0006747: [Bug] How to make a dizzy Morph Bounce - Mantis

Mantis Bugtracker

Viewing Issue Simple Details [ Jump to Notes ]

[ View Advanced ] [ Issue History ] [ Print ]

ID

Category

Severity

Reproducibility

Date Submitted

Last Update

0006747

[Squeak] Etoys

minor

sometimes

11-02-07 04:01

11-17-08 20:05

Reporter

wiz

View Status

public

Assigned To

Priority

normal

Resolution

open

Status

new

Product Version

3.8

Summary

0006747: [Bug] How to make a dizzy Morph Bounce

Description

For this one from Objects or Supplies bin
Get a Rectangle Morph
Get an ellipse morph
with the halo handles make the rectangle roomy enough for the ellipse.
move ellipse over rectangle
from the red halo menu embed the ellipse in the rectangle.

with the halo rotate the ellipse 90 degrees (approximately)
with the halo rotate the rectangle 90 degrees (approximately)

with the halo get a viewer for the ellipse
drag out the forward by 5 tiles and let them become a script.

run the script.
under the right circumstances the morph will start bounceing in place.

To help reproducability I will upload a OPLC etoys project from a fresh olpc etoys-dev-1513.

And a saved morph from 3dot10 version 7137 (when loaded rotate the rectangle 90 degrees. when squeak save the morph it missed the outer flex.)

Additional Information

Player forward is testing for the morph being within playfield boundries and bouncing when that fails.

The x and y's are all confused because of the layered flexing.

This is the first bug I've run across in looking for a set of ezekial tests (wheels within wheels).

I've marked this 3.8 because that branch is most likely interested in etoys bugs. However this is a current bug (or symtom of a bug) and needs to be fixed in 3dot10+ too.

dizzy and confused.003.pr [^] (30,525 bytes) 11-02-07 04:02

dizzyMorph.morph [^] (17,854 bytes) 11-02-07 04:03

rotation test probe.001.pr [^] (41,714 bytes) 11-05-07 00:38

Relationships

related to	0007112	new		Squeak 5.2	Embeding a NCPinMorph in a rotated morph move the pin to a incorrect position
child of	0005674	new		Squeak	Why doesn't heading = forwardDirection + rotationDegrees for all morphs all the time?

Notes
(0011416 - 867 - 1011 - 1011 - 1011 - 1011 - 1011) wiz 11-05-07 00:29	Ok it gets worse. For this one get a rectangle and an ellipse grow the rectange and embed the ellipse. Now get to handles (get ellipses and shrink them to handle size) Make a simple follow script for each handle do menu item bring to front set handles x to rectangles x set handles y to rectangles y and do the same for the other handle and the ellipse. set the scripts going the handle are over the centers of the rectangle and the ellipse get halos on the rectangle and click down on the blue halo handle the ellipses follower disappears or can be see up near the top left of the world. explanation when the rectange is flexed the ellipses x and y are relative to the rectangle local not world coordinates. This is a hard bug because it is inconsitant with the behavior of the embedded ellipse when the rectangle is not flexed.

(0011417 - 176 - 194 - 194 - 194 - 194 - 194) wiz 11-05-07 00:40	rotation test probe.001.pr uploaded. This is an olpc-etoys project (a fresh 1513 image was the base) which is set up to demonstrate the problem mentioned in the note above.

Issue History
Date Modified	Username	Field	Change
11-02-07 04:01	wiz	New Issue
11-02-07 04:02	wiz	File Added: dizzy and confused.003.pr
11-02-07 04:03	wiz	File Added: dizzyMorph.morph
11-05-07 00:29	wiz	Note Added: 0011416
11-05-07 00:38	wiz	File Added: rotation test probe.001.pr
11-05-07 00:40	wiz	Note Added: 0011417
11-05-07 03:22	wiz	Relationship added	child of 0005674
11-17-08 20:05	wiz	Relationship added	related to 0007112

Mantis 1.0.8[^]

Copyright © 2000 - 2007 Mantis Group

51 total queries executed.
33 unique queries executed.