You can subscribe to this list here.
2002 |
Jan
|
Feb
|
Mar
|
Apr
(39) |
May
(41) |
Jun
(20) |
Jul
(8) |
Aug
(15) |
Sep
(18) |
Oct
(15) |
Nov
(6) |
Dec
(48) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2003 |
Jan
(28) |
Feb
(6) |
Mar
(23) |
Apr
(13) |
May
(3) |
Jun
(12) |
Jul
(26) |
Aug
(1) |
Sep
|
Oct
(14) |
Nov
(2) |
Dec
(18) |
2004 |
Jan
(10) |
Feb
(5) |
Mar
(24) |
Apr
(7) |
May
(79) |
Jun
(25) |
Jul
|
Aug
(2) |
Sep
(11) |
Oct
(5) |
Nov
|
Dec
(1) |
2005 |
Jan
(1) |
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2006 |
Jan
(2) |
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2007 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
(1) |
Aug
|
Sep
(6) |
Oct
|
Nov
|
Dec
|
From: Ken A. <kan...@us...> - 2002-12-23 16:45:58
|
Update of /cvsroot/jscheme/jscheme/src/elf In directory sc8-pr-cvs1:/tmp/cvs-serv28243/src/elf Added Files: package.html Log Message: Add a package.html |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:33
|
Update of /cvsroot/jscheme/jscheme/src/dclass In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/dclass Modified Files: package.html Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:33
|
Update of /cvsroot/jscheme/jscheme/src/jscheme In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/jscheme Modified Files: Shebang.java Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:32
|
Update of /cvsroot/jscheme/jscheme/src/build In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/build Modified Files: jscheme-bootstrap.scm Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:32
|
Update of /cvsroot/jscheme/jscheme/src/elf In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/elf Modified Files: SchemeInvocationHandler.java Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:32
|
Update of /cvsroot/jscheme/jscheme/src/jschemeweb In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/jschemeweb Modified Files: SchemeServlet.java Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 16:43:32
|
Update of /cvsroot/jscheme/jscheme/src/jsint In directory sc8-pr-cvs1:/tmp/cvs-serv27506/src/jsint Modified Files: Pair.java Queue.java StaticReflector.java Log Message: Generate javadoc api documentation. |
From: Ken A. <kan...@us...> - 2002-12-23 15:06:04
|
Update of /cvsroot/jscheme/jscheme/src/build In directory sc8-pr-cvs1:/tmp/cvs-serv22548/src/build Modified Files: SchemeLite.java jscheme-bootstrap.scm Log Message: Have SchemeLite use the sourforge url for jscheme.jar. Have jscheme-bootstrap.scm make a jschemelite.jar. |
From: Ken A. <kan...@us...> - 2002-12-23 14:28:51
|
Update of /cvsroot/jscheme/jscheme/src/using In directory sc8-pr-cvs1:/tmp/cvs-serv9759/src/using Modified Files: run.scm Log Message: Change jscheme-bootstrap.scm to build jscheme using using/run.scm. |
From: Ken A. <kan...@us...> - 2002-12-23 14:28:51
|
Update of /cvsroot/jscheme/jscheme/src/jscheme In directory sc8-pr-cvs1:/tmp/cvs-serv9759/src/jscheme Modified Files: SchemeTests.scm Log Message: Change jscheme-bootstrap.scm to build jscheme using using/run.scm. |
From: Ken A. <kan...@us...> - 2002-12-23 14:28:51
|
Update of /cvsroot/jscheme/jscheme/src/jsint In directory sc8-pr-cvs1:/tmp/cvs-serv9759/src/jsint Modified Files: primitives.scm Log Message: Change jscheme-bootstrap.scm to build jscheme using using/run.scm. |
From: Ken A. <kan...@us...> - 2002-12-23 14:28:51
|
Update of /cvsroot/jscheme/jscheme/src/build In directory sc8-pr-cvs1:/tmp/cvs-serv9759/src/build Modified Files: bootstrap.scm compile.scm jscheme-bootstrap.scm Log Message: Change jscheme-bootstrap.scm to build jscheme using using/run.scm. |
From: Ken A. <kan...@bb...> - 2002-12-18 22:43:52
|
Just testing, sorry. I had a mail bounce yesterday. |
From: Ken A. <kan...@us...> - 2002-12-17 02:32:52
|
Update of /cvsroot/jscheme/jscheme/src/using In directory sc8-pr-cvs1:/tmp/cvs-serv21320 Modified Files: run.scm Log Message: Remove dependency on JDK 1.4 regular expressions. |
From: <da...@da...> - 2002-12-16 21:49:05
|
I am back at it again.. JScheme is a great way to explore the endless Java APIS. I just got some jscheme to play mp3s as well as transform XML, and load the DOM from a database.. So now I am trying figure out the simples way to display some images. It appears the normal java method is to override the paint() method. I looked at jlib but it did not mention images. What would be cool is to include some visual presenatations with the mp3s in a jar. Animation in jscheme?? |
