Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Topics - DMT

Pages: [1] 2 3
1
Suggestions and Requests / Drag & Drop to Reorder in Tree
« on: November 21, 2003, 05:57:30 am »
We have many sequences per use case, and we put these sequences under the use case. The problem is that because we can't create packages under the use case, we have a long list of sequences.

We'd like to use drag & drop to reorder so we don't have to create a sequence at the top of the list, then click the down button 20 times to get it placed in the correct location.

Thanks!
Dave

2
Suggestions and Requests / Don't Expand Dragged Nodes
« on: June 09, 2003, 09:37:42 am »
When I drag a classs from one package to another, it becomes expanded when dropped. This is pretty annoying with classes that have many attributes along with the associated property accessor methods.  I find myself dragging, dropping, clicking the "-" to collapse, dragging, dropping, etc.

This gets REALLY frustrating when dragging and dropping more than one or two classes between packages.  I perform this operation several times a day as I come up with better namespace breakdowns for my classes.

3
Suggestions and Requests / Sort List of Scenarios When Linking Notes
« on: July 01, 2003, 01:33:47 pm »
It would be great if the listview shown when selecting a scenario from a use case, when linking a note to a scenario, could be made sortable by clicking the column headers.

I add all scenarios to my sequence diagrams, and I like to keep them ordered alphabetically so I can easily find them.

4
Suggestions and Requests / Enter comments when creating properties
« on: June 26, 2003, 09:44:14 am »
When I'm adding attributes to a class, I typically will create the property accessor method right there by checking the Property box.  

This works great, except for the next step in my workflow--entering a comment for the property. It would be EXTREMELY helpful if you added a comment box to the dialog that pops up to create the property. This way, I wouldn't need to flip between the attributes and operations dialogs while I create my attributes.

5
Suggestions and Requests / WYS != WYG for Notes on Sequence Diagrams
« on: June 24, 2003, 07:31:17 am »
I list the use case text down the left-hand-side of my sequence diagrams, and place the appropriate messages for each part of the use case to the right of the text.

This works great on-screen, but when the diagram is printed, the printed version has the text compressed by about 20%, so none of the messages line up any longer.

Is there any way to address this?  I'd really like my printed layout to match what I see on-screen, truly giving me "WYSIWYG" layout capabilities.

Thanks!

6
Suggestions and Requests / ... Button Next to All Type Combo Boxes
« on: June 05, 2003, 09:26:57 am »
It would be extremely helpful to make it so all type combo boxes have the "..." button next to them to display the Select Type dialog.

I love this feature in the Attributes dialog, but it is sorely missed in the Operations dialog for the type for both return values and parameters.

I have many duplicates because my database tables show up in the list.

Alternatively, is there any way to specify that classes in a particular package should be hidden in the Type combo box?  That would be the PERFECT solution, since I don't want any of my tables to appear in that combo box.

7
Suggestions and Requests / Generation of Inner Classes
« on: June 03, 2003, 05:28:13 pm »
When I trigger code generation for a diagram that contains one or more C# inner classes, a message is displayed that says generation of inner classes is not currently supported.

I have two questions:
1. Are there plans to add support for this in the near future?
2. How have other people gotten around this in the meantime? Generate the class separately and copy/paste?

8
General Board / Branching the Model for New R&D Team
« on: April 30, 2010, 01:35:06 pm »
Here's the deal. I have a new team that I'm ramping up that's going to add a new area to our product. We're going to branch the source code so they can work in isolation for about 6 months to build the new area, without interrupting the core product's team ability to ship additional releases during that time.

My challenge is determining how to allow the new team to work on their own independent EA model for the product, and merge their changes back into the "real" EA model (the one for the core team) when they are ready to merge their new area of the product into the main team's code base.

Until that time, I don't want the main team confused by changes and additions made by the R&D team so I want to keep them isolated.

Does anyone have advice on the best way to handle this?
* Make a copy of the model, export to XMI when ready to merge, and do a diff and merge, and import the result?
* Use version control at the package level? (I've played around with this and the implementation leaves a lot to be desired when you're used to Visual Studio's integration with Team Foundation Server.)
* Use version control of an entire EAP model file?

Any sggestions would be greatly appreciated.

Thanks,
Dave

9
General Board / Source Safe - Configure for Entire Project?
« on: November 03, 2003, 08:50:05 am »
Now that Source Safe integration has been added, I intuitively expected to do something like the following, based upon my experience with Visual Studio:

1. Configure EA so it recognizes that I'm using Visual Source Safe, providing my user name, etc.
2. Right-click on the root node, and click something like, "Add Model to Source Control..."
3. Specify the name and location of the project in Source Safe.

