RE: Java security question
hi all, I have been reading this thread for a while now. I have a question about the java security in general and specifically caters to applets downloaded from the internet sites while browsing. People have raised security questions when users browse and download applets on their local computers. We have things called as hostile java applets etc. here is a reference.. www.rstcorp.com/hostile-applets/index.html BUT all these applets run on the JVM and they cannot access resources outside the JVM. so where is the security problem ? I have been at a loss to really get to the bottom of this. fine java applications can access system resources. but not applets. they do not have the rights to run beyond the JVM(sandbox). please correct me if i am wrong. thanks john --- Zack Grossbart <[EMAIL PROTECTED]> wrote: > Rajesh, > > The scheme you are discussing is very similar to > what Cisco does with a > lot of their network monitoring code. Cisco used > the actual software and > not the install. This is probably a better option, > given that the install > can also be tampered with or possibly reverse > engineered and rewritten. > However, there are some caveats. First, this > requires the machine have some > unique id. There are a couple of option there, but > the most popular is Mac > address. Some Cisco software uses IP address, but > this is prone to > difficulties given that the IP address of a given > machine is subject to > change. Second, this scheme requires a server to do > authentication. You > need to have some server authenticate or else it is > possible to break. > > So the bottom line is, any software you install on > a users machine is > theoretically something that the user could get at > and change. The only way > around this is to have a client and server > architecture, and have the actual > logic of the application on the server. However, in > real life the chances > of someone wanting to do this are pretty slim. > Consider it similar to a car > alarm. They don't make your car impenetrable to > thieves, but they do make > your car more difficult to steal. > > Zack > > > -Original Message- > From: Rajesh Nair [mailto:[EMAIL PROTECTED]] > Sent: Wednesday, April 19, 2000 4:22 PM > To: Gayathri Viswanathan; 'Zack Grossbart'; Gayathri > Viswanathan; > [EMAIL PROTECTED] > Subject: RE: Java security question > > > > > This would require a self-coded lock or something, I > presume. It's always > good to have obfuscation on the java class code. > Like Zack mentions once > its in somebody's hands, they could make changes. > If obfuscation is really as good as it sounds, > wouldn't it be possible to > limit the applet that has been installed once to > make sure it cannot be > copied onto another location? I mean, say your > applet has been > been installed on machine A. The applet is signed > and has access to > installed m/c. Applet during installation, > creates a lock that identifies this machine > uniquely. Person P is able to > make a small change say to logo > and sells it to Party Q. Party Q runs applet > install. Install knows it's > being dumped on another m/c. Install > spews scary legalise at Q and fails to install? > > > If the applet is being used like normal applets, it > would have access to m/c > that is serving it, right? > Does this sound even remotely fair to do? > > > > At 02:05 PM 04/19/2000 -0400, Gayathri Viswanathan > wrote: > >Zack, > > > >I have already signed my Java applet with a > certificate from Thawte. But I > >thought that > >this means that Thawte certifies that noone has > changed the jar file. But > >what if after > >accepting the certificate, some malicious user > wishes to change the > contents > >of the jar file > >by say changing some image files (used for > displaying logo) and then > signing > >it again and then > >selling it ? Would obfuscation help in this ? Can > obfuscation be used on > >applets ? > >Is there any other alternative ? > > > >Thanks. > > > >-- Gayathri > > > >-Original Message- > >From: Zack Grossbart [mailto:[EMAIL PROTECTED]] > >Sent: Wednesday, April 19, 2000 1:30 PM > >To: Gayathri Viswanathan; > [EMAIL PROTECTED] > >Subject: RE: Java security question > > > > > >Gayathri, > > > > Obfuscation would help prevent someone from > decompiling and > >understanding > >your code, but not from changing it. You should > sign your JAR file. Tools > >like Visual Cafe have this capability
accessing a web page via a proxy ?(how to)
i have a small java program that goes and accesses a web page. (used the URL class) the java program works fine, but how do i force it to use a proxy server. I mean when i run the java program via the command line it goes and fetches the http web page as mentioned in the arguments java test http://www.xyz.com one can configure one's web client browser to use the proxy server and all the accesses are routed via the proxy server. can someone let me know as to how to do i force my java program to use a proxy server ? the second problem is about the contents that are donwloaded. if it is a plain simple page with text and gif files it is fine. but if the page has some class files then these class files cause a lot of exceptions etc as the contents are being (tried to be interpreted by the via the command prompt) Any way this problem is secondary and the first one is how to route the requests via a proxy server. thanks john __ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
how to set proxy via the command line
i sent an earlier post as to how to set the proxy server via the command line. well i just found in the java.sun site and the command line parameters that need to set on a unix box are as follows java -Dhttp.proxyHost=proxyhost [-Dhttp.proxyPort=portNumber] URLReader question is that how do we enter this. via the command line... I get the following error if I enter it via the command line as above, Exception in thread "main" java.lang.NoClassDefFoundError: URLReader please let me know as to what is error earlier post that i posted --- john <[EMAIL PROTECTED]> wrote: > > i have a small java program that goes and accesses > a web page. (used the URL class) > the java program works fine, but how do i force it > to use a proxy server. I mean when i run the java > program via the command line it goes and fetches > the http web page as mentioned in the arguments > > java test http://www.xyz.com > > one can configure one's web client browser to use > the > proxy server and all the accesses are routed via the > > proxy server. > can someone let me know as to how to do i force > my java program to use a proxy server ? > > the second problem is about the contents that are > donwloaded. if it is a plain simple page with text > and > gif files it is fine. but if the page has some class > files then > these class files cause a lot of exceptions etc as > the > contents are being (tried to be interpreted by the > via the command prompt) > Any way this problem is secondary and the first one > is > how to route the requests via a proxy server. > > thanks > john > > __ > Do You Yahoo!? > Send instant messages & get email alerts with Yahoo! > Messenger. > http://im.yahoo.com/ > __ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Downloading web pages
