RFR: 8174735 Update JAX-WS RI integration to latest version

Wed Feb 15 13:58:23 UTC 2017

Hi,

comments wrt JSR-250 inline, also adding Linda who maintains it these 
days to the loop

On 2/15/17 2:36 PM, Roman Grigoriadi wrote:
>
> Adding Lukas
>
>
> On 02/15/2017 01:40 PM, Roman Grigoriadi wrote:
>>
>> Hi Lance,
>>
>> On 02/14/2017 09:28 PM, Lance Andersen wrote:
>>> Hi Roman,
>>>
>>> I made a pass through the changes.  And here are some general 
>>> comments which are mostly minor in nature….
>>>
>>> HTH
>>>
>>> Best
>>> Lance
>>> ----------------------
>>> JAXBRIContext.java:
>>> - The comment “If true, despite of the specification….”;  This does 
>>> not read well can this be clarified?
>> We have changed it to:
>> * If true and element namespace is not specified, namespace of parent 
>> element will be used.
>> * The default value is false.
>> Hope this is acceptable.
>>>
>>> Generated.java, PostConstruct.java, PreDestroy.java,Resource.java
>>>          -  Be nice to use {@code} vs <code></code> for readability
>>>          -  Do you really need <p></p>?  I am assuming there are no 
>>> doclint errors on Generated?
>>>          - Was there a CCC for the changes as there were some minor 
>>> semantic updates
>> These are from JSR 250 we don't own sources.

for the first two points - this is up to Linda, we - in JAX-WS are just 
consuming what JSR-250 provides, filtering out what is not needed and 
delivering the result into JDK; it does not make much sense for us to 
'alter' what we get (in terms of explicit code edits) ...

as for CCC - good question. Since there is no change in what has been 
delivered to JDK since its version 6, I believe that that was done in 
the past. All I know about 'CCC' is that in our sources I can see 
Miran's comment from 2013 'CCC would be created to include this into 
JDK?' related to the explicit exclusion of @javax.annotation.Priority 
from jsr-250.jar during jdk integration part. Do we have to go through 
'CCC' (whatever does it mean) even because of mostly javadoc related 
changes? If so, would it be possible to consider adding additional 
content into what's delivered into JDK? I've heard some comments that 
@Priority in particular would be quite useful for frameworks...

thanks,
--lukas

>>>
>>>
>>> MimeMultipart.java.:
>>>        - createMimeBodyPart @param len does not align properly
>>>
>>> SharedInputStream.java
>>>        - writeTo @param out does not align
>>>
>>> NameImpl.java
>>>        -  Some comments have a blank line before the params, others 
>>> do not.  It would be nice to be consistent
>> fixed
>>>
>>> ParserPool.java
>>>        - Can you explain the status regarding returnParser() and 
>>> https://jaxp.dev.java.net/issues/show_bug.cgi?id=59
>> removed //TODO and commented method, Aleksej will provide patch for 
>> resetting symbol table on parser.reset()
>>> SAAJUtil.java
>>>        - getSystemClassLoader, do you really need to check for a 
>>> SecurityManager?  Why not just AccessController.doPrivilege(could 
>>> also use a lambda also)
>> Removed SecurityManager check, lambda cannot be used without 
>> utilizing multirelease-jar for saaj, our min supported version is 1.7
>>>
>>> MethodUtil.java
>>>        - Looks like the alignment is off in the changes in invoke
>> There are 5 of them in the sync. I reformatted new ones. Can't see 
>> any wrong indentation in updated ones. Also fixed package in some 
>> comments/logs.
>>>
>>> ContextClassloaderLocal.java
>>>         -  getContextClassLoader could be changed to use a lambda 
>>> with AccessController.doPrivilege if you desire
>>>
>>> FactoryFinder.java
>>>        - getSystemProperty, it looks like you were using a lambda 
>>> and removed it for the AccessController.doPrivilege?
>> both no lambda for min target 1.7
>>>
>>> SecureLoader.java
>>>       - same comment as SAAJUtil.java
>> I don't know the reason why SecurityManager presence check is present 
>> there. Maybe performance? But there are 14 copies of SecureLoader 
>> class in JAXB, if we do remove unnecessary check we should replace 
>> all of them. Lets postpone it.
>>> NGCCRuntimeEx.java
>>>       - copyLocater, createValidationContext @return comment needs 
>>> checked
>> fixed
>>>
>>> package-info.java
>>>       - Seem the new files include a 2003/2004/1997, 2017, 
>>> copyright, should just be 2017 i believe
>> fixed
>>
>> we are working on a new webrev, hopefully will post it later today.
>>
>> Roman
>>
>>
>>>
>>>> On Feb 14, 2017, at 7:00 AM, Roman Grigoriadi 
>>>> <roman.grigoriadi at oracle.com <mailto:roman.grigoriadi at oracle.com>> 
>>>> wrote:
>>>>
>>>> Hi,
>>>>
>>>> Please review standalone JAXB/JAXWS changes, synced to jdk/jaxws repo.
>>>>
>>>> JBS: https://bugs.openjdk.java.net/browse/JDK-8174735
>>>> Webrev: 
>>>> http://cr.openjdk.java.net/~aefimov/jaxws-integrations/8174735/00/ 
>>>> <http://cr.openjdk.java.net/%7Eaefimov/jaxws-integrations/8174735/00/>
>>>>
>>>> You can find change list in the description of JBS and its linked 
>>>> issues. Most diffs are for SAAJ-RI related to JDK-8166745 - remove 
>>>> dependencies on jdks Xerces internal classes.
>>>>
>>>> Thanks,
>>>> Roman
>>>
>>> <http://oracle.com/us/design/oracle-email-sig-198324.gif>
>>> <http://oracle.com/us/design/oracle-email-sig-198324.gif><http://oracle.com/us/design/oracle-email-sig-198324.gif>
>>> <http://oracle.com/us/design/oracle-email-sig-198324.gif>Lance 
>>> Andersen| Principal Member of Technical Staff | +1.781.442.2037
>>> Oracle Java Engineering
>>> 1 Network Drive
>>> Burlington, MA 01803
>>> Lance.Andersen at oracle.com <mailto:Lance.Andersen at oracle.com>
>>>
>>>
>>>
>>
>