Flip Stretch issue

[Libbya]

Is there a simple solution to this issue?  See the attached block.  Use the stretch grip and see how the middle diamond remains properly formed.  Then use one of the two flip actions and go back and stretch it some more.  I understand why it behaves in that manner, but is there an easy solution that does not involve changing the basepoint of the stretch action or visibility states/lookups?

[jarredmonday]

For the moment...putting a visibility state in place of the flip action on the right side is the only thing I can come with right now.

[Libbya]

I have a solution using visibility states/lookups, I was just wanting to reduce the number of vis states in the finished block if possible.  This is part of a much more complex block and multiplies the vis states in the block by 4X.

[steven-g]

Flipping heck, I'm getting nowhere with this, I did manage to get it going with 2 flips driving the changes, but it's basically smoke and mirrors because it is visibility states and a lookup that do the actual work.

[ojuris]

I have little bit different solution

[ojuris]

https://dl.dropboxusercontent.com/u/401581/flip%20stretch-2.dwg

[steven-g]

Nice, here is a variation on the theme (why can't you turn off grips for a flip) anyway you get the idea, I'm starting to look at the options using a rotate parameter, maybe that will work ?

[Libbya · Accepted solution]

You guys are awesome.  Definitely some alternative thinking.  Very cool.

Steven-g, you can turn off the grip for a flip but you cannot chain them (without a lookup).  There have been a couple times that I've felt some pain about that, most notably with section line callout tools where I want both arrows to flip and a grip at each arrow.  Ah well, thankfully there are lookups.  🙂  Your previous version using visibility and lookups was what I had meant to post, but had forgotten to remove two of the 'arrows' from the flip selection sets.  That method is also what I was trying to avoid but it does work.

Ojuris, very cool concept and sorta fits the bill, but I still wasn't quite satisfied.  I knew that in the finished block I would be adding a stretch action to the top and bottom of the 'box' and so would be facing the same issue.

After a LOT of head scratching, sleep, etc... I came up with a fully workable solution.  It is the one on the left. 

I also included an early concept I tried yesterday before even posting the thread that really uses the same concept with stretch angles of 0 and 180, but the flip actions do not work on them.  WHY???  The one on the left using the added chained linear stretch and angles of 90 and 270 DOES work??  I guess cad figures 0 and 180 are equal, but 90 and 270 aren't??  Ah well, at the end, I'm glad for a solution...