That's it.  Now, this isn't EXACTLY what I'm seeing.  It appears that I need to configure source control for every single package in the entire project, including sub-packages.  It appears that I need to manually type in file names and versions.

It's really strange, and I'm not sure exactly why I would need to go through all of this.  One mouse click and one dialog should handle it.

Is there any other way to configure a model to use source control?  Is it really this difficult?  Or am I missing something.

Thanks in advance,

Dave

10
General Board / Hide Sequence Message Parameters
« on: September 23, 2003, 01:22:32 pm »
Is there any way to completely hide parameters for sequence messages? I thought I had figured out a way to do this at some point, but now I can't figure it out.

Thanks in advance,
Dave

11
General Board / Embed, as opposed to link file?
« on: September 19, 2003, 07:25:38 pm »
Does anyone know if the ability to embed a file inside the EA database has been added?

I'd like to be able to write my use cases (which can run up to 20 or 30 pages) in Word, but still model them in EA. I'd then attach the Word doc to the use case.

Needless to say, I don't want to just link to the doc; I want to have the doc accessible by any user viewing the model.

Thanks!
Dave

12
General Board / HELP! on UI Models
« on: August 18, 2003, 10:27:25 am »
My problem:

We're designing a Windows app that uses UI controls with specific behavior in multiple UI forms.  Most of the behavior is the same, regardless of where the control is used, but there are some exceptions to this rule. This means I need to be able to specify both "standard" behavior as well as "special" behavior.

Example: "State Text Box"
Standard requirements:
- Max length is 5 characters
- Forces characters to uppercase as they're entered

Special requirements:
- On this form, the State text box is disabled if the user has not yet entered the company name

Here are my ideas for possible solutions:
1. Either specify the "standard" requirements with the UI classes themselves, then add instances of the controls to a custom diagram that shows the mock-up of the UI.
   - Benefit: Since my problem fits nicely into a class/object paradigm, this solution would accomplish my goal well.
   - Drawback: I'm coupling the requirements to the implementation model. Theoretically, the UI mode shouldn't "know" about the implementation model.

2. Create special "UI requirements" classes, and do the same thing above.
  - Benefit: I keep the UI requirements decoupled from the implementation design
  - Drawback: Whenever I use the combo box to select a type for an attribute or method return type, I'll see all of my "requirements" classes

3. Create a completely separate model for the UI requirements, and use the method outlined in #2 above.
  - Benefit: I won't have my type lists being cluttered by strange types
  - Drawback: Now I need to manage two separate models, and flip between them to look at different types of information.

You know, the best solution would be if I had the ability, on a per-package basis to select an option that says, "Don't show classes in this package in type list." This would solve my problem.

Does anyone out there have any other ideas? What do you think is the best solution?

Thanks!
Dave

13
General Board / <br> in HTML Documentation
« on: August 15, 2003, 12:55:09 pm »
I have constraints on my entity properties that have carriage returns/line feeds in them. As I posted in other threads, I haven't had luck with the RTF documentation printing these correctly, so I tried the HTML documentation.

It APPEARS that it's trying to put them in, but they just show up as "<br>" in the HTML docs, as opposed to carriage returns.

Anyone have any idea how to fix this? Or is this simply a  bug in EA?

Thanks,
Dave

14
General Board / What does the red triangle mean?
« on: August 09, 2003, 04:50:07 am »
I have a custom diagram on which I have several entity classes. I have a small red triangle over the top center of one of them pointing down. I have no idea what this means.

Does anyone else know what this indicates?

Thanks,
Dave

15
General Board / How are you Handling Change?
« on: August 08, 2003, 08:04:34 am »
I'm wondering if I could get some general feedback about how others are handling changes to the model that they then need to disseminate for review and for updating the team.

My current thought goes something like the following for the initial pass:
1. Generate RTF documentation for version 1
2. Save as a Word document
3. Have it reviewed. The reviewer can use the track changes feature in Word to notate changes
4. Update the model
5. Re-generate the RTF doc
6. Check it into Source Safe, marking it as version 1.

For a revision, it would go something like:
1. Make changes in EA until you're ready for another pass to be reviewed and distributed.
2. Generate an RTF file
3. Overwrite the version 1 file, selecting Word's option to merge changes. This marks all changes as long as you have the track changes feature turned on.
4. Submit for review. The reviewer can just scroll through the parts that have changed.
5. Make edits from reviewer, check in, and distribute the doc. Again, all changes are marked and easily visible in the Word doc.
6. Accept all of the changes in the document before the next pass.

Does anyone have a better option than this?  This seems the only way to communicate each and every change for each new version of the document.

Thanks,
Dave

Pages: [1] 2 3