Book a Demo

Author Topic: Baseline merge  (Read 4646 times)

qwerty

  • EA Guru
  • *****
  • Posts: 13584
  • Karma: +397/-301
  • I'm no guru at all
    • View Profile
Baseline merge
« on: January 20, 2012, 10:56:23 pm »
I wonder why nobody else had stomach ache with this one so far: When you compare a model to XMI and merge a single element EA starts the whole model compare from the beginning - which takes quite a while when comparing complete models. I simply ask: why? The result of the merge should be that the merged element is no longer different and can simply be removed from the list. This should take only a millisecond.

q.

qwerty

  • EA Guru
  • *****
  • Posts: 13584
  • Karma: +397/-301
  • I'm no guru at all
    • View Profile
Re: Baseline merge
« Reply #1 on: January 31, 2012, 01:17:28 am »
No response. So either it's me being too dumb to use this tool or nobody is using it. Neither is advantageous.

q.

Geert Bellekens

  • EA Guru
  • *****
  • Posts: 13519
  • Karma: +573/-33
  • Make EA work for YOU!
    • View Profile
    • Enterprise Architect Consultant and Value Added Reseller
Re: Baseline merge
« Reply #2 on: January 31, 2012, 05:58:50 pm »
Sorry q.

I try to avoid merging in EA like the plague  :-X

Geert

Paolo F Cantoni

  • EA Guru
  • *****
  • Posts: 8626
  • Karma: +257/-129
  • Inconsistently correct systems DON'T EXIST!
    • View Profile
Re: Baseline merge
« Reply #3 on: January 31, 2012, 07:00:25 pm »
Quote
Sorry q.

I try to avoid merging in EA like the plague  :-X

Geert
But sometimes, like q, you still catch it...   :(

Paolo

Inconsistently correct systems DON'T EXIST!
... Therefore, aim for consistency; in the expectation of achieving correctness....
-Semantica-
Helsinki Principle Rules!

qwerty

  • EA Guru
  • *****
  • Posts: 13584
  • Karma: +397/-301
  • I'm no guru at all
    • View Profile
Re: Baseline merge
« Reply #4 on: February 03, 2012, 02:38:02 am »
Not much feedback still. In another thread a user also mourns the usability of the baseline merge. The reason is obvious. The iuse case Sparx has designed this feature is simply incorrect. One basic fault is that each single merge restarts the comparison (which is simply not necessary). Another one is that the result of the comparison shall be correct and useful. Now it's almost correct after a couple of bug fixes. Only one bug is still open where identical  double tags are marked as difference. There is space for more improvement if filtering was possible. That is to allow ignoring differences in certain tags like creation date, author, diagram position etc. (and not by restarting the comparison but directly on the result set).

I think there is potential in the baseline merge.

q.