Quick and dirty Matlab/Octave replacement using JScript

In my previous post I showed that it is possible to make a simple read-eval-print-loop (REPL) in JScript for the Windows command line. Today I show that it is also possible to add libraries to this REPL for quick access and experimentation. I have been adding functions to my JScript matrix library for work when I realized that I was quickly needing Matlab levels of functionality in order to get what I needed. Today I decided to try loading a Javascript library for numerical work into JScript. I was surprised how painless the process was!

The first thing I needed to do was write a WSF file to load the library with the jshell REPL I created in my last post. I then found a nice Javascript library online (http://numericjs.com/numeric/index.php) and integrated this library with my jshell.js REPL script into the WSF file. I called my file jlab.wsf.



   

Making new Windows commands using Windows Scripting Host and Jscript

I'm glad that all the Windows machines I use at home and at work come with at least one default programming language. There is a lot of momentum in the Javascript world now that HTML5 is taking hold and node.js is gaining in popularity. So I thought now would be a good time to learn some Javascript, since I'm starting to realize that it can be found EVERYWHERE.

I thought that a good way to learn more about Javascript would be to build some small tools for the machines I use at work. I've learned that Javascript on Windows machines come in at least 3 different flavors, Jscript hosted on the Windows Scripting Host, Jscript .NET compiled with the jsc compiler, and the Jscript running on Internet Explorer and Firefox. Not only that but doing things on Windows XP turns out to be slightly different than on Windows 7. Take for example embedding Jscript in a batch file so that I can call the tool from anywhere on the command line. What I found on StackOverflow was this cool calendar example:

REM Let's call this command "cal"

@set @junk=1 /*
@echo off
cscript //nologo //E:jscript %0 %*
goto :eof
*/
x = WScript.Arguments
Yr = x(0) ; Mo = x(1)

YS = "JanFebMarAprMayJunJulAugSepOctNovDec"
MN = Mo<1 || Mo>12 ? Mo : YS.substr(3*Mo-3, 3) // Month Name
WScript.echo(" ", Yr, "         ", MN)
WScript.echo(" Mo Tu We Th Fr Sa Su")
WD = new Date(Yr, Mo-1, 1).getDay() ;
if (WD==0) WD = 7 // Week Day Number of 1st
LD = new Date(Yr, Mo, 0).getDate() // Last Day of month
Wk = "" ; for (D=1 ; D < WD ; D++) Wk += "   "

for (D=1 ; D<=LD ; D++) {
  Wk = Wk + " " + (D<10 ? "0"+D : D) ; WD++
  if ((WD==8) || (D==LD)) { WScript.echo(Wk) ; WD = WD-7 ; Wk = "" }
  }

WScript.echo("        ------       ")

Calling this script in Windows XP works, but it doesn't work on Windows 7. More Googling showed that for Windows 7, I need to change the batch script code to the following:

@if (@CodeSection == @Batch) @then
@echo off
cscript //nologo //E:jscript %~f0 %*
goto :eof
@end

Then the script will work and testing with "cal 2014 11" outputs:

  2014           Nov
 Mo Tu We Th Fr Sa Su
                            01 02
 03 04  05 06  07 08 09
 10 11  12 13  14 15 16
 17 18  19 20  21 22 23
 24 25  26 27  28 29 30
---------------------------------

Another handy tool for Windows scripting would be a command line read-eval-print-loop similar to what Python and Lisp have. Jscript does not come with a REPL built in, but finding a basic example online was not hard. Here is the code that works good enough for my purposes.

@if (@CodeSection == @Batch) @then
@echo off
cscript //nologo //E:jscript %~f0 %*
goto :eof
@end

function hex(n) {
    if (n >= 0) {
        return n.toString(16);
    } else {
        n += 0x100000000;
        return n.toString(16);
    }
}
var scriptText;
var previousLine;
var line;
var result;
while(true) {
    WScript.StdOut.Write("jscript> ");
    if (WScript.StdIn.AtEndOfStream) {
        WScript.Echo("Bye.");
    break;
    }
    line = WScript.StdIn.ReadLine();
    scriptText = line + "\n";
    if (line === "") {
        WScript.Echo(
            "Enter two consecutive blank lines to terminate multi-line input.");
        do {
            if (WScript.StdIn.AtEndOfStream) {
                break;
            }
            previousLine = line;
            line = WScript.StdIn.ReadLine();
            line += "\n";
            scriptText += line;
        } while(previousLine != "\n" || line != "\n");
    }
    try {
        result = eval(scriptText);
    } catch (error) {
        WScript.Echo("0x" + hex(error.number) + " " + error.name + ": " +
            error.message);
    }
    if (result) {
        try {
            WScript.Echo(result);
        } catch (error) {
            WScript.Echo("<<>>");
    }
    }
    result = null;
}

 
One application I have in mind for using Jscript at work is as a quick and small footprint command line calculator. In the past I have used Matlab, Google Sheets, the basic Windows calculator, and Excel. Most of these are overkill for what I need to do, and Windows calculator is too basic. Hard-coding the calculations into the command line sounds a lot more convenient for the things I need to do. Here is calculation I have been doing a lot recently on Google Sheets. It realigns the offset for the camera I am looking through to the desired coordinate system.

