RFR (16): 8241003: Deprecate "denigrated" java.security.cert APIs that represent DNs as Principal or String objects
Sean Mullan
sean.mullan at oracle.com
Wed Aug 26 11:53:57 UTC 2020
On 8/21/20 2:01 PM, Xuelei Fan wrote:
> Looks fine to me.
>
> Just a trivial format comment. Some use link for "RFC 2253", some do
> not. It's OK. And it's good as well if you want to use a uniform style.
Good point. Several methods in X509CertSelector and X509CRLSelector
reference RFC 2253 several times. I have changed them to be consistent
so that the first reference in each method is a link to 2253, while the
rest of them are text.
--Sean
>
> Xuelei
>
> On 8/21/2020 9:24 AM, Sean Mullan wrote:
>> Ping ...
>>
>> On 8/7/20 10:01 AM, Sean Mullan wrote:
>>> Please review this change to deprecate the following APIs:
>>>
>>> java.security.cert.X509Certificate.getIssuerDN()
>>> java.security.cert.X509Certificate.getSubjectDN()
>>> java.security.cert.X509CRL.getIssuerDN()
>>> java.security.cert.X509CertSelector.setIssuer(String)
>>> java.security.cert.X509CertSelector.setSubject(String)
>>> java.security.cert.X509CertSelector.getIssuerAsString()
>>> java.security.cert.X509CertSelector.getSubjectAsString()
>>> java.security.cert.X509CRLSelector.addIssuerName(String)
>>>
>>> These APIs either take or return Distinguished Names as Principal or
>>> String objects which can cause issues due to loss of encoding
>>> information or differences when comparing names across different
>>> Principal implementations. All of them have alternative APIs which
>>> use X500Principal objects instead. They have long had warnings in the
>>> javadoc and have been discouraged from being used. There are no plans
>>> to remove the APIs at this time, as they have been in the platform
>>> for a long time and removing them would be a much higher
>>> compatibility risk.
>>>
>>> webrev: https://cr.openjdk.java.net/~mullan/webrevs/8241003/webrev.00/
>>> CSR: https://bugs.openjdk.java.net/browse/JDK-8250970
>>> bug: https://bugs.openjdk.java.net/browse/JDK-8241003
>>>
>>> --Sean
More information about the security-dev
mailing list