From: Ken A. <kan...@bb...> - 2002-12-13 01:09:27
|
The discussion about ANT made me write some code that i've been meaning to write for some time. JScheme has a new using/ directory which i hope will contain several real examples of using JScheme in a Java application. The first example is run.scm which provides a mini language for invoking external Processes and doing things like building Java applications portably. Here a link: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/jscheme/jscheme/src/using/run.scm?rev=1.1&content-type=text/vnd.viewcvs-markup Actually, if you download the source it should be viewable in your browser as well as executable, though it is JDK 1.4 dependant, so it is its own Javadoc. It isn't, yet, as capable as ant, but with a little more work i think it should be able to build your Java application. Let me know what's missing. While JScheme builds itself now, i'm hoping it will use this stuff in the future. Its implementation and documentation (~ 350) is about a third larger than the number of ant classes (~ 260). The truth will be if you can get your application building in under two days, or at least tell me what additional capabilites you need. There are some loose ends. For example, (files* directory type?) will return the files in directory that satisfy type?. Currently, these files are absolute and there is (rfiles*) to get relative to the directory. You can also relativize files to a higher level directory. I've been away from UNIX long enough to not remember what guidance it might have in this regard. I also hack over a windows PATH issue to get to cygwin people might be to provide better solutions too. Don't get me wrong, ant fills an important roll of being a portable build facility for a portable language. It's just nice when tools are useful, and small (easy to use and understand). I didn't know about jmake, i'll check it out. One important difference between ant and make is that make has 2 languages, a constraint language and a shell language to implement the update when the constraint is old. So if you know the shell interface to the commands you want to use, you only need to learn the constraint language (oh ya, whitespace is significant!). In ant, constraints are expressed in XML, and the update i also expressed as XML so knowing the shell invocation doesn't help, you have to read the documentation. k At 04:43 AM 12/10/2002, Andrew McDermott wrote: >Ken Anderson <kan...@bb...> writes: > >[...] > >>>Also, is src/build/bootstrap the only way (right way) to build the code? >>>I had to compile BacktraceException.java by hand and then run bootstrap >>>to re-build the jar file, but that somehow feels wrong. Has the project >>>considered Ant (http://jakarta.apache.org/ant/) as a build tool? >> >> I'm going to let you tickle a hot button. I don't mean to offend you, >> or anyone who uses ant. I'd like to start an honest discussion. To >> my mind, ant represents a BIG answer to the problem of make, just as >> JAVACC, or Lex (is there a Java version?) provide a BIG answer to >> parsing and lexing. I'm interested in smaller answers that are almost >> as capable that i can remember enough of that i don't have to go to >> the documentation. This makes me feel like i have more tools in my >> pocket, which feels better to me than going out and buying another >> book and reading for a while before i get the code i need. > >I can understand this a lot. I held off using ant for our internal >build mechanism (but for far too long). The initial learning curve >appeared to be *just* too big. However, about 4 months ago I bit the >bullet, and you know what, it took me about two full days to get >everything done. Period. Since then adding incremental changes has >been really easy. I really wish I hadn't deliberated on the decision to >switch for so long. > >So, despite the learning curve I would say it's worth the effort. The >payoff's get bigger the more you use it. I recently tried using >javamake (which has ant integration). This involved adding an >additional 1-liner to my build.xml, duplicating my compile task to use >javamake, and that was it. That's nice and simple integration, and >there have been a few other examples like this. But of course this only >works because folk are1 targeting ant. > >> What fells wrong about bootstrap? We build the .jar file and run unit > >Nothing per-se. But if I take a c/c++ tarball from the net, I pretty >much expect the following to work: ./configure; make; make install. >Most of the time I don't even read either the INSTALL or README files -- >I just expect this to be the de facto process. In the Java space I >believe ant is becoming the de facto build process. It would be nice if >you could download the latest+greatest jscheme and just type ant to get >a full build... > >Just my 2 cents... > >-- >andy |
From: Ken A. <kan...@bb...> - 2002-12-13 01:09:13
|
The discussion about ANT made me write some code that i've been meaning to write for some time. JScheme has a new using/ directory which i hope will contain several real examples of using JScheme. The first example is run.scm which provides a mini language for invoking external Processes and doing things like building Java applications portably. Here a link: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/jscheme/jscheme/src/using/run.scm?rev=1.1&content-type=text/vnd.viewcvs-markup Actually, if you download the source it should be viewable in your browser as well as executable, though it is JDK 1.4 dependant, so it is its own Javadoc. It isn't as capable as ant, but its implementation and documentation (~ 350) is about a third larger than the number of ant classes (~ 260). |
From: Ken A. <kan...@us...> - 2002-12-12 23:11:02
|
Update of /cvsroot/jscheme/jscheme/src/jsint In directory sc8-pr-cvs1:/tmp/cvs-serv24470/jsint Modified Files: Primitive.java Log Message: Add a new using/ directory which should have examples of real uses of JScheme. Add using/run.scm which is an example of portable shell scripting and application building. |
From: Ken A. <kan...@us...> - 2002-12-12 23:11:02
|
Update of /cvsroot/jscheme/jscheme/src/using In directory sc8-pr-cvs1:/tmp/cvs-serv24470/using Added Files: run.scm Log Message: Add a new using/ directory which should have examples of real uses of JScheme. Add using/run.scm which is an example of portable shell scripting and application building. |
From: Ken A. <kan...@us...> - 2002-12-12 23:05:27
|
Update of /cvsroot/jscheme/jscheme/src/using In directory sc8-pr-cvs1:/tmp/cvs-serv22881/using Log Message: Directory /cvsroot/jscheme/jscheme/src/using added to the repository |
From: Andrew M. <and...@wi...> - 2002-12-10 09:44:01
|
Ken Anderson <kan...@bb...> writes: [...] >>Also, is src/build/bootstrap the only way (right way) to build the code? >>I had to compile BacktraceException.java by hand and then run bootstrap >>to re-build the jar file, but that somehow feels wrong. Has the project >>considered Ant (http://jakarta.apache.org/ant/) as a build tool? > > I'm going to let you tickle a hot button. I don't mean to offend you, > or anyone who uses ant. I'd like to start an honest discussion. To > my mind, ant represents a BIG answer to the problem of make, just as > JAVACC, or Lex (is there a Java version?) provide a BIG answer to > parsing and lexing. I'm interested in smaller answers that are almost > as capable that i can remember enough of that i don't have to go to > the documentation. This makes me feel like i have more tools in my > pocket, which feels better to me than going out and buying another > book and reading for a while before i get the code i need. I can understand this a lot. I held off using ant for our internal build mechanism (but for far too long). The initial learning curve appeared to be *just* too big. However, about 4 months ago I bit the bullet, and you know what, it took me about two full days to get everything done. Period. Since then adding incremental changes has been really easy. I really wish I hadn't deliberated on the decision to switch for so long. So, despite the learning curve I would say it's worth the effort. The payoff's get bigger the more you use it. I recently tried using javamake (which has ant integration). This involved adding an additional 1-liner to my build.xml, duplicating my compile task to use javamake, and that was it. That's nice and simple integration, and there have been a few other examples like this. But of course this only works because folk are1 targeting ant. > What fells wrong about bootstrap? We build the .jar file and run unit Nothing per-se. But if I take a c/c++ tarball from the net, I pretty much expect the following to work: ./configure; make; make install. Most of the time I don't even read either the INSTALL or README files -- I just expect this to be the de facto process. In the Java space I believe ant is becoming the de facto build process. It would be nice if you could download the latest+greatest jscheme and just type ant to get a full build... Just my 2 cents... -- andy |
