Presenting... Mister Looper!  A loop-making script.  Say the name using Big Bird's voice, if you're over 30 but under 50.  :lol:

Yes, it's a script which makes loops.  This is something I've wanted for a long time, but everyone kept not making it.  :lol:  So here it is.

The script is attached above, and will also have its permanent home on the page linked below.  Change extension of attached file to .py, to run the script.

Images and basic instructions to follow.

http://www.the.cage.page.phantom3d.net/TDMT_Match/other_scripts/various.html

The script extrudes a loop around a path, wherever geometry intersects a plane which has been positioned by the user.

Basic instructions, also present in the comments at the top of the script:

Quote -
Mister Looper: a loop-making script

• Position the MR.LOOPER PLANE prop to define the plane in which your loop will be created.
• Reduce scale of plane as much as possible, for best results.
• If no vertices are found, try moving the plane closer to some vertices on your actors.  If only edges are intersected, it won't run.
• Move the RUN SCRIPT dial to run.
• Move the END SCRIPT dial to end.
• Change Sections setting to alter number of radial divisions of main loop.
• Change Slices setting to alter number of divisions in each cross-section ring of loop.
• Change Thickness setting to vary radius of cross-section rings.
• Set Apply smoothing to > 0.0 to smooth the resulting loop, <= 0.0 to disable smoothing.  Smoothing may shrink the loop slightly.

Possible bugs:

• Convex hull code may fail with some complex vertex selections, creating a tangle.  Try re-running with the plane scale reduced, to avoid this.
• UV mapping may not be correct.  U and V seem confused in some way which hasn't yet been sorted out.

As noted, I'm not sure the UV-mapping is quite correct, but I don't seem to be able to fix it just yet.  Hopefully I'll be able to update it with improved UV handling, soon.

Note also that the script uses a Poser scene event callback, so if this script is run while another script is already running a callback, there could be problems.

I hope someone can get some use from this script.

Enjoy!  :laugh: 

Thanks Cage! This will be usefull. I recently spent many hours creating a chain, very time consuming.

WOW! you're really on a roll.  :thumbupboth:

Thanks, I'll play with this and see what comes from it.

That looks very useful, indeed.

Hehehehe...brilliant! ;o)

Laurie

Thanks, guys.  :laugh:  I hope it can be useful, somehow.

The basic procedures used here can be adapted to generate a skin or envelope/cocoon around selected geometry, too.  I just need to figure out how to work the selection process.  :unsure:

If I'm on a roll, I'm still expecting that to end, pretty much any time now.  :lol:

I forgot to note in the basic instructions that this only works with visible actors.  If you want to exclude actors from contributing to the loop path, hide them before running the script.

very nice , thanks.

"If I'm on a roll, I'm still expecting that to end, pretty much any time now.  "

Would you like Butter or Jam on that Roll???
dph

Quote - "If I'm on a roll, I'm still expecting that to end, pretty much any time now.  "

Would you like Butter or Jam on that Roll???
dph

Butter might be slippery, and jam is too sticky & gooey.  Can I opt for apple butter?  The forgotten butter!  Comes from apple cows.  :lol:

(Apologies.  I'm a bit punchy tonight, if punchy is the word I want.  I have this fershlugginer tune by Vangelis stuck in my head.  :cursing:  :lol:)

"(Apologies.  I'm a bit punchy tonight, if punchy is the word I want.  I have this fershlugginer tune by Vangelis stuck in my head.    )"

Nothing I presume that a good cup of Turkish Coffee couldn't cure!!
dph
PS: Your Apple Cows sound like they may have had a bit too much of the Grape!!

I wonder if it can do French curves...;) d'like to make some Art Noveau frames for some ersazt Mucha..;)

Quote - Nothing I presume that a good cup of Turkish Coffee couldn't cure!!
dph
PS: Your Apple Cows sound like they may have had a bit too much of the Grape!!

Ooh.  Turkish coffee.  Sounds good.  Maybe Irish coffee.  :laugh:

None of the grapes here.  I'm just a bit loopy, maybe, after trying to sort out some of the complications involved with this script.  :lol:  Always try the five hardest ideas which are least likely to work before working out the best one, that's my motto.  Or maybe it should be.  :lol: 

Quote - I wonder if it can do French curves...;) d'like to make some Art Noveau frames for some ersazt Mucha..;)

I think a French curve involves some concave portions, places where the curve would sort of loop back on itself.  Is that correct?  This works with a 2D convex hull, to generate the loop path.  So any concave areas that emerge are the result of failures in the convex hull code, which apparently has some odd bugs.  I inherited that part from one of the Python code sites, so I'm not sure I can debug it effectively, unfortunately.  :unsure:  The convex hull code which is supposed to be superior wouldn't work for me.