@if (@CodeSection == @Batch) @then
@echo off
cscript //nologo //E:jscript %~f0 %*
goto :eof
@end

// realign.bat
// Jovan Trujill
// Arizona State University
// 11/04/2014
//
// Usage: realign XActual YActual XDesired YDesired XOffset_Old YOffset_Old
// This command will return the alignment offset coordinates needed to correct
// camera positioning problems on the Cherry prober system.
//

var X_Actual = 0.0;
var Y_Actual = 0.0;
var X_Desired = 0.0;
var Y_Desired = 0.0;
var dX = 0.0;
var dY = 0.0;
var X_Offset_Old = 0.0;
var Y_Offset_Old = 0.0;
var X_Offset_New = 0.0;
var Y_Offset_New = 0.0;

var args = WScript.Arguments;
if (args.length != 6) {
    WScript.echo("Usage: realign XActual YActual XDesired YDesired OldXOffset OldYOffset");
} else {
    X_Actual = parseInt(args(0));
    Y_Actual = parseInt(args(1));
    X_Desired = parseInt(args(2));
    Y_Desired = parseInt(args(3));
    X_Offset_Old = parseInt(args(4));
    Y_Offset_Old = parseInt(args(5));
   
    dX = X_Desired - X_Actual;
    dY = Y_Desired - Y_Actual;
   
    X_Offset_New = X_Offset_Old - dX;
    Y_Offset_New = Y_Offset_Old + dY;
   
    WScript.echo("X_Actual Y_Actual");
    WScript.echo(X_Actual + " " + Y_Actual);
    WScript.echo("X_Desired Y_Desired");
    WScript.echo(X_Desired + " " + Y_Desired);
    WScript.echo("X_Old_Offset Y_Old_Offset");
    WScript.echo(X_Offset_Old + " " + Y_Offset_Old);
    WScript.echo("X_New_Offset Y_New_Offset");
    WScript.echo(X_Offset_New + " " + Y_Offset_New);
}

 As the library of functions gets more complicated I will need to figure out a way to call JScript functions from other JScript programs. I think the most convenient way to do that will be:

/* include.js */
(function () {
    var L = {/* library interface */};
    L.hello = function () {return "greetings!";};
    return L;
}).call();

And then the main function calls "include.js" like this:

var Fs = new ActiveXObject("Scripting.FileSystemObject");
var Lib = eval(Fs.OpenTextFile("include.js", 1).ReadAll());
WScript.echo(Lib.hello()); /* greetings! */ 

 

 The example works but I have not tested it on anything more substantial.

The day of reckoning has come: Upgrading Excel 2003 VBA code to Excel 2010.

At work we have a few Excel spreadsheets which read data files and processes them into a specific worksheet format. One of the problems with upgrading to Excel 2010 has been that many of the VBA functions used in our macros have been removed. Here is one example of code that does not work in Excel 2010.

Excel 2003 Code:
Set fs = Application.FileSearch
    With fs
        .NewSearch
        .LookIn = vrtSelectedItem
        .SearchSubFolders = True
        .Filename = "Macro_I"
        .MatchExactly = False
        .FileType = msoFileTypeExcelWorkbooks
        If .Execute() > 0 Then
            For i = 1 To .FoundFiles.Count
                Call StuffCore(.FoundFiles(i), summaryType)
            Next i
        End If
    End With

The problem is that Application.FileSearch no longer exists in Excel 2010. Therefore a replacement for this function needed to be found. Others have had the same problem and posted a replacement function here.

The previous code snippet now becomes the following.

Excel 2010 Code:
Dim foundFiles() As FoundFileInfo
Dim foundFilesCount As Integer
searchPattern = "Macro_I*"
recursiveSearch = True
Dim boolFoundFiles As Boolean

boolFoundFiles = FindFiles(vrtSeclectedItem, foundFiles, foundFilesCount, searchPattern, recursiveSearch)

If boolFoundFiles = True Then
    For i = 1 To foundFilesCount
        With foundFiles(i)
            Call StuffCore(.sPath & .sName, summaryType)
        End With
    Next i
End If

For this code to work a new data type must be declared globally outside of all subroutines. This is the "FoundFileInfo" type and it is created with the following code.

FoundFileInfo type declaration:
 Type FoundFileInfo
    sPath As String
    sName As String
End Type

And the "FindFiles" function is defined as follows.

 FindFiles function definition:

Function FindFiles(ByVal sPath As String, _
    ByRef recFoundFiles() As FoundFileInfo, _
    ByRef iFilesFound As Integer, _
    Optional ByVal sFileSpec As String = "*.*", _
    Optional ByVal blIncludeSubFolders As Boolean = False) As Boolean
'
' FindFiles
' ---------
' Finds all files matching the specified file spec starting from the specified path and
' searches sub-folders if required.
'
' Parameters
' ----------
' sPath (String): Start-up folder, e.g. "C:\Users\Username\Documents"
'
' recFoundFiles (User-defined data type): a user-defined dynamic array to store the path
' and name of found files. The dimension of this array is (1 To nnn), where nnn is the
' number of found files. The elements of this array are:
'   .sPath (String) = File path
'   .sName (String) = File name
'
' iFilesFound (Integer): Number of files found.
'
' sFileSpec (String): Optional parameter with default value = "*.*"
'
' blIncludeSubFolders (Boolean): Optional parameter with default value = False
'   (which means sub-folders will not be searched)
'
' Return values
' -------------
' True: One or more files found, therefore
'   recFoundFiles = Array of paths and names of all found files
'   iFilesFound = Number of found files
' False: No files found, therefore
'   iFilesFound = 0
'
' Using the function (sample code)
' --------------------------------
'    Dim iFilesNum As Integer
'    Dim iCount As Integer
'    Dim recMyFiles() As FoundFileInfo
'    Dim blFilesFound As Boolean
'
'    blFilesFound = FindFiles("C:\Users\MBA\Desktop", _
'        recMyFiles, iFilesNum, "*.txt?", True)
'    If blFilesFound Then
'        For iCount = 1 To iFilesNum
'            With recMyFiles(iCount)
'                MsgBox "Path:" & vbTab & .sPath & _
'                    vbNewLine & "Name:" & vbTab & .sName, _
'                    vbInformation, "Found Files"
'            End With
'        Next
'    Else
'  MsgBox "No file(s) found matching the specified file spec.", _
'            vbInformation, "File(s) not Found"
'    End If
'
'
' Constructive comments and Reporting of bugs would be
' appreciated.

    Dim iCount As Integer           '* Multipurpose counter
    Dim sFileName As String         '* Found file name
    '*
    '* FileSystem objects
    Dim oFileSystem As Object, _
        oParentFolder As Object, _
        oFolder As Object, _
        oFile As Object

    Set oFileSystem = CreateObject("Scripting.FileSystemObject")
    On Error Resume Next
    Set oParentFolder = oFileSystem.GetFolder(sPath)
    If oParentFolder Is Nothing Then
        FindFiles = False
        On Error GoTo 0
        Set oParentFolder = Nothing
        Set oFileSystem = Nothing
        Exit Function
    End If
    sPath = IIf(Right(sPath, 1) = "\", sPath, sPath & "\")
    '*
    '* Find files
    'sPath = sPath & sFileSpec
    sFileName = Dir(sPath & sFileSpec, vbNormal)
    If sFileName <> "" Then
        For Each oFile In oParentFolder.Files
            If LCase(oFile.Name) Like LCase(sFileSpec) Then
                iCount = UBound(recFoundFiles)
                iCount = iCount + 1
                ReDim Preserve recFoundFiles(1 To iCount)
                With recFoundFiles(iCount)
                    .sPath = sPath
                    .sName = oFile.Name
                End With
            End If
        Next oFile
        Set oFile = Nothing         '* Although it is nothing
    End If
    If blIncludeSubFolders Then
        '*
        '* Select next sub-forbers
        For Each oFolder In oParentFolder.SubFolders
            FindFiles oFolder.Path, recFoundFiles, iFilesFound, sFileSpec, blIncludeSubFolders
        Next
    End If
    FindFiles = UBound(recFoundFiles) > 0
    iFilesFound = UBound(recFoundFiles)
    On Error GoTo 0
    '*
    '* Clean-up
    Set oFolder = Nothing   '* Although it is nothing
    Set oParentFolder = Nothing
    Set oFileSystem = Nothing

End Function

Another database stuffing macro needed a syntax change for the Sort method. The old code looked like this:

Old range sorting code:

' Define the range for the sort
Range("A" & SortStart & ":" & "BH" & SortEnd).Select
Selection.Sort Key1:=Range("B1"), Order1:=xlAscending, _
Header := xlNo, custom:=1, MatchCase:=False, _
Orientation:=xlTopToBottom, DataOption1:=xlSortNormal

New range sorting code:

With ActiveWorkbook.Worksheets(ActiveSheet.Name).Sort
   .SortFields.Clear
   .SortFields.Add Key:=Range("B1"), _
       SortOn:=xlSortOnValues, Order:=xlAscending, _
       DataOption:=xlSortNormal
    .SetRange Range("A" & SortStart & ":" & "BH" & SortEnd)
    .Header = xlNo
    .MatchCase = False
    .Orientation = xlTopToBottom
    .Apply
End With

Who knows what else we will need to fix. We have a dozen or so more macro to test on Excel 2010. I will keep taking notes here for future reference. 

Drawing lines in Gimp using Python-fu

Today at work I needed to create a few test images for debugging some image processing software. I needed to draw a few dark lines on a light background to check if the software will detect them correctly. I found out that Gimp can do this quite nicely for me.

import math
from gimpfu import *

rows = 640
cols = 480

def draw_vertical_lines(drawable,pixel,rows,cols,width):
    for i in range(1,rows):
        for j in range(1,cols,width):
            pdb.gimp_drawable_set_pixel(drawable,i,j,3,pixel)
    return

img = gimp.Image(cols,rows,RGB)
layer_one = gimp.Layer(img,"Layer1",cols,rows,RGB_IMAGE,100,NORMAL_MODE)
pdb.gimp_edit_fill(layer_one,BACKGROUND_FILL)
img.add_layer(layer_one,0)
pdb.gimp_edit_fill(layer_one,BACKGROUND_FILL)
gimp.Display(img)
pixel = [0.0,0.0,0.0,0.0]
draw_vertical_lines(layer_one, pixel, rows, cols, 25)

Today's script

Just another throw away script to get things done at work today:


#!/usr/bin/perl
use strict;
use warnings;

if ($#ARGV != 0) {
print "ARGV = ", $#ARGV, "\n";
print "Usage: FDC46CountLines.pl \n";
exit(1);
}

my $rootPath = $ARGV[0];
my @lotList = ();
my $dataFile = ();
my @dataLine = ();
my @X = ();
my @Y = ();
my @radius = ();
my $tempPath;
my $i;
my $j;
my $panel;
my $lot;
my $lineCount = 0;
my $outputFile = '';

# Generate list of subfolders
opendir DIR, $rootPath or die "Couldn't open dir $rootPath: $!\n";
my @folders = readdir DIR;
closedir DIR;

foreach (@folders) {
if (-d $rootPath . "\\" . $_) {
if ($_ eq ".") {}
elsif($_ eq "..") {}
else {
push(@lotList, $_);
}
}
}

my %lotHash = ();
my $currentLot;

#Generate list of lot folders
foreach (@lotList) {
$tempPath = $rootPath . "\\" . $_;
$currentLot = $_;
opendir DIR, $tempPath or die "Couldn't open dir $tempPath: $!\n";
@folders = readdir DIR;
closedir DIR;
# Generate list of process folders
foreach (@folders) {
if (-d $tempPath . "\\" . $_) {
if ($_ eq ".") {}
elsif($_ eq "..") {}
else {
push(@{$lotHash{$currentLot}}, $_);
}
}
}
}

@folders = ();
for my $key (keys %lotHash) {
for ($i=0; $i <= $#{$lotHash{$key}}; $i++) {
$tempPath = $rootPath . "\\" . $key . "\\" . $lotHash{$key}[$i];
$lot = $key;
$panel = $lotHash{$key}[$i];
print "tempPath = " . $tempPath . "\n";
print "lot = " . $lot . "\n";
print "panel = " . $panel . "\n";
$dataFile = $tempPath . "\\" . $lot . "_" . $panel . "_Notes.txt";
if (open(DATA, "<$dataFile")) {
while () {
@dataLine = split;
push(@X,$dataLine[3]);
push(@Y,$dataLine[4]);
push(@radius, sqrt($dataLine[3]**2+$dataLine[4]**2));
}
} else { print "dataFile does not exist"; }
@radius = sort(@radius);
$outputFile = $tempPath . "\\" . $lot . "_" . $panel . "_defectCount.txt";
open(OUTPUT, ">$outputFile") or die $!;
$lineCount = 0;
for ($j = 1; $j <= $#radius; $j++) {
if (($radius[$j] - $radius[$j-1]) > 100.0) {
print OUTPUT "$X[$j]\t$Y[$j]\n";
$lineCount += 1;
}
}
print OUTPUT "total unique defects were: $lineCount\n";
@dataLine = ();
@X = ();
@Y = ();
@radius = ();
close OUTPUT;

}
}



Learning assembly in my spare moments

The system:

Sparcstation 10 with a single 36 MHz Supersparc CPU and 16 MB RAM. OpenBSD 4.5 on a 1 GB hard disk. 

With this meager system I am learning about being efficient with my resources and learning assembly in order to still do something useful with the machine. The default OpenBSD install comes with their custom gcc compiler and perl 5.10. Perl runs very slow on this machine making CPAN use impossible without an upgrade. But through SSH I can comfortably write small programs to assist with my solutions to Project Euler. After installing GMP I can really get into number theory with the little box.

But the real fun comes with learning Sparc V8 assembly language. I'm finally at the point in assembly where I can modify the examples I read to do other things. And I used gdb for the first time too.

I'm already dreaming up of a way to translate one set of assembly instructions to another using perl. :-)

So what have I learned so far? Well I learned that every Sparc assembly file starts with a stack initialization. So you start with:

save %sp,-96,%sp

Fun times at Project Euler.

Felt like I needed to exercise my brain the other day so I went over to Project Euler to solve a few problems. It was problem #4. I tried to solve it on an old 3B2 computer running Unix System V with Perl 4.0. It didn't go too well. So I ran the same code on my MetaArpa account on SDF and solved the thing in less than a minute ha ha. I like the old system though because it forces me to realize just how bad my algorithms are when the site states that all problems should take less than a minute to solve with a normal computer. Fun day.

Fun with OpenVMS

I've had some discretionary time this work so I worked on a few projects that have been lingering in my mind for a while. I took out the old rx1620 and installed OpenVMS ia64 edition. It took me a while to remember that the serial console is needed to install something on this machine. Nothing comes out of the VGA output when you power on. Also there are two connectors labeled "console" in the back of the machine. Connecting to the normal looking RS-232 connector did not work either. Finally I tries the other connector with the special adapter that came with the machine. With a straight through RS-232 cable and the RS-232 connector on my PC I finally started seeing something.

I connected to the rx1620 using a PC with Solaris 10 installed. I modified the /etc/remote file to point the hardwire variable to /dev/term/a instead of b. running "tip hardwire" worked and I could  see that the machine was ready to install OpenVMS. 

The install process was lengthy and there were multiple questions interrupting the flow regarding DECnet setup, clustering, Kerberos, and other things. I didn't check if it would output to the VGA plug yet. 

