Changeset 711 for TI01-discovery

Timestamp:

27/03/06 17:25:14 (14 years ago)

Author:

lawrence

Message:

Moving towards NumSim V007

Location:

TI01-discovery/trunk/schema/numsim

Files:

• NumSim.xsd (modified) (6 diffs)
• NumSim2Html.py (modified) (1 diff)

TI01-discovery/trunk/schema/numsim/NumSim.xsd

@@ -1 +1 @@
 <?xml version="1.0" encoding="UTF-8"?>
-<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema">
+<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xlink="http://www.w3.org/1999/xlink"
+    xmlns:gco="http://www.isotc211.org/2005/gco" xmlns:gmd="http://www.isotc211.org/2005/gmd"
+    version="0071">
+    <xs:annotation>
+        <xs:documentation> Author: Bryan Lawrence Version date: March 27th, 2006 </xs:documentation>
+    </xs:annotation>
+    <xs:annotation>
+        <xs:documentation>The expectation is that this schema will be used initially by import into
+            DIF discovery records and eventually into ISO19115 based discovery records. While such
+            discovery records already include adequate possibilities to include free text
+            information and/or URI references which would allow such discrimination for a human
+            reader, they do not allow the construction of sensible automatic discriminants, which
+            would allow a software agent to identify, for example, all simulation datasets based on
+            the IPCC SRESA1 forcing specification by the HADCM3 model. Such a software agent is
+            necessary to provide meaningful discovery services on an archive holding multiple
+            simulation datasets, and thus, there is a requirement to build information structures to
+            support such software agents. The NumSim schema provides a minimum set of such metadata
+            structures aimed at supporting the widest possible set of simulation codes. It is
+            expected that more detailed discipline specific metadata schema will exist, such as the
+            Numerical Model Metadata Suite (hereafter the NMM). Where such schema and instances
+            exist, it is desirable that the NumSim instances are automatically generated from the
+            more detailed metadata. </xs:documentation>
+        <xs:documentation>The NumSim schema exists in a context where NumSim compliant xml metadata
+            is to included in a document conforming to an (extensible) parent schema. That parent
+            schema is assumed to have adequate tags to fully describe the data produced by the
+            simulation, the point of the NumSim metadata is to discriminate between such datasets,
+            as a consequence all dataset description is offloaded to the parent schema, but there
+            are some tags which might appear in both the parent and the NumSim schema. In such a
+            case, the scope of the tag in the NumSim schema is simply the scope of the parent
+            element, while the scope of the tag in the parent document should be assumed to be
+            either that of the parent element, or in regard to the data, not the simulation. (For
+            example, reference lists may appear in multiple places in a discovery document, but
+            their scope and applicability is related to where they appear. By doing this it is
+            possible to search for datasets created by simulations with a particular model component
+            which was described in a particular </xs:documentation>
+        <xs:documentation>The simulated element is intended to appear as part of a dataset
+            description, and is used to discriminate between differing simulations of the same
+            phenomena. There are essentially three different cases where it is expected be used, and
+            for which it has been designed: (1) Where the dataset consists of one realization of the
+            simulated phenomena (a single integration of a numerical code). (2) Where the dataset
+            consists of multiple realizations of the phenomena (an ensemble), and (3) Where the
+            dataset consists of a single realization of the simulated phenomena, but where that
+            realization has been generated from an ensemble average (an ensemble “climatology”). </xs:documentation>
+        <xs:documentation>There are a number of schema which are imported here, copies of which are
+            loaded into a directory of schema. These include xlinks.xsd and a number of draft
+            ISO19139 schema components. </xs:documentation>
+    </xs:annotation>
+    <xs:import namespace="http://www.w3.org/1999/xlink"
+        schemaLocation="../schemas.opengis.net/xlink/1.0.0/xlinks.xsd"/>
+    <xs:import namespace="http://www.isotc211.org/2005/gco"
+        schemaLocation="../iso19139/gml3.1.1version/gco/gco.xsd"/>
+    <xs:import namespace="http://www.isotc211.org/2005/gmd"
+        schemaLocation="../iso19139/gml3.1.1version/gmd/citation.xsd"/>
+    <xs:attributeGroup name="NS_link">
+        <xs:annotation>
+            <xs:documentation> The attributes are those of a hyperlink, with the addition of a role
+                which we might later want to control in a vocabulary</xs:documentation>
+        </xs:annotation>
+        <xs:attribute ref="xlink:href" use="required"/>
+        <xs:attribute ref="xlink:title" use="required"/>
+        <xs:attribute ref="xlink:role" use="optional"/>
+        <xs:attribute name="dateOfCitation" type="gco:Date_Type" use="optional"/>
+    </xs:attributeGroup>
     <xs:complexType name="NS_RelatedTo">
-        <xs:sequence>
-            <xs:element name="NS_URI" type="xs:anyURI"/>
+        <xs:annotation>
+            <xs:documentation> Used for any relationships.</xs:documentation>
+        </xs:annotation>
+        <xs:sequence>
             <xs:element name="NS_Relationship" type="xs:string"/>
         </xs:sequence>
