From _weijiang_ at yahoo.com Tue Oct 2 13:33:13 2007 From: _weijiang_ at yahoo.com (Wei Jiang) Date: Tue, 2 Oct 2007 13:33:13 -0700 (PDT) Subject: Older version of javac compiler source Message-ID: <496061.48466.qm@web52506.mail.re2.yahoo.com> Hi, Where can I find older version of javac compiler? I need javac 1.5. Any information would be appreciated. Thanks in advance. ____________________________________________________________________________________ Check out the hottest 2008 models today at Yahoo! Autos. http://autos.yahoo.com/new_cars.html From Jonathan.Gibbons at Sun.COM Tue Oct 2 15:58:19 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Tue, 02 Oct 2007 15:58:19 -0700 Subject: Older version of javac compiler source In-Reply-To: <496061.48466.qm@web52506.mail.re2.yahoo.com> References: <496061.48466.qm@web52506.mail.re2.yahoo.com> Message-ID: <4702CD0B.7010700@sun.com> The sources for JDK 5, including javac, are available under the JRL license at https://tiger.dev.java.net/ They are not available as part of OpenJDK. -- Jon Wei Jiang wrote: > Hi, > > Where can I find older version of javac compiler? I need javac 1.5. > Any information would be appreciated. Thanks in advance. > > > ____________________________________________________________________________________ > Check out the hottest 2008 models today at Yahoo! Autos. > http://autos.yahoo.com/new_cars.html > From Jonathan.Gibbons at Sun.COM Tue Oct 2 17:17:08 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Tue, 02 Oct 2007 17:17:08 -0700 Subject: replace assert by throw new AssertionError Message-ID: <4702DF84.9050309@sun.com> Team, javac currently contains assertion statements, and goes out of its way to ensure assertions are enabled. As such, it would seem better to replace "assert cond: expr" with "if (!cond) throw new AssertionError(expr)". Then we could remove the code from all the places where we force assertions to be enabled for javac. Does anyone have comments, especially any reasons why this would *not* be a good idea? -- Jon From aulakh3333 at gmail.com Tue Oct 2 18:02:41 2007 From: aulakh3333 at gmail.com (navroop singh) Date: Tue, 2 Oct 2007 21:02:41 -0400 Subject: replace assert by throw new AssertionError In-Reply-To: <4702DF84.9050309@sun.com> References: <4702DF84.9050309@sun.com> Message-ID: <78a803f70710021802y52f93b57w8d7e9a10d69e2e8d@mail.gmail.com> Hey guys Are you part of the sun JAVA group or Open JDK( developing own javac from ground up) ? Thanks On 10/2/07, Jonathan Gibbons wrote: > > Team, > > javac currently contains assertion statements, and goes out of its way > to ensure assertions are enabled. > As such, it would seem better to replace "assert cond: expr" with "if > (!cond) throw new AssertionError(expr)". > Then we could remove the code from all the places where we force > assertions to be enabled for javac. > Does anyone have comments, especially any reasons why this would *not* > be a good idea? > > -- Jon > -- thanks navroop singh -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071002/fa029828/attachment.html From Tom.Ball at Sun.COM Tue Oct 2 20:25:27 2007 From: Tom.Ball at Sun.COM (Tom Ball) Date: Tue, 02 Oct 2007 20:25:27 -0700 Subject: replace assert by throw new AssertionError In-Reply-To: <78a803f70710021802y52f93b57w8d7e9a10d69e2e8d@mail.gmail.com> References: <4702DF84.9050309@sun.com> <78a803f70710021802y52f93b57w8d7e9a10d69e2e8d@mail.gmail.com> Message-ID: <47030BA7.5000803@sun.com> The javac that is part of the Open JDK *is* the compiler that Sun's Java group created. AFAIK, all Sun javac work is open source now, so Jon was asking the Java compiler community both inside and outside Sun for design feedback (like he should now). I responded with a Glassfish comment not because that team should unilaterally decide an open source community question, but because they are just one of many appserver teams using javac within and outside of Sun and so a useful data point (and, Glassfish is open-source, too). I don't know of any Open JDK effort to develop a Java compiler from scratch. Do you? If so, please share the details with the other alias members. Tom navroop singh wrote: > Hey guys > > Are you part of the sun JAVA group or Open JDK( developing own javac > from ground up) ? > > > Thanks > > > On 10/2/07, *Jonathan Gibbons* > wrote: > > Team, > > javac currently contains assertion statements, and goes out of its way > to ensure assertions are enabled. > As such, it would seem better to replace "assert cond: expr" with "if > (!cond) throw new AssertionError(expr)". > Then we could remove the code from all the places where we force > assertions to be enabled for javac. > Does anyone have comments, especially any reasons why this would *not* > be a good idea? > > -- Jon > > > > > -- > thanks > navroop singh From _weijiang_ at yahoo.com Wed Oct 3 07:20:42 2007 From: _weijiang_ at yahoo.com (Wei Jiang) Date: Wed, 3 Oct 2007 07:20:42 -0700 (PDT) Subject: How to lean (debug) javac Message-ID: <992077.50286.qm@web52510.mail.re2.yahoo.com> Hi, I am trying to follow instructions to debug (to learn) javac. I downloaded openjdk 7 source at: openjdk-7-ea-src-b21-27_sep_2007.zip and unzip'd. (I can not find source code of javac for 1.6). 1. at http://www.ahristov.com/tutorial/java-compiler/netbeans.html. When I go to directory C:\openjdk7b21\langtools\make\netbeans\compiler and "Open Project", the project opened without error. But the project name remains as "compiler". I do not see the same screen as you showed. I found a build.xml and tried to run it. I got the following errors: >>>>>>>>>> C:\openjdk7b21\langtools\make\build.xml:315: The following error occurred while executing this line: C:\openjdk7b21\langtools\make\build.xml:650: Cannot locate bootstrap java: please set boot.java.home to its location BUILD FAILED (total time: 0 seconds) <<<<<<<<< 2. I tried at http://nb-openjdk.netbeans.org/get-and-build.html I nivagate directories and see: ../j2se/make/netbeans/ awt2d common j2se ... but there is no "compiler" directory. So I can not follow the instruction. Any thing I can do to follow an instruction to build-debug javac? Any information would be appreciated. Thanks in advance. ____________________________________________________________________________________ Shape Yahoo! in your own image. Join our Network Research Panel today! http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7 From Jonathan.Gibbons at Sun.COM Wed Oct 3 07:37:22 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Wed, 03 Oct 2007 07:37:22 -0700 Subject: How to lean (debug) javac In-Reply-To: <992077.50286.qm@web52510.mail.re2.yahoo.com> References: <992077.50286.qm@web52510.mail.re2.yahoo.com> Message-ID: Wei, We recently reorganized the source code and some of the documentation still needs to be updated. To fix the error regarding boot.java.home, create a file "build.properties" in the langtools directory, and in that, define boot.java.home to the version of java you wish to use to build the compiler. Regarding the instuctions at (2), these are out of date following the reorganization. The compiler can now be found under .../langtools/make/netbeans/compiler -- Jon On Oct 3, 2007, at 7:20 AM, Wei Jiang wrote: > Hi, > > I am trying to follow instructions to debug (to learn) javac. > I downloaded openjdk 7 source at: openjdk-7-ea-src-b21-27_sep_2007.zip > and unzip'd. (I can not find source code of javac for 1.6). > > 1. at http://www.ahristov.com/tutorial/java-compiler/netbeans.html. > > When I go to directory C:\openjdk7b21\langtools\make\netbeans\compiler > and "Open Project", the project opened without error. But the project > name remains as "compiler". I do not see the same screen as you > showed. I found a build.xml and tried to run it. I got the following > errors: > >>>>>>>>>>> > C:\openjdk7b21\langtools\make\build.xml:315: The following error > occurred while > executing this line: > C:\openjdk7b21\langtools\make\build.xml:650: Cannot locate > bootstrap java: > please set boot.java.home to its location > BUILD FAILED (total time: 0 seconds) > <<<<<<<<< > > > 2. I tried at http://nb-openjdk.netbeans.org/get-and-build.html > I nivagate directories and see: > ../j2se/make/netbeans/ > awt2d > common > j2se > ... > but there is no "compiler" directory. So I can not follow the > instruction. > > Any thing I can do to follow an instruction to build-debug javac? > Any information would be appreciated. Thanks in advance. > > > > > > ______________________________________________________________________ > ______________ > Shape Yahoo! in your own image. Join our Network Research Panel > today! http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7 > > From Jonathan.Gibbons at Sun.COM Wed Oct 3 09:19:13 