Deprecated. Core building blocks for the openMDM Business Logic and Web Frontend...

Clone this repo:
  1. c7163c7 Version 5.2.0M1 by Matthias Koller · 3 years, 11 months ago master 5.2.0M1
  2. 496e02c Merge branch 'dev' by Matthias Koller · 3 years, 11 months ago
  3. 5f2a0aa Fixed command to create file user in readme.md by Matthias Koller · 3 years, 11 months ago
  4. 59fb158 Improved exception message for failed imports by Matthias Koller · 3 years, 11 months ago
  5. fe01279 Added changes for 5.2.0M1 by Angelika Wittek · 3 years, 11 months ago

Documentation

Before you start, please take a look at the Getting Started Guide which can also be found in the Download area of the project and in the docs directory relative to this README.md

Minimum requirements

  • JDK 1.8.0_45

This project uses a Gradle Wrapper

Build dependencies

Before you can build and deploy the application, you have to checkout the following repositories in the same root directory as the current project

  • git clone https://git.eclipse.org/r/mdmbl/org.eclipse.mdm.api.base.git
  • git clone https://git.eclipse.org/r/mdmbl/org.eclipse.mdm.api.default.git
  • git clone https://git.eclipse.org/r/mdmbl/org.eclipse.mdm.api.odsadapter.git

You MUST have the following directory structure afterwards

.
├── org.eclipse.mdm.api.base
├── org.eclipse.mdm.api.default
├── org.eclipse.mdm.api.odsadapter
└── org.eclipse.mdm.nucleus

Now you can use gradlew install in the org.eclipse.mdm.nucleus directory, but you should configure the application first.

Build, deploy and configure the application

  1. Edit the org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/core/property.service.ts and set the data_host variable according to your deployment
    (This property will be used to create the REST URLs to communicate with the backend).

  2. Build the application with gradlew install.
    The command gradlew install executed at org.eclipse.mdm.nucleus creates a ZIP archive named build/distributions/openMDM_application-${version}.zip at org.eclipse.mdm.nucleus/build/distributions
    The ZIP archive contains the backend org.eclipse.mdm.nucleus-${version}.war and the configuration files in the configuration directory

  3. Check that the database for the preference service is running, otherwise start it with asadmin start-database as described in the Getting Started Guide.

  4. Deploy the WAR file org.eclipse.mdm.nucleuss-${version}.war on your application server. Make sure to deploy the WAR file with application name org.eclipse.mdm.nucleus, otherwise the LoginRealmModule is not able to lookup the ConnectorService EJB.
    Additionally in the following examples, we assume that the context root is also set to org.eclipse.mdm.nucleus.
    When deploying from command line you can use asadmin deploy --name org.eclipse.mdm.nucleus /path/to/org.eclipse.mdm.nucleus-${version}.war

  5. Copy the content of the extracted configuration folder to GLASSFISH_ROOT/glassfish/domains/domain1/config.
    There is also a system property org.eclipse.mdm.configPath, which can be used to redefine the location of the folder to another location.

  6. Edit the org.eclipse.mdm.connector/service.xml file to configure the data sources

  7. Configure a LoginModule with name MDMRealm (See section Configure LoginModule for details)

  8. Restart the application server

  9. Visit the main page of the client to make sure everything works fine.
    The main page of the client should be available under http://SERVER:PORT/{APPLICATIONROOT} (eg: http://localhost:8080/org.eclipse.mdm.nucleus)

Configure LoginModule

MDM 5 backend implements the delegated approach to roles and permissions, wherein the data sources (ASAM ODS server, PAK cloud) themselves can implement their own security scheme (which they already have) and then delegates the appropriate user data to the backends.

In the case of ASAM ODS servers this is done using a technical user and the for_user attribute on the established connection, so it is a form of ‘login on behalf’.

In the case of PAK Cloud this is done by passing the user name along with a http header X-Remote-User which is then used by PAK Cloud to establish the users roles (from an internal database or an external authentication provider).

