Proposed patch for review (JCTree.Tag)

Jonathan Gibbons jonathan.gibbons at oracle.com
Fri Nov 4 13:10:38 PDT 2011


Kumar,

I think hasTag is the appropriate method name, and I don't think there 
are enough uses to justify hasNotTag or whatever.

-- Jon

On 11/04/2011 12:59 PM, Kumar Srinivasan wrote:
> Vicente,
>
> Thank you for making these changes.
>
> One more nit, while I was looking at your changes for some other reason,
> it struck me that  the correct usage  would be  "have()" and 
> "haveNot()" vs. "has()"
> and "hasNot()".
>
> Thanks
> Kumar
>
>
>> Thank you all for your insightful comments!
>>
>> Vicente
>>
>>
>>
>> --- El *mié, 2/11/11, Kumar Srinivasan 
>> /<kumar.x.srinivasan at oracle.com>/* escribió:
>>
>>
>>     De: Kumar Srinivasan <kumar.x.srinivasan at oracle.com>
>>     Asunto: Re: Proposed patch for review (JCTree.Tag)
>>     Para: compiler-dev at openjdk.java.net
>>     Fecha: miércoles, 2 de noviembre, 2011 21:49
>>
>>     Nice work
>>
>>     2 cents,  but is it worthwhile to add a hasNotTag(ASSIGN) or
>>     noTag(ASSIGN) convenience method ?
>>     this will eliminate a few negative checks, ex:
>>
>>     Annotate.java.
>>     -if (args.length() == 1 && !args.head.hasTag(ASSIGN)) {
>>     +if (args.length() == 1 && args.head.hasNotTag(ASSIGN)) {
>>
>>     Kumar
>>
>>     > I have posted an updated patch from Vicente, available here:
>>     >
>>     > http://cr.openjdk.java.net/~jjg/6921494/webrev.01/
>>     <http://cr.openjdk.java.net/%7Ejjg/6921494/webrev.01/>
>>     >
>>     > -- Jon
>>     >
>>     > On 10/24/2011 02:39 AM, Maurizio Cimadamore wrote:
>>     >> On 24/10/11 06:25, Jonathan Gibbons wrote:
>>     >>> I've posted a patch from our newest contributor, Vicente
>>     Zaldivar.
>>     >>> The patch is a cleanup to the JCTree.tag values, to use enums
>>     >>> instead of small integers.
>>     >>>
>>     >>> You can see the patch here:
>>     >>> http://cr.openjdk.java.net/~jjg/6921494/webrev.00/
>>     <http://cr.openjdk.java.net/%7Ejjg/6921494/webrev.00/>
>>     >>>
>>     >>> Thanks to Vicente for the contribution. Review feedback here or
>>     >>> to Vicente, please.
>>     >>>
>>     >>> -- Jon
>>     >> Great work - couple of comments below:
>>     >>
>>     >> *) JCTree - I would add an import static as follows:
>>     >>
>>     >> import static com.sun.tools.javac.tree.JCTree.Tag.*
>>     >>
>>     >> This way you will get rid of all the qualified identifier
>>     (i.e. most
>>     >> of the code in JCTree should go back to its original form).
>>     >>
>>     >> The same applies to all files (I listed the files where there
>>     is at
>>     >> least more than a single qualified ident):
>>     >>
>>     >> -) Annotate.java
>>     >> -) Attr.java
>>     >> -) Check.java
>>     >> -) Flow.java
>>     >> -) Lower.java
>>     >> -) MemberEnter.java
>>     >> -) Resolve.java (a non-static import of JCTree.Tag would do fine)
>>     >> -) Gen.java
>>     >> -) JavacElements.java
>>     >> -) JavacParser.java
>>     >> -) Pretty.java
>>     >> -) TreeInfo.java
>>     >> -) TreeMaker.java (a non-static import of JCTree.Tag would do
>>     fine)
>>     >> -) ClassDocImpl.java
>>     >>
>>     >> Maurizio
>>     >
>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20111104/22193b75/attachment.html 


More information about the compiler-dev mailing list