thanks to all who responded to my thread for "how to set the proxy server via the command line". The problem was that proxy port no was 8080. ok. but this brings me to another halt ... (sort of) I am have written a small program which goes and downloads a web page given its url. The download is fine and works via proxy servers as well !! however i have two problems now. first when the download contains say some class files , the shell tries to interpret it. and there are lots of unicode characters appearing on the shell window etc. second the java program is pretty slow. ie the download takes a bit more time. I am writing a small tool so that i could download web pages automatically via a proxy server and stress the proxy server. I have a small loop which acts like a robot and picks out a few web pages and downloads them. This helps in analyzing performance of the web server. however the java execution is slow and this does not generate much of traffic via a proxy server. does anyone know how improve performance ? thanks all john --- Christopher Hinds <[EMAIL PROTECTED]> wrote: > Why is the proxyPort parm in those brackets? > > The command line should be: > java -Dhttp.proxyHost=proxyhost > -Dhttp.proxyPort=9 URLReader > > where proxyHost is the host name or IP address of > the proxy host > proxyPort is the port the the proxy is > listening on > > Cheers > Chris > > john wrote: > > > > i sent an earlier post as to how to set the proxy > > server via the command line. > > well i just found in the java.sun site and the > command > > line parameters that need to set on a unix box are > as > > follows > > > > > >java -Dhttp.proxyHost=proxyhost > > [-Dhttp.proxyPort=portNumber] URLReader > > > > question is that how do we enter this. via the > command > > line... I get the following error if I enter it > via > > the command line as above, > > > > Exception in thread "main" > > java.lang.NoClassDefFoundError: URLReader > > > > please let me know as to what is error > > > > > > earlier post that i posted > > > > --- john <[EMAIL PROTECTED]> wrote: > > > > > > i have a small java program that goes and > accesses > > > a web page. (used the URL class) > > > the java program works fine, but how do i force > it > > > to use a proxy server. I mean when i run the > java > > > program via the command line it goes and fetches > > > the http web page as mentioned in the arguments > > > > > > java test http://www.xyz.com > > > > > > one can configure one's web client browser to > use > > > the > > > proxy server and all the accesses are routed via > the > > > > > > proxy server. > > > can someone let me know as to how to do i force > > > my java program to use a proxy server ? > > > > > > the second problem is about the contents that > are > > > donwloaded. if it is a plain simple page with > text > > > and > > > gif files it is fine. but if the page has some > class > > > files then > > > these class files cause a lot of exceptions etc > as > > > the > > > contents are being (tried to be interpreted by > the > > > via the command prompt) > > > Any way this problem is secondary and the first > one > > > is > > > how to route the requests via a proxy server. > > > > > > thanks > > > john > > > > > > > __ > > > Do You Yahoo!? > > > Send instant messages & get email alerts with > Yahoo! > > > Messenger. > > > http://im.yahoo.com/ > > > > > > > __ > > Do You Yahoo!? > > Send instant messages & get email alerts with > Yahoo! Messenger. > > http://im.yahoo.com/ > > > > > > > -- > > To UNSUBSCRIBE, email to > [EMAIL PROTECTED] > > with a subject of "unsubscribe". Trouble? Contact > [EMAIL PROTECTED] > > > > > > > > > > __ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
bug in jdk1.1.5v7?
It seems as though the incorrect Exception is being thrown in jdk1.1.5v7. According to the Java documentation for java.net.Socket, -- public synchronized void setSoTimeout(int timeout) throws SocketException Enable/disable SO_TIMEOUT with the specified timeout, in milliseconds. With this option set to a non-zero timeout, a read() call on the InputStream associated with this Socket will block for only this amount of time. If the timeout expires, a java.io.InterruptedIOException is raised, though the Socket is still valid. The option must be enabled prior to entering the blocking operation to have effect. The timeout must be > 0. A timeout of zero is interpreted as an infinite timeout. --- But instead of an InterruptedIOException, I'm getting a SocketException with the message "Interrupted System Call." Can someone look in the source for the JDK to make sure that the proper exception is being thrown? I haven't found time to generate a test case on this yet, to prove it. -John _ "The human mind is a 400,000-year-old legacy application...and you expected to find structured programming?" -- Randall Davis, 1996 AAAI Pres. Address
Re: Does RMI work in jdk1.1.5v7/Red Hat 5.0?
Stephen Wynne wrote: > In message <[EMAIL PROTECTED]>, John Collins writes: > > However, I can't get past the attempt to bind to the rmi registry. This > code works on both NT and Solaris, and has been exercised fairly > extensively. > > John, > > I had no problems registering and running a simple RMI example on > 1.1.5v5. If you could send me a small RMI client/server pair that > demonstrates your problem, I'd check it out for you on my RH 5.0 > setup. > > I hesitate to make any promises because you've clearly done more work > on RMI than I have, and you would have tried all the obvious things. > However, I do have Java source here and can look at it from that > angle. OK. I'm still stuck. I've attached a tar file that exhibits the problem. It's about as simple as an RMI app can be. The problem: When I start up an RMI server, I get the following exception at the point where I do the Naming.rebind(name, server); (This is the printout from exception.printStackTrace();) --- java.rmi.ServerException: Server RemoteException; nested exception is: java.rmi.AccessException: Registry.rebind --- The environment: Red Hat 5.0/Pentium II, jdk1.1.5-v7. I have the same problem both with glibc-2.0.7-6 and glibc-2.0.7-7. I'm running the example locally, and it doesn't matter whether I'm logged in to the net or not. The code: There's an interface called jec.Server, a server class called jec.ServerImpl, and a client class called jec.ServerTest. I've supplied both the sources and the .class files. My test on NT involved moving over just the .class files from Linux, and that worked, so I deduce the problem is in the Linux runtime. To run the example. - Install the class files in the CLASSPATH, or install them and modify the CLASSPATH to see them. - Start the rmiregistry - In one shell, do java jec.ServerImpl - in another shell, do java jec.ServerTest The result should be that the line Result: gets printed out, after which the client exits. On Linux this isn't happening. So, now the question is, did I do something really stupid, or is there something wrong with my environment that I'm not seeing, or did I turn over a rock that needed looking under? Thanks. John Collins University of Minnesota [EMAIL PROTECTED] rmi-test.tar
Re: Does RMI work in jdk1.1.5v7/Red Hat 5.0?