Before the user is ‘logged in on behalf’ he is authenticated by a LoginModule within the Glassfish application server. There are different implementations available (e.g. LDAP, Certificate, JDBC, ...). To keep this guide simple, we will setup a FileRealm, which stores the user information in a flat file.

The following command will create a FileRealm with name MDMRealm that stores the users in a file called mdm-keyfile

asadmin create-auth-realm --classname com.sun.enterprise.security.auth.realm.file.FileRealm --property file=${com.sun.aas.instanceRoot}/config/mdm-keyfile:jaas-context=MDMRealm:assign-groups=Guest MDMRealm

To be able to login you need to explicitly add users to the MDMRealm. Here we add the user MdmUser

asadmin create-file-user --authrealmname MDMRealm --groups Admin:DescriptiveDataAuthor:Guest MdmUser

Currently three roles are supported by openMDM, adapt the users according to your needs

  • Admin: Administrator role allowed to access the Administration section in the web application
  • DescriptiveDataAuthor: Users allowed to edit context data
  • Guest: Default application users allowed to browse and read measurement data

Keep in mind that the roles are mainly used to enable/disable certain actions in the web application. Authorization in openMDM is still done via the delegation approach, e.g. the adapters are responsible for handling authorization on data.

Next you need to add the following snippet to your login.conf in ${com.sun.aas.instanceRoot}/config/

MDMRealm {
  com.sun.enterprise.security.auth.login.FileLoginModule required;
};

As a last step you have to provide the credentials of the technical user in adapter configuration in service.xml. For the ODS adapter you have to specify the parameters user and password of the technical user. For example

<service entityManagerFactoryClass="org.eclipse.mdm.api.odsadapter.ODSContextFactory">
    ...
	<param name="user">sa</param>
	<param name="password">sa</param>
	...
</service>

Make sure to restart Glassfish afterwards.

Remove MDMLoginRealm from previous versions

In versions 5.0.0M1, 0.10 and older the configuration of a custom login realm was necessary. If you configured your Glassfish server instance for one of these versions, you can remove the old configuration options and artifact, as they are no longer needed.

Steps to delete the old configuration and artifacts

  • Delete the JAR file org.eclipse.mdm.realm.login.glassfish-VERSION.jar from GLASSFISH_ROOT/glassfish/domains/domain1/lib
  • Open the Glassfish login configuration file at GLASSFISH_ROOT/glassfish/domains/domain1/config/login.conf
  • Delete custom MDM realm module entry to this config file
    MDMLoginRealm {
      org.eclipse.mdm.realm.login.glassfish.LoginRealmModule required;
    };
    
  • Remove the MDMLoginRealm by executing asadmin delete-auth-realm MDMLoginRealm or by deleting it in the Glassfish web console

Configure logging

MDM 5 uses SLF4J and Logback for logging. The default configuration file can be found at org.eclipse.mdm.nucleus/src/main/resources/logback.xml. It logs INFO level messages to mdm5.log in the logs folder of the Glassfish domain. If you want to customize logging, you can either edit the file within the WAR file or preferably provide your own logging configuration via system parameter in the JVM settings in Glassfish -Dlogback.configurationFile=/path/to/config.xml

Available REST URLs

Please use the generated OpenAPI Specification, which is generated at build time. The OpenAPI Specification is available in org.eclipse.mdm.nucleus/build/openapi/openapi.json or at runtime at http://{SERVER}:{PORT}/{APPLICATIONROOT}/openapi.json. Furthermore a Swagger UI is available at http://{SERVER}:{PORT}/{APPLICATIONROOT}/swagger.html

Structure of the URLs

  • SERVER the hostname of the Glassfish server
  • PORT the port of the Glassfish server
  • APPLICATIONROOT is the context root under which MDM is deployed
  • SOURCENAME is the source name the underlying data source