Overall it was a smooth first start. I still need to figure out if networking is set up correctly and why FTP is not running. I need to install the license PAK's and some useful programming languages too. One thing about this server though is that it is big and loud! I wish I could install it at work or something and run it 24x7 from there.

My first Arduino project

At work we print identification characters on the circuits we fabricate using a microscope and a matrix of ultraviolet LEDs. To do the same thing for our scaled up Gen II panels we needed to build another ID printer. This printer would shine ultraviolet light through an optics train to expose a patch of photoresist on the device during production.

We decided to control the 9x5 LED matrix using 14 digital outputs on the Arduino Micro because it simplified the hardware design.

/*
  RS232_MATRIX_CONTROL_v2p0
  Single character driver for LED matrix display. Current code is hardcoded to display 46 characters on a 9 row by 5 column matrix. 
  Each character is sent through serial terminal followed by a carriage return. 
  
  Jovan Trujillo
  Arizona State University
  03/27/2013
*/  
 
byte RN[] = {4,5,6,7,8,9,10,11,12};
byte CN[] = {23,22,21,20,19};
byte LED = 13;
byte strobe = 2;  // strobe rate for each row of matrix in milliseconds.
byte inByte = 0; // Input character received from serial port.
byte idx = 0;  // Generic index counter.
byte i = 0;    // Generic index counter.
int exposure = 250;  // Number of times entire character is displayed.
// Exposure time is calculated to be: strobe x # rows x exposure
byte rowcnt = 9;     // Number of rows to display for matrix arrays that have fewer than 9 rows.
// Character database contains the column bit patterns for each row in an array.
// Character database is hardcoded for a 9x5 LED matrix. 
byte D_0[] = {17,14,14,12,10,6,14,14,17};         // 1
byte D_1[] = {3,27,27,27,27,26,26,26,0};          // 2
byte D_2[] = {17,14,30,30,30,17,15,15,16};        // 3
byte D_3[] = {1,30,30,30,17,30,30,30,1};          // 4
byte D_4[] = {15,15,14,14,14,0,30,30,30};         // 5
byte D_5[] = {16,23,23,23,16,30,30,14,17};        // 6
byte D_6[] = {23,15,15,15,15,0,14,14,0};          // 7
byte D_7[] = {0,14,30,30,29,27,27,27,27};         // 8
byte D_8[] = {17,21,21,21,0,14,14,14,0};          // 9
byte D_9[] = {0,14,14,0,30,30,30,30,29};          // 10
byte D_A[] = {27,21,14,14,14,0,14,14,14};         // 11
byte D_B[] = {1,14,14,14,1,14,14,14,1};           // 12
byte D_C[] = {24,23,15,15,15,15,15,23,24};        // 13
byte D_D[] = {3,21,22,22,22,22,22,21,3};          // 14
byte D_E[] = {0,15,15,15,3,15,15,15,0};           // 15
byte D_F[] = {0,15,15,3,15,15,15,15,15};          // 16
byte D_G[] = {24,23,15,15,15,12,14,14,17};        // 17
byte D_H[] = {14,14,14,14,0,14,14,14,14};         // 18
byte D_I[] = {0,27,27,27,27,27,27,27,0};          // 19
byte D_J[] = {30,30,30,30,30,30,14,14,17};        // 20
byte D_K[] = {14,13,11,7,15,7,11,13,14};          // 21
byte D_L[] = {15,15,15,15,15,15,15,15,0};         // 22
byte D_M[] = {14,4,10,10,10,14,14,14,14};         // 23
byte D_N[] = {14,6,6,10,10,10,12,12,14};          // 24
byte D_O[] = {27,21,14,14,14,14,14,21,27};        // 25
byte D_P[] = {1,14,14,14,1,15,15,15,15};          // 26
byte D_Q[] = {30,26,22,14,14,14,10,9,22};         // 27
byte D_R[] = {1,14,14,1,15,7,11,13,14};           // 28
byte D_S[] = {17,14,15,23,27,29,30,14,17};        // 29
byte D_T[] = {0,10,27,27,27,27,27,27,27};         // 30
byte D_U[] = {14,14,14,14,14,14,14,14,0};         // 31
byte D_V[] = {14,14,14,14,14,14,14,21,27};        // 32
byte D_W[] = {14,14,14,10,10,10,10,10,21};        // 33
byte D_X[] = {14,14,14,21,27,21,14,14,14};        // 34
byte D_Y[] = {14,14,14,21,27,27,27,27,27};        // 35
byte D_Z[] = {0,30,30,21,27,21,15,15,0};          // 36
byte COLON[] = {31,17,17,17,31,17,17,17,31};      // 37
byte SEMICOLON[] = {17,17,17,31,17,17,17,25,23};  // 38
byte LESSTHAN[] = {30,29,27,23,15,23,27,29,30};   // 39
byte EQUAL[] = {31,31,0,31,31,31,0,31,31};        // 40
byte GRTRTHAN[] = {15,23,27,29,30,29,27,23,15};   // 41
byte QUESTION[] = {17,14,30,25,27,27,31,27,31};   // 42
byte D_AT[] = {31,0,14,8,10,8,15,0,31};           // 43
byte D_DASH[] = {31,31,31,0,0,0,31,31,31};        // 44
byte D_SPC[] = {31,31,31,31,31,31,31,31,31};      // 45
byte D_PER[] = {31,31,31,31,31,31,17,17,17};      // 46
byte D_BLK[] = {0,0,0,0,0,0,0,0,0};               // 47

char datalist[47];  // datalist holds an array of all available chracters to pattern match.
byte *decodelist[47]; // decodelist holds an array of memory references to the bit patterns for each character.
byte *refval = 0;  // refval contains a reference to the bit pattern for a specific character entered through the serial terminal.