Steven P. Heutinck wrote: > Yes, I can confirm it does work. Make sure you are trying to bind to IP > address of your primary network device i.e. eth0. > > Cheers, > Steve > > John Collins wrote: > > > >... I can't get past the attempt to bind to the rmi registry. This > > code works on both NT and Solaris, and has been exercised fairly > > extensively. I was hoping to be able to jettison NT and move over to > > Linux for further development. > > > > Here's what I've got: > > -Red Hat Linux 5.0. > > -libc 2.0.7 (I've tried downloading 2.0.7-7 a couple of times; it's > > been very slow.) > > -jdk1.1.5v7-glibc version > > > > I'm not sure how to proceed in debugging this. Here's the error message: > > > > --- > > java.rmi.ServerException: Server RemoteException; nested exception is: > > java.rmi.AccessException: Registry.rebind > > --- > > As far as I can figure out, I'm doing exactly the same thing here that I > > do on Solaris to run this thing. The only part of it that isn't scripted > > is the startup of the rmiregistry. On both systems I run this in my user > > account. Thanks, Steve - Others have reported that my code works on their systems. It still doesn't work on my system. What do you mean about "bind to IP address...". I don't see any options for binding to IP addresses in either the rmiregistry command or the Naming.rebind() method. It's in the rebind call that the problem occurs, and it's definitely a different message than the one that happens when I don't have rmiregistry running. John Collins
Re: RedHat 5.0 & JDK
Aleph One wrote: > Anyone had any luck running jdK 1.1.5v7 or v5 in a RH 5.0 box with the > latest glibc RPMs? I get a SIGSEGV with both deep inside glibc. The v7 works fine for me on RH 5.0. I use it every day, and I have a large RMI-based system that we run on NT and Solaris as well as Linux. It uses the latest ObjectStore PSE Pro package, which is 100% Java and seems to run without a hitch. I had to put in the latest glibc, but that's all. Also, to run RMI I had to make sure the hosts file and the HOSTNAME file agreed on what the name of this machine is. Keep at it; it should work. John Collins University of Minnesota
Re: Major Swing/Linux Project?
>>>>> "Cynbe" == Cynbe ru Taren <[EMAIL PROTECTED]> writes: [...] > Might be nice if -everything- were done through the > browser+[signed]applet mechanism and network protocol: This would allow > such Linux applications to be run from just about any Windows or Mac box, > and possibly some or all of the new dumbed-down browser-capable set-top > boxes due out soon. (Always nice to lead the technology curve a bit in > the design phase.) Hmm... What do the signed applets have to do with remote/networked administration? For purely local-only admin, it makes some sense but a server-based solution gives local and remote capabilities. The administration of the machine is much better handled through e.g., Java servlets running in e.g., Apache on each machine. You could then support either an HTML or a Java applet interface (or both). I'd like all of the configuration utilities provided by the RedHat control_panel plus SNMP, remote shutdown/reboot, file upload/download, network time management (clockspeed/ntp), distributed task execution, etc. :-) [...] > I'm likely to be doing something at least vaguely along these lines: I > want to use browser-based Swing widgets as the remote GUI client for a > distributed multi-user application server I'm developing. If a generic > mechanism were developed for allowing server-side apps to interact with > client-side Swing stuff, I might be interested in hopping on board... Um, er, RMI? Take care, John
Re: Major Swing/Linux Project?
>>>>> "David" == David Wilkinson <[EMAIL PROTECTED]> writes: > At 19:05 28/05/98 +, John Mitchell wrote: >> The administration of the machine is much better handled through e.g., >> Java servlets running in e.g., Apache on each machine. You could then >> support either an HTML or a Java applet interface (or both). > Hmm, how would that work? Your servlet is typically running as an > unprivileged user, but it would need to be root to perform most sysadmin > tasks. Indeed. That same problem is enjoyed if someone wants to do this sort of thing via signed applets. > Either the servlet would need to act as a wrapper to some suid programs > or your web server needs to run with root privileges - neither seems > particularly desirable. > I guess you could have a dedicated web server running as root configured > quite tightly to accept only a limited set of requests. I'm not sure how > authentication would work though... Administrator login/password iff connecting from local host. SSL communication channel built using both server & client certificates, administrator login/password if connecting remotely. Yes, I was thinking that these admin. servlets would be running in a separate instance of the web server (ala your Avenida) which only deals with admin. requests as opposed to actually using a separate complete instance of e.g., Apache. Take care, John
Java Web Server
>>>>> "Michael" == Michael D James <[EMAIL PROTECTED]> writes: [...] > I think I've heard discussion on this topic but I didn't pay attention > because I didn't realize I would need to know. Have you check the list archives? Hmm... Are the list archives even available since the java-linux.org fiasco? > Is anyone successfully running Java Web Server on Linux? > http://jserv.java.sun.com/products/webserver/features/index.html#cross > makes a big deal about it being supported across many platforms but the > system requirements don't list Linux as one of the supported operating > systems. > If not, does anyone recommend another web server supporting SSL on Linux > (such as Stronghold or Raven)? First, it's no big whup to run the base JWS on Linux. I posted my changes to run it under RedHat Linux to the JWS mailing list a 5 or 6 months ago to get the native code to switch uid, etc. to work. Check out the archives for the Java Web Server mailing list. The biggest PITA is that Sun won't support SSL on any platforms but Solaris and NT due to various things like their licensing deal with RSA. All of my entreaties for at least some sort of decent documentation so that we can hook in our own SSL support have been totally ignored. I haven't tried servlet support in Apache in a long time so I can't help you much there but if you're already using Apache then you should definitely check that out first and then the JRun servlet-runner from Live Software (particuarly if you want support). Hope this helps, John
Re: Java Web Server
>>>>> "Michael" == Michael D James <[EMAIL PROTECTED]> writes: [...] > Apache-SSL looks like a good choice since it's free, open source, we > already know Apache, and it would be easy to get servlets working on it. > The only problem is my boss doesn't believe its kosher. The Raven > website says: > Q: I was wondering what might make your SSL module for Apache superior to > the already freely available patches such as Apache-SSL with SSLeay? > R: U.S. Law requires that we use cryptography algorithms supplied by RSA > corporation. >Raven provides an affordable way for you to legally use Apache with > SSL in the U.S. > This implies there would be something illegal about using Apache-SSL. Is > this a scare tactic? Our application is within the US only. It's a patent issue. RSA (the company) owns the patents to RSA (the public-key crypto) and so to legally use RSA crypto you need to purchase a license. Hence Stronghold and, it seems also Raven. Yet another reason to vote Libertarian. Usual disclaimers apply that I'm not a lawyer, this is *not* legal advice, get your own, competent, professional legal counsel. Take care, John
Re: RH5.0 and JDK1.1.?