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Wed, 03 Oct 2007 09:19:13 -0700 Subject: Mac usage Message-ID: OK, here's a question I probably shouldn't ask because I'm not sure I want to know the answer ... OpenJDK is currently targetted at the Solaris, Windows and Linux platforms. Mac is not a supported platform for OpenJDK. But the langtools component is pure Java and has had a somewhat more liberal notion of platform. But, in trying out langtools on the Mac, I've run into some issues, such as the different organization Apple's Java product, most recently, the names of the subdirectories under ${JAVA_HOME}. So, I'm wondering how much interest/demand there is for working on the compiler on a Mac platform. I know there are folk out there that like the Mac -- I'm beginning to like it too, apart from hiccups like this. Depending on the response, I'll be able to gauge the priority of fixes for the issues that arise. -- Jon From nathan.ryan at solidwaretechnologies.com Wed Oct 3 09:28:25 2007 From: nathan.ryan at solidwaretechnologies.com (Nathan D. Ryan) Date: Wed, 3 Oct 2007 10:28:25 -0600 Subject: Mac usage In-Reply-To: References: Message-ID: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> I'd be in for that. I run Parallels on my PowerBook with OS X, Windows, and Linux. It's frustrating being able to successfully test things in Windows and Linux but having them fail in OS X due to the (let's be honest, goofy) layout and naming of the Java environment. Nathan On Oct 3, 2007, at 10.19, Jonathan Gibbons wrote: > OK, here's a question I probably shouldn't ask because I'm not sure > I want to know the answer ... > > OpenJDK is currently targetted at the Solaris, Windows and Linux > platforms. Mac is not a supported platform for OpenJDK. But the > langtools component is pure Java and has had a somewhat more > liberal notion of platform. > > But, in trying out langtools on the Mac, I've run into some issues, > such as the different organization Apple's Java product, most > recently, the names of the subdirectories under ${JAVA_HOME}. > > So, I'm wondering how much interest/demand there is for working on > the compiler on a Mac platform. I know there are folk out there > that like the Mac -- I'm beginning to like it too, apart from > hiccups like this. Depending on the response, I'll be able to gauge > the priority of fixes for the issues that arise. > > -- Jon > From robilad at kaffe.org Wed Oct 3 09:31:49 2007 From: robilad at kaffe.org (Dalibor Topic) Date: Wed, 03 Oct 2007 18:31:49 +0200 Subject: Mac usage In-Reply-To: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> References: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> Message-ID: <4703C3F5.8060000@kaffe.org> Nathan D. Ryan wrote: > I'd be in for that. I run Parallels on my PowerBook with OS X, Windows, > and Linux. It's frustrating being able to successfully test things in > Windows and Linux but having them fail in OS X due to the (let's be > honest, goofy) layout and naming of the Java environment. Sounds like a nice area to solicit contributions to me (only running OS X at work, myself). cheers, dalibor topic From Jonathan.Gibbons at Sun.COM Wed Oct 3 09:41:19 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Wed, 03 Oct 2007 09:41:19 -0700 Subject: Mac usage In-Reply-To: <4703C3F5.8060000@kaffe.org> References: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> <4703C3F5.8060000@kaffe.org> Message-ID: <98784D7E-5288-4C11-9F3A-BFCE5BF9B6EF@sun.com> To be clear, the build itself already works on a Mac. The problem right now is jtreg, which thinks there is a "bin/" directory. Unfortunately, jtreg is not (yet) open source (we're working on it!) so I can't solicit contributions, but fortunately, I know the owner of jtreg quite well ... -- Jon On Oct 3, 2007, at 9:31 AM, Dalibor Topic wrote: > Nathan D. Ryan wrote: >> I'd be in for that. I run Parallels on my PowerBook with OS X, >> Windows, >> and Linux. It's frustrating being able to successfully test things in >> Windows and Linux but having them fail in OS X due to the (let's be >> honest, goofy) layout and naming of the Java environment. > > Sounds like a nice area to solicit contributions to me (only > running OS > X at work, myself). > > cheers, > dalibor topic From ted at tedneward.com Wed Oct 3 11:06:29 2007 From: ted at tedneward.com (Ted Neward) Date: Wed, 3 Oct 2007 11:06:29 -0700 Subject: Mac usage In-Reply-To: References: Message-ID: <071d01c805e8$21920b70$64b62250$@com> Most of the speakers on the No Fluff Just Stuff tour use Macs; I'd say that supporting the Mac is, for us, a high priority. Having said that, I don't want to speak for them; I'll forward this on to them and let them sing out to you directly. Ted Neward Java, .NET, XML Services Consulting, Teaching, Speaking, Writing http://www.tedneward.com > -----Original Message----- > From: compiler-dev-bounces at openjdk.java.net [mailto:compiler-dev- > bounces at openjdk.java.net] On Behalf Of Jonathan Gibbons > Sent: Wednesday, October 03, 2007 9:19 AM > To: compiler-dev at openjdk.java.net > Subject: Mac usage > > OK, here's a question I probably shouldn't ask because I'm not sure I > want to know the answer ... > > OpenJDK is currently targetted at the Solaris, Windows and Linux > platforms. Mac is not a supported platform for OpenJDK. But the > langtools component is pure Java and has had a somewhat more liberal > notion of platform. > > But, in trying out langtools on the Mac, I've run into some issues, > such as the different organization Apple's Java product, most > recently, the names of the subdirectories under ${JAVA_HOME}. > > So, I'm wondering how much interest/demand there is for working on > the compiler on a Mac platform. I know there are folk out there that > like the Mac -- I'm beginning to like it too, apart from hiccups like > this. Depending on the response, I'll be able to gauge the priority > of fixes for the issues that arise. > > -- Jon > > No virus found in this incoming message. > Checked by AVG Free Edition. > Version: 7.5.488 / Virus Database: 269.13.39/1045 - Release Date: > 10/2/2007 6:43 PM > No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.488 / Virus Database: 269.14.0/1046 - Release Date: 10/3/2007 10:08 AM From alexdmiller at yahoo.com Wed Oct 3 11:23:02 2007 From: alexdmiller at yahoo.com (Alex Miller) Date: Wed, 3 Oct 2007 11:23:02 -0700 (PDT) Subject: Mac usage Message-ID: <59801.13485.qm@web32101.mail.mud.yahoo.com> Would definitely be nice to have things work on the Mac if possible. Alex Miller http://tech.puredanger.com ----- Original Message ---- From: Ted Neward To: Jonathan Gibbons ; compiler-dev at openjdk.java.net Sent: Wednesday, October 3, 2007 1:06:29 PM Subject: RE: Mac usage Most of the speakers on the No Fluff Just Stuff tour use Macs; I'd say that supporting the Mac is, for us, a high priority. Having said that, I don't want to speak for them; I'll forward this on to them and let them sing out to you directly. Ted Neward Java, .NET, XML Services Consulting, Teaching, Speaking, Writing http://www.tedneward.com > -----Original Message----- > From: compiler-dev-bounces at openjdk.java.net [mailto:compiler-dev- > bounces at openjdk.java.net] On Behalf Of Jonathan Gibbons > Sent: Wednesday, October 03, 2007 9:19 AM > To: compiler-dev at openjdk.java.net > Subject: Mac usage > > OK, here's a question I probably shouldn't ask because I'm not sure I > want to know the answer ... > > OpenJDK is currently targetted at the Solaris, Windows and Linux > platforms. Mac is not a supported platform for OpenJDK. But the > langtools component is pure Java and has had a somewhat more liberal > notion of platform. > > But, in trying out langtools on the Mac, I've run into some issues, > such as the different organization Apple's Java product, most > recently, the names of the subdirectories under ${JAVA_HOME}. > > So, I'm wondering how much interest/demand there is for working on > the compiler on a Mac platform. I know there are folk out there that > like the Mac -- I'm beginning to like it too, apart from hiccups like > this. Depending on the response, I'll be able to gauge the priority > of fixes for the issues that arise. > > -- Jon > > No virus found in this incoming message. > Checked by AVG Free Edition. > Version: 7.5.488 / Virus Database: 269.13.39/1045 - Release Date: > 10/2/2007 6:43 PM > No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.488 / Virus Database: 269.14.0/1046 - Release Date: 10/3/2007 10:08 AM From robilad at kaffe.org Wed Oct 3 11:33:18 2007 From: robilad at kaffe.org (Dalibor Topic) Date: Wed, 03 Oct 2007 20:33:18 +0200 Subject: Mac usage In-Reply-To: <98784D7E-5288-4C11-9F3A-BFCE5BF9B6EF@sun.com> References: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> <4703C3F5.8060000@kaffe.org> <98784D7E-5288-4C11-9F3A-BFCE5BF9B6EF@sun.com> Message-ID: <4703E06E.8040009@kaffe.org> Jonathan Gibbons wrote: > To be clear, the build itself already works on a Mac. > > The problem right now is jtreg, which thinks there is a "bin/" directory. > Unfortunately, jtreg is not (yet) open source (we're working on it!) so > I can't > solicit contributions, but fortunately, I know the owner of jtreg quite > well ... Going a bit off-topic here: do you have some kind of vague ETA on jtreg's source code? cheers, dalibor topic From pdoubleya at gmail.com Wed Oct 3 11:34:39 2007 From: pdoubleya at gmail.com (Patrick Wright) Date: Wed, 3 Oct 2007 20:34:39 +0200 Subject: Mac usage In-Reply-To: References: Message-ID: <64efa1ba0710031134i312e7048lb42b5864aca9875d@mail.gmail.com> +1 from me, even if it just means compiler support to start with. The long release cycle from Apple is a killer, I have a MacBook with dual boot (and a VMWare image) with Ubuntu just to work with 6.0 features. Patrick From Jonathan.Gibbons at Sun.COM Wed Oct 3 11:44:10 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Wed, 03 Oct 2007 11:44:10 -0700 Subject: Mac usage In-Reply-To: <64efa1ba0710031134i312e7048lb42b5864aca9875d@mail.gmail.com> References: <64efa1ba0710031134i312e7048lb42b5864aca9875d@mail.gmail.com> Message-ID: <4703E2FA.5010006@sun.com> You can download and build the compiler as a jar file even now. Download the langtools code, open the NetBeans compiler project and build it. You can use Apple's Java 5 to build the OpenJDK compiler. Note that this does not give you the full native launcher, which still lives in and is built by the main JDK build. But you can run this compiler in a number of ways, such as "java -jar javac.jar", or via a small shell script that emulates javac, or by putting the compiler jar file on your JVM bootclasspath. However, to bring this back on topic, I was asking about the desire to use a Mac from those developers who might be interested in _working_ on the compiler, as compared to just using it. -- Jon Patrick Wright wrote: > +1 from me, even if it just means compiler support to start with. The > long release cycle from Apple is a killer, I have a MacBook with dual > boot (and a VMWare image) with Ubuntu just to work with 6.0 features. > > > Patrick > From Jonathan.Gibbons at Sun.COM Wed Oct 3 11:51:19 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Wed, 03 Oct 2007 11:51:19 -0700 Subject: Mac usage In-Reply-To: <4703E06E.8040009@kaffe.org> References: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> <4703C3F5.8060000@kaffe.org> <98784D7E-5288-4C11-9F3A-BFCE5BF9B6EF@sun.com> <4703E06E.8040009@kaffe.org> Message-ID: <4703E4A7.6040505@sun.com> Given we don't have a jtreg-dev alias (yet?) I'll answer your question here. The code is basically ready to go. We are just working through the internal processes to get the paperwork done. So, on the one hand, I'd say, "real soon now", and on the other, I'd caution you to be patient, as would anyone who knows what it's like dealing with process. For what it's worth, you can get some/most of the code even now, since jtreg is just a set of extensions to the JavaTest harness, also known as JT Harness. This part of the code is already available at http://jtharness.dev.java.net/ -- Jon Dalibor Topic wrote: > Jonathan Gibbons wrote: > >> To be clear, the build itself already works on a Mac. >> >> The problem right now is jtreg, which thinks there is a "bin/" directory. >> Unfortunately, jtreg is not (yet) open source (we're working on it!) so >> I can't >> solicit contributions, but fortunately, I know the owner of jtreg quite >> well ... >> > > Going a bit off-topic here: do you have some kind of vague ETA on > jtreg's source code? > > cheers, > dalibor topic > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071003/e37f671c/attachment.html From alexdmiller at yahoo.com Wed Oct 3 11:56:49 2007 From: alexdmiller at yahoo.com (Alex Miller) Date: Wed, 3 Oct 2007 11:56:49 -0700 (PDT) Subject: Mac usage Message-ID: <221870.45115.qm@web32114.mail.mud.yahoo.com> Yes, interested in working on it. Isn't that the intent of this compiler-**dev** list? :) Alex Miller http://tech.puredanger.com ----- Original Message ---- From: Jonathan Gibbons To: Patrick Wright Cc: compiler-dev at openjdk.java.net Sent: Wednesday, October 3, 2007 1:44:10 PM Subject: Re: Mac usage You can download and build the compiler as a jar file even now. Download the langtools code, open the NetBeans compiler project and build it. You can use Apple's Java 5 to build the OpenJDK compiler. Note that this does not give you the full native launcher, which still lives in and is built by the main JDK build. But you can run this compiler in a number of ways, such as "java -jar javac.jar", or via a small shell script that emulates javac, or by putting the compiler jar file on your JVM bootclasspath. However, to bring this back on topic, I was asking about the desire to use a Mac from those developers who might be interested in _working_ on the compiler, as compared to just using it. -- Jon Patrick Wright wrote: > +1 from me, even if it just means compiler support to start with. The > long release cycle from Apple is a killer, I have a MacBook with dual > boot (and a VMWare image) with Ubuntu just to work with 6.0 features. > > > Patrick > From robilad at kaffe.org Wed Oct 3 12:09:12 2007 From: robilad at kaffe.org (Dalibor Topic) Date: Wed, 03 Oct 2007 21:09:12 +0200 Subject: Mac usage In-Reply-To: <4703E4A7.6040505@sun.com> References: <16D1AD0C-4FA0-423B-B651-937B28184A77@solidwaretechnologies.com> <4703C3F5.8060000@kaffe.org> <98784D7E-5288-4C11-9F3A-BFCE5BF9B6EF@sun.com> <4703E06E.8040009@kaffe.org> <4703E4A7.6040505@sun.com> Message-ID: <4703E8D8.2030008@kaffe.org> Jonathan Gibbons wrote: > Given we don't have a jtreg-dev alias (yet?) I'll answer your question > here. > The code is basically ready to go. We are just working through the internal > processes to get the paperwork done. So, on the one hand, I'd say, > "real soon now", and on the other, I'd caution you to be patient, as > would anyone > who knows what it's like dealing with process. > Thanks Jon, that's great news! cheers, dalibor topic From charleskailunglee at gmail.com Wed Oct 3 12:10:38 2007 From: charleskailunglee at gmail.com (Charles Lee) Date: Wed, 3 Oct 2007 20:10:38 +0100 Subject: Mac usage In-Reply-To: <221870.45115.qm@web32114.mail.mud.yahoo.com> References: <221870.45115.qm@web32114.mail.mud.yahoo.com> Message-ID: Yea so +1 for me too... I run mac for everything i do. is a nice thought that I can code to the latest with a awesome OS Thanks Charlie On 3 Oct 2007, at 19:56, Alex Miller wrote: > Yes, interested in working on it. Isn't that the intent of this > compiler-**dev** list? :) > > Alex Miller > http://tech.puredanger.com > > ----- Original Message ---- > From: Jonathan Gibbons > To: Patrick Wright > Cc: compiler-dev at openjdk.java.net > Sent: Wednesday, October 3, 2007 1:44:10 PM > Subject: Re: Mac usage > > > You can download and build the compiler as a jar file even now. > Download the > langtools code, open the NetBeans compiler project and build it. You > can use > Apple's Java 5 to build the OpenJDK compiler. > > Note that this does not give you the full native launcher, which still > lives in and > is built by the main JDK build. But you can run this compiler in a > number of ways, > such as "java -jar javac.jar", or via a small shell script that > emulates > javac, or > by putting the compiler jar file on your JVM bootclasspath. > > However, to bring this back on topic, I was asking about the desire to > use a Mac > from those developers who might be interested in _working_ on the > compiler, as > compared to just using it. > > -- Jon > > > > > > > > Patrick Wright wrote: >> +1 from me, even if it just means compiler support to start with. The >> long release cycle from Apple is a killer, I have a MacBook with dual >> boot (and a VMWare image) with Ubuntu just to work with 6.0 features. >> >> >> Patrick >> > > > > From jesse.glick at sun.com Wed Oct 3 14:10:02 2007 From: jesse.glick at sun.com (Jesse Glick) Date: Wed, 03 Oct 2007 17:10:02 -0400 Subject: replace assert by throw new AssertionError In-Reply-To: <4702DF84.9050309@sun.com> References: <4702DF84.9050309@sun.com> Message-ID: Jonathan Gibbons wrote: > javac currently contains assertion statements, and goes out of its > way to ensure assertions are enabled. Why? Wouldn't non-javac developers just want it to run as fast as possible and trust that it is not buggy? You can always use -J-ea to enable assertions if you want to be sure, right? -J. -- jesse.glick at sun.com netbeans.org ant.apache.org hudson.dev.java.net http://google.com/search?q=e%5E%28pi*i%29%2B1 From forax at univ-mlv.fr Wed Oct 3 15:19:18 2007 From: forax at univ-mlv.fr (=?UTF-8?B?UsOpbWkgRm9yYXg=?