Environment

  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/localizations

Business objects

The following parameters are recurring throughout the different URLs

  • ID the identifier of the business object

  • CONTEXTTYPE is one of

    • unitundertest
    • testsequence
    • testequipment
  • DATATYPE is one of

    • STRING
    • STRING_SEQUENCE
    • DATE, DATE_SEQUENCE
    • BOOLEAN
    • BOOLEAN_SEQUENCE
    • BYTE
    • BYTE_SEQUENCE
    • SHORT
    • SHORT_SEQUENCE
    • INTEGER
    • INTEGER_SEQUENCE
    • LONG
    • LONG_SEQUENCE
    • FLOAT
    • FLOAT_SEQUENCE
    • DOUBLE
    • DOUBLE_SEQUENCE
    • BYTE_STREAM
    • BYTE_STREAM_SEQUENCE
    • FLOAT_COMPLEX
    • FLOAT_COMPLEX_SEQUENCE
    • DOUBLE_COMPLEX
    • DOUBLE_COMPLEX_SEQUENCE
    • FILE_LINK
    • FILE_LINK_SEQUENCE
  • FILTERSTRING is a String defining a filter. For example Test.Name eq "t*" filters for all tests which names begin with t.
    Strings should be quoted with ". " characters within strings have to be escaped with a backslash. For backward compatibility ' is also allowed to quote strings, but may be blocked in URLs in some environments, thus " should be preferred.

  • REMOTE_PATH is the remote path of a file link as it is returned in the attributes of type FILE_LINK and FILE_LINK_SEQUENCE
    Make sure to properly URL escape the value of the remote path. Especially slashes have to be escaped with %2F.

Examples

Most of the Business objects support the following calls (examples for TestStep)

  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps?filter={FILTERSTRING}
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps/searchattributes
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps/localizations
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps/{ID}
  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps/{ID}/contexts/{CONTEXTTYPE}

For TestStep and Measurement it is also possible to receive files

  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/teststeps/{ID}/files/{REMOTE_PATH}

Query endpoint

  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/query
    Example: curl -POST -H "Content-Type: application/json" -d '{"resultType": "test", "columns": ["Test.Name", "TestStep.Name"], "filters": { "sourceName": "SOURCENAME", "filter": "Test.Id gt 1", "searchString": ""}}'http://sa:sa@localhost:8080/org.eclipse.mdm.nucleus/mdm/query

  • http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/suggestions
    Example: curl -POST -H "Content-Type: application/json" -d '{"sourceNames": ["SOURCENAME"], "type": "Test", "attrName": "Name"}' http://sa:sa@localhost:8080/org.eclipse.mdm.nucleus/mdm/suggestions

ATFX Import

POST: http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/environments/{SOURCENAME}/import

Imports an ATFX file into the specified environment. The ATFX file must conform to the MDM 5 data model.

TODO Templates

Three content types are accepted

  • application/xml: The ATFX file is sent in the body of the POST request.

  • multipart/form-data: The ATFX file and its binary files are provided as multipart form data. The field name represents the original file name and the field value the contents of the file.
    The ATFX file is detected by its file extension ‘atfx’ or the media type ‘application/xml’. Only one ATFX file can be sent at a time.

  • application/zip: The ATFX file and its component files are provided as a zip file. The ATFX file is detected by its file extension ‘atfx’. Only one ATFX file can be included in the zip file.

The POST request returns a JSON-Object with the following properties

  • state: Status of the import. Either OK or FAILED.
  • message: Error message in case the import fails.
  • stacktrace: Stacktrace in case the import fails.

The returned HTTP status is either 200, if the imported succeeded or 400, if the import failed.

ATFX Export

POST: http://{SERVER}:{PORT}/{APPLICATIONROOT}/mdm/export

Accepts a shopping basket XML file and exports the contained elements into an ATFX file. Currently it is only supported to export objects from one environment.

Returned content types in response:

  • application/xml: The response body contains the exported ATFX file

  • application/zip: The response contains a zip file with the exported ATFX file and accompanying component files.