Pete Hardie wrote: > Hello all, > > As I seem to have been unsubscribed during the last (non) swap of sites, > I think I > missed on some of the news. What is the status of the JDK ports and > RedHat 5.0? I have > found enough freeware needing stuff I didn't include in my last RH4.2 > installation > that I am looking at an upgrade to 5.0, but I also want a working JDK. I'm using the 1.1.5v7 glibc version very successfully on RH 5.0. > > > Also, is there a working Netscape for RH5.0? > My copy is 4.05, and it works fine. John Collins University of Minnesota
Re: feedback on JDE's
Jerry Solomon x2761 wrote: > > Hello List, > > I have spent the past several weeks trying out two commercially available > JDE's, Simplicity (datarepresentations.com) and SuperMojo (penumbrasoftware.com). > Although I have not done an exhaustive evaluation, I am coming to the > conclusion that these (and probably other) JDE's are not terribly useful > for more than "toy" projects. I would greatly appreciate hearing from > anyone who has actually used either of these products to produce real, > substantial project code, and what their experiences might have been. > Thanks in advance, I've done fairly substantial stuff with JBuilder and Visual Cafe on NT, and with emacs/JDE on linux. I tried SuperMojo, and it was never able to absorb my existing, working code, so I gave up. I tried CoffeeShop, and never got it to work properly. I would like an interface builder for prototyping; maybe SuperMojo would be OK for that, but I'm really very happy with emacs/JDE for the big stuff. The only thing I miss from JBuilder is the debugger. The jdk debugger is a bit crufty in comparison, but at least it does work. The other thing I'm experimenting with is Together/J (www.oisoft.com). I think for server-side stuff, and about 50% of client-side stuff, a UML tool is much more useful than a GUI builder. Unfortunately, Together/J is VERY expensive for the commercial version. John Collins
Re: Is there a JDBC-ODBC bridge for Linux?
>>>>> "Rick" == Rick Stone <[EMAIL PROTECTED]> writes: [...] > We have Empress RDBMS installed, which is lovely and runs on Linux as > well as other Unix flavors. They supply an ODBC server which listens to > connections from ODBC drivers which they also supply. But there is no > native Java support. So if we had a JDBC-ODBC bridge on Linux, we could > connect to the Empress ODBC server using JDBC. > Empress even gives this as their Java solution, so I didn't think to > question it until I realized that Blackdown's 1.1.6 jdk linux port > doesn't come with the bridge. > Which is a drag. Nobody should be using the JDBC-ODBC bridge for anything. It was a hack to get people to play with JDBC while the vendors/etc. could write real JDBC drivers for the various DBs. So, I suggest that you: * Beat on Empress to make a JDBC driver (or contract someone to write it for them :-). * Or, if you want a free DB which has a free JDBC driver you may want to check out the usual suspects (PostgresSQL, mSql, mySql, etc.). * Or, if you want a commercial DB with a JDBC driver, check out Solid DB. They also still seem to have some special deal for Linux users (but I don't recall if that includes the JDBC driver or just the DB itself). Hope this helps, John
Re: debugging an applet
Of all the debuggers I've come across for Java (most suck unmentionable things), Sun's JavaWorkshop seems to be the best. It's written in Java, so expect it to crash a fair amount; but it allows you to get waist deep into the various stack frames of all your threads, inside a nice gui. Problem? Sun puts it out for Windows and Solaris. I've heard mention that it works under Linux with a patch from "SuSE", but that some features don't work in such operation. As I do development on NT/Linux at work, I use it on NT. There's a trial version to be had (JWS): http://shop.sun.com/cgi-bin/statetrack/parts/796-0044-01/index.html?*:19980716004945~17703&*:1 On Wed, 15 Jul 1998, Brad Giaccio wrote: > Date: Wed, 15 Jul 1998 13:44:31 -0400 (EDT) > From: Brad Giaccio <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Subject: debugging an applet > Resent-Date: Wed, 15 Jul 1998 13:43:44 -0400 > Resent-From: [EMAIL PROTECTED] > > well I've seen a number of you mention debuggers for java apps but I seem > to be going in to deadlock somewhere in my applet and have already spent > three days searching I 've. Does anyone one have a good debugger for > applets it doesn't need to be robust basically what I need is to run the > applet in a debugger, then lock it up and have the debugger tell me where > all the threads are so I can find my deadlock. > > Yes I've tried > jdb and jikes with sun.applet.AppletViewer but when I do this > appletviewer dumps core on the first line of my init which is super(). > But by itself it doesn't dump core. > > Hope someone can help, > > Thanks, > Brad > > -- > --- There are two kinds of knowledge, you either know the answer or > you know where to find it > -Kane, Johnson, and anonymous > _ "The human mind is a 400,000-year-old legacy application...and you expected to find structured programming?" -- Randall Davis, 1996 AAAI Pres. Address
Re: migration to linux
