dev@glassfish.java.net

Re: [Fwd: sso-enabled and access-logging-enabled attribute changed recently ?]

From: Lloyd Chambers <Lloyd.Chambers_at_Sun.COM>
Date: Fri, 16 Oct 2009 08:38:50 -0700

I think we're creating a mess by accepting so many values (as a
general principle). It's sloppy. We should stick to exactly what
Boolean.valueOf() returns.

Why not "vrai" and "nyet" also? Once the line is crossed, english-
centric seems unfair when I don't speak english.

Lloyd

On Oct 16, 2009, at 6:11 AM, Justin Lee wrote:

> Agreed.
>
> Jan Luehe wrote:
>> Anissa,
>>
>> On 10/15/09 18:34, Anissa Lam wrote:
>>>
>>> I didn't hear any response from the admin alias . resend to the
>>> dev alias.
>>> I need to know if GUI should change this to 'true|false|
>>> inherit' or keep it as 'on|off|inherit'. Actually i have
>>> 'temp' change that to make the verifier happy, but not sure if
>>> thats the right thing, or masking another bug.
>>
>> I did some research and found that the change was made in response to
>> https://glassfish.dev.java.net/issues/show_bug.cgi?id=10113
>>
>> The web container uses ConfigBeansUtilities#toBoolean to evalute
>> these props,
>> which interprets "true", "on", "yes", and "1" as enabled.
>>
>> Therefore, my preference would be to accept "true|on|false|off|
>> inherit' as legal values.
>>
>>
>> Jan
>>
>>
>>>
>>> thanks
>>> Anissa.
>>>
>>> ------------------------------------------------------------------------
>>>
>>> Subject:
>>> sso-enabled and access-logging-enabled attribute changed recently ?
>>> From:
>>> Anissa Lam <Anissa.Lam_at_sun.com>
>>> Date:
>>> Tue, 13 Oct 2009 17:25:59 -0700
>>> To:
>>> admin_at_glassfish.dev.java.net
>>>
>>> To:
>>> admin_at_glassfish.dev.java.net
>>>
>>>
>>> Hi,
>>>
>>> In promoted build 67, GUI is able to pass in "on/off/inherit" for
>>> sso-enabled and access-logging-enabled attribute in virtual server.
>>> But today, I notice that backend is throwing an error. It says:
>>>
>>> *Constraints for this bean violated. Message = ssoEnabled must
>>> match "(true|false|inherit)"
>>> *same for accessLoggingEnabled.
>>>
>>> Is this a bug in the verifier or this is an intended change ?
>>> I hope it is not the latter. I think its really too late in the
>>> development cycle to make change like that, and *without* sending
>>> out notification. It will affect the console, doc and testing.
>>>
>>> thanks
>>> Anissa.
>>>
>>> ------------------------------------------------------------------------
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
>>> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>

Lloyd Chambers
lloyd.chambers_at_sun.com
GlassFish Team