Merge pull request #56 from ModECI/test_xml3

Latest XML export/import changes

examples/neuroml2/NeuroML2.md

@@ -13,7 +13,21 @@ Some description...
   <tr>
     <td><b>xmlns</b></td>
     <td>str</td>
-    <td><i>Schema for NeuroML 2, usually http://www.neuroml.org/schema/neuroml2</i></td>
+    <td><i>Default namespace for the NeuroML file, usually http://www.neuroml.org/schema/neuroml2</i></td>
+ </tr>
+
+
+  <tr>
+    <td><b>xmlns_xsi</b></td>
+    <td>str</td>
+    <td><i>Namespace for XMLSchema-instance</i></td>
+ </tr>
+
+
+  <tr>
+    <td><b>xmlns_loc</b></td>
+    <td>str</td>
+    <td><i>Specifies location of the main namespace</i></td>
  </tr>
 
 
@@ -24,14 +38,14 @@ Some description...
   <tr>
     <td><b>izhikevich2007Cells</b></td>
     <td><a href="#izhikevich2007cell">izhikevich2007Cell</a></td>
-    <td><i></i></td>
+    <td><i>The izhikevich2007Cells</i></td>
   </tr>
 
 
   <tr>
     <td><b>pulseGenerators</b></td>
     <td><a href="#pulsegenerator">pulseGenerator</a></td>
-    <td><i></i></td>
+    <td><i>The pulse current generators</i></td>
   </tr>
 
 

examples/neuroml2/NeuroML2.rst

@@ -5,22 +5,24 @@ Some description...
 
 **Allowed parameters**
 
-===============  ===========  ====================================================================
+===============  ===========  ======================================================================================
 Allowed field    Data Type    Description
-===============  ===========  ====================================================================
+===============  ===========  ======================================================================================
 **id**           str          The id of the NeuroML 2 document
-**xmlns**        str          Schema for NeuroML 2, usually http://www.neuroml.org/schema/neuroml2
-===============  ===========  ====================================================================
+**xmlns**        str          Default namespace for the NeuroML file, usually http://www.neuroml.org/schema/neuroml2
+**xmlns_xsi**    str          Namespace for XMLSchema-instance
+**xmlns_loc**    str          Specifies location of the main namespace
+===============  ===========  ======================================================================================
 
 **Allowed children**
 
-=======================  ============================================  ====================
+=======================  ============================================  ============================
 Allowed child            Data Type                                     Description
-=======================  ============================================  ====================
-**izhikevich2007Cells**  `izhikevich2007Cell <#izhikevich2007cell>`__
-**pulseGenerators**      `pulseGenerator <#pulsegenerator>`__
+=======================  ============================================  ============================
+**izhikevich2007Cells**  `izhikevich2007Cell <#izhikevich2007cell>`__  The izhikevich2007Cells
+**pulseGenerators**      `pulseGenerator <#pulsegenerator>`__          The pulse current generators
 **networks**             `network <#network>`__                        The networks present
-=======================  ============================================  ====================
+=======================  ============================================  ============================
 
 ==================
 izhikevich2007Cell

examples/neuroml2/NeuroML2.specification.json

@@ -8,17 +8,25 @@
             },
             "xmlns": {
                 "type": "str",
-                "description": "Schema for NeuroML 2, usually http://www.neuroml.org/schema/neuroml2"
+                "description": "Default namespace for the NeuroML file, usually http://www.neuroml.org/schema/neuroml2"
+            },
+            "xmlns_xsi": {
+                "type": "str",
+                "description": "Namespace for XMLSchema-instance"
+            },
+            "xmlns_loc": {
+                "type": "str",
+                "description": "Specifies location of the main namespace"
             }
         },
         "allowed_children": {
             "izhikevich2007Cells": {
                 "type": "izhikevich2007Cell",
-                "description": ""
+                "description": "The izhikevich2007Cells"
             },
             "pulseGenerators": {
                 "type": "pulseGenerator",
-                "description": ""
+                "description": "The pulse current generators"
             },
             "networks": {
                 "type": "network",