Example: File download

  • Login
    curl -XPOST --cookie-jar cookie -H "Content-Type: application/x-www-form-urlencoded" -d j_username=sa -d j_password=sa http://localhost:8080/org.eclipse.mdm.nucleus/j_security_check

  • Request a test and extract the remotePath from the first file reference
    FILE_PATH="$(curl --cookie cookie -s http://localhost:8080/org.eclipse.mdm.nucleus/mdm/environments/MDMNVH/teststeps/10 | jq -r '.data[0].attributes[] | select (.name | contains("MDMLinks")) | .value[0].remotePath')"

  • Urlescape remotePath if necessary
    FILE_PATH="$(echo $FILE_PATH | python -c 'import sys,urllib;print urllib.quote(sys.stdin.read().strip(), "")')"

  • Request the file content
    curl --cookie cookie http://localhost:8080/org.eclipse.mdm.nucleus/mdm/environments/MDMNVH/teststeps/10/files/$FILE_PATH

Preference Service

Preference service stores its data to a relational database. The database connection is looked up by JNDI and the JNDI name and other database relevant parameters are specified in src/main/resources/META-INF/persistence.xml. The default JNDI name for the JDBC resource is set to jdbc/openMDM. This JDBC resource and its dependent JDBC Connection Pool has to be created and configured within the Glassfish web administration console or through asadmin command line tool.

Furthermore the schema has to be created in the configured database. Therefore database DDL scripts are available for PostgreSQL and Apache Derby databases in the folder schema/org.eclipse.mdm.preferences of the distribution. Other databases supported by EclipseLink may also work, but is up to the user to adapt the DDL scripts.

Examples

  • Receive values
    curl -GET -H "Content-Type: application/json" http://localhost:8080/org.eclipse.mdm.nucleus/mdm/preferences?scope=SYSTEM&key=ignoredAttributes

  • Set value
    curl -PUT -H "Content-Type: application/json" -d '{"scope": "SYSTEM", "key": "ignoredAttributes", "value": "[\"*.MimeType\"]"}' http://localhost:8080/org.eclipse.mdm.nucleus/mdm/preferences

  • Delete entry
    curl -DELETE http://localhost:8080/org.eclipse.mdm.nucleus/mdm/preferences/ID

FreeTextSearch

Configuration

  1. Start ElasticSearch. ElasticSearch can be downloaded at https://www.elastic.co/products/elasticsearch. The minimum supported Elasticsearch version is 7. For testing purpose, it can be simply started by executing bin/run.sh

  2. Edit the configuration (global.properties) to fit your environment. You need an ODS Server which supports Notifications. All fields have to be there, but can be empty. However certain ODS Servers ignore some parameters (e.g. PeakODS ignores pollingIntervall since it pushes notifications).

  3. Start the application. At the first run it will index the database. This might take a while. After that MDM registers itself as NotificationListener and adapts all changes one-by-one.

Run on dedicated server

The Indexing is completely independent from the searching. So the Indexer can be freely deployed at any other machine. In the simplest case, the same steps as in Configuration have to be done. The application can then be deployed on any other machine. All components besides the FreeTextIndexer and its dependencies are not user. Those can be left out, if desired.

Connector and service.xml

The service.xml contains all information necessary for the Connector-Service to connect to the available datasources/adapter instances. Since this information includes secret information like passwords, it is possible to provide lookups, which gives you the possibility to specify tokens as references to properties defined elsewhere.

There are different lookups available

  • sys: Looks up variables defined as system properties
  • env: Looks up variables defined as environment variables

Example:

<param name="password">${env:odsPassword}</param>

Known issues

If you run into java.lang.ClassNotFoundException: javax.xml.parsers.ParserConfigurationException not found by org.eclipse.persistence.moxy this is a bug described in https://bugs.eclipse.org/bugs/show_bug.cgi?id=463169 and https://java.net/jira/browse/GLASSFISH-21440. This solution is to replace GLASSFISH_HOME/glassfish/modules/org.eclipse.persistence.moxy.jar with this file http://central.maven.org/maven2/org/eclipse/persistence/org.eclipse.persistence.moxy/2.6.1/org.eclipse.persistence.moxy-2.6.1.jar

