Project

General

Profile

Bug #325

Functions with floating point results lock up AOZ. Also, Integer functions return floating point results!

Added by Brian Flanagan almost 4 years ago. Updated over 3 years ago.

Status:
Feedback
Priority:
Normal
Target version:
-
Start date:
03/31/2020
Due date:
% Done:

0%

Estimated time:
Affected version:

Description

No error displays, but AOZ locks up.
You can't even break out with Ctrl-C!

Example 1:

' 
' This fails
'
Function "mult#", _num1#, _num2#
End Function ( _num1# * _num2# )

Result is that AOZ locks up at black screen. Can't break out.

Example 2: (Even the simplest floating point function will fail at runtime.)

'
' Return self
'
Function "slf#", _num#
End Function ( _num# )

Example 3:

'
' This integer function works (but gets wrong result)
'
Function "mult", _num1#, _num2#
End Function ( _num1# * _num2# )

Result: (What the floating point function SHOULD have returned!)

29.20999999999997

Result SHOULD be:

29

Perhaps something is reversed between the integer and floating point results?

#1

Updated by Francois Lionet almost 4 years ago

  • Status changed from New to Resolved
  • Assignee set to Francois Lionet
  • Target version set to 0.9.8.1

There were problems in Functions written in Basic, AND for the moment (I can correct this but it needs a bit of work), you CANNOT use # in the name of a function: the transpiler thinks that you are using an array and generates an "Array not dimensioned" error...

THIS works now:

Print slf( 2.2 )
Wait key

Function "slf", _num#
End Function ( _num# )

#2

Updated by Brian Flanagan almost 4 years ago

  • Status changed from Resolved to Feedback

re-tested in 0.9.8.1 Test 2 Yes, the example you gave works, but I have questions:

  1. Are you planning on making the # work for floating point functions in the future?

  2. Wouldn't your (temporary?) solution cause a problem if you want to define a function with Integer results?
    I realize I could always just use:

End Function(Int(result))

...to work-around , but for consistency, shouldn't we let the function's type define the results?

#3

Updated by Brian Flanagan over 3 years ago

  • Affected version changed from 0.9.7 to 0.9.9.3

Re-tested in 0.9.9.3. Problems still exist.

#4

Updated by Brian Flanagan over 3 years ago

  • Target version deleted (0.9.8.1)
  • Affected version changed from 0.9.9.3 to 0.9.9.4-RC1

Re-tested in 0.9.9.4-RC1
Although it no longer locks up AOZ, it still gets the non-dimensioned array error.

Also available in: Atom PDF