Quote - (Apologies.  I'm a bit punchy tonight, if punchy is the word I want.  I have this fershlugginer tune by Vangelis stuck in my head.  :cursing:  :lol:)

Vangelis will have that effect. Something about the cadence and repetitiveness... (running along the beach, now)

GREAT script, btw... and of course, one can save results to pp2 or whatever! How cool is this! Looks low-res, too, another bonus. Well done, Cage!

My sister was over for the holiday, earlier, and she was watching that movie about the runners.  Regular ear-worm, that theme song.  Sob.

The resolution can be set to suit your needs, using the "Sections" and "Slices" settings.  (I used the same terminology that Wings 3D uses in its options for creating a torus.)

I've taken a closer look at the results with textures, and the UV's are coming out worse than I thought (see attached - textures should be plain old tiles, not a mess).    Something's quite wrong with them.  Sorry, folks.  I didn't test enough and I posted too soon.  :blushing:

I'll try to fix it first thing tomorrow, and put up a corrected version.

Danged UV's confuse me every time.  :lol:

You're right, mate... we'll wait patiently. Good on you for sharing this!

I'll try to get the per-ring UV handling set up tomorrow.  Looks like the texsets array is not being constructed properly, so the indices are snarled up in surprising ways.  The results look okay in UV Mapper if you just glance, which is all I did earlier.    Look more closely and the flaws are evident.

Very nicely done. Appreciate the well commented script.

Thanks.

Apple Butter with Cinnamon, or Apple Butter with Nutmeg???
dph
PS: Are these Apple Cows Macintosh proficient, are are they only Lisa compatible???

Cage,

you said something about a cocoon, could this be adapted to create scull and face caps for hair props? If so it would be extremely usefull.

Brillant, absolutely brillant!

Your script sounds like a great thing. I tried mr_looper4e.py in P6, unfortunatly it does not seem to work for me. Is this a version limitation, or am I doing something wrong? The prop itself seems to function OK, but when I turn RUN SCRIPT to one, there is no indication that anything is happening. Same when I set END SCRIPT. No indication that python is running, no error messages, no new props. Am I supposed to set RUN SCRIPT to zero before setting END SCRIPT to one? Do I need to leave an interval between setting RUN SCRIPT and END SCRIPT, if so how long? Should I expect to see a python window? Any thoughts?

Quote - Your script sounds like a great thing. I tried mr_looper4e.py in P6, unfortunatly it does not seem to work for me. Is this a version limitation, or am I doing something wrong? The prop itself seems to function OK, but when I turn RUN SCRIPT to one, there is no indication that anything is happening. Same when I set END SCRIPT. No indication that python is running, no error messages, no new props. Am I supposed to set RUN SCRIPT to zero before setting END SCRIPT to one? Do I need to leave an interval between setting RUN SCRIPT and END SCRIPT, if so how long? Should I expect to see a python window? Any thoughts?

You're using Poser 6.  Hmm.  I've tested this with Posers 7 and 8 and found it working.  I would assume that if I'm using some Python method not available to Poser 6, there would be an error.  Hmm.

You shouldn't need any interval.  You can just delete the prop if the script won't run.  That will kill the callback, assuming the callback is running for you.

You didn't receive a printed statement that no vertices were found?  It doesn't sound like it.  There should only be a Python console if the plane has only intersected edges (or nothing), and hasn't found any vertices with which to work.

Are you running any other Python script which use a callback, when trying to run this?

I'll try to prepare a script version which will print statements about which step in the process has been reached.  If you run that, it should help identify where things are falling down.

Sorry for the trouble, Les.  I hope I can get this working for you.

Quote - Very nicely done. Appreciate the well commented script.

Thanks.

Thanks, Phil.  There's apparently a bit of confusion in Poserland about using matrices in PoserPython, so I tried to comment it well, since I'd figured out how to make them work.  I thought it might be useful for someone, that way.  You don't need those lessons, however.  :lol:

I just wish the danged UV's weren't screwy with this release. 

Quote - Cage,

you said something about a cocoon, could this be adapted to create scull and face caps for hair props? If so it would be extremely usefull.

My thought was to create a series of 2D convex hulls as cross-sections around a geometry.  So the idea would basically give a re-meshed 3D convex hull.  I'm not sure it would prove useful for something like skullcaps, which would presumably require some kind of careful design, and would need to be able to fit the scalp, whereas a convex hull would wrap over the ears, say, rather than working around them.

Possibly something in the process could help lead in the direction you suggest.  I'll think about that.  Hmm.

Quote - Apple Butter with Cinnamon, or Apple Butter with Nutmeg???
dph
PS: Are these Apple Cows Macintosh proficient, are are they only Lisa compatible???

Golly.  Maybe cinnamon?  I don't know.  I'd have to try both, to decide.  I didn't realize there were so many options.  :laugh:

Hurm.  I tried plugging them into the computer, these cows, and they weren't Windows compatible.  So I know that much.  I don't have a Mac handy with which to test.  If the choices were Lisa and Homer, I'd go with Homer.  Possibly they're actually Linux cows, pretending to be Apple cows, for some reason?  :unsure:

@lesbentley:

Try rotating the plane on any axis, to see if that changes anything.  Possibly this is a flaw in the vertex-selection process I'm using.  I'd expect that case to generate a console message, but it's probably something that should be tried, at least....

I'll set up the special-edit version for you.

It should print something like this, if all goes well (actors will vary):

Quote - Created plane
Callback started
Starting callback
Callback is running
Callback is running
Callback is running
Starting loop generation
Making loop
Getting actors
tongue2:1
rPinky2:1
chest:1
lToe:1
rEye:1
lIndex2:1
newloop
rThigh:1
head:1
lFoot:1
lInstep:1
rInstep:1
abdomen:1
hip2:1
lThumb1:1
lRing2:1
lMid1:1
rToe:1
lShin:1
rBigToe1:1
rForeArm:1
lThumb3:1
lEye:1
lBigToe2:1
rShldr:1
lRing3:1
lThumb2:1
lIndex1:1
newloop 1
lPinky3:1
lMid3:1
rMid1:1
lHand:1
lRing1:1
rRing1:1
lIndex3:1
rBigToe2:1
rHand:1
lBigToe1:1
rIndex3:1
rShin:1
rMid3:1
rRing3:1
jawUpper:1
lPinky2:1
lForeArm:1
rRing2:1
rThumb2:1
lCollar:1
rMid2:1
rFoot:1
lThigh:1
tongue1:1
rThumb3:1
rPinky3:1
hip:1
rIndex2:1
lPinky1:1
jawLower:1
lMid2:1
rThumb1:1
rPinky1:1
rIndex1:1
rCollar:1
neck:1
lShldr:1
Found vertices
Convex hull generated
Perimeter markers placed
Ring vertices placed
Verts, polys, sets created
UV mapping applied
New Geometry() instantiated
Prop created
Applying smoothing

I recall that someone had trouble with the use of poser. in another script, when using Poser 6.  I'm wondering it P6 may be having trouble here:
if( (event & poser.kEventCodePARMCHANGED) != 0):

If so, the script would seem to do nothing, as it apparently is.  If this is the case, I'll have to try to figure out how to properly invoke a poser typecode for older versions.  Hmm. 

they make changes to the python in poser a lot. the script that makes injection channels works in 7 and 8 to a degree in 6 but not at all in 5..

Quote - they make changes to the python in poser a lot. the script that makes injection channels works in 7 and 8 to a degree in 6 but not at all in 5..

I know they've changed the Python version, for P7+, but that shouldn't affect anything here.  I don't think I'm trying to do anything which couldn't be done in Python 2.1 or 2.2, or whatever P5 uses.  I didn't realize they'd made any drastic changes to the existing Poser Python procedures.  Huh.

I've Checked the Poser 5 PPy documentation, and it looks like the callback procedure should work in Poser 6.  So that shouldn't be the problem.

I've deliberately created a bad array to send to the geometry-creation step, and it looks like Poser may either seem to do nothing or may crash if it has a problem at that point.  No error messages are generated.  I'm not sure why it only crashes sometimes.  Hmm.  Possibly there's actually a problem in the geometry generation.