=) Date: Thu, 04 Oct 2007 00:19:18 +0200 Subject: replace assert by throw new AssertionError In-Reply-To: References: <4702DF84.9050309@sun.com> Message-ID: <47041566.2080104@univ-mlv.fr> Jesse Glick a ?crit : > Jonathan Gibbons wrote: >> javac currently contains assertion statements, and goes out of its >> way to ensure assertions are enabled. > > Why? Wouldn't non-javac developers just want it to run as fast as > possible and trust that it is not buggy? You can always use -J-ea to > enable assertions if you want to be sure, right? jesse, it seems that currently the compiler doesn't rely on the command line and directly calls ClassLoader.setPackageAssertionStatus(). > > -J. > R?mi From peter at ahe.dk Wed Oct 3 15:25:00 2007 From: peter at ahe.dk (=?ISO-8859-1?Q?Peter_Ah=E9?=) Date: Wed, 3 Oct 2007 15:25:00 -0700 Subject: replace assert by throw new AssertionError In-Reply-To: References: <4702DF84.9050309@sun.com> Message-ID: Some parts of javac use this (in my opinion, very bad) coding idiom: assert false; When javac is launched from elsewhere than the command-line, without assertions enabled, it may skip happily over this and compile a piece of text that does not represent a program written in the Java programming language. This is a conformance violation which is bad. Even worse is that the compiler may seem to work but generate bad class files. Even when assertions are enabled, the user will get a lousy error message and have no clue how to change the source code to get past it. That is almost as great a sins as producing bad class files. Ideally, all assertions should be turned into explicit exceptions that included all the details needed to give a proper diagnostic at a higher level. What I'm saying is that this is much better: Bug.java:14: internal error; please report this bug at http://java.sun.com/webapps/bugreport s(t); ^ Than this: An exception has occurred in the compiler (1.5.0_02). Please file a bug at the Java Developer Connection (http://java.sun.com/webapps/bugreport) after checking the Bug Parade for duplicates. Include your program and the following diagnostic in your report. Thank you. java.lang.AssertionError at com.sun.tools.javac.comp.Lower.access(Lower.java:958) at com.sun.tools.javac.comp.Lower.access(Lower.java:1051) at com.sun.tools.javac.comp.Lower.makeOwnerThis(Lower.java:1382) at com.sun.tools.javac.comp.Lower.accessBase(Lower.java:877) at com.sun.tools.javac.comp.Lower.access(Lower.java:1037) at com.sun.tools.javac.comp.Lower.visitIdent(Lower.java:2762) at com.sun.tools.javac.tree.Tree$Ident.accept(Tree.java:1009) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitApply(Lower.java:2487) at com.sun.tools.javac.tree.Tree$Apply.accept(Tree.java:819) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.tree.TreeTranslator.visitExec(TreeTranslator.java:227) at com.sun.tools.javac.tree.Tree$Exec.accept(Tree.java:734) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.tree.TreeTranslator.translate(TreeTranslator.java:54) at com.sun.tools.javac.tree.TreeTranslator.visitBlock(TreeTranslator.java:145) at com.sun.tools.javac.comp.Lower.visitBlock(Lower.java:2944) at com.sun.tools.javac.tree.Tree$Block.accept(Tree.java:540) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.tree.TreeTranslator.visitMethodDef(TreeTranslator.java:129) at com.sun.tools.javac.comp.Lower.visitMethodDefInternal(Lower.java:2281) at com.sun.tools.javac.comp.Lower.visitMethodDef(Lower.java:2200) at com.sun.tools.javac.tree.Tree$MethodDef.accept(Tree.java:482) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitClassDef(Lower.java:2003) at com.sun.tools.javac.tree.Tree$ClassDef.accept(Tree.java:438) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitNewClass(Lower.java:2340) at com.sun.tools.javac.tree.Tree$NewClass.accept(Tree.java:845) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1907) at com.sun.tools.javac.comp.Lower.boxArgs(Lower.java:2530) at com.sun.tools.javac.comp.Lower.visitApply(Lower.java:2433) at com.sun.tools.javac.tree.Tree$Apply.accept(Tree.java:819) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.tree.TreeTranslator.visitExec(TreeTranslator.java:227) at com.sun.tools.javac.tree.Tree$Exec.accept(Tree.java:734) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitMethodDefInternal(Lower.java:2244) at com.sun.tools.javac.comp.Lower.visitMethodDef(Lower.java:2200) at com.sun.tools.javac.tree.Tree$MethodDef.accept(Tree.java:482) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitClassDef(Lower.java:2003) at com.sun.tools.javac.tree.Tree$ClassDef.accept(Tree.java:438) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.visitClassDef(Lower.java:2003) at com.sun.tools.javac.tree.Tree$ClassDef.accept(Tree.java:438) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1895) at com.sun.tools.javac.comp.Lower.translate(Lower.java:1915) at com.sun.tools.javac.comp.Lower.translateTopLevelClass(Lower.java:3081) at com.sun.tools.javac.main.JavaCompiler.compile(JavaCompiler.java:496) at com.sun.tools.javac.main.Main.compile(Main.java:592) at com.sun.tools.javac.main.Main.compile(Main.java:544) at com.sun.tools.javac.Main.compile(Main.java:67) at com.sun.tools.javac.Main.main(Main.java:52) Cheers, Peter On 10/3/07, Jesse Glick wrote: > Jonathan Gibbons wrote: > > javac currently contains assertion statements, and goes out of its > > way to ensure assertions are enabled. > > Why? Wouldn't non-javac developers just want it to run as fast as > possible and trust that it is not buggy? You can always use -J-ea to > enable assertions if you want to be sure, right? > > -J. > > -- > jesse.glick at sun.com netbeans.org ant.apache.org hudson.dev.java.net > http://google.com/search?q=e%5E%28pi*i%29%2B1 > > From neal at gafter.com Fri Oct 5 11:59:17 2007 From: neal at gafter.com (Neal Gafter) Date: Fri, 5 Oct 2007 11:59:17 -0700 Subject: Mac usage In-Reply-To: References: Message-ID: <15e8b9d20710051159v2f373ec7ue826d087332f16f7@mail.gmail.com> On 10/3/07, Jonathan Gibbons wrote: > > So, I'm wondering how much interest/demand there is for working on > the compiler on a Mac platform. I know there are folk out there that > like the Mac -- I'm beginning to like it too, apart from hiccups like > this. Depending on the response, I'll be able to gauge the priority > of fixes for the issues that arise. I use the Mac exclusively for my javac development. I'd really like to be able to run the regression tests before distributing fixes or prototype language features. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071005/7c1a265f/attachment.html From neal at gafter.com Fri Oct 5 12:02:36 2007 From: neal at gafter.com (Neal Gafter) Date: Fri, 5 Oct 2007 12:02:36 -0700 Subject: Bug in switch around Enums? In-Reply-To: <824b0d840709291142p7be9f836y899d7d0fe8ef9d38@mail.gmail.com> References: <05fb01c8027f$658ab370$30a01a50$@com> <22ec15240709290315g44e8136dx437e5b6f4b1b496b@mail.gmail.com> <46FE6AE0.7020108@sun.com> <824b0d840709291142p7be9f836y899d7d0fe8ef9d38@mail.gmail.com> Message-ID: <15e8b9d20710051202y11edb3bbi6cacc11fd070a154@mail.gmail.com> On 9/29/07, Brian Slesinsky wrote: > > Sure, support for constant expressions that return an enum seems like a > win. > > It would also solve a different gotcha. This doesn't work: > > void doSwitch(MyEnum x) { > switch (x) { > case MyEnum.FOO: > // do something > } > } I'd be interested to see what modifications to the name lookup rules you propose. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071005/f896718d/attachment.html From John.Rose at Sun.COM Fri Oct 5 14:02:06 2007 From: John.Rose at Sun.COM (John Rose) Date: Fri, 05 Oct 2007 14:02:06 -0700 Subject: Mac usage In-Reply-To: <64efa1ba0710031134i312e7048lb42b5864aca9875d@mail.gmail.com> References: <64efa1ba0710031134i312e7048lb42b5864aca9875d@mail.gmail.com> Message-ID: On Oct 3, 2007, at 11:34 AM, Patrick Wright wrote: > +1 from me, even if it just means compiler support to start with. The > long release cycle from Apple is a killer, I have a MacBook with dual > boot (and a VMWare image) with Ubuntu just to work with 6.0 features. I use a VMWare Solaris image for similar reasons. (See http://wikis.sun.com/display/SunOnMac/Home for details.) But I'd prefer to do my pure-java development work on the native (host) Mac OS X system, running Netbeans or some such. -- John From andras.gerlits at gmail.com Sat Oct 13 08:26:49 2007 From: andras.gerlits at gmail.com (Andras Gerlits) Date: Sat, 13 Oct 2007 16:26:49 +0100 Subject: Introduction and RFC Message-ID: Hi all, I'm new to the list, so I'd like to say hello first. I am a developer working almost exclusively with Java and related technologies. The reason I joined this list is that I would like some opinions on the following proposal (some first-level review by folks much knowledgable about the java compiler than myself): The following examples demonstrate the usage of a new, generic wildcard which I always thought was a missing feature of the generics framework and which would allow _some_ basic level of polymorphism when programming with generics. The wildcard is represented by a tilde '~', meaning the first encapsulating class of the instance implementing the method. It might only be allowed for top-level classes, (for the sake of simplicity) but all these are just ideas for which I am fishing for thoughts. /** * ~ stands as a generic wildcard for the class it was invoked on. * */ public abstract class Animal implements Cloneable { protected static final int NUMBER_OF_DESCENDENTS = 10; public List<~> children() { //create list for(NUMBER_OF_DESCENDENTS) { list.add(this.clone()); } //return list } } public class Dog extends Animal { //specific implementation, does not override children } Any thoughts would be appreciated. Thanks, Andras Gerlits Valid examples of usage: //Any code using API Dog dog = new Dog(); List puppies = dog.children(); Animal dogButDeclaredAsAnimal = new Dog(); List babies = dogButDeclaredAsAnimal.children(); Invalid example of usage: Animal dogButDeclaredAsAnimal = new Dog(); List babies = dogButDeclaredAsAnimal.children(); From peter at ahe.dk Mon Oct 15 12:47:06 2007 From: peter at ahe.dk (=?ISO-8859-1?Q?Peter_Ah=E9?=) Date: Mon, 15 Oct 2007 12:47:06 -0700 Subject: Introduction and RFC In-Reply-To: References: Message-ID: On 10/13/07, Andras Gerlits wrote: > The reason I joined this list is that I would like some opinions on > the following proposal (some first-level review by folks much > knowledgable about the java compiler than myself): Unfortunately, that is not the purpose of this list. If you *implement* your feature as an extension of javac, then your message could be relevant for this list. In order for me not to be a complete bozo about your off-topic message, allow me to redirect your attention to this: http://blogs.sun.com/ahe/entry/selftypes The page includes a link to an RFE for this, and the comment section of that particular bug is an appropriate forum for discussing this subject further. Cheers, Peter From Kelly.Ohair at Sun.COM Mon Oct 15 10:02:45 2007 From: Kelly.Ohair at Sun.COM (Kelly O'Hair) Date: Mon, 15 Oct 2007 10:02:45 -0700 Subject: [Fwd: Patch to build workspace langtools on Mac OS X 10.4 for b22] Message-ID: <47139D35.6070508@sun.com> Forwarding to javac alias. -kto -------- Original Message -------- Subject: Patch to build workspace langtools on Mac OS X 10.4 for b22 Date: Sun, 14 Oct 2007 23:35:16 -0700 (PDT) From: David Shao To: build-dev at openjdk.java.net Attached is a trivial patch to build.xml to get workspace langtools for b22 to at least compile on Mac OS X 10.4, using the default Java 1.5 and using apache-ant 1.7.0 from MacPorts. On my system, the check for the bootstrap java would fail. I am a novice at Ant, but from my reading of the manual, I changed the usage of task available so that if the file on the path created from concatenating property and marker was found, a marker property would be set, then as was done with the property test of macrodef check, I simply check for whether the marker property was not set to decide whether to fail. My description is longer than the actual changes. I do not know if using task available with no property set within a conditional is an Ant idiom that is supposed to work, but that is all that prevented initial building for me. I have not tested the actual functionality of the built langtools. ____________________________________________________________________________________ Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC -------------- next part -------------- A non-text attachment was scrubbed... Name: openjdk_b22.diff Type: application/octet-stream Size: 834 bytes Desc: not available Url : http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071015/202a43f6/openjdk_b22.diff From eranyar at gmail.com Thu Oct 18 05:16:40 2007 From: eranyar at gmail.com (Eran Yarkon) Date: Thu, 18 Oct 2007 14:16:40 +0200 Subject: Learning the javac compiler Message-ID: Hi, I need to apply several changes to javac and would like to learn the compiler beforehand. What is the best way to do it? Is there any kind of tutorial or user manual or should I read the documentation (where is it located?) Thanks, Keren. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071018/9cff4e36/attachment.html From forax at univ-mlv.fr Thu Oct 18 05:45:11 2007 From: forax at univ-mlv.fr (=?ISO-8859-1?Q?R=E9mi_Forax?=) Date: Thu, 18 Oct 2007 14:45:11 +0200 Subject: Learning the javac compiler In-Reply-To: References: Message-ID: <47175557.1000505@univ-mlv.fr> Eran Yarkon a ?crit : > Hi, > > I need to apply several changes to javac and would like to learn the > compiler beforehand. What is the best way to do it? > Is there any kind of tutorial or user manual or should I read the > documentation (where is it located?) > > Thanks, > Keren. There are not much info, some are gather in this page: http://openjdk.java.net/groups/compiler/ There are small tutorials: http://www.ahristov.com/tutorial/java-compiler.html But it's not sufficient, so you have to checkout the source :( the SVN repository is located here: https://openjdk.dev.java.net/svn/openjdk/jdk/trunk/langtools If the changes you have planed to introduce can interrest other people, you provide it as a branch of the kitchen sink language (after code review). https://ksl.dev.java.net/ cheers, R?mi PS: ping jonathan, please, update the ksl source tree to use langtools workspace. From ted at tedneward.com Thu Oct 25 00:32:32 2007 From: ted at tedneward.com (Ted Neward) Date: Thu, 25 Oct 2007 00:32:32 -0700 Subject: Introduction and RFC In-Reply-To: References: Message-ID: <00f501c816d9$3657b0d0$a3071270$@com> Interesting--which list *would* be the proper place to discuss language proposals? Is there one? (Personally, I thought it was this one.) Along those same lines, is there a reason for the segregation? Is there really that much of a semantic difference between the two topics that we would want to keep those two topics distinct and apart? (I'm not criticizing, just trying to understand the thought process.) Ted Neward Java, .NET, XML Services Consulting, Teaching, Speaking, Writing http://www.tedneward.com > -----Original Message----- > From: compiler-dev-bounces at openjdk.java.net [mailto:compiler-dev- > bounces at openjdk.java.net] On Behalf Of Peter Ah? > Sent: Monday, October 15, 2007 12:47 PM > To: Andras Gerlits > Cc: compiler-dev at openjdk.java.net > Subject: Re: Introduction and RFC > > On 10/13/07, Andras Gerlits wrote: > > The reason I joined this list is that I would like some opinions on > > the following proposal (some first-level review by folks much > > knowledgable about the java compiler than myself): > > Unfortunately, that is not the purpose of this list. If you > *implement* your feature as an extension of javac, then your message > could be relevant for this list. > > In order for me not to be a complete bozo about your off-topic > message, allow me to redirect your attention to this: > http://blogs.sun.com/ahe/entry/selftypes > The page includes a link to an RFE for this, and the comment section > of that particular bug is an appropriate forum for discussing this > subject further. > > Cheers, > Peter > > No virus found in this incoming message. > Checked by AVG Free Edition. > Version: 7.5.488 / Virus Database: 269.14.11/1071 - Release Date: > 10/15/2007 6:48 AM > No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.503 / Virus Database: 269.15.10/1091 - Release Date: 10/24/2007 2:31 PM From peter at ahe.dk Thu Oct 25 09:25:33 2007 From: peter at ahe.dk (=?ISO-8859-1?Q?Peter_Ah=E9?