Smalltalk › Cincom › VisualWorks

[Bug] RBCodehighlighting

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

4 messages Options

Reinout Heeck-2

[Bug] RBCodehighlighting

The method
Icon class>>figure: colorImage transparentPixels: pixels does not
colorize properly in our image, whole swaths are red.

Replacing code like
"height-1" by
"height- 1" makes the red go away...

R
-

Adrian Kuhn-3

Re: [Bug] RBCodehighlighting

I also observed this bug several times.

AA

On 13 Mar 2007, at 10:32 , Reinout Heeck wrote:

> The method
> Icon class>>figure: colorImage transparentPixels: pixels does not
> colorize properly in our image, whole swaths are red.
>
> Replacing code like
> "height-1" by
> "height- 1" makes the red go away...
>
>
> R
> -

Steven Kelly

RE: [Bug] RBCodehighlighting

In reply to this post by Reinout Heeck-2

If you are talking about window icons on screen, IIRC correctly this is
a Windows thing. You have to make sure that the first colour in the
colour palette is the transparent one. Actually, #cleanFigure is already
trying to do this.

The height-1 bit is interesting: "height-1" and "height- 1" are both
legal and bytecode-identical?!? The bytecode in 7.4.1 for this method is
correct, and unchanged by the "height- 1" change, so it's hard to guess
what might be happening.

If you are talking about printing to a HostPrinter, VW doesn't support
masked images, and in some common cases the masked part shows up red
IIRC.

Steve

> -----Original Message-----
> From: Adrian Kuhn [mailto:[hidden email]]
> Sent: 13. maaliskuuta 2007 16:03
> To: Reinout Heeck
> Cc: 'VWNC'
> Subject: Re: [Bug] RBCodehighlighting
>
> I also observed this bug several times.
>
> AA
>
> On 13 Mar 2007, at 10:32 , Reinout Heeck wrote:
>
> > The method
> > Icon class>>figure: colorImage transparentPixels: pixels does not
> > colorize properly in our image, whole swaths are red.
> >
> > Replacing code like
> > "height-1" by
> > "height- 1" makes the red go away...
> >
> >
> > R
> > -

Reinout Heeck-2

Re: [Bug] RBCodehighlighting

Steven Kelly wrote:

> If you are talking about window icons on screen, IIRC correctly this is
> a Windows thing. You have to make sure that the first colour in the
> colour palette is the transparent one. Actually, #cleanFigure is already
> trying to do this.
>
> The height-1 bit is interesting: "height-1" and "height- 1" are both
> legal and bytecode-identical?!? The bytecode in 7.4.1 for this method is
> correct, and unchanged by the "height- 1" change, so it's hard to guess
> what might be happening.
>
> If you are talking about printing to a HostPrinter, VW doesn't support
> masked images, and in some common cases the masked part shows up red
> IIRC.
>

Neither ;-)

I was talking about how the *source* of that particular method does not
get colorized properly in the browser.
I suspect a bug in the SmaCC language definition used in the
RBCodeHighlighting package.

R
-

> Steve
>
>
>
>> -----Original Message-----
>> From: Adrian Kuhn [mailto:[hidden email]]
>> Sent: 13. maaliskuuta 2007 16:03
>> To: Reinout Heeck
>> Cc: 'VWNC'
>> Subject: Re: [Bug] RBCodehighlighting
>>
>> I also observed this bug several times.
>>
>> AA
>>
>> On 13 Mar 2007, at 10:32 , Reinout Heeck wrote:
>>
>>
>>> The method
>>> Icon class>>figure: colorImage transparentPixels: pixels does not
>>> colorize properly in our image, whole swaths are red.
>>>
>>> Replacing code like
>>> "height-1" by
>>> "height- 1" makes the red go away...
>>>
>>>
>>> R
>>> -
>>>
>
>
>