Dale & Karen Nicholson wrote: > > I want to instal Redhat linux to run in dual boot mode with win95 and > need the capacity to use the JDK with both. > > I am a developer in a HP-UX and SCO unix environment at work using > mostly unibasic, C, & two proprietary EIS languages (very little linux, > only EIS' modified version for Digitized voice) but have been developing > applets and some simple java programs at home for my own personal > development and web design. I want the power of UNIX at home but am > afraid of switching without getting some more detailed knowledge first. > Unfortunately I have no experience in compiling and configuring an OS so > I thought I'd try one of the more popular binaries available. > > Any help would be very much appreciated, either with a better flavor of > linux to use or with how to port the JDK to linux. I use RedHat 5.0 with Steve Byrne's jdk 1.1.5 port. I haven't upgraded to 1.1.6 yet; I've been too busy writing code and keeping up with other stuff. I regularly use my linux box, Solaris, and WinNT. The Linux platform has become my primary development environment. I am not an experienced Unix sysadmin, but I've managed to stumble around and get RH working just fine, without an inordinate amount of effort. I find it much more transparent than NT; when things go wrong, it's a lot easier to figure out why, and to fix it. On the other hand, linux is a LOT more stable than NT, particularly for a developer. I've tried a number of IDE packages for java on linux, and I've settled on the JDE package for emacs. I'm using RMI and Object Design's PSE Pro extensively, and I find I need the power of gnu make to construct my system; the basic jdk tools just don't cut it, nor do any of the tools built into the big expensive IDE's. I don't do too much GUI work, but when I do there are several tools that are pretty good for that. Often I end up using JBuilder on the NT box to whip up a front end, then turn around to the linux box and polish it up and add the RMI stuff and the real logic. Linux runs Samba out-of-the-box, so it's a file and print server for the NT box. You can either just use the same directories on both sides, or you can use cvs on the linux side to manage your code. By the way, linux can also be set up as a masquerading firewall, so both linux and NT can see out through the same ppp link when I'm dialed in. Good luck. John Collins
Re: Printing
The program works fine under Windows and Solaris. I have not personally tested it under Linux. Does the Linux JDK support printing? The error message seems to imply that it didn't come from the source in from the AWT book. Is this something you added, or does it come from the Linux port? I'm no longer on the Java-Linux mailing list, so I'm not sure if anyone else answered this for you already. Regarding printing in the JDK 1.2 API, it is much easier/cleaner to print, especially multiple pages. In fact, you can even print w/o the dialog box coming up. J Thanks for your message at 04:34 PM 8/12/98 -0700, Laura L. Evangelista: >Sir, > >I tried to run your sample application "TestPrint" but it >wouldn't print anything ... The print dialog shows up. But when I >choose "Print" from the dialog, a "Print Error" dialog shows up >with the message: > >"Could not execute print >command: [Ljava.lang.String;@80cbe12" > >This is the first time I tried the printing feature in Java >(I also tested other sample applications, but they all wouldn't >work) ... The printer (HP LaserJet 4L) is from within the (local) >network ... I am using RedHat Linux 4.2, JDK 1.1.3 ... > >I hope you can help me on this. > >Thank you. > >-- Laura > >Tidbits: >What can you say about the Printing API in JDK 1.2? > John Zukowski Focus on Java Guide / http://java.miningco.com [EMAIL PROTECTED] | [EMAIL PROTECTED] Author Java AWT Reference / Borland's JBuilder No Experience Required MageLang Institute - http://www.magelang.com Will hack Java code / book for food
Re: Printing
Thanks for your message at 10:51 AM 8/14/98 -0700, Laura L. Evangelista: >John Zukowski wrote: > >> Does the Linux JDK support printing? > >If you mean it comes with the JDK, yes ... I'm using JDK 1.1.3. That's not what I am asking. The Linux JDK is a port. They have to add some of their own stuff to make it work. It isn't the real thing from Sun. Did the people who created the port, add the printing support. > >> The error message seems to imply that it didn't come from the source in >> from the AWT book. Is this something you added, or does it come from the >> Linux port? > >The error did not come from the source. This error dialog is part of the >Print Dialog that comes up when the getPrintJob() method is called ... (I did >not change anything in the source code.) ... Right there, pjob becomes null, >not giving me the chance to get into the loop following it ... That sounds like whoever created the port didn't add support for printing then, if the JDK/JRE itself is throwing the exception. J John Zukowski Focus on Java Guide / http://java.miningco.com [EMAIL PROTECTED] | [EMAIL PROTECTED] Author Java AWT Reference / Borland's JBuilder No Experience Required MageLang Institute - http://www.magelang.com Will hack Java code / book for food
JDK 1.1.6 Version 4a and VolanoMark 2.0
Thanks for the great port in Version 4a!
It fixes that socket timeout bug I reported here back in October 1997
against JDK 1.1.3. I was unable to reproduce the problem at all outside
of the VolanoMark benchmark or our VolanoChat product, but I'm grateful
someone figured it out.
Using the same VolanoMark 2.0.0 Build 137 that I used in my August
JavaWorld article, I get the new scores shown below -- a 37% performance
improvement over JDK 1.1.6 Version 2! From the stress tests I've run,
it seems to be very stable as well.
VolanoMark 2.0.0 Build 137 Loopback Test
Java virtual machine Scores Average (best 2 of 3)
- -
Linux JDK 1.1.6 v2230, 234, 233 234
Linux JDK 1.1.6 v4a 317, 321, 319 320 (+ 37%)
For the details on the test, see:
http://www.javaworld.com/javaworld/jw-08-1998/jw-08-volanomark.html
If anyone on the Blackdown Java-Linux Porting Team wants a copy of
VolanoMark 2.0, please let me know.