=) Date: Thu, 25 Oct 2007 09:25:33 -0700 Subject: Introduction and RFC In-Reply-To: <00f501c816d9$3657b0d0$a3071270$@com> References: <00f501c816d9$3657b0d0$a3071270$@com> Message-ID: Ted, I don't see how you ever reached the conclusion that this list was for discussion of language features. Both Jon and I have been clear about this from the start. We open sourced javac (and the JDK implementation), not its specification. Unfortunately, most people who propose a language feature take no time to get themselves up to speed on the history, or the computer science behind language design. As a former Sun employee, I have experienced the inflow of, to put it mildly, below par language feature proposals. Sun receives a lot of these, and they are a huge distraction. When we decided to open the our mailing lists, one of our primary concerns was that it would drown in language feature proposals, counter proposals, and endless discussions about those proposals. If that were to happen, this list would be useless. You are free to setup a mailing list for discussions of language features; I would not be interested in subscribing, and I suspect many of my former colleagues feel the same way. Eventually, I hope that Sun will be able to open their bug database, in which case you should be able to subscribe directly to the java/specification category. For now you will have to make a Google Alert on "java/specification site:bugs.sun.com". Cheers, Peter On 10/25/07, Ted Neward wrote: > Interesting--which list *would* be the proper place to discuss language > proposals? Is there one? (Personally, I thought it was this one.) > > Along those same lines, is there a reason for the segregation? Is there > really that much of a semantic difference between the two topics that we > would want to keep those two topics distinct and apart? (I'm not > criticizing, just trying to understand the thought process.) > > Ted Neward > Java, .NET, XML Services > Consulting, Teaching, Speaking, Writing > http://www.tedneward.com > > > > -----Original Message----- > > From: compiler-dev-bounces at openjdk.java.net [mailto:compiler-dev- > > bounces at openjdk.java.net] On Behalf Of Peter Ah? > > Sent: Monday, October 15, 2007 12:47 PM > > To: Andras Gerlits > > Cc: compiler-dev at openjdk.java.net > > Subject: Re: Introduction and RFC > > > > On 10/13/07, Andras Gerlits wrote: > > > The reason I joined this list is that I would like some opinions on > > > the following proposal (some first-level review by folks much > > > knowledgable about the java compiler than myself): > > > > Unfortunately, that is not the purpose of this list. If you > > *implement* your feature as an extension of javac, then your message > > could be relevant for this list. > > > > In order for me not to be a complete bozo about your off-topic > > message, allow me to redirect your attention to this: > > http://blogs.sun.com/ahe/entry/selftypes > > The page includes a link to an RFE for this, and the comment section > > of that particular bug is an appropriate forum for discussing this > > subject further. > > > > Cheers, > > Peter > > > > No virus found in this incoming message. > > Checked by AVG Free Edition. > > Version: 7.5.488 / Virus Database: 269.14.11/1071 - Release Date: > > 10/15/2007 6:48 AM > > > > No virus found in this outgoing message. > Checked by AVG Free Edition. > Version: 7.5.503 / Virus Database: 269.15.10/1091 - Release Date: 10/24/2007 > 2:31 PM > > > From robilad at kaffe.org Thu Oct 25 09:47:37 2007 From: robilad at kaffe.org (Dalibor Topic) Date: Thu, 25 Oct 2007 18:47:37 +0200 Subject: Introduction and RFC In-Reply-To: <00f501c816d9$3657b0d0$a3071270$@com> References: <00f501c816d9$3657b0d0$a3071270$@com> Message-ID: <4720C8A9.1090809@kaffe.org> Ted Neward wrote: > Interesting--which list *would* be the proper place to discuss language > proposals? Is there one? (Personally, I thought it was this one.) > I think the kitchen sink language project is the venue you are looking for: https://ksl.dev.java.net/ cheers, dalibor topic From Jonathan.Gibbons at Sun.COM Thu Oct 25 13:45:01 2007 From: Jonathan.Gibbons at Sun.COM (Jonathan Gibbons) Date: Thu, 25 Oct 2007 13:45:01 -0700 Subject: KSL updated to OpenJDK langtools b22 Message-ID: <4721004D.6030100@sun.com> Team, I've updated KSL with OpenJDK langtools b22: i.e. KSL now contains all of OpenJDK langtools instead of (just) the compiler. As such, it contains javadoc, javah, javap, etc. If you consider adding new features to the language, you might want to consider updating javadoc and other tools as well. (See Joe's blog here: http://blogs.sun.com/darcy/entry/so_you_want_to_change ). I've updated the sun/ branch and merged the update into trunk/ as well. Lots of files got changed here: all of the SCCS keywords got removed from all the files, since we're about to convert to Mercurial and leave SCCS behind. And, FYI, the next update may be big as well, since the JDK team has been busy cleaning up whitespace, converting tabs to spaces, and removing trailing spaces as well. -- Jon From freds at jfrog.org Fri Oct 26 04:34:56 2007 From: freds at jfrog.org (Frederic Simon) Date: Fri, 26 Oct 2007 13:34:56 +0200 Subject: Introduction and RFC In-Reply-To: <4720C8A9.1090809@kaffe.org> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> Message-ID: I really thought that the KSL was created exactly for that: Filtering language proposals. But there are no mailing lists for KSL pure, and the Bug database should not be populated with KSL noise. About KSL, in my experience adding some code to javac (pure implementation) to support small language proposal, is a lot faster and cheaper than: - Evaluate the coherence/readability - Evaluate its usefulness (in writing code and API) - Evaluate its impact on current API - Evaluate the risk impact on the javac and JVM So, the KSL is great. Have it, play with it, throw it away (and "may be", "sometimes", "rarely", "occasionally": keep it). And it does not have to be Sun employees doing the steps. For me, once a language proposal and RFE entry starts to get momentum (votes and so on), so the team leaders as decided in the GB (Sun for the moment) can get involved and evaluate the next steps. The KSL for me is "Extreme Agility", and luxury of having the implementation before deciding if you need it or not. So, please keep the spirit of it, it's good for everyone. On 10/25/07, Dalibor Topic wrote: > > Ted Neward wrote: > > Interesting--which list *would* be the proper place to discuss language > > proposals? Is there one? (Personally, I thought it was this one.) > > > I think the kitchen sink language project is the venue you are looking > for: https://ksl.dev.java.net/ > > cheers, > dalibor topic > -- http://freddy33.bglogspot.com/ http://www.jfrog.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071026/dbf063d1/attachment.html From robilad at kaffe.org Fri Oct 26 06:21:01 2007 From: robilad at kaffe.org (Dalibor Topic) Date: Fri, 26 Oct 2007 15:21:01 +0200 Subject: Introduction and RFC In-Reply-To: References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> Message-ID: <4721E9BD.4060205@kaffe.org> Frederic Simon wrote: > I really thought that the KSL was created exactly for that: Filtering > language proposals. But there are no mailing lists for KSL pure, and > the Bug database should not be populated with KSL noise. I think that's https://ksl.dev.java.net/servlets/SummarizeList?listName=language the bug database is at https://ksl.dev.java.net/servlets/ProjectIssues cheers, dalibor topic From scolebourne at btopenworld.com Fri Oct 26 07:35:31 2007 From: scolebourne at btopenworld.com (Stephen Colebourne) Date: Fri, 26 Oct 2007 15:35:31 +0100 Subject: KSL [was Re: Introduction and RFC] In-Reply-To: References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> Message-ID: <4721FB33.50003@btopenworld.com> The KSL is great in theory, but it has yet to prove its value in practice. As far as I know, there have been no changes committed to KSL that separate it from javac. In addition, KSL has a high barrier of entry (in terms of fulfilling all the many things that compiler writers and language designers should do, including documentation, spec writing and tests). Of course, none of this is wrong, it just may signal that 'we the community' need to create another project separate from KSL where there is a much lower barrier of entry, but as a result a higher risk that the compiler/resulting code will be broken. ie. a place where ideas can actually be investigated. Stephen PS. This discussion about KSL probably should be on the KSL list, but thread-wise it makes sense here for now. PPS.I should also note that I originally signed up here as the KSL project webpages told me to do so, so the fact that there is now a KSL mailing list is rather a surprise. Frederic Simon wrote: > I really thought that the KSL was created exactly for that: Filtering > language proposals. But there are no mailing lists for KSL pure, and the > Bug database should not be populated with KSL noise. > About KSL, in my experience adding some code to javac (pure > implementation) to support small language proposal, is a lot faster and > cheaper than: > - Evaluate the coherence/readability > - Evaluate its usefulness (in writing code and API) > - Evaluate its impact on current API > - Evaluate the risk impact on the javac and JVM > > So, the KSL is great. Have it, play with it, throw it away (and "may > be", "sometimes", "rarely", "occasionally": keep it). And it does not > have to be Sun employees doing the