examples/neuroml2/NeuroML2.specification.yaml

@@ -6,14 +6,20 @@ neuroml:
             description: The id of the NeuroML 2 document
         xmlns:
             type: str
-            description: Schema for NeuroML 2, usually http://www.neuroml.org/schema/neuroml2
+            description: Default namespace for the NeuroML file, usually http://www.neuroml.org/schema/neuroml2
+        xmlns_xsi:
+            type: str
+            description: Namespace for XMLSchema-instance
+        xmlns_loc:
+            type: str
+            description: Specifies location of the main namespace
     allowed_children:
         izhikevich2007Cells:
             type: izhikevich2007Cell
-            description: ''
+            description: The izhikevich2007Cells
         pulseGenerators:
             type: pulseGenerator
-            description: ''
+            description: The pulse current generators
         networks:
             type: network
             description: The networks present

examples/neuroml2/TestNeuroML.xml

@@ -1,5 +1,5 @@
 <?xml version="1.0" ?>
-<neuroml xmlns="http://www.neuroml.org/schema/neuroml2" id="TestNeuroML">
+<neuroml xmlns="http://www.neuroml.org/schema/neuroml2" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" id="TestNeuroML" xsi:schemaLocation="http://www.neuroml.org/schema/neuroml2 https://raw.github.com/NeuroML/NeuroML2/development/Schemas/NeuroML2/NeuroML_v2.3.xsd">
     <izhikevich2007Cell id="izh2007RS0" C="100pF" v0="-60mV" k="0.7nS_per_mV" vr="-60mV" vt="-40mV" vpeak="35mV" a="0.03per_ms" b="-2nS" c="-50.0mV" d="100pA"/>
     <pulseGenerator id="pulseGen_0" delay="100ms" duration="800ms" amplitude="0.07 nA"/>
     <network id="IzNet">

examples/neuroml2/neuroml2_spec.py

@@ -101,14 +101,26 @@ class neuroml(Base):
 
     Args:
         id: The id of the NeuroML 2 document
-        xmlns: Schema for NeuroML 2, usually http://www.neuroml.org/schema/neuroml2
+        xmlns: Default namespace for the NeuroML file, usually http://www.neuroml.org/schema/neuroml2
+        xmlns_xsi: Namespace for XMLSchema-instance
+        xmlns_loc: Specifies location of the main namespace
+        izhikevich2007Cells: The izhikevich2007Cells
+        pulseGenerators: The pulse current generators
         networks: The networks present
     """
 
     id: str = field(validator=instance_of(str))
+
     xmlns: str = field(
         validator=instance_of(str), default="http://www.neuroml.org/schema/neuroml2"
     )
+    xmlns_xsi: str = field(
+        validator=instance_of(str), default="http://www.w3.org/2001/XMLSchema-instance"
+    )
+    xmlns_loc: str = field(
+        validator=instance_of(str),
+        default="http://www.neuroml.org/schema/neuroml2 https://raw.github.com/NeuroML/NeuroML2/development/Schemas/NeuroML2/NeuroML_v2.3.xsd",
+    )
 
     izhikevich2007Cells: List[izhikevich2007Cell] = field(factory=list)
     pulseGenerators: List[pulseGenerator] = field(factory=list)
@@ -187,3 +199,8 @@ class neuroml(Base):
         yy = yaml.dump(doc_dict, indent=4, sort_keys=False)
         print(yy)
         d.write(yy)
+
+    from modelspec.utils import load_xml
+
+    new_neuroml = load_xml("hello_world_neuroml.net.nml")
+    print(new_neuroml)

src/modelspec/base_types.py

@@ -125,7 +125,6 @@ def to_xml(self) -> str:
             )
         from modelspec.utils import build_xml_element
 
