kicad-developers team mailing list archive

[Cirilo Bernardo <cirilo_bernardo@xxxxxxxxx>]

----- Original Message -----
> From: jp charras <jp.charras@xxxxxxxxxx>
> To: Cirilo Bernardo <cirilo_bernardo@xxxxxxxxx>; "kicad-developers@xxxxxxxxxxxxxxxxxxx" <kicad-developers@xxxxxxxxxxxxxxxxxxx>
> Cc: 
> Sent: Monday, February 10, 2014 5:55 PM
> Subject: Re: [Kicad-developers] PATCH: partial fix to bug 1255937
> 
> Le 09/02/2014 20:55, Cirilo Bernardo a écrit :
>>  ----- Original Message -----
>> 
>>>  From: jp charras <jp.charras@xxxxxxxxxx>
>>>  To: kicad-developers@xxxxxxxxxxxxxxxxxxx; cirilo_bernardo@xxxxxxxxx
>>>  Cc: 
>>>  Sent: Monday, February 10, 2014 3:19 AM
>>>  Subject: Re: [Kicad-developers] PATCH: partial fix to bug 1255937
>>> 
>>>  Le 09/02/2014 09:06, Cirilo Bernardo a écrit :
>>> 
>>>>   The attached patch is a partial fix to bug 1255937.
>>>> 
>>>>   Symptom: Pad outlines (arcs) in the back silk and mask are 
> rendered 
>>>  incorrectly
>>>>   (arc drawn in wrong direction) when a mirrored image is printed.
>>>> 
>>>>   This patch works by checking the Device Origin and if the Y 
> coordinate is 
>>>  not
>>>>   0 it is assumed that the X axis is reversed. I would appreciate 
> any 
>>>  comments
>>>>   on this proposed solution; I don't know the code well enough 
> to know if 
>>>  it is
>>>>   possible to have a non-zero Y origin in another possibly future 
> context
>>>>   (for example, drawing on screen with the X orientation reversed).
>>>> 
>>>>   Unfortunately this patch does not quite close the bug yet; there 
> are other
>>>>   related artifacts which I am still investigating.
>>>> 
>>>>   The patch is against rev. 4679
>>>> 
>>>>   - Cirilo
>>> 
>>>  You could try something like this:
>>> 
>>>      int slx = DC->DeviceToLogicalX( 1 ) - DC->DeviceToLogicalX( 0 
> );
>>>      int sly = DC->DeviceToLogicalY( 1 ) - DC->DeviceToLogicalY( 0 
> );
>>>      bool mirrored = (slx > 0 && sly < 0) || (slx < 0 
> && 
>>>  sly > 0);
>>> 
>>>  to know if the DC is mirrored.
>>> 
>>>  -- 
>>>  Jean-Pierre CHARRAS
>>> 
>> 
>>  Thanks Jean-Pierre,
>> 
>>   I'll test that and also check the fix for the mirrored title frame 
> since it is the same issue.
>> 
>>  - Cirilo
> 
> The title frame is not mirrored.
> 
> However the board itself is mirrored from the X axis.
> It could be better to mirror it from the Y axis (like in plot mirror).
> 
> Thanks.
> 
> 
> -- 
> Jean-Pierre CHARRAS
> 

OK, I'll work on mirroring from the Y axis so that it is consistent with the plot functions; this will also fix the other complaint in bug 1255937 about mirroring on the X axis.

The frame was originally mirrored as well, but a previous patch fixed that; however mirroring on the Y axis means I have to look at that code again. This is not a problem since I already know what to expect there.

- Cirilo