I'll await further information from Les before delving into the possible problem any more.

wasn't refering to your script, was refering to the one that showed up in the python folder of P7 or P6 that creates morph channels. it works in both but differently.

it doesn't work in P5 either.  loads but doesn't run. 
you can't run a script from a pose file in P5...maybe it's got something to do with that?
works great in PP2010 though!

do i run it for each individual loop or is there a way to do multiples?
& could Mr.Looper be a different shape than a plane?

-jan-

Quote - wasn't refering to your script, was refering to the one that showed up in the python folder of P7 or P6 that creates morph channels. it works in both but differently.

Okay.  Thank you.  :thumbupboth:

I've tested the script in Poser 5, and it runs without difficulty, so apparently the errors aren't related to the version.

I have noted a couple of cases where Poser didn't seem to respond to the "RUN SCRIPT" dial, in my tests today.  I deleted the prop and ran the script again (copy the prop transforms to the clipboard and apply when running again, if desired), and there weren't any problems.  I'm not sure why this happened.  It might be some obscure error in my code, or it might be one of Poser's many oddities.  :unsure:

Quote - it doesn't work in P5 either.  loads but doesn't run. 
you can't run a script from a pose file in P5...maybe it's got something to do with that?
works great in PP2010 though!

do i run it for each individual loop or is there a way to do multiples?
& could Mr.Looper be a different shape than a plane?

-jan-

Ooh.  We cross-posted.    I wasn't responding to you, with my last post.

I did test in Poser 5, however, and it ran for me without difficulty.    Which isn't a good sign, when debugging.  :scared:

Did you receive any errors when trying to run using Poser 5?

Multiple loops would require multiple defined planes, or else they'd all end up being the same loop.  As such, I hadn't considered any sort of batch handling.  but I suppose one could create a stack of loops, at regular intervals, using the same plane orientation.  Hmm.  So: currently there's no batch handling, but you've just made me think about possibly developing some.  :laugh:

A different shape than a plane?  Currently, no.  I'm having trouble trying to imagine anything useful resulting if I tried to unwrap a sphere and use its surface as the basis for the process which uses the plane.  You might get some very strange shapes.  :lol:  I suppose it could theoretically be done, using parametric coordinates or something, but I'm not sure the result would be useful.  I'll have to think about that.  :laugh:

i'll have to try it in P5 again then.

in PP10, if i tried to have more than 1 instance of Mr. Looper in the scene, none would run til i deleted & started over. 
made some cool bracelets for Antonia before i crashed the program....takes all the pain out of maneuvering them into place too.  sweeeet!

cross-crossposted.

no errors in my P5; it behaved like Les described.  it may be my version.  i never use P5 anymore anyhow but many still do. 

Quote - cross-crossposted.

no errors in my P5; it behaved like Les described.  it may be my version.  i never use P5 anymore anyhow but many still do. 

Even though P7 is available at $40...

& had i not accidentally won PP2010 i'd be grabbing that offer just to have a backup copy--P7 is a great leap forward from P5.

Quote - in PP10, if i tried to have more than 1 instance of Mr. Looper in the scene, none would run til i deleted & started over. 
made some cool bracelets for Antonia before i crashed the program....takes all the pain out of maneuvering them into place too.  sweeeet!

I think if you end up running more than one, only the most recent will work.  If I'm not mistaken, the callback being used by the script will be, er, usurped by a new version of the script if an old one is still running.  My experiences while testing seem to bear this out, but I am actually guessing.  Heh.  :laugh:

If that happens, though, you should just be able to delete all of the plane props and run the script again.

Did the script crash PP2010?  If so, under what circumstances?

Quote - no errors in my P5; it behaved like Les described.  it may be my version.  i never use P5 anymore anyhow but many still do. 

How odd.  I ran the script repeatedly in Poser 5, v 5.0.4.325.  

I think I'm on-track to fixing these UV problems, so hopefully necklaces with better UV-mapping can be made using the script, pretty soon....

i manage to crash 2010 at least once daily--it seems to be random so may be my elderly computer being senile.  but yes i crashed it twice while the script was active.  no error message other than 'so sorry we can't go on like this & Poser must close'.....again i think it's my hardware not your script.

better uv's will be most welcome.  they're ok as is for procedurals but textures would suck.

could i make a flat cylinder instead of a torus? (thinking choker necklace vs round chain)

A flat cylinder?  Do you mean, like a disk?    That might be an interesting variation on what's done here.

Do you mean something more like a heightened torus?  The skinning variant script I mentioned earlier might achieve something like that, although without the inner geometry.

