[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gEDA-user: gerbv and eagle - crashed - retested again
Holger Waechtler wrote:
> Stefan Petersen wrote:
>
>> Hello Alvin!
>>
>> On Sun, 28 Sep 2003, Alvin Oga wrote:
>>
>>
>>> i've regenerated and confirmed that gerbv-0.13 segfaults on
>>> "gerberauto" files
>>> output from eagle
>>> # gerbv .../LedAssy-2.1/crashbot.gbr
>>> Segmentation fault
>>>
>>> gerbv-0.14 shows the following error message which is probably right
>>> # gerbv-0.14 .../LedAssy-2.1/crashbot.gbr
>>>
>>> | /tmp/gerb/LedAssy-2.1/crashbot.gbr: Parse error:
>>> | * Missing format
>>> | * Missing apertures/drill sizes
>>> |
>>> | You probably tried to read an RS-274D file, which gerbv doesn't
>>> support
>>>
>>> good .. looks like the segfult is fixed in gerbv-0.14 for
>>> "gerberauto" outputs
>>> from eagle ...
>>
>>
>>
>> I get very worried when it segfaults, but that seems to be that old
>> problem. Great to hear that it seems fixed and sorry for the problems
>> caused by 0.13!
>>
>>
>>
>>> but, it still errors out with links to the real drill files
>>> ( ln -s led-assy.dri crashbot.dri
>>> ( ln -s led-assy.drl crashbot.drl
>>> ( ln -s led-assy.dri crashbot.dri
>>
>>
>>
>> So the drillfiles causes segfault? Or are they just "missdrawn"? You have
>> used the scripts that Holger pointed out, since gerbv doesn't like Eagle
>> default drill files?
>
>
> Hi Stefan,
>
> when Eagle was used with a metric grid I had to slightly modify
> eagle2exc.pl:
>
> I added
>
> print OUTFILE "INCH,LZ\n";
>
> right after
>
> print OUTFILE "M48\n";
>
> and replaced
>
> while (<INFILE>)
> {
> print OUTFILE $_;
> }
>
> by
>
> while (<INFILE>)
> {
> if (/^X(\d+)Y(\d+)$/) {
> printf OUTFILE "X%.03fY%.03f\n",$1/1000,$2/1000;
> } else {
> print OUTFILE $_;
> }
> }
>
> to ensure correct fractional numbers.
>
> Some Windows Viewers don't use inch by default and I had problems with
> the files generated by the original version when Eagle wrote files in
> the metric setting.
Errmmm, sorry -- if I remember correctly it was not a metric grid but an
inch grid. I believe the original script worked on mil grids. I'm not
sure right now if drill coordinates need to get scaled for mm grids.
Holger