OS.bee Documentation for Designer

From OS.bee documentation
Revision as of 10:00, 28 March 2018 by Yang (Talk | contribs) (OS.bee Documentation for Designer)

Jump to: navigation, search

OS.bee Documentation for Designer

Here are frequently asked questions from designers which are not mentioned by other Documentation. In this Page, you could find the answer to your question.

Get Started

Pitfalls with new Eclipse installations

Be aware that when installing a new Eclipse environment to look at the preferences for DS Annotations and check the box "Generate descriptors from annotated sources" as OS.bee makes heavy use of automatically generated component descriptors in the OSGI-INF directory. It is unchecked by default for incomprehensible reasons.

Ds annotations.png

Don't forget to set your target platform correctly as described in the installation guide.

Eclipse Installation / Installation SWF / New Project from GIT

Question:

Using Eclispe Neon, execution of Installation Software-Factory as described in the Installation notes, Connect to a GIT Archiv, After Building Workspace the application is not valid (see Screen-Shot) Try to clean the Project was not successful.

File:Xx.png
Figure x: xxx

Answer:

The version of the installed Software Factory and the version needed for the project do not match.

Please install the appropriate Software Factory version.

cvs2app - question regarding ENUM types

Question:

cvs2app is mentioned in the documentation "App up in 5 minutes" and gives the possibility to create one app directly out of the csv file. One of the first steps is to create an entity, which is generated based on the information on the first line (which contains the column names). In consequence I have 2 questions regarding ENUMS:

  • Is it possible to create an entity ENUM out the csv - FILE?
  • Is it possible to use an existing ENUM entity during the creation of the app?

For example:

in the entity there is already a definition:

 enum type_ps {
            PROCESS_DESCRIPTION, ORGANISATIONAL
}

and the csvfile looks as follows:

ticket_type_number;ticket_type_description;ticket_type_ps_type
1;CRS handling;PROCESS_DESCRIPTION
2;Administrative;ORGANISATIONAL
3;Delivery package;ORGANISATIONAL
4;Software behaviour;PROCESS_DESCRIPTION

Answer:

Yes it is possible. When using the latest version (from feb 2018), it is possible to supply various meta-information to each column. One meta-info is the hint to the application builder that this column is meant to be a ENUM. By default it wouldn't be possible to guess that fact.

Launch Application from Eclipse (very slow)

Question:

When starting the Application from within the Eclipse it took very long time until the application is up. Are there some settings to be controlled?

Answer:

If you experience very slow performance with Eclipse itself as well as the application you launch from Eclipse it might be a good idea to check the virus scanner you have installed. Some virus scanners check all the files inside the Eclipse installation directory, the Eclipse workspace and the GIT repository which might lead to extreme slow performance. Ask your Administrator how to avoid this.


Structure of the documentation page

In the OS.bee Software Factory Documentation page, there are 3 Headlines used to structure the page:

  1. OS.bee DSL Documentation
  2. Other OS.bee-Specific Solutions
  3. OS.bee Third-Party Software Solutions

At the end of Chapter one there are some helpful hints to work with eclipse. To start with Eclipse and the SWF these hints could be very useful.

  • One more hint: to Use STRG-Shift-O to organize the import inside the DSL.

Modeling

I18N.properties (Reorganization of obsoleted values)

Question:

Each DSL has its own I18N.properties to translate the values. The property file includes values that are no longer used in the Modell-Description. (For example based on an addition correction of orthography)

The update of the properties seems to be one-way. Is there a function to reorganize the values (with the target to drop all obsoleted values)?

Answer:

No, there is no way to delete unused entries for security reasons. Translations to foreign languages are expensive. The modeler is responsible to delete obsoleted entries.

Display values from ENUMS in Table

Question:

We are using a perspective with the combination of table and dialog to maintain master data. The dialog contains combo-boxes to select values based on ENUMS. The selected value is displayed in the dialog, but is not displayed in the table. Is there a way to display the values of the selected ENUM in the table, with the objective to use the filter?

Answer:

This is not yet implemented but on schedule.

References to images

Question:

Using for example MENU.DSL Documentation:

package <package name>[{ 
expandedImage <expandedImage String>

Keyword expandedImage and collapsedImage is used to define the menu image when the menu expanded or collapsed. Where can we find the valid names of the Images to use in the String?

Answer:

The images can be selected by using hitting <CTRL>+<space> right behind the keyword. This will lead you to the image picker. Double-click on "Select icon...“ , then you can select the image from the image pool. The corresponding name will be added to the model. If you want to add own images to the pool follow How to add pictures.

Perspective with master/slave relationship

Question: We created a perspective with a "master" table and 5 related "slave" tables. The target is to select one row in the master table and automatically filter the data rows in the slave tables related to the selected data. Where can we find information about the way to realize this goal?


Answer: With "master" you probably mean an entity that has relative to the "slave" an one2many relationship. Views of their underlying tables will synchronize if they share a filter with the selection of another table. The filter must use the same entity as the entity of a table that selected a row. So every datamart of a "slave" must have a condition in a many2one join where the "one" side is the entity that should synchronize while selecting rows. In effect the selected row tries to change all filters to the same ID if the same entity is used. This applies to all open views in all open perspectives. In datamartDSL, using the keyword "filtered", the result in the application is a combo-box in the header of the table, which allows to select a value which is used as a filter of all tables of the perspective which have use the same condition; using the keyword “selected”, the result will be a list-box.

File:Xx.png
filtered (optional) => Combo-box
File:Xx.png
selected => List-box

Perspective use border to show boundary for each sash

Question: Is it possible to show border between each sash-container used in a perspective? From our point of view the arrangement is not pretty clear to the user. It could be helpful to get a clearer view to show (optional) real visible border between sash-container and even between parts. for example:

  • green border between parts of a sash-container
  • blue border between sash-container

Is this already possible?

Answer: For the moment there are no plans to colorize borders by means of grammar keywords. It is possible by changing the CSS for the currently selected theme.


Core Dev

Apps