Messages

1

Suggestions and Requests / Globally hide parentheses?

« on: February 16, 2009, 12:51:02 am »

I was very happy to see a new option (I think...) in EA 7.5 beta to hide the empty parentheses on methods without parameters.  That's particularly nice for properties where those parentheses just looked silly!  

But I'd really like to globally enable this option.  Is that already possible, and if so how?  Otherwise, I would submit this as a feature request.

2

Suggestions and Requests / Re: Hide operations by name

« on: April 07, 2005, 03:21:31 am »

Thanks for the link, I didn't think to look in the General board...  

So there's currently no way to persist custom stereotypes in code.  Like the other poster I'd definitely welcome some kind of "magic comments" or the like to make this possible, although personally I'd be just as happy if I could simply provide a list of "don't show" names to EA.

3

Suggestions and Requests / Hide operations by name

« on: April 07, 2005, 01:10:15 am »

I'm using EA as a documentation tool. I'm creating UML diagrams from reverse-engineered C# source code.

Now I often have to hide operations that I don't want to show up in a specific diagram. As far as I can tell, EA offers two ways to automate operation visibility in a diagram: by code visibility (public, private, ...) or by stereotypes.

Code visibility is way too coarse for me, so I'm using custom "pseudo-stereotypes" to hide the operations that I don't want to show up.

However, EA drops any stereotypes I've added when the same code is reverse-engineered the next time. So I have to go through all my classes and re-apply those stereotypes, which is quite a tedious job!

Is there any way to provide EA with a list of names of class operations that should be hidden in a specific diagram, perhaps even with wildcards? Or is there a way to somehow embed my pseudo-stereotypes in the code?

If there isn't I'd like to supply such a feature (either or both of the two ideas) as an enhancement request. It would greatly simplify reverse engineering for me. Thanks!

4

General Board / Re: Free (as in beer) proprietary tools ...

« on: November 18, 2005, 03:32:34 am »

Visual C++ Express actually contains the full C/C++ compiler which can target both .NET and native code. You do need to download the Platform SDK separately but otherwise it appears to be the real thing. I was surprised to find out about that, too...

5

General Board / Re: Diagram icons missing in PostScript/PDF output

« on: October 26, 2005, 11:34:15 pm »

Well, that was quick! Sparx Support has replied to my bug report -- it's a known bug which they hope to fix in the near future. Meanwhile, the workarounds are creating an RTF report for the diagram, or saving as an image.

6

General Board / Re: Diagram icons missing in PostScript/PDF output

« on: October 26, 2005, 06:05:29 am »

Thanks, I wasn't aware of the dedicated bug report page and thought the Sparx folks would monitor this forum anyway. I've submitted a bug report now.

I think it's rather unlikely that I'm the only one who ever tried to create a PDF file from a package diagram, though, and I can't quite imagine what would be special about my system to produce such an odd bug...

7

General Board / Diagram icons missing in PostScript/PDF output

« on: October 26, 2005, 02:29:42 am »

(Looks like nobody's reading the Suggestions forum, so I'll repost here...)

When you create a package diagram of a namespace, EA will put small icons next to all elements that differentiate classes, interfaces, and nested namespaces.

Those icons also appear when saving to a graphics file. They are replaced with monochrome icons when printing to a HP PCL printer.

However, when printing to a PostScript printer or to Adobe's  PDF creator, those icons are completely missing. Only the text appears.

The workarounds are not satisfactory: exporting to a graphics file either renders the text as bitmaps (= ugly) or shows the icons on a black background within the yellow/gray component boxes (= also ugly).

I can't imagine that this should be hard to fix, given that EA already has dedicated for-print icons which are correctly used when printing to PCL printers.

So could you please fix this, so that we get proper icons in PostScript and PDF output? Thanks.

8

General Board / Re: Preserving Stereotypes when reverse-engineerin

« on: April 13, 2005, 01:27:19 am »

I guess the silence means "no"...

9

General Board / Re: Preserving Stereotypes when reverse-engineerin

« on: April 10, 2005, 10:00:08 am »

Or how about using the .NET attributes for this purpose when reverse-engineering .NET code?

As I said in another thread, I'm concerned with controlling operation visibility in r-e'd code. I just discovered that EA can r-e .NET attributes just fine. I could simply create a "HideInDiagram" custom attribute, and label my methods accordingly in the code.

Unfortunately, EA handles attributes as tagged values, and there's no way to control diagram visibility with tagged values...

But I also would be happy with a very simple reverse-engineering option: "Keep custom stereotypes (y/n)". Or with a simple list of names to show/hide class members in diagrams.

Simon, could you please tell us whether you intend to do anything about this issue -- manual control of the reverse-engineering process?  It's really important for those of us who use EA to document code, as opposed to creating it.

10

Bugs and Issues / Re: Parameterized Interfaces?

« on: May 12, 2010, 05:01:23 pm »

Good to know, but that doesn't really help me.  I use EA only for documentation, i.e. reverse-engineering, and I don't want to manually recreate those classes/interfaces after every code import.

11

Bugs and Issues / Parameterized Interfaces?

« on: May 11, 2010, 08:56:08 pm »

Does EA support interfaces with type parameters in class diagrams?  That is, the little box in the top right corner that says "T" (or whatever) for the generic type argument.

When I reverse-engineer code from C# this little box correctly appears for generic classes, but not for generic interfaces.  There does not appear to be an option for type parameters in the property dialog for these interfaces, either.

If this is indeed not possible, could you please add this functionality?  Thanks.

12

Bugs and Issues / Re: Hide initial field values?

« on: February 17, 2009, 08:49:26 pm »

Yes, please!  I would even say that such an option should default to hiding the initial value since it's rarely useful -- typically the diagram just shows the first couple of character followed by an ellipsis anyway.

13

Bugs and Issues / Re: Hide initial field values?

« on: February 16, 2009, 07:12:22 pm »

Open your diagram and invoke the Properties dialog (F5). Click the Features tab and clear the Show Property String option. This should do what you need.

Thanks for your reply, but unfortunately this option does not do what I need.

"Show Property String" only controls display of the {readOnly} modifier, not of the initial property value.  The initial value is always shown, regardless of whether this option is checked or not.  I don't really care about the {readOnly} tag, it's the initial value I want to hide...

Any other ideas?

14

Bugs and Issues / Hide initial field values?

« on: February 16, 2009, 12:47:38 am »

When reverse-engineering C# source code that contains fields with initialization expressions, EA always shows these expressions whenever the fields themselves are shown.  This is also the case in EA 7.5 beta.

Example: public static readonly MyField = new MyType(...);

In this case the whole "new MyType(...)" expression (or as much as will fit in the box) always shows up in any diagram that shows MyField.

Question: Is there an option somewhere to suppress this display?  Currently, my recourse is to simply delete this expression when generating UML diagrams for documentation, but of course that's rather laborious.