What device are you using?
Typical low-end devices have poor focus, which is crucial for
reading the fine lines of a 1D bar-code.
When you have a smaller amount of data to encode a bar-code may end
up with thicker lines, which may contribute to the fact, that you're
successful in reading shorter bar-codes. When the bar-code is larger
the lines become thinner. Occasionally you might end up with a
FALSE-POSITIVE read which will show different text (I get typically
false-positive UPC-E reads). That might account for the occasional
'wrong' read.
Also make sure when you use CODE-128 with digits only to correctly
specify 128-C mode to take advantage of the higher capacity for that
case. Using 128-A or 128-B will result in larger bar-code for the same
content.
We're using extensively Code128 with 20 digits (SSCC) printed on
adhesive labels, and we have serious problems with low-end devices. We
were forced to add a back-up Data-Matrix on the label to alleviate
this issue a bit.
2013/4/16 Amal Somaweera <
am...@ceynox.com>:
> I tried to Scan code 128 barcode with character length 22 using Zxing
> Android. it is not scanning sometimes scanning not getting correct string.
> what is the problem zxing with code 128 and character length more than 22?
> (it is working fine for length 18)