Thanks again,
John Neffenger
Re: JDK 1.1.6 Version 4a and VolanoMark 2.0
Hi Dan, > Thanks John for updating us with the new results! > Great, we're now neck and neck with that speed demon, MRJ 2.0 :-) It looks as if FreeBSD's JDK 1.1.6 port is the same speed (for VolanoMark) as their JDK 1.1.5, but they also fixed that nasty socket timeout bug I was hitting on both Linux and FreeBSD. Microsoft's and Tower's latest releases are coming up a bit slower than before, but not enough to matter. Actually, the connection scalability on Linux -- not speed -- is the bottleneck for now. Even after I rebuilt my Red Hat Linux 4.2 kernel for 1024 file descriptors per process, I was unable to get over about 250 connections to the VolanoChat server. TowerJ 2.1.2 could get up to about 350 connections on the same machine. Is there a limit on file descriptors built into the Linux Java VM or its libraries somewhere? I followed the instructions at: http://www.volano.net/guide/linux.html I haven't tried more than 200 connections on Red Hat Linux 5.1 yet. The default limit is increased to 1024 in the kernel for version 2.1, right? > Does the the Volano benchmark run > with the TYA JIT (http://www.tya.home.ml.org)? I haven't tried it yet. > Or for that matter with Kaffe (http://www.kaffe.org)? Kaffe can't run VolanoMark (nor our product, VolanoChat), but Tim Wilkinson at Transvirtual Technologies is working on it. John Neffenger
Test - ignore:)
Just testing before I write a nice long mail all about firewall tunneling:) -- John Baker, Software Engineer, Java coder, Salad sarnie lover. Work: (01203) 562000 ext 4153 Home: (01203) 601890
URLConnection
Hiya,
I've been messing about with firewall tunneling recently, but am having
massive problems with URLConnection. It seems that there is one combination
of sequence of events and streams used which actually makes it work.. take
the Post method for example.. the online docs say the client should do the
following:
private final String serverURLin = new String("http://localhost:9876/");
private URLConnection urlConClientIn;
private PrintStream clientOutStream;
urlConClientOut = (new URL(serverURLin)).openConnection();
urlConClientOut.setDoOutput(true);
urlConClientOut.setUseCaches(false);
clientOutStream = new PrintStream( urlConClientOut.getOutputStream() );
clientOutStream.println("Linux");
clientOutStream.println("Is");
clientOutStream.println("Lovely");
clientOutStream.close();
Yet this doesnt seem to send anything to the server. Now I know the server
is running ok because by entering http://localhost:9876/ in netscape, it
connects and the server gets all the header information from the url
request.. so what's wrong with that client?
Here is the server code I have:
private Socket socketIn;
private DataInputStream sInStream;
sInStream = new DataInputStream
(new BufferedInputStream(socketIn.getInputStream()));
int contentLength = 0;
String inStr;
// Read the headers
do
{
inStr = sInStream.readLine();
System.out.println(inStr);
// if inStr has Content-length, read the
// length into a variable here
if (inStr.startsWith("Content-length:"))
{
String s =
inStr.substring(16, inStr.length());
System.out.println("Got clength "+s);
contentLength = Integer.parseInt(s.trim());
}
} while (!inStr.equals (""));
But it always gets 'stuck' at the sInStream.readLine() !
If anyone has any pointers, or working code which does roughly the same
as this, please help!
Ta,
John Baker
--
John Baker, Software Engineer, Java coder, Salad sarnie lover.
Work: (01203) 562000 ext 4153
Home: (01203) 601890
build of jdk1.1.6 v2 failed
My attempts to build jdk1.1.6 v2 fail because the file java_g can't be found. Does anyone know what I'm doing wrong? If this is the wrong forum for a newbie question, I appologize. -John C
Re: Re[2]: Can Java Catch Signals ?
On Mon, Sep 28, 1998 at 10:00:45AM -0400, Nelson Minar wrote: > >What I would like is just a very simple SignalCatcher for SIGTERM, > >SIGQUIT, SIGHUP and SIGINT: just four signals. I want to catach them > >in Java, and write the capture to a systems log. So that at least > >overnight personnel would know how my java batcher died . > > Someone wrote just that for Linux recently, it was posted to this > list. I don't know what, though. Hopefully the list is archived somewhere? http://interstice.com/~kevinh/projects/javasignals/ -John -- John Baker, Software Engineer, Java coder, Salad sarnie lover. Work: (01203) 562000 ext 4153 Home: (01203) 601890
Re: Experiences with jdk and lesstif?
Juergen Kreileder wrote: > > > Jason Dillon writes: > > Jason> I had lots of trouble with popup menus using lesstif. I > Jason> was never able to resolve this issue, so I had to stop > Jason> using it. I run into similar problems with netscape. When > Jason> you right-click, the menu pops up but does not return > Jason> control... ever (well after a kill it does). > > The lesstif people fixed this a week ago. > > Juergen
Re: redhat RPM & bin license.
Uncle George wrote: > > Bec of the limited space & bandwidth, I was thinking of providing a > redhat rpm for the various alpha platforms ( 21064, 21164, and the > 21164a processors ). I'd like to create the rpm, and submit the rpm to > redhat for their inclusion into the jewel set distribution. > Am i violating the bin license, by doing so. I suppose if someone else > were to package the bins, and submitted the RPM to redhat they would be > safe, and unencumbered by the license. > gatTAKE ME OFF YOUR MAIL LIST
Re: Trouble installing JDK 1.1.6
David Buddrige wrote: > > Hi all, > > I have recently downloaded JDK 1.1.6 from > ftp.progsoc.uts.edu.au/pub/Linux/java to run on my Redhat 5.1 system... > I've un-tar'ed it into /usr/local/jdk1.1.6, and then added the following > lines to my /etc/profile file > > PATH="$PATH:/usr/X11R6/bin:/usr/local/jdk1.1.6/bin" > export JAVA_HOME="/usr/local/jdk1.1.6 > export CLASSPATH="/usr/local/jdk1.1.6/lib/classes.zip" > > I then wrote a very basic "hello world" program, and attempted to > compile it, using javac... however I get the message: > > No library path set. > > Can anyone tell me what I have done wrong? I've been searching through > the archives here, and reading through the readme file, however I can't > see what I've missed... > > Ta > > David Buddrige. > [EMAIL PROTECTED] take me off your mail list
Re: Netscape 4.05 support Java 1.1??
