
-    verify_ssl=False  # For self-signed certificates
+    use_https=True
 )
 ```
 
@@ -237,7 +250,7 @@ client = ONVIFClient(
 )
 ```
 
-**No Caching (Testing):**
+**No Caching and No Zeep Patching (Testing):**
 ```python
 from onvif import ONVIFClient, CacheMode
 
@@ -296,7 +309,6 @@ if client.xml_plugin:
     client.xml_plugin.clear_history()
 ```
 
-> [!NOTE]
 > **XML Capture Plugin Methods:**
 > - `last_sent_xml` - Get the last SOAP request XML
 > - `last_received_xml` - Get the last SOAP response XML
@@ -308,8 +320,11 @@ if client.xml_plugin:
 > - `save_to_file(request_file, response_file)` - Save XML to files
 > - `clear_history()` - Clear captured history
 
+</details>
+
+<details>
+<summary><b>Production Configuration</b></summary>
 
-**Production Configuration:**
 ```python
 from onvif import ONVIFClient, CacheMode
 
@@ -320,13 +335,14 @@ client = ONVIFClient(
     username="admin",
     password="secure_password",
     timeout=15,
-    cache=CacheMode.ALL,        # Maximum performance
+    cache=CacheMode.ALL,        # Maximum performance (default)
     use_https=True,             # Secure communication
     verify_ssl=True,            # Verify certificates (default)
     apply_patch=True,           # Enhanced parsing (default)
     capture_xml=False           # Disable debug mode (default)
 )
 ```
+</details>
 
 ### Notes
 
@@ -598,8 +614,6 @@ This project is an **independent open-source implementation** of the [ONVIF](htt
 
 If you require certified ONVIF-compliant devices or clients, please refer to the official [ONVIF conformant product list](https://www.onvif.org/conformant-products/). For authoritative reference and the latest official ONVIF specifications, please consult the [ONVIF Official Specifications](https://www.onvif.org/profiles/specifications/specification-history/).
 
-Use of this library is at your own risk. The authors and contributors assume no liability for any damages, direct or indirect, arising from its use.
-
 ## License
 
 This project is licensed under the MIT License. See [LICENSE](./LICENSE.md) for details.

{onvif_python-0.0.7 → onvif_python-0.0.8}/README.md

@@ -2,14 +2,11 @@
 
 [![License](https://img.shields.io/badge/License-MIT-blue)](https://github.com/nirsimetri/onvif-python?tab=MIT-1-ov-file)
 [![DeepWiki](https://img.shields.io/badge/DeepWiki-AI%20Wiki-orange)](https://deepwiki.com/nirsimetri/onvif-python)
-[![Release](https://img.shields.io/badge/Release-v0.0.7-red?logo=archive)](https://github.com/nirsimetri/onvif-python/releases)
+[![Release](https://img.shields.io/badge/Release-v0.0.8-red?logo=archive)](https://github.com/nirsimetri/onvif-python/releases)
 <br>
-[![PyPI](https://img.shields.io/badge/PyPI-0.0.7-yellow?logo=archive)](https://pypi.org/project/onvif-python/)
+[![PyPI](https://img.shields.io/badge/PyPI-0.0.8-yellow?logo=archive)](https://pypi.org/project/onvif-python/)
 [![Downloads](https://img.shields.io/pypi/dm/onvif-python?label=PyPI%20Downloads)](https://clickpy.clickhouse.com/dashboard/onvif-python)
 
-Are you having trouble finding a Python ONVIF library that supports your device?  
-Are you confused about the compatibility of the various ONVIF versions, which are updated every six months?
-
 **This project provides a comprehensive and developer-friendly Python library for working with ONVIF-compliant devices.** It is designed to be reliable, easy to integrate, and flexible enough to support a wide range of ONVIF profiles and services.  
 