If you run into java.lang.ClassNotFoundException: com.fasterxml.jackson.module.jaxb.JaxbAnnotationIntrospector not found by com.fasterxml.jackson.jaxrs.jackson-jaxrs-json-provider you have to download http://central.maven.org/maven2/com/fasterxml/jackson/module/jackson-module-jaxb-annotations/2.5.1/jackson-module-jaxb-annotations-2.5.1.jar and put in the GLASSFISH_HOME/glassfish/domains/domain1/autodeploy/bundles folder

Client preferences

The applications preferences are managed in the administration section. This section can be accessed via the Administration button in the main navigation bar or via http://localhost:8080/org.eclipse.mdm.nucleus/administration

A preference is a pair of a unique key and a value. The key is composed of a prefix defining the purpose of the preference followed by an arbitrary but unique identifier string. It is recommended to choose the identifier the same as the preferences name field, in case there is one. The value holds the preference's data in a JSON string.

The following preferences, sorted by their scope, can be set

  • User

    • Basket
    • View
    • Filter
  • System

    • Node provider
    • Shopping basket file extensions
  • Source

    • Ignored attributes

However, it might be necessary to reload the application before a newly defined preference is available or any changes on an existing preferences are applied.

WARNING: Corrupted preferences can result in malfunctions of the application.

User scope

A user scoped preference's area of effect is limited to the logged in user. All user scoped preferences can also be set in dialogs in the main application.

  1. Basket
    Basket preferences keys must start with the prefix basket.nodes.
    This preference has the fields items and name and holds all the information for saved baskets.

    • The field items holds an array of MDMItems, providing the relevant information of a related node, i.e. source, type and id.
    • The field name defines the name, which is provided in the main application to load this basket.

    Example

    {
      "items": [{"source":"MDMNVH","type":"Test","id":38}],
      "name": "basketExample"
    }
    
  2. View
    View preferences keys must start with the prefix tableview.view.
    This preference has the fields columns and name and holds the layout information for the tables displaying the search results and the basket nodes.

    • The field columns holds an array of ViewColumn objects. A ViewColumn is an Object with the fields type, name, sortOrder and an optional field style.
    • The field type can be set to all available MDM data types, i.e. Project, Pool, Test, TestStep, Measurement, ChannelGroup, Channel.
    • The field name field specifies an attribute, which must be an searchable attribute for the given type.
    • The field sortOrder can be set by the number 1 (ascending), -1 (descending), or null (unsorted). Only one column of the array can have a non-null value sortOrder at a time. The ViewColumn's style element can hold any CSS-style object. However, it is supposed to contain only the columns width. The column order in the array is identically with the appearance in the table.
    • The field name defines the name, which is provided in the main application to load this view.

    Example

    {
     "columns": [
       {
         "type": "Test",
         "name": "Id",
         "style": {"width":"75px"},
         "sortOrder": null
       }
     ],
     "name": "viewExample" 
    }
    
  3. Filter
    Filter preferences keys must start with the prefix filter.nodes.
    This preference has the fields conditions, name, environments, resultType and fulltextQuery.
    It provides the information for the attribute based / advanced search.

    • The field conditions holds an array of Condition objects. A Condition specifies a search condition for attribute based search. It consists of the fields type, name, operator, value and valueType. The Condition's type can be set to all available MDM data types, i.e. Project, Pool, Test, TestStep, Measurement, ChannelGroup, Channel.
      • The Condition's name field specifies an attribute, which must be an searchable attribute for the given type.
      • The Condition's operator field, holds on of the following numbers: 0(=), 1(<), 2(>), 3(like).
      • The Condition's value field holds a string array containing input for the attribute based search.
      • The Condition's resultType field should match the type corresponding to the attribute specified in the 'name filed, e.g. string, date or long.
    • The field name defines the name, which is provided in the main application to load this filter.
    • The field environments holds an string array with the names of the sources that should be included in the search.
    • The field resultType can be set to all available MDM data types (see above). Only nodes of this type will be included in the search.
    • The field fulltextQuery holds a string containing full text search input.