steps. For me, once a language > proposal and RFE entry starts to get momentum (votes and so on), so the > team leaders as decided in the GB (Sun for the moment) can get involved > and evaluate the next steps. > > The KSL for me is "Extreme Agility", and luxury of having the > implementation before deciding if you need it or not. > > So, please keep the spirit of it, it's good for everyone. > > On 10/25/07, *Dalibor Topic* > wrote: > > Ted Neward wrote: > > Interesting--which list *would* be the proper place to discuss > language > > proposals? Is there one? (Personally, I thought it was this one.) > > > I think the kitchen sink language project is the venue you are looking > for: https://ksl.dev.java.net/ > > cheers, > dalibor topic > > > > > -- > http://freddy33.bglogspot.com/ > http://www.jfrog.org/ From Tom.Ball at Sun.COM Fri Oct 26 08:15:28 2007 From: Tom.Ball at Sun.COM (Tom Ball) Date: Fri, 26 Oct 2007 08:15:28 -0700 Subject: KSL [was Re: Introduction and RFC] In-Reply-To: <4721FB33.50003@btopenworld.com> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721FB33.50003@btopenworld.com> Message-ID: <47220490.8070206@sun.com> Stephen Colebourne wrote: > In addition, KSL has a high barrier of entry (in terms of fulfilling all > the many things that compiler writers and language designers should do, > including documentation, spec writing and tests). That's why I like it -- most software developers think they are world-class language designers, but the truth is only a very small minority are (and I'm not in that group). Features can be easily added to an untyped or loosely-typed language, but most additions to the Java language require a rigorous examination of their impact on issues like type-safety and security. Few people take these requirements very seriously when throwing out suggestions, but for those who do want the level of rigor that working with javac provides, KSL is the place to be. There are other aliases for people who just want to intellectually joust; personally, I think those sorts of discussions belong instead in a pub with beer for everyone. > Of course, none of this is wrong, it just may signal that 'we the > community' need to create another project separate from KSL where there > is a much lower barrier of entry, but as a result a higher risk that the > compiler/resulting code will be broken. ie. a place where ideas can > actually be investigated. My guess is that it would be easier to hack BeanShell or Groovy, as their type-safety requirements appear to be lower. I certainly could be wrong about that, of course. Tom > Stephen > PS. This discussion about KSL probably should be on the KSL list, but > thread-wise it makes sense here for now. > > PPS.I should also note that I originally signed up here as the KSL > project webpages told me to do so, so the fact that there is now a KSL > mailing list is rather a surprise. > > > Frederic Simon wrote: >> I really thought that the KSL was created exactly for that: Filtering >> language proposals. But there are no mailing lists for KSL pure, and >> the Bug database should not be populated with KSL noise. >> About KSL, in my experience adding some code to javac (pure >> implementation) to support small language proposal, is a lot faster >> and cheaper than: >> - Evaluate the coherence/readability >> - Evaluate its usefulness (in writing code and API) >> - Evaluate its impact on current API >> - Evaluate the risk impact on the javac and JVM >> >> So, the KSL is great. Have it, play with it, throw it away (and "may >> be", "sometimes", "rarely", "occasionally": keep it). And it does not >> have to be Sun employees doing the steps. For me, once a language >> proposal and RFE entry starts to get momentum (votes and so on), so >> the team leaders as decided in the GB (Sun for the moment) can get >> involved and evaluate the next steps. >> >> The KSL for me is "Extreme Agility", and luxury of having the >> implementation before deciding if you need it or not. >> >> So, please keep the spirit of it, it's good for everyone. >> >> On 10/25/07, *Dalibor Topic* > > wrote: >> >> Ted Neward wrote: >> > Interesting--which list *would* be the proper place to discuss >> language >> > proposals? Is there one? (Personally, I thought it was this one.) >> > >> I think the kitchen sink language project is the venue you are >> looking >> for: https://ksl.dev.java.net/ >> >> cheers, >> dalibor topic >> >> >> >> >> -- >> http://freddy33.bglogspot.com/ >> http://www.jfrog.org/ From freds at jfrog.org Fri Oct 26 14:33:47 2007 From: freds at jfrog.org (Frederic Simon) Date: Fri, 26 Oct 2007 23:33:47 +0200 Subject: Introduction and RFC In-Reply-To: <4721E9BD.4060205@kaffe.org> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721E9BD.4060205@kaffe.org> Message-ID: Thanks, I did not see this mailing list and bug report before (and I looked for it). So, now I hope they will start to be used for language feature instead of the compiler-dev mailing list. So, see you soon on language-help at ksl.dev.java.net Strange name, but let's do with it... On 10/26/07, Dalibor Topic wrote: > > Frederic Simon wrote: > > I really thought that the KSL was created exactly for that: Filtering > > language proposals. But there are no mailing lists for KSL pure, and > > the Bug database should not be populated with KSL noise. > I think that's > https://ksl.dev.java.net/servlets/SummarizeList?listName=language > > the bug database is at https://ksl.dev.java.net/servlets/ProjectIssues > > cheers, > dalibor topic > -- http://freddy33.bglogspot.com/ http://www.jfrog.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071026/982ce342/attachment.html From peter at ahe.dk Sun Oct 28 14:12:23 2007 From: peter at ahe.dk (=?ISO-8859-1?Q?Peter_Ah=E9?=) Date: Sun, 28 Oct 2007 14:12:23 -0700 Subject: Introduction and RFC In-Reply-To: <4721E9BD.4060205@kaffe.org> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721E9BD.4060205@kaffe.org> Message-ID: KSL was created as a place for people who do dip into the code of javac to modify its behavior. If modifying javac scares you off, then neither this list nor KSL is what you are looking for. Cheers, Peter On 10/26/07, Dalibor Topic wrote: > Frederic Simon wrote: > > I really thought that the KSL was created exactly for that: Filtering > > language proposals. But there are no mailing lists for KSL pure, and > > the Bug database should not be populated with KSL noise. > I think that's > https://ksl.dev.java.net/servlets/SummarizeList?listName=language > > the bug database is at https://ksl.dev.java.net/servlets/ProjectIssues > > cheers, > > dalibor topic > From neal at gafter.com Mon Oct 29 10:21:22 2007 From: neal at gafter.com (Neal Gafter) Date: Mon, 29 Oct 2007 10:21:22 -0700 Subject: KSL [was Re: Introduction and RFC] In-Reply-To: <4721FB33.50003@btopenworld.com> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721FB33.50003@btopenworld.com> Message-ID: <15e8b9d20710291021i7a35f880s25b5eb2c6a8da06a@mail.gmail.com> After the migration to mercurial is complete, I am considering creating an open-source project at openjavac.net as a feeder project into KSL, with a hopefully lower barrier to entry. The idea is to provide a centralized place to host mercurial patchsets for proposals until they're more cooked. On 10/26/07, Stephen Colebourne wrote: > > The KSL is great in theory, but it has yet to prove its value in > practice. As far as I know, there have been no changes committed to KSL > that separate it from javac. > > In addition, KSL has a high barrier of entry (in terms of fulfilling all > the many things that compiler writers and language designers should do, > including documentation, spec writing and tests). > > Of course, none of this is wrong, it just may signal that 'we the > community' need to create another project separate from KSL where there > is a much lower barrier of entry, but as a result a higher risk that the > compiler/resulting code will be broken. ie. a place where ideas can > actually be investigated. > > Stephen > PS. This discussion about KSL probably should be on the KSL list, but > thread-wise it makes sense here for now. > > PPS.I should also note that I originally signed up here as the KSL > project webpages told me to do so, so the fact that there is now a KSL > mailing list is rather a surprise. > > > Frederic Simon wrote: > > I really thought that the KSL was created exactly for that: Filtering > > language proposals. But there are no mailing lists for KSL pure, and the > > Bug database should not be populated with KSL noise. > > About KSL, in my experience adding some code to javac (pure > > implementation) to support small language proposal, is a lot faster and > > cheaper than: > > - Evaluate the coherence/readability > > - Evaluate its usefulness (in writing code and API) > > - Evaluate its impact on current API > > - Evaluate the risk impact on the javac and JVM > > > > So, the KSL is great. Have it, play with it, throw it away (and "may > > be", "sometimes", "rarely", "occasionally": keep it). And it does not > > have to be Sun employees doing the steps. For me, once a language > > proposal and RFE entry starts to get momentum (votes and so on), so the > > team leaders as decided in the GB (Sun for the moment) can get involved > > and evaluate the next steps. > > > > The KSL for me is "Extreme Agility", and luxury of having the > > implementation before deciding if you need it or not. > > > > So, please keep the spirit of it, it's good for everyone. > > > > On 10/25/07, *Dalibor Topic* > > wrote: > > > > Ted Neward wrote: > > > Interesting--which list *would* be the proper place to discuss > > language > > > proposals? Is there one? (Personally, I thought it was this one.) > > > > > I think the kitchen sink language project is the venue you are > looking > > for: https://ksl.dev.java.net/ > > > > cheers, > > dalibor topic > > > > > > > > > > -- > > http://freddy33.bglogspot.com/ > > http://www.jfrog.org/ > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20071029/21d03d9c/attachment.html From scolebourne at btopenworld.com Mon Oct 29 11:49:10 2007 From: scolebourne at btopenworld.com (Stephen Colebourne) Date: Mon, 29 Oct 2007 18:49:10 +0000 Subject: KSL [was Re: Introduction and RFC] In-Reply-To: <15e8b9d20710291021i7a35f880s25b5eb2c6a8da06a@mail.gmail.com> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721FB33.50003@btopenworld.com> <15e8b9d20710291021i7a35f880s25b5eb2c6a8da06a@mail.gmail.com> Message-ID: <47262B26.6040302@btopenworld.com> That sounds exactly like what I think we're missing ATM :-) Stephen Neal Gafter wrote: > After the migration to mercurial is complete, I am considering creating > an open-source project at openjavac.net as a > feeder project into KSL, with a hopefully lower barrier to entry. The > idea is to provide a centralized place to host mercurial patchsets for > proposals until they're more cooked. > > On 10/26/07, *Stephen Colebourne* > wrote: > > The KSL is great in theory, but it has yet to prove its value in > practice. As far as I know, there have been no changes committed to KSL > that separate it from javac. > > In addition, KSL has a high barrier of entry (in terms of fulfilling > all > the many things that compiler writers and language designers should do, > including documentation, spec writing and tests). > > Of course, none of this is wrong, it just may signal that 'we the > community' need to create another project separate from KSL where there > is a much lower barrier of entry, but as a result a higher risk that the > compiler/resulting code will be broken. ie. a place where ideas can > actually be investigated. > > Stephen > PS. This discussion about KSL probably should be on the KSL list, but > thread-wise it makes sense here for now. > > PPS.I should also note that I originally signed up here as the KSL > project webpages told me to do so, so the fact that there is now a KSL > mailing list is rather a surprise. > > > Frederic Simon wrote: > > I really thought that the KSL was created exactly for that: Filtering > > language proposals. But there are no mailing lists for KSL pure, > and the > > Bug database should not be populated with KSL noise. > > About KSL, in my experience adding some code to javac (pure > > implementation) to support small language proposal, is a lot > faster and > > cheaper than: > > - Evaluate the coherence/readability > > - Evaluate its usefulness (in writing code and API) > > - Evaluate its impact on current API > > - Evaluate the risk impact on the javac and JVM > > > > So, the KSL is great. Have it, play with it, throw it away (and "may > > be", "sometimes", "rarely", "occasionally": keep it). And it does > not > > have to be Sun employees doing the steps. For me, once a language > > proposal and RFE entry starts to get momentum (votes and so on), > so the > > team leaders as decided in the GB (Sun for the moment) can get > involved > > and evaluate the next steps. > > > > The KSL for me is "Extreme Agility", and luxury of having the > > implementation before deciding if you need it or not. > > > > So, please keep the spirit of it, it's good for everyone. > > > > On 10/25/07, *Dalibor Topic* > > >> wrote: > > > > Ted Neward wrote: > > > Interesting--which list *would* be the proper place to discuss > > language > > > proposals? Is there one? (Personally, I thought it was > this one.) > > > > > I think the kitchen sink language project is the venue you > are looking > > for: https://ksl.dev.java.net/ > > > > cheers, > > dalibor topic > > > > > > > > > > -- > > http://freddy33.bglogspot.com/ > > http://www.jfrog.org/ > > From peter.kitt.reilly at gmail.com Mon Oct 29 13:28:09 2007 From: peter.kitt.reilly at gmail.com (Peter Reilly) Date: Mon, 29 Oct 2007 20:28:09 +0000 Subject: KSL [was Re: Introduction and RFC] In-Reply-To: <47262B26.6040302@btopenworld.com> References: <00f501c816d9$3657b0d0$a3071270$@com> <4720C8A9.1090809@kaffe.org> <4721FB33.50003@btopenworld.com> <15e8b9d20710291021i7a35f880s25b5eb2c6a8da06a@mail.gmail.com> <47262B26.6040302@btopenworld.com> Message-ID: On 10/29/07, Stephen Colebourne wrote: > That sounds exactly like what I think we're missing ATM :-) > Stephen > > Neal Gafter wrote: > > After the migration to mercurial is complete, I am considering creating > > an open-source project at openjavac.net as a > > feeder project into KSL, with a hopefully lower barrier to entry. The > > idea is to provide a centralized place to host mercurial patchsets for > > proposals until they're more cooked. Awsome, http://gafter.blogspot.com/2007/10/java-closures-first-prototype.html will be part of this !? Peter > > > > On 10/26/07, *Stephen Colebourne* > > wrote: > > > > The KSL is great in theory, but it has yet to prove its value in > > practice. As far as I know, there have been no changes committed to KSL > > that separate it from javac. > > > > In addition, KSL has a high barrier of entry (in terms of fulfilling > > all > > the many things that compiler writers and language designers should do, > > including documentation, spec writing and tests). > > > > Of course, none of this is wrong, it just may signal that 'we the > > community' need to create another project separate from KSL where there > > is a much lower barrier of entry, but as a result a higher risk that the > > compiler/resulting code will be broken. ie. a place where ideas can > > actually be investigated. > > > > Stephen > > PS. This discussion about KSL probably should be on the KSL list, but > > thread-wise it makes sense here for now. > > > > PPS.I should also note that I originally signed up here as the KSL > > project webpages told me to do so, so the fact that there is now a KSL > > mailing list is rather a surprise. > > > > > > Frederic Simon wrote: > > > I really thought that the KSL was created exactly for that: Filtering > > > language proposals. But there are no mailing lists for KSL pure, > > and the > > > Bug database should not be populated with KSL noise. > > > About KSL, in my experience adding some code to javac (pure > > > implementation) to support small language proposal, is a lot > > faster and > > > cheaper than: > > > - Evaluate the coherence/readability > > > - Evaluate its usefulness (in writing code and API) > > > - Evaluate its impact on current API > > > - Evaluate the risk impact on the javac and JVM > > > > > > So, the KSL is great. Have it, play with it, throw it away (and "may > > > be", "sometimes", "rarely", "occasionally": keep it). And it does > > not > > > have to be Sun employees doing the steps. For me, once a language > > > proposal and RFE entry starts to get momentum (votes and so on), > > so the > > > team leaders as decided in the GB (Sun for the moment) can get > > involved > > > and evaluate the next steps. > > > > > > The KSL for me is "Extreme Agility", and luxury of having the > > > implementation before deciding if you need it or not. > > > > > > So, please keep the spirit of it, it's good for everyone. > > > > > > On 10/25/07, *Dalibor Topic* > > > > >> wrote: > > > > > > Ted Neward wrote: > > > > Interesting--which list *would* be the proper place to discuss > > > language > > > > proposals? Is there one? (Personally, I thought it was > > this one.) > > > > > > > I think the kitchen sink language project is the venue you > > are looking > > > for: https://ksl.dev.java.net/ > > > > > > cheers, > > > dalibor topic > > > > > > > > > > > > > > > -- > > > http://freddy33.bglogspot.com/ > > > http://www.jfrog.org/ > > > > >