From: Ken A. <kan...@bb...> - 2002-12-10 01:00:51
|
Sorry, the email escaped prematurely. What i ment to say was: If you'd like to provide an ant script for building JScheme, i'd be glad to try to do the same thing in JScheme. I think the comparison could be interesting. Thanks for your bug report! k At 06:48 PM 12/9/2002, Ken Anderson wrote: >At 05:35 PM 12/9/2002, Kyle R . Burton wrote: >>Not being 100% familiar with the code-base, I'm not sure if this 'fix' is >>the right approach. >> >>I was trying to track down a bug in some of my Java code being called >>from Scheme -- the final line in the Backtrace was just >>java.lang.NullPointerException (not very informative). After digging in >>the JScheme sources a bit, can we just add a >> >> this.exception.printStackTrace(s); > >If you do >(set! jsint.BacktraceException.printJavaTrace$ #t) >You should get backtraces, but i presume you're already doing that. > >Could you send me the actual bactraceException.java file you are using so i can compare it in EMACS. I haven't learned to use patch sucessfullly yet. >Though i think you've identified a problem - one should never println a backtrace. > >>To the else block in printStackTrace so the Java part of the stack trace >>is dumped as well as the scheme part(s)? >> >>A patch that seems to work is attached. > > > >>Also, is src/build/bootstrap the only way (right way) to build the code? >>I had to compile BacktraceException.java by hand and then run bootstrap >>to re-build the jar file, but that somehow feels wrong. Has the project >>considered Ant (http://jakarta.apache.org/ant/) as a build tool? > >I'm going to let you tickle a hot button. I don't mean to offend you, or anyone who uses ant. I'd like to start an honest discussion. To my mind, ant represents a BIG answer to the problem of make, just as JAVACC, or Lex (is there a Java version?) provide a BIG answer to parsing and lexing. I'm interested in smaller answers that are almost as capable that i can remember enough of that i don't have to go to the documentation. This makes me feel like i have more tools in my pocket, which feels better to me than going out and buying another book and reading for a while before i get the code i need. > >What fells wrong about bootstrap? >We build the .jar file and run unit tests because we expected people to just take the .jar and let JScheme start infesting their Java application. >It is also the minimum we do when we check in new software. > >We could provide more "make targets" which would you like? > >Ant is a nice tool for building Java applications portably, but it requires knowing XML and reading a lot of documentation. > >We build JScheme using JScheme because it is portable, and a programming language that can be used for scripting, so it can do things finding what .java you depend on and which need to be recompiled easily. > >In our LL2 Talk: http://ll2.ai.mit.edu/talks/bbnll2/ we present a simple application building example where we compile all the .java files in about 70 lines of code. This is less than most ant build.xml scripts that i've seen. > >This doesn't do everything ant can do, but at BBN i use a larger system, maybe 900 lines, that does much of what ant does, including invoking ant tasks if the behavior you want isn't accessible any other way. > >If you feel like pushing me > > > >------------------------------------------------------- >This sf.net email is sponsored by:ThinkGeek >Welcome to geek heaven. >http://thinkgeek.com/sf >_______________________________________________ >Jscheme-devel mailing list >Jsc...@li... >https://lists.sourceforge.net/lists/listinfo/jscheme-devel |
From: Ken A. <kan...@bb...> - 2002-12-09 23:49:05
|