    Example

    { "conditions": [
        {
          "type": "Test",
          "attribute": "Name",
          "operator": 0,
          "value": [],
          "valueType":"string"
        }
      ],
      "name": "filterExample",
      "environments": ["sourceName"],
      "resultType": "Test",
      "fulltextQuery": ""
    }
    

System scope

System scoped preference are applied globally.

  • Node provider
    The navigation tree structure can be defined via a node provider. The default node provider is set in org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/navigator/defaultnodeprovider.json.
    It is recommended not to change the default node provider. Instead new node providers can be added as preferences.
    Their keys must start with the prefix nodeprovider.. Once a custom node provider is supplied it can be selected in the dropdown menu in the navigation tree header.

    • Structure

      • First layer/root nodes
        In the node provider each layer of nodes of the navigation tree is defined in a nested object.
        The first layer of nodes, is always the environment level. This is necessary due to the provided endpoints.
        The first layer consists of the fields name, type, and children.

        • The field name sets the name, which is displayed in the application to select the corresponding node provider.
        • The field type defines the data type of the nodes, which is always Environments on the first layer.

        The next layer of nodes are defined via the field children.

      • Children
        A child object consists of the fields type, query, and children.

        • The field type sets the data type of this layer of nodes. It can be set to all available MDM data types, i.e. Project, Pool, Test, TestStep, Measurement, ChannelGroup, Channel.
        • The filed query holds the URL to load this layer of nodes. The first part of the query for each child object should be / plus the type of the layer in small letters followed by an s.
          For a nested child objected a filter is attached to the query in the form: ?filter=<parentType>.Id eq {<parentType>.Id}. The placeholder has to be replaced by the actual parent type (set in the field type in the parent child or root layer, see example below).
          At runtime the curly braces will be replaced by the Id of the parent node. Further child objects, and thereby more sublayers, can be nested via the field children.

        The children field does not need to be set for the last layer of nodes.

    • Examples

      • Minimal node provider

        { "name": "My name to display", "type": "Environment"}  
        
      • Node provider with two child layers

         {
            "name": "My name to display",
            "type": "Environment",
            "children": {
              "type": "Project",
              "query": "/projects",
              "children": {
                "type": "Pool",
                "query": "/pools?filter=Project.Id eq {Project.Id}"
              }
            }
          }
        
  • Shopping basket file extensions

    When downloading the contents of a shopping basket, a file with extension mdm is generated.
    Additional file extensions can be adding by poviding a preference with key shoppingbasket.fileextensions.
    Here you can define a list of objects with attributes label and extension.
    For example: [ { "label": "MyTool", "extension": "mdm-mytool" }, { "label": "OtherTool", "extension": "mdm-other" } ].
    If MyTool has a file handler registered for the extension mdm-mytool, the application will be launched if the browser automatically opens the file after download.

Source scope

Source scoped preferences are applied at any user, but are limited to the specified source. The source can be specified in the Add Preference or Edit Preference dialog.

  • Ignored Attributes
    The ignore attributes preference must have the exact key ignoredAttributes. An identifier must not be added.
    The preference specifies all attributes, which are supposed to be ignored in the detail view. The preference is a simple JSON string holding a list of attributes in the form {“.”}.
    The placeholders and have to be replaced by the actual type and name of the attribute which should be ignored, respectively.