+        <xs:attributeGroup ref="NS_link"/>
     </xs:complexType>
     <xs:complexType name="NS_Description" mixed="true">
@@ -14 +79 @@
     </xs:complexType>
     <xs:complexType name="NS_ReferenceList">
-        <xs:sequence>
+        <xs:choice>
             <xs:element name="NS_Reference" type="xs:string" minOccurs="1" maxOccurs="unbounded"/>
-        </xs:sequence>
+            <xs:element name="NS_Citation" type="gmd:CI_Citation_Type" minOccurs="1"
+                maxOccurs="unbounded"/>
+        </xs:choice>
     </xs:complexType>
     <xs:complexType name="NS_ControlledParameters">
@@ -132 +199 @@
     <xs:complexType name="NS_EnsembleDetails">
         <xs:annotation>
-            <xs:documentation>Should only appear if the simulated data is from an ensemble of
-                simulations. Statistics should be set true if the original individual simulations
-                are not available. </xs:documentation>
-        </xs:annotation>
-        <xs:sequence>
-            <xs:element name="NS_EnsembleDescription" type="NS_Description" minOccurs="0" maxOccurs="1"/>
+            <xs:documentation>If not present, the default assumption should be that the simulation
+                is a single realization. If present, the attribute can identify that the data
+                represents an ensemble climatology by setting the Boolean EnsembleStatistics to
+                True, otherwise the default is that the data are multiple realizations. This element
+                may include multiple EnsembleType elements, which are drawn from the EnsembleTypes
+                controlled vocabulary: o Initial Condition o Perturbed Physics o Arbitrary
+                Collection o Single Mode o Multiple Model o Multiple Forcing o GrandEnsemble (any
+                combination of the others, in which case the exact combination should be recorded in
+                the EnsembleDescription element - itself including both a textural component and an
+                optional related URI for more information). If only one integration is included,
+                just one model element should be used. Whether an ensemble or a single integration
+                is described by this element, multiple boundary and initial conditions elements are
+                allowed (in the case of single integrations to support multiple components of the
+                model having different initial and/or boundary conditions, and for ensembles to
+                potentially include multiple boundary conditions associated with multiple forcings).
+                (Note that more complex ensemble statistics, such as, for example probability
+                distribution functions, are likely to be more fully described by the data
+                attributes, this attribute is simply used to aid users in finding ensemble averages
+                and avoid them being mistaken for instances and used inappropriately).
+            </xs:documentation>
+        </xs:annotation>
+        <xs:sequence>
+            <xs:element name="NS_EnsembleDescription" type="NS_Description" minOccurs="0"
+                maxOccurs="1"/>
             <xs:element name="NS_EnsembleType" type="NS_EnsembleTypes" minOccurs="1"
                 maxOccurs="unbounded"/>
@@ -144 +229 @@
     </xs:complexType>
     <xs:complexType name="NS_Model">
+        <xs:annotation>
+            <xs:documentation> The NS_Model complex type is used to both as the type of the overall
+                model description (forming a compulsory element within the NumSIm element itself),
+                and as the type of any sub component of that model. </xs:documentation>
+        </xs:annotation>
         <xs:sequence>
             <xs:element name="NS_Name" type="xs:string" minOccurs="1" maxOccurs="1"/>
@@ -151 +241 @@
                 maxOccurs="unbounded"/>
             <xs:element name="NS_Description" type="NS_Description" minOccurs="0" maxOccurs="1"/>
-            <xs:element name="NS_RelatedModel" type="NS_RelatedTo" minOccurs="0" maxOccurs="1"/>
+            <xs:element name="NS_Resolution" type="NS_Description" minOccurs="0" maxOccurs="1"/>
+            <xs:element name="NS_RelatedModel" type="NS_RelatedTo" minOccurs="0"
+                maxOccurs="unbounded"/>
             <xs:element name="NS_References" type="NS_ReferenceList" minOccurs="0" maxOccurs="1"/>
             <xs:element name="NS_Component" type="NS_Model" minOccurs="0" maxOccurs="unbounded"/>
@@ -159 +251 @@
         <xs:sequence>
             <xs:element name="NS_Description" type="NS_Description"/>
-            <xs:element name="NS_Model" type="NS_Model" minOccurs="0" maxOccurs="unbounded"/>
+            <xs:element name="NS_Model" type="NS_Model" minOccurs="1" maxOccurs="unbounded"/>
             <xs:element name="NS_BoundaryCondition" type="NS_BoundaryCondition" minOccurs="0"
                 maxOccurs="unbounded"/>

TI01-discovery/trunk/schema/numsim/NumSim2Html.py

@@ -3 +3 @@
 # Start with methods for handling the various types we've defined and what they should look like
 #
+#  See useful ideas at http://effbot.org/zone/element-iterparse.htm#plist
+
+class ElementWrapper:
+        '''Return the character text tab from any subelement (returns None for
+        missing attributes/subelements unless they begin with __).'''
+        #From http://online.effbot.org/2003_07_01_archive.htm#element-tricks
+        def __init__(self, element):
+            self._element = element
+        def __getattr__(self, tag):
+            if tag.startswith("__"): raise AttributeError(tag)
+            return self._element.findtext(tag)
+            
 def hyperlink(target,name):
         return '<a href="%s">%s</a>'%(target,name)
+
+unmarshallers = {
+        # simple HTML
+        "anyURI": lambda x: hyperlink(x)
+            
+            
+
 
 def RelatedTo(element,linkName='hyperlink',title=False):