 **[ONVIF](https://www.onvif.org) (Open Network Video Interface Forum)** is a global standard for the interface of IP-based physical security products, including network cameras, video recorders, and related systems.  
@@ -113,7 +110,9 @@ Explore more advanced usage and service-specific operations in the [`examples/`]
 
 The `ONVIFClient` class provides various configuration options to customize the connection behavior, caching strategy, security settings, and debugging capabilities. Below is a detailed description of all available parameters:
 
-### Basic Parameters
+
+<details>
+<summary><b>Basic Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
@@ -122,7 +121,10 @@ The `ONVIFClient` class provides various configuration options to customize the
 | `username` | `str` | ✅ Yes | - | Username for device authentication (use digest authentication) |
 | `password` | `str` | ✅ Yes | - | Password for device authentication |
 
-### Connection Parameters
+</details>
+
+<details>
+<summary><b>Connection Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
@@ -130,33 +132,45 @@ The `ONVIFClient` class provides various configuration options to customize the
 | `use_https` | `bool` | ❌ No | `False` | Use HTTPS instead of HTTP for secure communication |
 | `verify_ssl` | `bool` | ❌ No | `True` | Verify SSL certificates when using HTTPS (set to `False` for self-signed certificates) |
 
-### Caching Parameters
+</details>
+
+<details>
+<summary><b>Caching Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
 | `cache` | `CacheMode` | ❌ No | `CacheMode.ALL` | WSDL caching strategy (see [Cache Modes](#cache-modes) below) |
 
-### Feature Parameters
+</details>
+
+<details>
+<summary><b>Feature Parameters</b></summary>
 
 | Parameter | Type | Required | Default | Description |
 |-----------|------|----------|---------|-------------|
-| `apply_patch` | `bool` | ❌ No | `True` | Enable zeep patching for better xsd:any field parsing and automatic flattening |
+| `apply_patch` | `bool` | ❌ No | `True` | Enable zeep patching for better `xsd:any` field parsing and automatic flattening |
 | `capture_xml` | `bool` | ❌ No | `False` | Enable XML capture plugin for debugging SOAP requests/responses |
 
-### Cache Modes
+</details>
+
+<details>
+<summary><b>Cache Modes</b></summary> 
 
 The library provides four caching strategies via the `CacheMode` enum:
 
 | Mode | Description | Best For | Startup Speed | Disk Usage | Memory Usage |
 |------|-------------|----------|---------------|------------|--------------|
-| `CacheMode.ALL` | In-memory + disk cache (SQLite) | Production servers, multi-device apps | ⚡⚡⚡ Fast | 💾 High | 🧠 High |
-| `CacheMode.DB` | Disk cache only (SQLite) | Batch jobs, CLI tools | ⚡⚡ Medium | 💾 Medium | 🧠 Low |
-| `CacheMode.MEM` | In-memory cache only | Short-lived scripts, demos | ⚡⚡ Medium | 💾 None | 🧠 Medium |
-| `CacheMode.NONE` | No caching | Testing, debugging | ⚡ Slow | 💾 None | 🧠 Low |
+| `CacheMode.ALL` | In-memory + disk cache (SQLite) | Production servers, multi-device apps | Fast | High | High |
+| `CacheMode.DB` | Disk cache only (SQLite) | Batch jobs, CLI tools | Medium | Medium | Low |
+| `CacheMode.MEM` | In-memory cache only | Short-lived scripts, demos | Medium | None | Medium |
+| `CacheMode.NONE` | No caching | Testing, debugging | Slow | None | Low |
 
 **Recommendation:** Use `CacheMode.ALL` (default) for production applications to maximize performance.
 
-### Usage Examples
+</details>
+
+<details>
+<summary><b>Usage Examples</b></summary>
 
 **Basic Connection:**
 ```python
@@ -177,8 +191,7 @@ client = ONVIFClient(
     "admin", 
     "password",
     timeout=30,
-    use_https=True,
-    verify_ssl=False  # For self-signed certificates
+    use_https=True
 )
 ```
 
@@ -196,7 +209,7 @@ client = ONVIFClient(
 )
 ```
 
-**No Caching (Testing):**
+**No Caching and No Zeep Patching (Testing):**
 ```python
 from onvif import ONVIFClient, CacheMode
 
@@ -255,7 +268,6 @@ if client.xml_plugin:
     client.xml_plugin.clear_history()
 ```
 
-> [!NOTE]
 > **XML Capture Plugin Methods:**
 > - `last_sent_xml` - Get the last SOAP request XML
 > - `last_received_xml` - Get the last SOAP response XML
@@ -267,8 +279,11 @@ if client.xml_plugin:
 > - `save_to_file(request_file, response_file)` - Save XML to files
 > - `clear_history()` - Clear captured history
 
+</details>
+
+<details>
+<summary><b>Production Configuration</b></summary>
 
-**Production Configuration:**
 ```python
 from onvif import ONVIFClient, CacheMode
 
@@ -279,13 +294,14 @@ client = ONVIFClient(
     username="admin",
     password="secure_password",
     timeout=15,
-    cache=CacheMode.ALL,        # Maximum performance
+    cache=CacheMode.ALL,        # Maximum performance (default)
     use_https=True,             # Secure communication
     verify_ssl=True,            # Verify certificates (default)
     apply_patch=True,           # Enhanced parsing (default)
     capture_xml=False           # Disable debug mode (default)
 )
 ```
+</details>
 
 ### Notes
 
@@ -557,8 +573,6 @@ This project is an **independent open-source implementation** of the [ONVIF](htt
 
 If you require certified ONVIF-compliant devices or clients, please refer to the official [ONVIF conformant product list](https://www.onvif.org/conformant-products/). For authoritative reference and the latest official ONVIF specifications, please consult the [ONVIF Official Specifications](https://www.onvif.org/profiles/specifications/specification-history/).
 
-Use of this library is at your own risk. The authors and contributors assume no liability for any damages, direct or indirect, arising from its use.
-
 ## License
 
 This project is licensed under the MIT License. See [LICENSE](./LICENSE.md) for details.

{onvif_python-0.0.7 → onvif_python-0.0.8}/onvif/services/__init__.py

@@ -1,3 +1,5 @@
+# onvif/services/__init__.py
+
 from .devicemgmt import Device
 from .events.events import Events
 from .events.pullpoint import PullPoint

{onvif_python-0.0.7 → onvif_python-0.0.8}/onvif/utils/__init__.py

@@ -1,3 +1,5 @@
+# onvif/utils/__init__.py
+
 from .wsdl import ONVIFWSDL
 from .exceptions import ONVIFOperationException
 from .zeep import ZeepPatcher

{onvif_python-0.0.7 → onvif_python-0.0.8}/onvif/utils/xml_capture.py

@@ -2,6 +2,7 @@
 
 from zeep import Plugin
 from lxml import etree
+from xml.dom import minidom
 
 
 class XMLCapturePlugin(Plugin):
@@ -35,12 +36,43 @@ class XMLCapturePlugin(Plugin):
         self.last_operation = None
         self.history = []  # Store all requests/responses
 
+    def _format_xml(self, element):
+        """
+        Format XML element with proper indentation using minidom.
+
+        Args:
+            element: lxml Element to format
+
+        Returns:
+            str: Pretty-printed XML string
+        """
+        try:
+            # Convert lxml element to string
+            xml_string = etree.tostring(element, encoding="unicode")
+            # Parse with minidom and pretty print
+            dom = minidom.parseString(xml_string)
+            # Use toprettyxml with proper indentation
+            pretty_xml = dom.toprettyxml(indent="  ", encoding=None)
+            # Remove extra blank lines and XML declaration
+            lines = [line for line in pretty_xml.split("\n") if line.strip()]
+            # Remove XML declaration line if present
+            if lines and lines[0].startswith("<?xml"):
+                lines = lines[1:]
+            return "\n".join(lines)
+        except Exception as e:
+            # Fallback to lxml if minidom fails
+            return etree.tostring(element, pretty_print=True, encoding="unicode")
+
     def egress(self, envelope, http_headers, operation, binding_options):
         """Called before sending the SOAP request"""
-        # Serialize XML
-        self.last_sent_xml = etree.tostring(
-            envelope, pretty_print=self.pretty_print, encoding="unicode"
-        )
+        # Serialize XML with proper pretty printing
+        if self.pretty_print:
+            self.last_sent_xml = self._format_xml(envelope)
+        else:
+            self.last_sent_xml = etree.tostring(
+                envelope, pretty_print=False, encoding="unicode"
+            )
+
         self.last_operation = operation.name
 
         # Store in history
@@ -57,10 +89,13 @@ class XMLCapturePlugin(Plugin):
 
     def ingress(self, envelope, http_headers, operation):
         """Called after receiving the SOAP response"""
-        # Serialize XML
-        self.last_received_xml = etree.tostring(
-            envelope, pretty_print=self.pretty_print, encoding="unicode"
-        )
+        # Serialize XML with proper pretty printing
+        if self.pretty_print:
+            self.last_received_xml = self._format_xml(envelope)
+        else:
+            self.last_received_xml = etree.tostring(
+                envelope, pretty_print=False, encoding="unicode"
+            )
 
         # Store in history
         self.history.append(

{onvif_python-0.0.7 → onvif_python-0.0.8}/onvif/utils/zeep.py

@@ -63,6 +63,93 @@ class ZeepPatcher:
         except ValueError:
             return val
 
+    @staticmethod
+    def _parse_element_recursive(element):
+        """
+        Recursively parse an XML element and its children into a dictionary.
+
+        Handles:
+        - Elements with attributes only (converted to dict)
+        - Elements with children only (recursively parsed)
+        - Elements with BOTH attributes AND children (merged into single dict)
+        - Elements with text content only (type-converted)
+
+        Args:
+            element: lxml Element to parse
+
+        Returns:
+            Dict, parsed value, or None
+        """
+        result = {}
+
+        for child in element:
+            child_qname = QName(child.tag)
+            child_name = child_qname.localname
+            child_has_attrib = bool(child.attrib)
+            child_has_children = bool(list(child))
+
+            if child_has_attrib and child_has_children:
+                # Element has BOTH attributes AND children - merge them
+                child_dict = {
+                    k: ZeepPatcher.parse_text_value(v) for k, v in child.attrib.items()
+                }
+                # Recursively parse children and merge into the same dict
+                nested = ZeepPatcher._parse_element_recursive(child)
+                if nested:
+                    child_dict.update(nested)
+                result[child_name] = child_dict
+            elif child_has_attrib:
+                # Element with attributes only
+                result[child_name] = {
+                    k: ZeepPatcher.parse_text_value(v) for k, v in child.attrib.items()
+                }
+            elif child_has_children:
+                # Element has nested children only
+                result[child_name] = ZeepPatcher._parse_element_recursive(child)
+            else:
+                # Element only has text content
+                result[child_name] = ZeepPatcher.parse_text_value(child.text)
+
+        return result if result else None
+
+    @staticmethod
+    def _zeep_object_to_dict(obj):
+        """
+        Convert zeep object to dictionary, including manually added attributes.
+
+        Args:
+            obj: Zeep object or primitive value
+
+        Returns:
+            Dict or primitive value
+        """
+        if obj is None:
+            return None
+        if isinstance(obj, (str, int, float, bool)):
+            return obj
+        if isinstance(obj, list):
+            return [ZeepPatcher._zeep_object_to_dict(item) for item in obj]
+        if isinstance(obj, dict):
+            return {k: ZeepPatcher._zeep_object_to_dict(v) for k, v in obj.items()}
+
+        # Handle zeep objects with __values__
+        if hasattr(obj, "__values__"):
+            result = {}
+            for key, val in obj.__values__.items():
+                if not key.startswith("_"):  # Skip private fields like _value_1
+                    result[key] = ZeepPatcher._zeep_object_to_dict(val)
+            return result
+
+        # Handle objects with __dict__
+        if hasattr(obj, "__dict__"):
+            result = {}
+            for key, val in obj.__dict__.items():
+                if not key.startswith("_"):
+                    result[key] = ZeepPatcher._zeep_object_to_dict(val)
+            return result
+
+        return obj
+
     @staticmethod
     def flatten_xsd_any_fields(obj, _visited=None):
         """
@@ -147,10 +234,24 @@ class ZeepPatcher:
                         if should_flatten:
                             # This is truly a wrapper - flatten by copying fields to parent
                             # But still set the tag_name field to the inner_content
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(inner_content, "__values__") or hasattr(
+                                inner_content, "__dict__"
+                            ):
+                                inner_content = ZeepPatcher._zeep_object_to_dict(
+                                    inner_content
+                                )
                             values[tag_name] = inner_content
                             # Don't copy fields up to parent - keep them in the structured object
                         else:
                             # Not a wrapper - just set the field directly
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(inner_content, "__values__") or hasattr(
+                                inner_content, "__dict__"
+                            ):
+                                inner_content = ZeepPatcher._zeep_object_to_dict(
+                                    inner_content
+                                )
                             if tag_name in values and values[tag_name] is None:
                                 values[tag_name] = inner_content
                             elif tag_name not in values:
@@ -161,6 +262,10 @@ class ZeepPatcher:
                             if key.startswith("_"):
                                 continue
 
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(val, "__values__") or hasattr(val, "__dict__"):
+                                val = ZeepPatcher._zeep_object_to_dict(val)
+
                             if key in values and values[key] is None:
                                 values[key] = val
                             elif key not in values:
@@ -207,9 +312,23 @@ class ZeepPatcher:
 
                         if should_flatten:
                             # Set the field to the structured content
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(inner_content, "__values__") or hasattr(
+                                inner_content, "__dict__"
+                            ):
+                                inner_content = ZeepPatcher._zeep_object_to_dict(
+                                    inner_content
+                                )
                             setattr(obj, tag_name, inner_content)
                         else:
                             # Not a wrapper - just set the field directly