// the setup routine runs once when you press reset:
void setup() {  
  // Initialize datalist with valid characters.  
  datalist[0] = '0';
  datalist[1] = '1';
  datalist[2] = '2';
  datalist[3] = '3';
  datalist[4] = '4';
  datalist[5] = '5';
  datalist[6] = '6';
  datalist[7] = '7';
  datalist[8] = '8';
  datalist[9] = '9';
  datalist[10] = 'A';
  datalist[11] = 'B';
  datalist[12] = 'C';
  datalist[13] = 'D';
  datalist[14] = 'E';
  datalist[15] = 'F';
  datalist[16] = 'G';
  datalist[17] = 'H';
  datalist[18] = 'I';
  datalist[19] = 'J';
  datalist[20] = 'K';
  datalist[21] = 'L';
  datalist[22] = 'M';
  datalist[23] = 'N';
  datalist[24] = 'O';
  datalist[25] = 'P';
  datalist[26] = 'Q';
  datalist[27] = 'R';
  datalist[28] = 'S';
  datalist[29] = 'T';
  datalist[30] = 'U';
  datalist[31] = 'V';
  datalist[32] = 'W';
  datalist[33] = 'X';
  datalist[34] = 'Y';
  datalist[35] = 'Z';
  datalist[36] = ':';
  datalist[37] = ';';
  datalist[38] = '<';
  datalist[39] = '=';
  datalist[40] = '>';
  datalist[41] = '?';
  datalist[42] = '@';
  datalist[43] = '-';
  datalist[44] = ' ';
  datalist[45] = '.';
  datalist[46] = '*';
  
  // Initialize decodelist with references to the bit patterns for valid characters.
  decodelist[0] = D_0;
  decodelist[1] = D_1;
  decodelist[2] = D_2;
  decodelist[3] = D_3;
  decodelist[4] = D_4;
  decodelist[5] = D_5;
  decodelist[6] = D_6;
  decodelist[7] = D_7;
  decodelist[8] = D_8;
  decodelist[9] = D_9;
  decodelist[10] = D_A;
  decodelist[11] = D_B;
  decodelist[12] = D_C;
  decodelist[13] = D_D;
  decodelist[14] = D_E;
  decodelist[15] = D_F;
  decodelist[16] = D_G;
  decodelist[17] = D_H;
  decodelist[18] = D_I;
  decodelist[19] = D_J;
  decodelist[20] = D_K;
  decodelist[21] = D_L;
  decodelist[22] = D_M;
  decodelist[23] = D_N;
  decodelist[24] = D_O;
  decodelist[25] = D_P;
  decodelist[26] = D_Q;
  decodelist[27] = D_R;
  decodelist[28] = D_S;
  decodelist[29] = D_T;
  decodelist[30] = D_U;
  decodelist[31] = D_V;
  decodelist[32] = D_W;
  decodelist[33] = D_X;
  decodelist[34] = D_Y;
  decodelist[35] = D_Z;
  decodelist[36] = COLON;
  decodelist[37] = SEMICOLON;
  decodelist[38] = LESSTHAN;
  decodelist[39] = EQUAL;
  decodelist[40] = GRTRTHAN;
  decodelist[41] = QUESTION;
  decodelist[42] = D_AT;
  decodelist[43] = D_DASH;
  decodelist[44] = D_SPC;
  decodelist[45] = D_PER;
  decodelist[46] = D_BLK;
  
  // Initialize serial and wait for port to open:
  Serial.begin(9600);

  Serial.println("Started...\n");
   
  // initialize the digital pin as an output.
  pinMode(4, OUTPUT);
  pinMode(5, OUTPUT);
  pinMode(6, OUTPUT);
  pinMode(7, OUTPUT);
  pinMode(8, OUTPUT);
  pinMode(9, OUTPUT);     
  pinMode(10, OUTPUT);
  pinMode(11, OUTPUT);
  pinMode(12, OUTPUT);
  pinMode(19, OUTPUT);
  pinMode(20, OUTPUT);
  pinMode(21, OUTPUT);
  pinMode(22, OUTPUT);
  pinMode(23, OUTPUT);
  
  // Initialize LED matrix to OFF state. This will have to change for different array dimensions.
  for (i=0; i<=8; i++) {
    digitalWrite(RN[i], HIGH);
  }
  for (i=0; i<=4; i++) {
    digitalWrite(CN[i], LOW);
  }

}

// the loop routine runs over and over again forever:
void loop() {
  // read from port 3
  // TODO: Need to shut off serial port if using pins 0 and 1 for digital I/O

  while (Serial.available() > 0) {
    inByte = Serial.read();
  
  // Find serial input character in datalist and set refval to bit pattern reference based on datalist index.
  if (Serial.read() == '\r') {
  idx = exposure;
  for (i=0; i<=46; i++) {
    if (datalist[i] == inByte) {
       idx = i;
    }
  }
  refval = decodelist[idx];
  if (idx != exposure) {
    idx = 0;
  }
  
// Write bit pattern to LED matrix. 

  while (idx < exposure) {
    for (i=0; i<rowcnt; i++) {
        digitalWrite(RN[i], LOW);
        digitalWrite(CN[4], !bitRead(refval[i],4));
        digitalWrite(CN[3], !bitRead(refval[i],3));
        digitalWrite(CN[2], !bitRead(refval[i],2));
        digitalWrite(CN[1], !bitRead(refval[i],1));
        digitalWrite(CN[0], !bitRead(refval[i],0));
        delay(strobe);
        digitalWrite(RN[i],HIGH);
      }
    digitalWrite(RN[i], HIGH);
    digitalWrite(CN[0], LOW);
    digitalWrite(CN[1], LOW);
    digitalWrite(CN[2], LOW);
    digitalWrite(CN[3], LOW);
    digitalWrite(CN[4], LOW);    
  idx = idx + 1;
  }
  Serial.println(inByte, DEC);
  }
  }
}

Data file conversion script

It's been a while since I have posted anything to this blog. In my memory the last use of Perl for work was a file parser that did a coordinate transform on one of the fields. I think it's some of the more complicated regular expression applications I have made so far.

The file takes defectivity data generated by an Orbotech display circuit scanner and converts it to a format that is compatible with a TNP laser tool. With the TNP laser we can move to the defective coordinates found by the Orbotech and attempt to salvage the circuit using laser ablation.

# Orbotech_to_TNP_Converter_20111121.pl
#
# Jovan Trujillo
# Arizona State University
# 11/21/2011
#
# This script formats an Orbotech file for use with the TNP Laser system.
#

use strict;

if ($#ARGV != 0) {
print "ARGV = ", $#ARGV, "\n";
print "Usage: Orbotech_to_TNP_Converter_20111121.pl \n";
exit(1);
}

my $rootPath = $ARGV[0];
my @lotList;
my @dataFile;
my $dataLine;
my $tempPath;

# Generate list of subfolders
opendir DIR, $rootPath or die "Couldn't open dir $rootPath: $!\n";
my @folders = readdir DIR;
closedir DIR;

foreach (@folders) {
if (-d $rootPath . "\\" . $_) {
if ($_ eq ".") {}
elsif($_ eq "..") {}
else {
push(@lotList, $_);
}
}
}

my %lotHash = ();
my $currentLot;

#Generate list of lot folders
foreach (@lotList) {
$tempPath = $rootPath . "\\" . $_;
$currentLot = $_;
opendir DIR, $tempPath or die "Couldn't open dir $tempPath: $!\n";
@folders = readdir DIR;
closedir DIR;
# Generate list of process folders
foreach (@folders) {
if (-d $tempPath . "\\" . $_) {
if ($_ eq ".") {}
elsif($_ eq "..") {}
else {
push(@{$lotHash{$currentLot}}, $_);
}
}
}
}

my $lot = "NULL";
my $process_step = "NULL";
my $data_file = "NULL";
my $panel_type = "";
my $panel_id = "";
my $operator_id = "";
my $scan_status = "";
my $queue = "";
my $go_repeats = "";
my $nogo_repeats = "";
my $num_repeats = "";
my $image_file_prefix = "";
my $slot_no = "";
my $glass_size_X = "";
my $glass_size_Y = "";
my $board_center_X = "";
my $board_center_Y = "";
my $panel_status = "NOT CLASSIFIED";
my $align_targets = "";
my $num_defects = "";
my $rep_no = "NULL";
my $X_coord = "NULL";
my $Y_coord = "NULL";
my $size = "NULL";
my $retype = "NULL";
my $type = "NULL";
my $cell_zone = "NULL";
my $size_X = "NULL";
my $size_Y = "NULL";
my $class = "NULL";
my $imagesuffix = "NULL";
my $row_cell_num = "NULL";
my $col_cell_num = "NULL";
my $i = 0;
my $output_file;
my @defect_list;
my $defect_string;

@folders = ();
for my $key (keys %lotHash) {
for ($i=0; $i <= $#{$lotHash{$key}}; $i++) {
$tempPath = $rootPath . "\\" . $key . "\\" . $lotHash{$key}[$i];
$lot = $key;
$process_step = $lotHash{$key}[$i];
opendir DIR, $tempPath or die "Couldn't open dir $tempPath: $!\n";
@folders = readdir DIR;
closedir DIR;
# Open each data file and parse for valid Orbotech Data
foreach (@folders) {
$tempPath = $rootPath . "\\" . $key . "\\" . $lotHash{$key}[$i] . "\\" . $_;
$data_file = $_;
open(DATA, "<$tempPath");

# Open report output file
$output_file = "TNP_" . $data_file;
$tempPath = $rootPath . "\\" . $key . "\\" . $lotHash{$key}[$i] . "\\" . $output_file;
open (REPORT, ">$tempPath") or die "Could not open $tempPath: $!\n";

