Posted: July 28 2006 at 8:03am | IP Logged

We are using the 128 barcodes and have some confusion as to the encoding functions.  From the descriptions it would seem that you could always use Code128Auto function since it automatically switches to the right character set, but in practice we find that submitting numerical values to both the Code128C functions and Code128Auto functions return different results (even on the Web Beta on Morovia's website). 

Why does this happen and when should one really use the Code128Auto function?

Thanks for clearing this up.

Posted: July 28 2006 at 9:34am | IP Logged

Code128 standard defines three sub character set - A, B and C. In theory, the encoder should be able to select the character set based on the data and switch from one to another. However, when doing that, you lose the human readable of each charcter, since the human readable is tied to the character set.

Code128C only encodes numeric data with even-length. It achieves the best length by combining two adjacent digits into one character. On the contrary, Code128Auto has to determine the character set one by one, therefore it does not always produce the best results. Although you get different barcodes, the decode result is the same. - i.e. the same data can have different code128 representations.

If you always encode even-length numeric data, just use Code128C. This is also the most widely used scenario, because it produces the shortest barcode. However, if you are enconding mixed data, you have to use Code128Auto (or Code128Ex).