Best way to define the data dictionary for CenterStage aspect

In anticipation of the migration to CenterStage, we are looking for some suggestions to what is the best way to define the data dictionary for aspect.

From what we learned, Even though CenterStage provides full data dictionary support for aspect, the Composer 6.6 does not provide the full support for defining the data dictionary for aspect. Composer 6.7 is supposed to support data dictionary for aspect.

If we plan to utilize the data dictionary for aspect today(before 6.7 release), what is the best way to make sure the DQL script created compatible with Composer 6.7 in the future?

Find more posts tagged with

Note:

If there are Accepted Answers, those will be shown by default. You can switch to 'All Replies' by selecting the tab below.

Accepted answers

agoodale

Perhaps you could give an example of what kind of incompatibility you are worried about? Composer 6.7 is providing a GUI that will generate the same DQL that you would write by hand. Ultimately, the DQL is setting data dictionary information for the aspect type; this functionality is not changed in 6.7.

I think once Composer 6.7 is out, your DQL script is simply no longer needed. I don't see how any incompatibility could be created.

All comments

agoodale

I think once Composer 6.7 is out, your DQL script is simply no longer needed. I don't see how any incompatibility could be created.

pwarren

On the face of it I would agree with Andrew. Your script installs information into the data dictionary. If it supports this, Composer 6.7 would perform the equivalent function or modify your existing data.

Because we will be working on the same canvass I cannot see any issues. Unless you are doing something specific?

hao.feng

Thank you both for sharing the insights. This really helps us in making the decision to go ahead to adopt the data dictionary approach.

The incompability items(It may have been addressed by the previous reply) are :

If I create the the data dictionary for aspect using the DQL, will the composer 6.7 be able to import the aspect artifactscorrectly along with the data dictionary created by DQL (Import artifacts from docbase)?

For example, if I create a data dictionary using the following DQL,

ALTER ASPECT test_aspect_dic MODIFY "aspect_attrib2" (SET label_text = NULL, SET help_text = NULL, SET comment_text = NULL, SET category_name = NULL, SET not_null_msg = NULL, SET format_pattern = NULL, SET format_pattern_tag = NULL, TRUNCATE val_constraint_msg, MAPPING TABLE (VALUE = 'dataValue' DISPLAY = 'displayValue' COMMENT = '', VALUE = 'dataValue1' DISPLAY = 'displayValue1' COMMENT = ''

Could it be correctly formatted as the following after the import?

If not, this would require all the aspect data dictionary defined by DQL to be recreated using the composer from scratch.

pwarren

Hao,

I have afew follow-on questions.

- What is this effort a part of?

- Why would you need to import this aspect back into a composer project?

- When and IF composer supported this would you not just make the equivalent set of changes in your composer project and re-install it as an upgrade?

Cheers

_Paul

hao.feng

Paul, Thanks for the followup.

In order to support the upcoming organization consoliation, we are refactoring our platform to be aspect driven. We have developed a home grown GUI based on GXT and hope to migrate to centerstage when the version 1.2 is released. So we plan to support data dictionary in our custom GUI app for now. This would ensure the easier transition to Centerstage in the future.

when composer 6.7 is released, it will be great if we could simply import the aspect along with its data dictionary information rather than redefine all of them from scratch.

After all, the change is the only constant. There is always the need to add some new attributes to the existing aspect. We would hope to import the existing aspect into composer project and simply make the data dictionary change using the new capability of composer 6.7 (compared to continue to make the Data Dic change using DQL.)