    Example:
    ["*.MimeType", "TestStep.Sortindex"]

Create a module for the Web application

Any MDM module needs to be a valid Angular module aka NgModule.
A NgModule consists of the module definition, components, services and other files that are in the scope of the module. The component can hold any content. The component must be declared in a module definition to grant accessibility in the rest of the application.
All related files should be stored in a new module subfolder in the app folder org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app (eg. org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/new-module)

Example module

An example for a new module can be found at org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/example-module

Creating a MDM module

  1. Create a new folder eg. org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/new-module

  2. Create an Angular component (eg. mdm-new.component.ts) inside that new folder

     import {Component} from '@angular/core';
     @Component({template: '<h1>Example Module</h1>'})
     export class MDMNewComponent {}
    

    A component is defined in a Typescript file with the @Component() decorator.
    Any HTML content can be provided here in an inline template or via a link to an external HTML resource. Thereafter the component itself, which is supposed to hold any logic needed, is defined and exported.
    For more details see https://angular.io/guide/architecture-components.

  3. Create a minimal NgModule (eg. mdm-new.module.ts ) inside that new folder

     import { NgModule } from '@angular/core';
     import { MDMCoreModule } from '../core/mdm-core.module';
     import { MDMNewComponent } from './mdm-new.component';
     @NgModule({imports: [MDMCoreModule], declarations: [MDMNewComponent]})
     export class MDMNewModule {}
    

    The imports array holds all modules from the application needed in this MDM module. It should always hold the MDMCoreModule, which provides basic functionalities.
    On the other hand a NgModule grants accessibility of components of this module in other directives (including the HTML template) within the module (in a declaration array) or even in other parts of the application (in an export array).
    For more details see https://angular.io/guide/architecture-modules.

Embedding a module (no lazy loading)

To embed this new module in MDM you have to register this module in the MDMModules Module.

  • Import the new module at org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/modules/mdm-modules.module.ts

  • Register a route to the new module at org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/modules/mdm-modules-routing.module.ts

    { path: 'new', component: MDMNewComponent}
    
  • Furthermore you have to define a display name for the registered route in the links array in org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/modules/mdm-modules.component.ts

{ path: 'new', name: 'New Module' }

For further information refer to the Angular documentation:

Lazy loading and routing module

For lazy-loading (recommended in case there is a high number of modules) embedding of the module is slightly different.

  { path: 'example', loadChildren: '../example-module/mdm-example.module#MDMExampleModule'}

Additionally, a NgModule, the so called routing module (eg. mdm-new-routing.module.ts), is needed to provide the routes to this modules components.

  const moduleRoutes: Routes = [{ path: '', component: MDMExampleComponent }];
  @NgModule({imports: [RouterModule.forChild(moduleRoutes)], exports: [RouterModule]})
  export class MDMExampleRoutingModule {}

Filerelease module

The filerelease module is stored in the following folder org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/filerelease

It can be embedded as any other module described above.

{ path: 'filerelease', component: MDMFilereleaseComponent }
{ name: 'MDM Files', path: 'filerelease'}

Adding a module to the detail view (eg. filerelease module)

To make the filerelease module available in the detail view it needs to be imported in the corresponding MDM Module org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/details/mdm-detail.module.ts
Thereafter, the MDMFilereleaseCreateComponent can be imported to the org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/details/components/mdm-detail-view/mdm-detail-view.component.ts.
Then the following has to be added to the org.eclipse.mdm.nucleus/org.eclipse.mdm.application/src/main/webapp/src/app/details/components/mdm-detail-view/mdm-detail-view.component.html file:

  <mdm-filerelease-create [node]=selectedNode [disabled]="isReleasable()"></mdm-filerelease-create>

It should be located right after the add to basket button:

<div class="btn-group pull-right" role="group">
  <button type="button" class="btn btn-default" (click)="add2Basket()" [disabled]="isShopable()">In den Warenkorb</button>
  <mdm-filerelease-create [node]=selectedNode [disabled]="isReleasable()"></mdm-filerelease-create>
</div>

Copyright and License

Copyright (c) 2015-2018 Contributors to the Eclipse Foundation

See the NOTICE file(s) distributed with this work for additional
information regarding copyright ownership.

This program and the accompanying materials are made available under the
terms of the Eclipse Public License v. 2.0 which is available at
http://www.eclipse.org/legal/epl-2.0.

SPDX-License-Identifier: EPL-2.0