tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Konstantin Kolinko <knst.koli...@gmail.com>
Subject Re: svn commit: r960864 - /tomcat/tc6.0.x/trunk/STATUS.txt
Date Wed, 07 Jul 2010 10:36:07 GMT
2010/7/6 Mark Thomas <markt@apache.org>:
> On 06/07/2010 12:40, kkolinko@apache.org wrote:
>>
>> Author: kkolinko
>> Date: Tue Jul  6 10:40:16 2010
>> New Revision: 960864
>>
>> URL: http://svn.apache.org/viewvc?rev=960864&view=rev
>> Log:
>> vote
>>
>> Modified:
>>     tomcat/tc6.0.x/trunk/STATUS.txt
>>
>> Modified: tomcat/tc6.0.x/trunk/STATUS.txt
>> URL:
>> http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/STATUS.txt?rev=960864&r1=960863&r2=960864&view=diff
>>
>> ==============================================================================
>> --- tomcat/tc6.0.x/trunk/STATUS.txt (original)
>> +++ tomcat/tc6.0.x/trunk/STATUS.txt Tue Jul  6 10:40:16 2010
>> @@ -154,11 +154,14 @@ PATCHES PROPOSED TO BACKPORT:
>>  * Fix https://issues.apache.org/bugzilla/show_bug.cgi?id=49551
>>    Allow default context.xml file to be specified using an absolute path
>>    http://svn.apache.org/viewvc?rev=960701&view=rev
>> -  +1: markt
>> +  +1: markt, kkolinko
>>    -1:
>>
>>  * Fix https://issues.apache.org/bugzilla/show_bug.cgi?id=49217
>>    Prevent use of Java keywords in EL identifiers
>>    http://svn.apache.org/viewvc?rev=960711&view=rev
>>    +1: markt
>> -  -1:
>> +  -1: kkolinko: r960711 is a testcase, the fix is in r960712
>> +   Regarding r960712 I am -1 because 1) the list of reserved keywords
>> must be shorter,
>> +   as I mentioned in Comment #3 for the issue and 2) it might break
>> existing applications,
>
> That analysis is not correct. Java identifier is defined by the Java
> language spec, not the EL spec.
>

I still think that

1) There is no direct example/evidence in the EL spec itself,
and I understand that TCK passes without a fix for BZ 49217,
because we have released 6.0.26 and 7.0.0

2) There is a separate chapter "Reserved Words" in EL spec,
which explicitly lists words that are forbidden as identifiers.

It is not mentioned there that Java language keywords are
also reserved.

It explicitly mentions "instanceof" which is already a Java keyword,
as well as mentions "true", "false" and "null".

3) I agree, that it is possible to read the spec in your way,
so it makes sense to have such a check,
but I think that there must be a way to turn it off.


Re:
  https://issues.apache.org/bugzilla/attachment.cgi?id=25726
The patch is incomplete. The "SKIP_IDENTIFIER_CHECK" property is mentioned
in documentation only. It is not used in the code.


>> +   so I think it must be possible to turn this check off.
>
> Agreed. I'll propose a back-port with an option to disable once I have
> implemented all of the restrictions in the JLS.
>
> Mark
>
>

Best regards,
Konstantin Kolinko

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tomcat.apache.org
For additional commands, e-mail: dev-help@tomcat.apache.org


Mime
View raw message