-        # root = ET.Element("modelspec")
         root = build_xml_element(self)
 
         xml_string = ET.tostring(
@@ -170,12 +169,38 @@ def from_bson(cls, bson_str: str) -> "Base":
     @classmethod
     def from_xml(cls, xml_str: str) -> "Base":
         """Instantiate a Base object from an XML string"""
-        from modelspec.utils import element_to_dict, handle_id, convert_values
+        from modelspec.utils import (
+            elementtree_element_to_dict,
+            handle_xml_dict_id,
+            convert_xml_dict_values,
+            process_xml_namespace,
+        )
+        import re
+
+        # When the to_xml() method is used it messes up the string therefore,
+        # it is necessary to convert it into an elementree object then decode into a string.
+        xml_string_a = ET.fromstring(xml_str)
+        xml_string_b = ET.tostring(xml_string_a).decode()
+
+        # while trying to obtain a useable xml structure, using the conversion above it acquires
+        # some unusual string element that sometimes can be incremental from either :ns0 to :nsX or ns0: to nsX:.
+        # Using the regex expression pattern catches it in any form and removes it from the xml string structure.
+        ns_prefix_pattern = r"(ns\d+:|:ns\d+)"
+        cleaned_xml = re.sub(ns_prefix_pattern, "", xml_string_b).strip()
+
+        # For the xml to be useable in modelspec unnecessary string elements which only serve as asthetics for the xml must
+        # be removed when converting to a dict, the process_xml_namespaes function does just that.
+        removed_namespaces = process_xml_namespace(cleaned_xml)
+
+        # process_xml_namespace function returns an elementtree object which can be directly worked upon by the elementtree_element_to_dict
+        # function, this returns a python dictionary
+        data_dict = elementtree_element_to_dict(removed_namespaces)
+
+        # This strips every instance of 'id' from the resulting dictionary structure
+        removed_id = handle_xml_dict_id(data_dict)
 
-        root = ET.fromstring(xml_str)
-        data_dict = element_to_dict(root)
-        removed_id = handle_id(data_dict)
-        converted_to_actual_val = convert_values(removed_id)
+        # XML conversions do not returns exact values, instead all values are returned as a string, this reassigns their actual values
+        converted_to_actual_val = convert_xml_dict_values(removed_id)
 
         return cls.from_dict(converted_to_actual_val)
 
@@ -377,15 +402,38 @@ def from_xml_file(cls, filename: str) -> "Base":
         Returns:
             A modelspec Base for this XML.
         """
-        from modelspec.utils import element_to_dict, handle_id, convert_values
+        from modelspec.utils import (
+            elementtree_element_to_dict,
+            handle_xml_dict_id,
+            convert_xml_dict_values,
+            process_xml_namespace,
+        )
+        import re
 
         with open(filename) as infile:
-            tree = ET.parse(infile)
-            root = tree.getroot()
+            tree = ET.parse(infile)  # Parse the XML file into an ElementTree object
+            root = tree.getroot()  # Get the root element
+
+        # This defines regular expressions to match the namespace patterns to be removed
+        ns_prefix_pattern = r"(ns\d+:|:ns\d+)"
+
+        # Converts the loaded xml into a string and removes unwanted string values ':ns0' to :ns∞ and 'ns0:' to ns∞:
+        # They prevent the xml from loading correctly
+        xml_string = ET.tostring(root).decode()
+        cleaned_xml = re.sub(ns_prefix_pattern, "", xml_string).strip()
+
+        # Removes xmlns, xmlns:xsi and xsi:schemaLocation from the xml structure for conversion
+        # it passes an element tree object to the elementtree_element_to_dict function
+        removed_namespaces = process_xml_namespace(cleaned_xml)
+
+        # Converts the resulting xml stripped of xmlns, xmlns:xsi and xsi:schemaLocation into a dict
+        data_dict = elementtree_element_to_dict(removed_namespaces)
+
+        # Removes every key having 'id' and replaces it with it's value
+        removed_id = handle_xml_dict_id(data_dict)
 
-        data_dict = element_to_dict(root)
-        removed_id = handle_id(data_dict)
-        converted_to_actual_val = convert_values(removed_id)
+        # Values are returned as strings after conversion, this corrects them to their actual values
+        converted_to_actual_val = convert_xml_dict_values(removed_id)
         return cls.from_dict(converted_to_actual_val)
 
     def get_child(self, id: str, type_: str) -> Any: