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 - Uffe

Pages: 1 ... 5 6 [7] 8 9 ... 18
Suggestions and Requests / Add-In commands in toolbar
« on: December 17, 2013, 03:47:06 am »
In Tools - Customize, you can add a new toolbar and drop commands into it. Cool.

I'd like an AddIn-specific toolbar, in one of two ways. Either

1) Add to the Commands - Categories list the currently loaded Add-Ins, and for each such category add its currently available menu items to the Commands list, then automatically enable/disable the toolbar when the corresponding Add-In is enabled/disabled using some black magic, or

2) Add a new profile category "toolbar" to the MDG Technology thingy, which lets the MDG Technician specify an icon, a tooltip and an Add-In function to be called using some other black magic. Just simple buttons, no fancy dropdowns or such.

Actually, the second one doesn't sound all that black now that I think about it. It's pretty much on a par with the rest of the MDG Techsorcery.

What do you other high priests of the dark MDG circle say?

Suggestions and Requests / Skip GUI customization on first start
« on: November 27, 2013, 07:03:12 pm »
Hi all,

The first time EA is started, before you get to pick a project, it invites you to customize the GUI.
While well-intentioned, this actually tends to cause newbies more problems than it solves because they'll pick something that sounds simple and then poof, something they need is missing from the menus and there is not hint as to why.
We've seen several posts on this forum related to this issue, and for someone in my role (providing support to a large organization) it's something that has to be dealt with every time there's a batch of new users (yes the information is on the Intranet but who has time to read that? I wanna get stuck in!)

SO - I suggest you skip the first-start non-optional GUI customization popup, and replace it with a prominent button on the start page.

This way, new users will still get the option very quickly, but they will at least have a chance to look at the GUI they're customizing first.



Suggestions and Requests / Unlock button in Select <Item> dialog
« on: November 22, 2013, 09:04:32 pm »
Hi there!

The Select <Item> dialog, aka Construct Picker in the API, allows you to select a specific type/stereotype element or create a new one.

What it doesn't do is allow you to lock the package you need in order to create the element in the right place. If you try and the package is locked, EA flashes a "You do not have the required permissions to perform this operation" message.

So, either a button to unlock (there's space) or an option to unlock from the message box would be nice.



Suggestions and Requests / Model Validation: Fix
« on: November 14, 2013, 09:17:51 pm »
Hej hullo hi,

An option to right-click a model validation output line and select Fix would be very useful for easily-remedied validation failures.

This could be implemented as a set of broadcasts similar to EA_OnRun*Rule(), except they would include the identity of the rule which triggered the output, and return a boolean stating whether or not a fix was applied.

For extra brownie points, a Suggest option could also be included, which would trigger a similar broadcast which would return a string with the suggested fix, which EA would then display in a dialog along with buttons to Apply and Skip.



Suggestions and Requests / Hierararchical Manage Locks for the love of...
« on: October 24, 2013, 05:48:43 pm »

- EA Support, Uffe.
- Yes, hello, so-and-so is off sick today. Could you unlock such-and-such a package?
- I can release all so-and-so's locks, is that OK?
- No, just package such-and-such please.
- OK, give me a couple of hours.

Could we please get either a tree view in the Manage Locks dialog, or a button Unlock Recursive, or a pop-up "Do you wish to unlock child elements as well?"

I mean OK, fine, I'm a consultant, I get paid by the hour. But I'd rather be developing custom profiles or a cool Add-In than unravelling a flattened-out package hierarchy to unlock stuff selectively.
Hell, I'd rather be battling with the RTF template editor...!

Suggestions and Requests / Custom metatypes in Audit Settings
« on: November 13, 2013, 10:21:05 pm »

It would be very helpful to be able to select custom (MDG) metatypes in addition to the standard UML ones in the Audit Settings - Audit Options - Customize - Types dialog.

This way, you could choose to audit changes to classes in, say, a domain model without drowning in changes to implementation classes, or system-level but not subsystem-level use cases.



Suggestions and Requests / "Paste Element": show name; all current
« on: October 16, 2013, 10:17:23 pm »

The Paste Element dialog asks you how to paste an element; as link, instance, etc. The Options button allows you to change the default behaviour for this dialog.

How about adding a checkbox "Do this for all current items"?

This would only be enabled if you were pasting multiple elements, and would cause all elements to be pasted the same way (suppressing the subsequent Paste Element dialogs).

In addition, when pasting multiple elements (and not checking the hypothetical box), it'd be useful to see the element's type and name in the dialog, maybe in the title bar: "Paste Class MyClass".


Suggestions and Requests / Linked Feature in tooltip
« on: October 29, 2013, 12:43:08 am »

Link to Element Feature is a cool function which is occasionally extremely useful indeed. It may not (correct me if I'm wrong) be strictly UML compliant, but what's that between good friends eh?

One thing which would make it even more useful is if the linked feature, if any, were shown in the connector tooltip, which currently shows "A > B" for a connector from A to B. "A.attr > B.op" would be great.



Suggestions and Requests / Command Sets in MDG Technologies
« on: October 28, 2013, 07:20:10 pm »

Command sets are really cool, but in order to realize their potential as a productivity enhancer we need to get them into MDG Technologies.

Nobody who runs a three-person programming/modelling team will care, but those of us who support hundreds of users including business analysts, project managers and other non-IT-nerd-type people will thank you forever.



Suggestions and Requests / Command Sets: Security override
« on: October 28, 2013, 07:16:00 pm »

As noted in, it is quite possible to configure yourself into a corner by selecting the wrong command set: pick one that doesn't include Security, log into a security-enabled project, and you're stuck.

Would it be possible to override the command set and enable the security commands if the project has security enabled?

It shouldn't be a problem from the data integrity perspective: it will still be the user's permissions which determine whether or not they can actually unlock stuff.



Suggestions and Requests / Request: "Unlock" button in Properties dialog
« on: August 06, 2013, 12:25:03 am »
Hey hi,

Just what it sounds like - a button to unlock the current element. There's plenty of space in the dialog, and it's annoying to have to close the dialog, right-click and select Apply Lock, then open the dialog and find the right tab again.
Could even be a Lock/Unlock toggle.


Suggestions and Requests / Traceability: hide backwards link
« on: May 24, 2013, 04:19:45 pm »
Hi all,

When you're traversing a model structure using the Traceability window, you get something like this:
Code: [Select]
    links to
            links from
You can't expand the inner "A" node, since that would lead to a loop in the tree.

I'd like an option to hide backwards links, so that the inner "A" node would not be shown. If that would turn the "links from" node into a leaf, that node should be hidden too.

Sound good?




The specific types of connectors aren't shown in the Traceability window. This means dependency-based connectors get lumped together as "depends on / needed by" and association-based ones become "links to / links from."

Having the option to display connector stereotypes would help a LOT. The name could be useful too, as could source/target roles and guards, but the stereotype is the important one.

In this example, A has a normal dependency to B and a trace to C:
Code: [Select]
    depends on
        C («trace»)

Of course, it would also be nice to be able to separate the link nodes based on stereotype. We'd then get something like:
Code: [Select]
    depends on
    depends on («trace»)
But that's a change in how the window works and I'm sure not everyone would agree with me on that one.

But I am absolutely positive that everyone in the entire wide world agrees with the first suggestion.



Suggestions and Requests / Traceability: show stereotypes
« on: May 25, 2012, 10:42:25 pm »

The traceability window is a very handy little tool but it uses its own connector labels, which can be confusing and sometimes downright misleading.

Consider for instance two components, where one has a «use» and the other an «abstraction» connector to the same interface.
Both these relationships will be reported as "needed by", since they are stereotyped dependencies. In the «use» case this might be permissible, but «abstraction» is quite a different kettle of barramundi.

This would be resolved if the connector stereotypes were shown, as in "needed by («use»)" and "needed by («abstraction»)". This solution will also work for third-party profiles, which an effort to come up with separate descriptions for all UML connector types would not.

Any takers?



Suggestions and Requests / Link Note to connector note
« on: May 05, 2012, 12:58:04 am »
Hey hi ho,

I can create a Note and link it to an Element Note.
I can create a Note and link it to a Connector Tag.
Is there a particular reason why I can't create a Note and link it to a Connector Note?

Only it'd be really really useful.
As I'm sure you'll all agree....... Right? :)


Pages: 1 ... 5 6 [7] 8 9 ... 18