The current script can't do either of these, because I didn't think of that.  :lol:  Something along these lines might be possible, though, using similar techniques.

Many good ideas coming out.  Thank you.  :laugh:

i mean something more like the heightened torus but i can see uses for either shape.  the skinning variant sounds promising too.  i made a big body bag in an early experiment & ran that through the clothroom, but it was too los-res to do much draping.  so i went back to jewelry.  this is a nifty script.

I also added some duplicate vertex position screening to the convex hull part of the process.  From what I've read online, the hull code I've used may encounter unspecified problems when finding multiple vertices at the same coordinates.  I suspect this is what creates the "tangled" paths which have been showing up sometimes in my tests.  Hopefully the screening will remove that problem.

Sorry for the bugs in the initial script.  This one is pretty much what I set out to make.  The only thing left to do is work out the welding of the UV's, which are currently split for every polygon.

So.  Five days in the making, anyways.  :lol:

Quote - i mean something more like the heightened torus but i can see uses for either shape.  the skinning variant sounds promising too.  i made a big body bag in an early experiment & ran that through the clothroom, but it was too los-res to do much draping.  so i went back to jewelry.  this is a nifty script.

A body bag?  How did you do that? 

Thank you for calling it nifty!  :laugh:  I've suggested the idea to a couple of people over the past few years, because I wasn't quite capable enough to pull it off, at the time.  Someone missed the chance to charge me $$$ for this script.  :lol:  Oh well.  It's been fun and I've learned a lot, and I hope you folks can use it.

I'll see if I can work out UV welding, then try moving on to attempt the skinning/cocooning idea I've mentioned.

Hi Cage,

The thread has gotten a lot longer since the last time I was her. I haven't read the intervening posts yet. I will do that after posting this.

Q: You didn't receive a printed statement that no vertices were found?
A: No.

Q: Are you running any other Python script which use a callback, when trying to run this?
A: No, not running any other py scripts.

Quote: "I'll try to prepare a script version which will print statements about which step in the process has been reached.  If you run that, it should help identify where things are falling down."

Thanks, I very much appreciate the edffort you are puting in to help me. I ran the new print statements version. Not much more joy that before.

Created plane
Callback started
Starting callback
Callback is running
Callback is running
Callback is running
Callback is running
[etc, etc]

I just get a long procession of "Callback is running", nothing else after the first three lines.

Thanks, Les.  That helps isolate where the process is failing.  The callback works and querying the callback works, but it apparently isn't recognizing the dials when they're moved.

The attached adds a couple of additional print statements, which might help narrow things down further.  I hate to keep putting you through this, but if you can run the test script and report the output, I may be able to fix this.  I have a couple of ideas, but I can't test them here.  The script is running on my Poser 5 installation, oddly. 

It's kind of a long shot, but have you altered the default dial sensitivity (assuming one can even do that) for Poser?  What units do you have Poser set up to use?  I'm set up for Poser Native Units.  I doubt that's the trouble, but possibly all dial responses are recalibrated somehow if PNU's aren't being used.

Quote - It's kind of a long shot, but have you altered the default dial sensitivity (assuming one can even do that) for Poser?

If it can be done, I don't know how.

Quote - What units do you have Poser set up to use?

I'm using PNU.

Will run the new script and report back shortly.

This time I tried  'mr_looper4e_edit.py' on a cylinder prop. First try I got this.

Created plane
Callback started
Starting callback
Callback is running
RUN SCRIPT 0.0
END SCRIPT 0.0
[more of the same]

I have been typing the "1" rather than spinning the dial. If I type the number in nothing happens. I decided to try dialling it in. That worked to some degree, but it was not possible to get exactly 1.0 on the dial. I set the min and max for the dial, then turned on Use Limits. I got this.

Created plane
Callback started
Starting callback
Callback is running
RUN SCRIPT 0.0
END SCRIPT 0.0
Callback is running
RUN SCRIPT 0.0
END SCRIPT 0.0
Callback is running
RUN SCRIPT 0.0
END SCRIPT 0.0
Callback is running
RUN SCRIPT 1.0
Starting loop generation
Making loop
Getting actors
cyl_1
Found vertices
Convex hull generated
Perimeter markers placed
Ring vertices placed
Verts, polys, sets created
UV mapping applied
New Geometry() instantiated

In spite of the encouraging message "New Geometry() instantiated", I see no new props in the document window, or in the dropdown Props list, and MR Looper is still just a plain square.