Code 39 (also known as USS Code 39, Code 3 of 9) is the first alpha-numeric symbology developed to be used in non-retail environment. It is widely used to code alphanumeric information, such as the model number etc. It is designed to encode 26 upper case letters A-Z, 10 digits 0-9 and 7 special characters - hyphen (-), period (.), dollar sign ($), forward slash (/), plus sign (+), percent (%) as well as the space character.

Although you can put as many characters as you can, in practice you can not encode many characters. Many barcode scanners have a short scan range at 3 inches.

Each Code 39 barcode must begin and end with special bar/space patterns. They are often referred as start/stop character. In the font, the start/stop character is mapped to the asterisk (*). You can also use left square bracket [ and right square bracket ] to produce them.

Creating a Code39 barcode is a simple task. Just enclose your data with asterisks and print the whole string with a C39 font, you get a barcode. For example, let's say that the data to be encoded is:

PN99018

The barcode string for printing as a Code39 barcode would be:

*P99018*

The below escape sequences print the barcode, and the human readable text below the barcode:

(0Y<ESC>(s0p8h12*PN99018*
<ESC>(3@PN99018

C Code

The following code snippet demonstrates the barcode printing in C language.

int PrintPartNumberBarcode(char* pPartNo, FILE* prn)
{
    /*select MRV Code39MA Height 12 points) */
    fprintf(prn, "\x1B(0Y\x1B(s12.00H"); 
    fprintf(prn, "*%s\n", pPartNo);
}

BASIC Code

The following BASIC function returns the same barcode string.

Function GetBarcodeString(partno as string) as String
    Dim str as string
    str = "*" & partno & "*"
    GetBarcodeString=str
End Function

Code39 standard also specifies a way to print characters outside the native Code 39 character set. It does so by encoding these characters with two native Code 39 characters. Four characters, the percent sign (%), dollar ($), slash (/) and plus sign (+) are used as leading characters.

Since these leading characters are also in the native Code 39 character set, the barcode reader is unable to determine whether a barcode is Code 39 extended or just normal code 39. For example, a barcode reader may read the extended code 39 barcode abc as +A+B+C. Fortunately, most scanners allow you configure if you expect a Code 39 extended symbol (sometimes called Code 39 Full ASCII), or just a normal code 39 symbol.

When you need to encode characters outside the native set, use the table below. For example, the sequence for lower case letter a is +A.

UPC-A is widely used in the United States and Canada to identify retail products at the time of checkout. A UPC-A symbol has a fixed length of 12 digits, with the last digit served as a checksum. The first digit is called “number system”, followed by the manufacturer code and product code. The Uniform Code Council has the authority assigning manufacturer code. The remaining digits, the product code, is assigned by the manufacturer. This arrangement ensures that every retail proudct has its unqiue identification number.

The UPC-A encoding scheme is relativley complicated. A digit has two different encodings depending on the position of the digit. A complete barcode string contains 15 characters, divided into 7 parts: ^[3]

The left guard and right guard have the same encoding patterns. They can be printed using asterisk, left and right parenthesis or brackets: *, ( and ), or [ and ]. The center guard can be printed using either a vertical bar symbol | or the hyphen -.

The patterns for the left halve digits are represented by 0-9 respectively; and the right halve digits are mapped to A-J:

For example, suppose we wish to encode the number 543000186706. The following three lines produce identical barcodes:

    *543000|BIGHAG*
    (543000-BIGHAG)
    [543000|BIGHAG] 

C Code

The following C code illustrates the check digit calculation and the barcode string generation.

static char barcode_string[20];
int upc_a_check_digit(char* data)
{
    int sum, check_digit;
    char* p;
    sum=0;
    sum = (data[0]+data[2]+data[4]+data[6]+data[8]+data[10])*3 +
          (data[1]+data[3]+data[5]+data[7]+data[9]+data[11]);
    check_digit = 10-(sum %10 );
    if (check_digit==10 ) check_digit=0;
    return check_digit;
}

char* upc_barcode_string(char* data)
{
    char* p;
    int i;
    int check_digit;

    memset(barcode_string, 0, sizeof(barcode_string));
    check_digit=upc_a_check_digit(data);
    barcode_string[0]='[';
    for(i=0; i<6; i++)
        barcode_string[i+1]=data[i];
    barcode_string[7]='|';
    for(i=6; i<10; i++)
        barcode_string[i+2]=data[i]-'0'+'A';
    barcode_string[13]=check_digit +'A';
    barcode_string[14]=']';
    return barcode_string;
}

BASIC Code

Function upc_a_check_digit(data as String) As Integer
    Dim sum as Integer
    Dim i As Integer
    
    sum=0
    for i=1 to 11 Step 2
        sum = sum + Val(Mid(data,I,1))
    Next i
    sum = sum * 3
    For i=2 to 11 Step 2
        sum=sum+Val(Mid(data, I, 1))
    Next i
    sum = 10 - (10 Mod 10)
    if ( sum=10 ) Then sum=0 End IF
    upc_a_check_digit=sum;
End Function
Function upc_barcode_string(data as String) As String
    Dim check_digit As Integer
    Dim I As Integer
    Dim barcode_string As String
    Dim myChar as Integer
    
    barcode_string="["
    barcode_string = barcode_string & Left$(data, 6) & "|"
    for i=7 to 11
        myChar = Val(Mid(data,I,1))
        myChar = myChar + Asc("A")
        barcode_string=barcode_string & chr(myChar)
    Next i
    myChar = upc_a_check_digit(data)
    barcode_string = barcode_string & chr(myChar) & "]"
    upc_barcode_string=barcode_string
End Function 

The POSTNET (POSTal Numeric Encoding Technique) bar code type was developed by the U. S. Post Office to encode a delivery address, which can be one of three forms (1) 5-digit ZIP; (2) 5-digit ZIP+4 code and (3) 11-digit delivery point code. A valid POSTNET barcode comprises either 32 bars, 52 bars or 62 bars (including frame bars and check digit, see below).

POSTNET is a numeric symbology. Different from other symbologies, POSTNET is a height-modulated symbology which encodes the data in the height of the barcode instead of the width.

Figure 5.1. POSTNET barcode structure

char zipcode[20], barcodestring[20]; 
char* p; 
int sum, check_digit; 
strcpy(zipcode, "918011234"); 
sum=0; 
p=zipcode; 
while (*p!=NULL) {
    sum += *p-'0'; 
    p++; 
} 
check_digit = 10 - (sum%10); 
if (check_digit==10 ) check_digit=0; 
printf("*%s%d*", zipcode, check_digit);

Dim i As Integer
Dim charToEncode As String
Dim checkSum  As Integer
Dim checkDigit As String
Dim strZipCode As String
strZipCode = "918011234";

For i = 1 To Len(inpara)
    charToEncode = Mid(strZipCode, i, 1)
    Postnet = Postnet + charToEncode
    checkSum = checkSum + Val(charToEncode)
Next i
checkSum = checkSum Mod 10
If checkSum <> 0 Then checkSum = 10 - checkSum
checkDigit = Chr(checkSum + Asc("0"))

Included in this Morovia PCL Bar codes & More package are the PCL bitmap fonts for printing OCR-A and OCR-B characters. The mrvocra.sfp prints OCR-A size I characters while mrvocrb.sfp for OCR-B size I characters. Both fonts have a point size of 12 and a fixed pitch of 10 cpi (Character per inch).

Figure 5.2. OCR-A characters

Figure 5.3. OCR-B characters

For other OCR character sizes, see http://www.morovia.com/font/ocr.asp.