@dataFile = ;
foreach $dataLine (@dataFile) {
# print "dataLine: $dataLine\n";
if ($dataLine =~ /panel_type\s*=\s*"([\w\s]*)"/) {
$panel_type = $1;
}
if ($dataLine =~ /panel_id\s*=\s*"([\w\s]*)"/) {
$panel_id = $1;
}
if ($dataLine =~ /operator_id\s*=\s*"([\w\s]*)"/) {
$operator_id = $1;
}
if ($dataLine =~ /queue\s*=\s*"([\w\s]*)"/) {
$queue = $1;
}
if ($dataLine =~ /scan_status\s*=\s*"([\w\s]*)"/) {
$scan_status = $1;
}
if ($dataLine =~ /go_repeats\s*=\s*"([\w\s]*)"/) {
$go_repeats = $1;
}
if ($dataLine =~ /nogo_repeats\s*=\s*"([\w\s]*)"/) {
$nogo_repeats = $1;
}
if ($dataLine =~ /image_file_prefix\s*=\s*"([\w\s\.]*)"/) {
$image_file_prefix = $1;
}
if ($dataLine =~ /slot_no\s*=\s*([\d\.\-]*)/) {
$slot_no = $1;
}
if ($dataLine =~ /num_repeats\s*=\s*([\d\.\-]*)/) {
$num_repeats = $1;
}
if ($dataLine =~ /glass_size.x\s*=\s*([\d\.\-]*)/) {
$glass_size_X = $1;
}
if ($dataLine =~ /glass_size.y\s*=\s*([\d\.\-]*)/) {
$glass_size_Y = $1;
}
if ($dataLine =~ /board_center.x\s*=\s*([\d\.\-]*)/) {
$board_center_X = $1;
}
if ($dataLine =~ /board_center.y\s*=\s*([\d\.\-]*)/) {
$board_center_Y = $1;
}
if ($dataLine =~ /panel_status\s*=\s*"([\w\s]*)"/) {
$panel_status = $1;
}
if ($dataLine =~ /align_targets\s*=\s*\(([\d\.\-]*)\s*([\d\.\-]*)\)\s*\(([\d\.\-]*)\s*([\d\.\-]*)\)\s*\(([\d\.\-]*)\s*([\d\.\-]*)\)\s*\(([\d\.\-]*)\s*([\d\.\-]*)\)/ ) {
$align_targets = "(" . $2 . " " . -$1 . ")(" . $4 . " " . -$3 . ")(" . $6 . " " . -$5 . ")(" . $8 . " " . -$7 . ")";
}
if ($dataLine =~ /num_defects\s*=\s*([\d\.\-]*)/) {
$num_defects = $1;
}
if ($dataLine =~ /([\d\.\-]*)\s*,\s*([\d\.\-]*)\s*,\s*([\d\.\-]*)\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\[\]\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*"*([\w\+\-\.]*)"*\s*,\s*([\d\.]*)\s*,\s*([\d\.]*),?/) {
$rep_no = $1;
$X_coord = $3;
$Y_coord = -$2;
$size = $4;
$retype = $5;
$type = $6;
$cell_zone = $7;
$size_X = $8;
$size_Y = $9;
$class = $10;
$imagesuffix = $11;
$row_cell_num = $12;
$col_cell_num = $13;
$defect_string = $rep_no . ", " . $X_coord . ", " . $Y_coord . ", \"" . $size . "\", \"" . $retype . "\", \"" . $type . "\", \"" . $cell_zone . "\", \"" . $size_X . "\", \"" . $size_Y . "\", \"" . $class . "\", \"" . $imagesuffix . "\", " . $row_cell_num . ", " . $col_cell_num . ", \n";
push(@defect_list, $defect_string);
}
}
print REPORT "panel_type =\"$panel_type\"\n";
print REPORT "panel_id =\"$panel_id\"\n";
print REPORT "operator_id =\"$operator_id\"\n";
print REPORT "queue =\"$queue\"\n";
print REPORT "scan_status =\"$scan_status\"\n";
print REPORT "go_repeats =\"$go_repeats\"\n";
print REPORT "nogo_repeats =\"$nogo_repeats\"\n";
print REPORT "image_file_prefix =\"$image_file_prefix\"\n";
print REPORT "slot_no =$slot_no\n";
print REPORT "num_repeats =$num_repeats\n";
print REPORT "glass_size.x =$glass_size_X\n";
print REPORT "glass_size.y =$glass_size_Y\n";
print REPORT "board_center.x =$board_center_X\n";
print REPORT "board_center.y =$board_center_Y\n";
print REPORT "panel_status =\"$panel_status\"\n";
print REPORT "align_targets =$align_targets\n";
print REPORT "num_defects =$num_defects\n";
print REPORT "defect_list =\"#Rep no. X coord. Y coord. Size-X Size-Y Class ImageSuffix Row_Cell_Num Col_Cell_Num\"\n";
foreach (@defect_list) {
print REPORT $_;
}
$panel_type = "";
$panel_id = "";
$operator_id = "";
$queue = "";
$scan_status = "";
$go_repeats = "";
$nogo_repeats = "";
$image_file_prefix = "";
$slot_no = "";
$num_repeats = "";
$glass_size_X = "";
$glass_size_Y = "";
$board_center_X = "";
$board_center_Y = "";
$panel_status = "NOT CLASSIFIED";
$align_targets = "";
$num_defects = "";
@defect_list = ();
close(DATA);
close(REPORT);
}
@folders = ();
}
}
print "Data conversion complete!\n";