ODB2GBR on Linux

This page will explain how to use ODB2GBR on Linux in engine mode and to produce the Gerber files and the job deck needed by NETEX-G. We will take the user through a simple example.

The Example Array.tgz

The example we will use is a small ODB++ file called array.tgz. It has two STEPS - the ARRAY step and the 220-01720-05a step. For net tracing one does not want the array step as we need only a single circuit.

When you first open this file using the Valor viewer, you will have to choose which STEP to open. This is what the Valor viewer's dialog looks like:

We will choose 220-01720-05a. We now can see the layout:

Layers

The viewer also shows us the layers defined in the ODB++ file.

Running the odb2gbr Engine from the Command Line

If you want to run the odb2gbr conversion from a command line then you must know something about the ODB++ file you wish to convert. You need to know:

• The name of the STEP to convert
• The layer or layers you wish to convert
• All of the conversion parameters you want to use.

All of this information must be encoded into an XML file. A sample XML file is shown below:

<?xml version="1.0" encoding="UTF-8"?>
<odb2gbr version="" xmlns=""
xmnls:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="">

<inputFilePath>/home/cad/odb2gbr/examples/array.tgz</inputFilePath>  The input file name and path
<workDirPath>/home/cad/odb2gbr/examples</workDirPath>                working directory
<keepTemp>0</keepTemp>                                               0=don't keep temp files 1=keep temp files
<nonStd>1</nonStd>                                                   don't know what this does
<threadNum>2</threadNum>                                             number of concurrent threads
<inputArgs>-dbg</inputArgs>                                          any special input arguments 

<outputFilePath>/home/cad/odb2gbr/examples</outputFilePath>          output file name and path.
<stepName>array</stepName>                                           name of the STEP to process
<layerName>comp_+_top</layerName>                                    layer to convert
<layerName>smt</layerName>                                           layer to convert
<layerName>top</layerName>                                           layer to convert
<layerName>bottom</layerName>                                        layer to convert
<layerName>smb</layerName>                                           layer to convert
<layerName>comp_+_bot</layerName>                                    layer to convert
<layerName>ncroute_path</layerName>                                  layer to convert
<layerName>drill</layerName>                                         layer to convert

<arcResolutionDeg>9.0</arcResolutionDeg>                          arc resolution when booleanizing custom apts
<arcChordError>0.0</arcChordError>                                chord error to use when booleanizing custom apts
<scale>1.0</scale>                                                user defined scaling
<units>1</units>                                                  output units 1=   0=  
<outScale>0</outScale>                                            Output scale factor 
<showProgress>1</showProgress>                                    display progress 1=yes, 0=no
<butting>0</butting>                                              use butting polygons for output

<g2kVer>0</g2kVer>
<g2kScale>2</g2kScale>
<g2kBreakSR>0</g2kBreakSR>
<g2kAnchor>0</g2kAnchor>
<g2kOffset>0</g2kOffset>
<g2kOrder>0</g2kOrder>
<g2kXAnchor>0.0</g2kXAnchor>
<g2kYAnchor>0.0</g2kYAnchor>
<g2kXOffset>0.0</g2kXOffset>
<g2kYOffset>0.0</g2kYOffset>

<outputArgs></outputArgs>

</odb2gbr>

Running in Batch Mode

Once the .xml file has been constructed you can run in batch or script mode. For example from the example directory:

$ ../bin/odb2gbr array.xml [enter]

What Does the odb2gbr Script Do?

when you invoke odb2gbr you are running a script. Let's see what it does.

#! /bin/bash

ODBINSTDIR=/home/cad/odb2gbr/bin      directory where the program executables reside

LIB64=                                these lines determine whether to use the 32 or 64 bit
ENG=${ODBINSTDIR}/odb2gbrDlg64        engine based on what libraries are present
LNXBIT=`getconf LONG_BIT`             returns either 32 or 64 depending on the OS

echo
if [ "$LNXBIT" = "64" ]               If OS is 64 bit, then use the 64 bit engine     
then
  if [ -x $ENG ]
  then
    LIB64=${ODBINSTDIR}/lib64         64 bit libraries are in the subdirectory lib64  
    echo executing script $ENG
  else
    ENG=${ODBINSTDIR}/odb2gbrDlg32
    echo executing script $ENG
  fi
else
  ENG=${ODBINSTDIR}/odb2gbrDlg32      OS is not 64 bit, then use the 32 bit engine
    echo executing script $ENG
fi

echo

defaults=`find $HOME/.artwork -name odb2gbr.Xdefaults -newer $ENG`

if [ "$defaults" = "" ]
then

if [ -f $HOME/.artwork/odb2gbr.Xdefaults ]
then
echo
echo 
echo renaming potentially stale \"$HOME/.artwork/odb2gbr.Xdefaults\" to \"$HOME/.artwork/odb2gbr.Xdefaults.old\"
echo
echo
mv $HOME/.artwork/odb2gbr.Xdefaults $HOME/.artwork/odb2gbr.Xdefaults.old
fi
fi

export LD_LIBRARY_PATH=${ODBINSTDIR}:${LIB64}:/usr/local/lib:$LD_LIBRARY_PATH

$ENG $*                                run the engine with the argument (which is our .xml file)