Mail Archives: geda-user/2023/10/02/04:20:19
Hello Karl,
Op 02-10-2023 om 00:35 schreef karl AT aspodata DOT se [via geda-user AT delorie DOT com]:
> Richard Rasker:
> ...
>> - When specifying width in millimeters, things work up to 73
>> millimeters but throw a sort of overflow error above that value:
>>
>> $ pcbsilk -W 74mm testlogo.jpg
>> double free or corruption (!prev)
>> /home/richard/scripts/pcbsilk: line 36: 15520 Aborted
>> (memory dumped) pstoedit -q -f "pcbfill" -ssp $silkname.eps
>> $silkname.pcb
> ...
>
> Just tested with your latest pcbsilk and i don't get any errors
> except that the output width is slightly in error.
[snip]
> What version of pstoedity do you have, perhaps try with a later
> version. I have:
> $ pstoedit 2>&1 | head -1
> pstoedit: version 3.75 / DLL interface 108 (built: Dec 18 2020 - release build - g++ 7.4.0 - 64-bit) : Copyright (C) 1993 - 2020 Wolfgang Glunz
Strange, I can't reproduce the error any more, regardless whether I use
the "pcb" or "pcbfill" option. I also have the same pstoedit version
3.75 / DLL interface 108.
I do recall however running into some safety issues when experimenting
with different ways to create EPS files, tinkering with ImageMagick 6
policy settings just prior to triggering those errors, but I never kept
any of those changes. Ah well ...
[second mail]
> To be uptodate, the pcb line should be something like:
> Layer(5, "top silk", "silk)
In my pcb files, Layer 5 is bottom silk, and Layer 6 is top silk. But
this is easily adjusted.
> $ head -10 < TestLogo.pcb
> PCB["" 850000 1100000]
>
> Grid[1000.000000 0 0 0]
>
> Layer(1 "poly")
> (
> Polygon("clearpoly")
> (
> [2304472 -3448985]
> [2303472 -3445930]
That's weird, I only get positive Y values here. Anyway, I might give it
a shot using grep, cut, sed and sort to manipulate the X and Y values --
even though I think it's going to be pretty inefficient.
Thanks again,
Regards,
Richard
- Raw text -