On 2 Oct 1998, Chi-Ming Yang wrote: > Hi there, > I got netscape 4.05 that came with RedHat 5.1. Does it support Java 1.1? I >tried the following and got errors: > I have JDK1.1.6 installed. Will that help netscape? > If you had JDK 1.1.6 installed when you got these errors, clearly not. You shouldn't use any netscape prior to 4.06: 4.04 has a problem with some web sites (mine couldn't tell me about the then new 4.3 Gb removable syquest disk, 4,05 has problems with forms and I gather all prior to 4.06 have security problems with mail. > Thanks > > http://www.javasoft.com/applets/jdk/1.1/demo/NervousText/index.html > > #Applet exception: error: java.lang.NoSuchMethodError: > java.awt.Component: method addMouseListener > (Ljava/awt/event/MouseListener;) V not found > at NervousText.init(NervousText.java:63) > at netscape.applet.DeriveAppletFrame.run(DerivedAppletFrame.java) > at java.lang.Thread.run(Thread.java) Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Kaffe
On Fri, 2 Oct 1998, Didier Boucard [98-99] wrote: > Hello, > > I've tried Kaffe under linux to see if it worked fine. I've found no > problem to run jdk classes, but it's really slow: on the class I've > tested, it's about 4 times slower than under the jdk 1.1.6 from > blackdown. > > What I would know is if something exists to make it faster ? > Is Guavac a good (faster) compiler? The compilers won't make much difference to runtime speed. The fastest compiler I've seen is jikes (www.alphaworks.ibm.com) which is about an order of magnitude faster than javac and the compiled code of the one program I test it on is a little faster. jikes is now my preferred compiler. IBM has some other nice addons too including a GUI debugger written in java. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Java and CVS
On Mon, 5 Oct 1998, Travis Shirk wrote: > Hello All, > > This post does not have anything to do with Linux, but some of you > may be able to help since you're Unix users. > > --- > > I'm having some problems using CVS with a pretty large Java API. The problem > is pretty basic, I want my CVS repository to contain only java source files. > This is necessary because (1) I don't want CVS to try and merge .class > files and (2) people working with the repository should need to do a cvs add > on only the source file and not the new .class file. My problem is > with dependencies. Many of the classes depend on other classes that > may not have been created yet in the 'make all' build process. I've played > around with javac -depend but I don't think it likes how I organize > my source environment. I have all .java files under a src directory > in each subpackage directory. So if I have a package called foo.bar > the directory hierarchy would look like this: > > foo/bar contains the .class files > and foo/bar/src contains the .java files > > I refuse to mix the .java and .class files and think it is ridiculous how > java tools expect this. > > Dependencies are not a problem once all the .class files are around, but > when a user checks out a new repository I don't want them to get any .class > files. Basically, the first thing that should be done after checking out the > repository is a 'make all' so that all the .class files are built. > > Has anyone ever solved this problem. This is the first time I've really > wished for header files, because C/C++ does not have this problem during > compile time. javac: invalid flag: -? use: javac [-g][-O][-debug][-depend][-nowarn][-verbose][-classpath path][-nowrite][-deprecation][-d dir][-J] file.java... [summer@possum summer]$ jikes -? use: jikes [-g][-O][-debug][-depend][-nowarn][-verbose][-classpath path][-nowrite][-d dir] [++][+C][+E][+F][+M][+R][+V] file.java ... (C) Copyright IBM Corp. 1997, 1998. - Licensed Materials - Program Property of IBM - All Rights Reserved. use: jikes [-classpath path][-d dir][-debug][-depend][-deprecation] [-g][-nowarn][-nowrite][-O][-verbose][+1.0][+$][++][+B][+E][+F][+M][+P] file.java... -classpath path use path for CLASSPATH -d dirwrite class files in directory dir These are the java compilers I use here. Both support -d to put the class files somewhere else altogether. Does this help? Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Problem installing JDK
On Mon, 5 Oct 1998, Bordelon wrote: > This is a multi-part message in MIME format. > > --=_NextPart_000_0006_01BDF03C.B5AF23E0 > Content-Type: text/plain; > charset="iso-8859-1" > Content-Transfer-Encoding: quoted-printable > > --=_NextPart_000_0006_01BDF03C.B5AF23E0 > Content-Type: text/html; > charset="iso-8859-1" > Content-Transfer-Encoding: quoted-printable > > > > Must we have it in duplicate? Do us a favour and drop the HTML. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Still trying to build jdk1.1.6 v2
Hi, I'm still trying to build jdk1.1.6 from source. I have the executables (v2 for libc5) installed and they work. I applied the diffs in the mirror common directories to the source I got from sun. I set ALT_BOOTDIR to the top of the jdk1.1.6 executables (a problem I had earlier). The build fails about five minutes in because javaString.h uses java_lang_String.h, which isn't there. Any ideas? -John C
Re: Java app without X installed
On Tue, 6 Oct 1998, Steve Byrne wrote: > > Aw, come on, man! RTFM! I covered this in great detail; you should take the Well I for one don't know which frigging document you're talking about. I've just installed the jre on a system without X. To make the jre even run I found it necessary to install a part of XFree even though I don't want the gui. I can see no mention of this in any of the documentation included with the package jre1.1.6-v4a-i386-libc5.tar.gz. Exercise patience and courtesy in your replies. You too were a beginner once. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Java app without X installed
