Notice: This Wiki is now read only and edits are no longer possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.
PTP/designs/resource manager xsd
The following is a draft of the proposed Resource Manager XML Schema (XSD). It has been annotated internally where I felt some explanation necessary.
Note that I have not attempted to figure out yet what the monitoring part will look like or what the control and monitoring sections may need to interact; it may also be the case that some things inside the control node should come up a level.
Revised version 1 (17/01/2011):
Revised version 2 (19/01/2011): I have adjusted the schema to accommodate the points raised by our discussions on ptp-dev. More tomorrow, I'm sure. -alr
Revised version 3 (23/01/2011): A few more changes I discovered needed to be made as I work through the XML instance for PBS. -alr
Revised version 4 (24/01/2011): More tweaks. -alr
<?xml version="1.0" encoding="UTF-8"?> <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" elementFormDefault="qualified"> <xs:element name="resource-manager"> <xs:annotation> <xs:documentation> It is assumed that the xml instance of the schema will bear a name which sufficiently distinguishes it from others of a similar type; e.g., pbs-torque-v_2.3.7_abe is specific to an installation on the host abe, ll-v_4.0 suits all installations of LoadLeveler version 4, etc. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element ref="control" /> <xs:element ref="monitor" /> </xs:sequence> <xs:attribute name="name" type="xs:string" /> </xs:complexType> </xs:element> <xs:element name="control"> <xs:annotation> <xs:documentation> The implementation will construct a map serving as the resource manager "environment"; these are dereferenced in the configuration file via ${name} or ${name.part}; in addition to the properties specified at the top level of the control tree, all attributes and parsers will be mapped for reference. A number of runtime variables such as remote host name, user name, etc., will also be included. The top level of the tree represents the available commands/methods on all resource managers, places for defining stdout or stderr parsers, for defining the structure of a batch script to be used, for defining (job) attributes, and finally for describing how to construct the UI (Launch Configuration Tab). </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element minOccurs="0" maxOccurs="unbounded" ref="property" /> <xs:element minOccurs="0" ref="on-start-up" /> <xs:element minOccurs="0" ref="discover-attributes" /> <xs:element name="run-commands"> <xs:complexType> <xs:choice minOccurs="1" maxOccurs="3"> <xs:element ref="run-interactive" /> <xs:element ref="run-batch" /> <xs:element ref="run-debug" /> </xs:choice> </xs:complexType> </xs:element> <xs:element ref="cancel-job" /> <xs:element minOccurs="0" ref="suspend-job" /> <xs:element minOccurs="0" ref="resume-job" /> <xs:element minOccurs="0" ref="on-shut-down" /> <xs:element minOccurs="0" ref="commands" /> <xs:element minOccurs="0" ref="parsers" /> <xs:element minOccurs="0" ref="script" /> <xs:element ref="attribute-definitions" /> <xs:element ref="launch-tab" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="property"> <xs:annotation> <xs:documentation> The value is any object type; these are set at runtime. </xs:documentation> </xs:annotation> <xs:complexType> <xs:attribute name="name" type="xs:string" /> </xs:complexType> </xs:element> <xs:element name="on-start-up"> <xs:annotation> <xs:documentation> Arbitrary commands which need to be run on startup. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="discover-attributes"> <xs:annotation> <xs:documentation> A command for retrieving (if possible) attributes defined by the service dynamically. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="run-interactive"> <xs:annotation> <xs:documentation> The command(s) for running interactive jobs. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="run-batch"> <xs:annotation> <xs:documentation> The command(s) for running jobs through the scheduler. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="run-debug"> <xs:annotation> <xs:documentation> The command(s) for running debug jobs. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="cancel-job"> <xs:annotation> <xs:documentation> The command(s) for cancelling or killing jobs. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="suspend-job"> <xs:annotation> <xs:documentation> The command(s) for suspending jobs. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="resume-job"> <xs:annotation> <xs:documentation> The command(s) for restarting suspended jobs. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="on-shut-down"> <xs:annotation> <xs:documentation> Arbitrary commands which need to be run on startup. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command-ref" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="commands"> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="command" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="parsers"> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="stream-parser" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="script"> <xs:complexType> <xs:sequence> <xs:element minOccurs="0" ref="directive-definitions" /> <xs:element minOccurs="0" ref="environment-definitions" /> <xs:element minOccurs="0" ref="pre-execute-commands" /> <xs:element ref="execute-command" /> <xs:element minOccurs="0" ref="post-execute-commands" /> </xs:sequence> <xs:attribute name="prefix" use="required" /> <xs:attribute name="shell" use="required" /> </xs:complexType> </xs:element> <xs:element name="directive-definitions"> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="directive-definition" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="directive-definition"> <xs:annotation> <xs:documentation> 'ValueFrom' refers to properties in the resource manager environment. </xs:documentation> </xs:annotation> <xs:complexType mixed="true"> <xs:attribute name="valueFrom" /> </xs:complexType> </xs:element> <xs:element name="environment-definitions"> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="environment-definition" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="environment-definition"> <xs:annotation> <xs:documentation> 'ValueFrom' refers to properties in the resource manager environment. </xs:documentation> </xs:annotation> <xs:complexType mixed="true"> <xs:attribute name="valueFrom" /> </xs:complexType> </xs:element> <xs:element name="pre-execute-commands"> <xs:complexType> <xs:sequence> <xs:element name="command" maxOccurs="unbounded" type="xs:string" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="execute-command"> <xs:complexType> <xs:sequence> <xs:element ref="command" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="post-execute-commands"> <xs:complexType> <xs:sequence> <xs:element name="command" type="xs:string" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="attribute-definitions"> <xs:complexType> <xs:sequence> <xs:element name="command" maxOccurs="unbounded" type="xs:string" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="job-attribute"> <xs:complexType> <xs:sequence> <xs:element name="description" type="xs:string" /> <xs:element name="tooltip" minOccurs="0" type="xs:string" /> <xs:element name="choice" minOccurs="0" type="xs:string"> <xs:annotation> <xs:documentation> A a comma-delimited set of values. </xs:documentation> </xs:annotation> </xs:element> <xs:element name="defaultValue" minOccurs="0" type="xs:string" /> <xs:element minOccurs="0" ref="validator" /> </xs:sequence> <xs:attribute name="id" use="required" type="xs:string" /> <xs:attribute name="name" use="required" type="xs:string" /> <xs:attribute name="type" use="required" type="xs:string" /> <xs:attribute name="basic" type="xs:boolean"> <xs:annotation> <xs:documentation> Indicates if this attribute is part of a basic (required) set, or if it is an (advanced) option. </xs:documentation> </xs:annotation> </xs:attribute> <xs:attribute name="minValue" type="xs:int" /> <xs:attribute name="maxValue" type="xs:int" /> </xs:complexType> </xs:element> <xs:element name="command-ref" type="xs:string" /> <xs:element name="parser-ref" type="xs:string" /> <xs:element name="command"> <xs:complexType mixed="true"> <xs:sequence> <xs:element maxOccurs="unbounded" ref="arglist" /> <xs:element minOccurs="0" maxOccurs="2" ref="parser-ref" /> </xs:sequence> <xs:attribute name="name" type="xs:string" /> <xs:attribute name="displayStdout" type="xs:boolean" /> <xs:attribute name="displayStderr" type="xs:boolean" /> </xs:complexType> </xs:element> <xs:element name="arglist"> <xs:complexType mixed="true"> <xs:sequence> <xs:element maxOccurs="unbounded" ref="arg" /> </xs:sequence> <xs:attribute name="dynamicAppend" type="xs:boolean"> <xs:annotation> <xs:documentation> Iterate over all dynamic attributes, appending the sequence of args for each attribute. </xs:documentation> </xs:annotation> </xs:attribute> </xs:complexType> </xs:element> <xs:element name="arg" type="xs:string"> <xs:annotation> <xs:documentation> The text of the argument will be parsed for references to the resource manager environment; e.g., ${a}, v=${v}, etc. By convention, ${@name} and ${@value} will refer to a given dynamic attribute name and value (of undefined position i in the list). </xs:documentation> </xs:annotation> </xs:element> <xs:element name="stream-parser"> <xs:annotation> <xs:documentation> These are attached to the stdout or stderr streams in order to capture the output of the command and add values into the resource manager environment. It is understood that a parser tokenization is a list of matching tokens. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element name="parse-lines" type="xs:string"> <xs:annotation> <xs:documentation> Which lines of output should be examined. Indicated by single numerical value, comma-delimited list, or colon-delimited range; by convention, N = total number of lines returned. The parser will detach itself from the stream once the last line indicated here is seen. </xs:documentation> </xs:annotation> </xs:element> <xs:element maxOccurs="unbounded" ref="token" /> <xs:element maxOccurs="unbounded" ref="put" /> </xs:sequence> <xs:attribute name="name" use="required" type="xs:string" /> <xs:attribute name="stderr" type="xs:boolean" /> </xs:complexType> </xs:element> <xs:element name="put"> <xs:annotation> <xs:documentation> Value assigned to the name is the token with the given index in the returned list; a put without index assigns the entire list to the name. </xs:documentation> </xs:annotation> <xs:complexType> <xs:attribute name="name" use="required" type="xs:string" /> <xs:attribute name="index" type="xs:integer" /> </xs:complexType> </xs:element> <xs:element name="token"> <xs:complexType> <xs:sequence> <xs:element minOccurs="0" ref="token" /> </xs:sequence> <xs:attribute name="delim" use="required" /> <xs:attribute name="return" use="required" type="xs:string"> <xs:annotation> <xs:documentation> Indicates which segments generated by the tokenization should be returned; can be a single numerical value, comma-delimited list, or colon-delimited range; by convention, N = total number of segments created. </xs:documentation> </xs:annotation> </xs:attribute> </xs:complexType> </xs:element> <xs:element name="validator"> <xs:annotation> <xs:documentation> A regular expression or an acl used to validate an attribute value. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:choice> <xs:element name="regex" type="xs:string" /> <xs:element name="acl" type="xs:string" /> </xs:choice> <xs:element name="error-message" minOccurs="0" type="xs:string" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="launch-tab"> <xs:annotation> <xs:documentation> This is a recursive description of the main UI element associated with the control part, the Launch Tab. NOTE: I do not as yet know how many of the underlying widget options need to be exposed here for effective building in the createControl method. What follows is minimal. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element maxOccurs="unbounded" ref="tab-controller" /> </xs:sequence> <xs:attribute name="advancedModeEnabled" type="xs:boolean"> <xs:annotation> <xs:documentation> If this is set to true, a tab-controller which allows the user to browse the local file system and/or workspace to open and edit a custom batch script will be created; this will include copy and renaming buttons. </xs:documentation> </xs:annotation> </xs:attribute> </xs:complexType> </xs:element> <xs:element name="tab-controller"> <xs:annotation> <xs:documentation> as in: org.eclipse.ptp.rm.ui.launch.ExtendableRMLaunchConfigurationDynamicTab; a top-level "switch" between views. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element name="title" type="xs:string" /> <xs:element minOccurs="0" maxOccurs="unbounded" ref="group" /> <xs:element minOccurs="0" maxOccurs="unbounded" ref="tab-folder" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="tab-folder"> <xs:complexType> <xs:sequence> <xs:element name="title" type="xs:string" /> <xs:element maxOccurs="unbounded" ref="tab-item" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="tab-item"> <xs:complexType> <xs:sequence> <xs:element name="title" type="xs:string" /> <xs:element minOccurs="0" maxOccurs="unbounded" ref="group" /> <xs:element minOccurs="0" maxOccurs="unbounded" ref="tab-folder" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="group"> <xs:annotation> <xs:documentation> We may want a few tweaks here, such as borders, etc. Dynamic attribute means this group gets rebuilt if there is a button allowing the user to select displayed attributes. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element name="title" type="xs:string" /> <xs:choice> <xs:choice maxOccurs="unbounded"> <xs:element ref="group" /> <xs:element ref="tab-folder" /> <xs:element ref="widget" /> </xs:choice> <xs:element ref="discovered-attributes" /> </xs:choice> </xs:sequence> <xs:attribute name="columns" type="xs:integer" /> <xs:attribute name="dynamic" type="xs:boolean" /> </xs:complexType> </xs:element> <xs:element name="discovered-attributes"> <xs:annotation> <xs:documentation> Indicates that all the attributes defined at runtime be placed in this group. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element name="single-table" minOccurs="0"> <xs:annotation> <xs:documentation> The presence of this element indicates all attributes should be placed in a single table; otherwise, attributes are mapped serially to widgets by type. </xs:documentation> </xs:annotation> <xs:complexType mixed="true"> <xs:attribute name="checkbox" type="xs:boolean" /> <xs:attribute name="editable" type="xs:boolean" /> </xs:complexType> </xs:element> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="widget"> <xs:complexType> <xs:sequence> <xs:element name="label" minOccurs="0" type="xs:string" /> <xs:element name="content"> <xs:annotation> <xs:documentation> The text of this element will be parsed for references to the resource manager environment; e.g., ${a}, version-${b}, etc. </xs:documentation> </xs:annotation> <xs:complexType> <xs:sequence> <xs:element ref="arglist" /> </xs:sequence> </xs:complexType> </xs:element> <xs:element name="saveValueTo"> <xs:annotation> <xs:documentation> The text of this element will be the name of a property </xs:documentation> </xs:annotation> </xs:element> </xs:sequence> <xs:attribute name="type" default="text"> <xs:annotation> <xs:documentation> Custom, fixed widgets: 1. select-attributes: push button which brings up wizard exposing attribute names with checkboxes, allowing dynamic groups to be rebuilt. 2. show-script: push button which displays the batch script as it would currently be submitted. </xs:documentation> </xs:annotation> <xs:simpleType> <xs:restriction base="xs:string"> <xs:enumeration value="text" /> <xs:enumeration value="push-button" /> <xs:enumeration value="radio-button" /> <xs:enumeration value="spinner" /> <xs:enumeration value="checkbox" /> <xs:enumeration value="combo" /> <xs:enumeration value="browse-button" /> <xs:enumeration value="table" /> <xs:enumeration value="select-attributes" /> <xs:enumeration value="show-script" /> </xs:restriction> </xs:simpleType> </xs:attribute> <xs:attribute name="read-only" type="xs:boolean" /> </xs:complexType> </xs:element> <xs:element name="monitor"> <xs:annotation> <xs:documentation> I have for the moment left this undefined. </xs:documentation> </xs:annotation> </xs:element> </xs:schema>
I have not yet decided whether it will be simpler (or more efficient) to use JAXB runtime to build the actual instance objects from valid instances of the XML, or whether we just might want to parse it in using a DOM parser and then pull from it as necessary.