At 05:35 PM 12/9/2002, Kyle R . Burton wrote: >Not being 100% familiar with the code-base, I'm not sure if this 'fix' is >the right approach. > >I was trying to track down a bug in some of my Java code being called >from Scheme -- the final line in the Backtrace was just >java.lang.NullPointerException (not very informative). After digging in >the JScheme sources a bit, can we just add a > > this.exception.printStackTrace(s); If you do (set! jsint.BacktraceException.printJavaTrace$ #t) You should get backtraces, but i presume you're already doing that. Could you send me the actual bactraceException.java file you are using so i can compare it in EMACS. I haven't learned to use patch sucessfullly yet. Though i think you've identified a problem - one should never println a backtrace. >To the else block in printStackTrace so the Java part of the stack trace >is dumped as well as the scheme part(s)? > >A patch that seems to work is attached. >Also, is src/build/bootstrap the only way (right way) to build the code? >I had to compile BacktraceException.java by hand and then run bootstrap >to re-build the jar file, but that somehow feels wrong. Has the project >considered Ant (http://jakarta.apache.org/ant/) as a build tool? I'm going to let you tickle a hot button. I don't mean to offend you, or anyone who uses ant. I'd like to start an honest discussion. To my mind, ant represents a BIG answer to the problem of make, just as JAVACC, or Lex (is there a Java version?) provide a BIG answer to parsing and lexing. I'm interested in smaller answers that are almost as capable that i can remember enough of that i don't have to go to the documentation. This makes me feel like i have more tools in my pocket, which feels better to me than going out and buying another book and reading for a while before i get the code i need. What fells wrong about bootstrap? We build the .jar file and run unit tests because we expected people to just take the .jar and let JScheme start infesting their Java application. It is also the minimum we do when we check in new software. We could provide more "make targets" which would you like? Ant is a nice tool for building Java applications portably, but it requires knowing XML and reading a lot of documentation. We build JScheme using JScheme because it is portable, and a programming language that can be used for scripting, so it can do things finding what .java you depend on and which need to be recompiled easily. In our LL2 Talk: http://ll2.ai.mit.edu/talks/bbnll2/ we present a simple application building example where we compile all the .java files in about 70 lines of code. This is less than most ant build.xml scripts that i've seen. This doesn't do everything ant can do, but at BBN i use a larger system, maybe 900 lines, that does much of what ant does, including invoking ant tasks if the behavior you want isn't accessible any other way. If you feel like pushing me |
From: Kyle R . B. <mo...@vo...> - 2002-12-09 22:35:52
|
Not being 100% familiar with the code-base, I'm not sure if this 'fix' is the right approach. I was trying to track down a bug in some of my Java code being called from Scheme -- the final line in the Backtrace was just java.lang.NullPointerException (not very informative). After digging in the JScheme sources a bit, can we just add a this.exception.printStackTrace(s); To the else block in printStackTrace so the Java part of the stack trace is dumped as well as the scheme part(s)? A patch that seems to work is attached. Also, is src/build/bootstrap the only way (right way) to build the code? I had to compile BacktraceException.java by hand and then run bootstrap to re-build the jar file, but that somehow feels wrong. Has the project considered Ant (http://jakarta.apache.org/ant/) as a build tool? Thank you for your time. Kyle R. Burton -- ------------------------------------------------------------------------------ Wisdom and Compassion are inseparable. -- Christmas Humphreys mo...@vo... http://www.voicenet.com/~mortis ------------------------------------------------------------------------------ |
From: Ken A. <kan...@us...> - 2002-12-07 15:57:19
|
Update of /cvsroot/jscheme/jscheme/src/jsint In directory sc8-pr-cvs1:/tmp/cvs-serv5472/src/jsint Modified Files: primitives.scm Added Files: Values.java Log Message: Finally add Derek Upham's values and call-with-values. |