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(inner_content, "__values__") or hasattr(
+                                inner_content, "__dict__"
+                            ):
+                                inner_content = ZeepPatcher._zeep_object_to_dict(
+                                    inner_content
+                                )
                             if (
                                 hasattr(obj, tag_name)
                                 and getattr(obj, tag_name) is None
@@ -223,6 +342,10 @@ class ZeepPatcher:
                             if key.startswith("_"):
                                 continue
 
+                            # Convert zeep object to dict to preserve manually added attributes
+                            if hasattr(val, "__values__") or hasattr(val, "__dict__"):
+                                val = ZeepPatcher._zeep_object_to_dict(val)
+
                             if hasattr(obj, key) and getattr(obj, key) is None:
                                 setattr(obj, key, val)
                             elif not hasattr(obj, key):
@@ -293,10 +416,42 @@ class ZeepPatcher:
 
             tag_name = QName(xmlelement.tag).localname
             children = list(xmlelement)
+
             if children and schema:
                 child_result = {}
+
+                # If xmlelement itself has attributes, add them first
+                if xmlelement.attrib:
+                    child_result.update(
+                        {
+                            k: ZeepPatcher.parse_text_value(v)
+                            for k, v in xmlelement.attrib.items()
+                        }
+                    )
+
                 for child in children:
                     child_qname = QName(child.tag)
+                    child_localname = child_qname.localname
+
+                    # If child has attributes, parse manually to preserve them
+                    # Schema parsing often loses attributes not in schema definition
+                    if child.attrib:
+                        parsed = {}
+                        # Add attributes first
+                        parsed.update(
+                            {
+                                k: ZeepPatcher.parse_text_value(v)
+                                for k, v in child.attrib.items()
+                            }
+                        )
+                        # Then add nested children
+                        nested = ZeepPatcher._parse_element_recursive(child)
+                        if nested:
+                            parsed.update(nested)
+                        child_result[child_localname] = parsed
+                        continue
+
+                    # No attributes - try schema parsing first
                     try:
                         xsd_el = schema.get_element(child_qname)
                         val = xsd_el.parse_xmlelement(
@@ -305,28 +460,28 @@ class ZeepPatcher:
                         child_result[child_qname.localname] = val
                     except Exception:
                         # If schema lookup fails, try to parse manually
-                        # Check if child has attributes (common in ONVIF capabilities)
+                        parsed = {}
+
+                        # Parse child's own attributes first (if any)
                         if child.attrib:
-                            # Parse attributes as a dict
-                            attr_dict = {
-                                k: ZeepPatcher.parse_text_value(v)
-                                for k, v in child.attrib.items()
-                            }
-                            child_result[child_qname.localname] = attr_dict
-                        elif list(child):
-                            # Has nested children
-                            nested = {
-                                QName(sub.tag).localname: ZeepPatcher.parse_text_value(
-                                    sub.text
-                                )
-                                for sub in child
-                            }
-                            child_result[child_qname.localname] = nested
-                        else:
-                            # Only has text content
-                            child_result[child_qname.localname] = (
-                                ZeepPatcher.parse_text_value(child.text)
+                            parsed.update(
+                                {
+                                    k: ZeepPatcher.parse_text_value(v)
+                                    for k, v in child.attrib.items()
+                                }
                             )
+
+                        # Parse child's nested children (if any)
+                        nested = ZeepPatcher._parse_element_recursive(child)
+                        if nested:
+                            parsed.update(nested)
+
+                        # If no attributes and no children, try text content
+                        if not parsed:
+                            parsed = ZeepPatcher.parse_text_value(child.text)
+
+                        child_result[child_qname.localname] = parsed
+
                 parsed_result[tag_name] = child_result
             else:
                 parsed_result[tag_name] = ZeepPatcher.parse_text_value(xmlelement.text)