Re: Jigsaw + Web Start

5.3/com/jgoodies/looks/windows/WindowsLookAndFeel.java?av=f#71 Best regards, Andrej Golovnin

Re: 8154956: Module system implementation refresh (4/2016)

tion: Is it guaranteed that entry.getFileName().toString always returns a value in lowercase? Best regards, Andrej Golovnin

Re: 8154956: Module system implementation refresh (4/2016)

modules. You will get an > unrecognized module error with `java -modulepath foo.JAR ..` for example. > You'll see the same in the tools. This has not changed in this refresh. Ah, good to know! :-) Thanks for the info, Alan! Best regards, Andrej Golovnin

Re: Speeding up jmod

InputStream per specification. Best regards, Andrej Golovnin diff -r 7210b5dbd92f src/jdk.jlink/share/classes/jdk/tools/jmod/JmodTask.java --- a/src/jdk.jlink/share/classes/jdk/tools/jmod/JmodTask.java Sat Apr 30 16:41:08 2016 -0700 +++ b/src/jdk.jlink/share/classes/jdk/tools/jmod/JmodTask.java Sun May

Re: Speeding up jmod

duplicate searching for module info when hashing of modules is required. - uses String#lastIndexOf(int) in toPackageName(ZipEntry) instead of String#lastIndexOf(String). - avoids checking for every directory entry in a JarFile if it ends with “module-info.class”. Best regards, Andrej Golovnin diff

Re: 8154956: Module system implementation refresh (4/2016)

g dots Maybe the regular expressions in the above lines should be precompiled in static final fields (Pattern objects are immutable and thread safe) to improve the performance of the #deriveModuleDescriptor(JarFile)-method. Best regards, Andrej Golovnin

Re: Review Request JDK-8160286: jmod hash is creating unlinkable modules.

it into an Optional. So why it is needed to wrap moduleName into Optional? Best regards, Andrej Golovnin On Fri, Jan 13, 2017 at 6:13 AM, Mandy Chung <mandy.ch...@oracle.com> wrote: > Updated webrev: >http://cr.openjdk.java.net/~mchung/jdk9/webrevs/8160286/webrev.01/ > > I did

Re: RFR: 8171400: Move checking of duplicate packages in the boot layer to link time

ther); 324 } 325 } 326 } 327 } I’m sorry for the late review but I think the indentation in the lines 314-327 is not correct. Best regards, Andrej Golovnin > On 19 Dec 2016, at 21:38, Claes Redestad <claes.redes...@oracle.com> wrote:

Minor performance improvement to java.lang.ModuleLayer.findModule(String name)

ll values: 846return layers() 847 .skip(1) // skip this layer 848 .map(l -> l.nameToModule.get(name)) 849 .filter(Objects::nonNull) 850 .findAny(); The suggested change is attached as diff. Best regards, Andrej G

Re: Review Request JDK-8192945: Need stable sort for MODULES entry in the release file

nion 8kb is to small. 16kb or even 32kb would be much better. Best regards, Andrej Golovnin On Wed, Dec 6, 2017 at 6:50 PM, mandy chung <mandy.ch...@oracle.com> wrote: > Okay this is a better version (traversing the graph keyset directly): > > http://cr.openjdk.java.net/~mch

Re: Minor performance improvement to java.lang.ModuleLayer.findModule(String name)

ueries, so that we can profile and get a tab on where improvements matter > most. Yes, that would be nice. Best regards, Andrej Golovnin

Re: Review Request JDK-8192945: Need stable sort for MODULES entry in the release file

the TreeSet in the method addNode(ResourcePoolModule). A little bit unrelated to your changes: 110 Deque visited = new LinkedList<>(); 111 Deque done = new LinkedList<>(); visited and done should be of type Set to avoid linear searching in the lines 114, 128, 129. Best

Re: ModuleDescriptor#toString(Set<>, String) may produce unexpected results.

> Thanks, this should be using toLowerCase(Locale). I'll create a bug for > this. Thanks, Alan! Best regards, Andrej Golovnin

ModuleDescriptor#toString(Set<>, String) may produce unexpected results.

Locale) with Locale.ROOT as the argument. The suggested change is attached as diff. Best regards, Andrej Golovnin diff --git a/src/java.base/share/classes/java/lang/module/ModuleDescriptor.java b/src/java.base/share/classes/java/lang/module/ModuleDescriptor.java --- a/src/java.base/share/classes/java/lang/