[
https://issues.apache.org/jira/browse/VALIDATOR-330?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13885444#comment-13885444
]
Sebb commented on VALIDATOR-330:
--------------------------------
In theory it is quite easy to check for invalid check digits (must be numeric,
and in range 02-98)
Some countries use fixed numbers for the check digits (e.g. because their
national code already includes a similar check digit scheme).
Unfortunately it appears that Jordan has decided to use 99 as the fixed numbers
- this would require special treatment.
Further research is needed to determine whether there are any other anomalies
(e.g. non-numeric check digits?).
Also, should the validation method treat "impossible" check digits as merely
incorrect, or should they be treated as invalid characters, in which case an
Exception is thrown?
> IbanCheckDigit.isValid() returns True for some invalid IBANs
> ------------------------------------------------------------
>
> Key: VALIDATOR-330
> URL: https://issues.apache.org/jira/browse/VALIDATOR-330
> Project: Commons Validator
> Issue Type: Bug
> Components: Routines
> Affects Versions: 1.4.0 Release
> Reporter: Peter Shawe
> Assignee: Sebb
>
> For example, isValid() returns True for both of these IBANs;
> "IE01AIBK93118702569045" and "IE98AIBK93118702569045". The "IE98" version is
> the correct one (confirmed with online checkers, which also fail the "IE01"
> version).
> calculate() correctly returns "98". As a workaround I'm calling calculate()
> and comparing the result with the checksum in the original IBAN.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