On Wed, 7 Oct 1998, Steve Byrne wrote: > John Summerfield writes: > > On Tue, 6 Oct 1998, Steve Byrne wrote: > > > > > > > > Aw, come on, man! RTFM! I covered this in great detail; you should take the > > > > Well I for one don't know which frigging document you're talking about. > > I guess README.linux *could* be named more obviously :-) > > > I've just installed the jre on a system without X. To make the jre even run > > I found it necessary to install a part of XFree even though I don't want > > the gui. > > Fixed in v5. > > > I can see no mention of this in any of the documentation included with the > > package jre1.1.6-v4a-i386-libc5.tar.gz. > > Look again in README.linux. > > > Exercise patience and courtesy in your replies. You too were a beginner > > once. > > Yes. And I didn't ask for help before exhausting the other options, if I asked > for help at all. I unpacked the tarball into /usr/java. [summer@os2 summer]$ locate READ | grep java /usr/java/jre/lib/i686/green_threads/README.versions /usr/java/jre/README I'm not going to read all the replies to this thread. Let me say this: I've spent a considerable amount of time helping others over several years. My web pages were at one time taking 10 000+ hits per month when hosted at another IAP. don't know what traffic they take now: I don't use OS/2 much now and don't provide much OS/2 help now either. I stand by my comments: I think Steve was rude. And the doc he says to read ARE NOT in the package I installed. I'd have checked first: I check that my answers are correct when offering advice: if not then I preface my remarks with "I think" or similar indications of uncertainty. In the above case, I'd have 1 Checked that the document exists 2 Mentioned it by name 3 Probably given a short summary. I always quote enough that the reply carries enough question to make sense alone. I always bear in mind that there are likely to be others with similar questions and try to get those too. If anyone wishes to argue my point further, you'll need evidence. Search usenet archives: you'll need to go back a fair way as I've not been very active since August last year. I don't wish to diminish Steve's work: my point is that in this case he could have done better. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Java app without X installed
On Wed, 7 Oct 1998 [EMAIL PROTECTED] wrote: > > On Wed, 7 Oct 1998, Steve Byrne wrote: > > > John Summerfield writes: > > > On Tue, 6 Oct 1998, Steve Byrne wrote: > > > > > > > > > > > Aw, come on, man! RTFM! I covered this in great detail; you should take the > > > > > > Well I for one don't know which frigging document you're talking about. > > > > I guess README.linux *could* be named more obviously :-) > > > RTFM means manpage -- a README file is not a manpage. Plus, isn't it > conceivable that with so many docs, someone could miss something? the M means Manual. The same term is used in other circles where manpages are unheard of. More loosely it refers to any documentation and I for one DO read it. The only README in the package I mentioned refers to Windows and Solaris. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: stop the abonnement
On Thu, 8 Oct 1998, Levente Farkas wrote: > karl, > IMHO it'd be good to add a very easy procmail command for this list, > if a mail contain the word "unsubscribe" forward to the list manager. > we may lose some mail with this, but who wanna write the word "unsubscribe" > on this list (and can aviod a few dozen of such mails). > thanks. A line (or better, paragraph) like this is fair game unsubscribe but if there's any other text, white space excluded, it should NOT lead to the sender being unsubscribed. It would be a bit rough being unsubscribed on account of this paragraph. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: WINNERS
On Fri, 9 Oct 1998 [EMAIL PROTECTED] wrote: If the people running this list can't deal with spam, can you at least leave the headers in so those of us who want to can? Please? > > ** 89% BASEBALL PLAY-OFFS > > ** 68% COLLEGE FOOTBALL > > ** 67% PRO FOOTBALL > > ** CALL FOR TODAY'S WINNING PLAYS! > > ** VEGAS SMART MONEY...LATE INFORMATION PLAYS >DAILY1-900-726-9056 $25/CALL. > > > > > > > > > > > Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Arbitrary Precision Arithmetic?
On Fri, 9 Oct 1998, Dimitrios Vyzovitis wrote: > Hi, > Has anyone implemented any arbitrary precision arithmetic routines in > java? Do you mean someting different from BigDecimal (which is included)? BigDecimal allows numbers of any size with any number of decimal places. It's a rotten way to do decimal arithmetic (overheads MUST be horrendous) but nice when you really want those big numbers. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Zero sized display panes from recompiled jdk
Im getting very small display panes which cannot be resized from my recompiled jdk1.1.6v5. The down-loaded jdk works fine. I suppose i'm not loading lessTif correctly. Any ideas about where to start? -John Campbell
Problems with SuperMojo
Has anyone gotten SuperMojo to work? I've got RedHat 5.0, jdk1.1.5v7. Lots of other Java stuff works. The two things that aren't working are SuperMojo (brand new, version 1.3) and Together/J (also brand new, version 2.0). In both, they start up and some things work, but I can't initiate a new project. SuperMojo puts up a 0X0 window that, when stretched out, is empty and unresponsive, and won't respond to window manager close requests. Together/J puts up a huge, undecorated window in a different virtual screen (below and to the right) from the one I'm working in. It also has no contents and is unresponsive. Any experiences or ideas? Thanks. John Collins
Re: Zero sized display panes from recompiled jdk
>>>> Michael Sinz wrote >John Campbell wrote: >>Im getting very small display panes which cannot be resized >>from my recompiled jdk1.1.6v5. >This is one of the LessTif problems that I know of. From the last >LessTif build I did it still was a problem. Are you your'e saying is that there is no work-around?
IBM & Linux
BM's not an obvious place for linux software, but it's worth a visit. O've found a selection of Java stuff there: jikes - Java compiler, about an order of magnitude faster than javac on my test. I use it almost exclusively. Jikes debugger - GUI written in Java, POP3, SMTP and FTP addons. I just visited http://www.ibm.com/hci/ and finished up at http://www.software.ibm.com/ucd/ - there are some free downloads there too, some useful to java programmers. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Does Sun/JavaSoft support Java on Linux?
On Fri, 23 Oct 1998, Baron Roberts wrote: > > Hi All, > > I just read the java-linux FAQ but need to get a > bit more clarification. As of today, does Sun/JavaSoft > officially support Java and its APIs on Linux? I No > tried to follow the link in question 2.9 but was > asked for a user ID and password. > So register. Registration is free. It makes them happy. You are welcome to report Java problems that are not specific to Linux. At least. I feel as welcome there as anywhere on the www. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: nusty linux exception
On Fri, 23 Oct 1998, Pinca George wrote:
> hello world,
>
> I made a program useing symantec's classes.zip.
>
> It works fine on windows bat when I move it to Linux I get a nusty
> NullPointerException exception on the line :
>
>
> searchButton.setImageURL(symantec.itools.net.RelativeURL.getURL("xx.jpg"));
>
> The image is there
>
> Does anyone ,have a clue ?
Has searchButton been initialised properly? I often get these
NullPointerExceptions when I try to access a method an object doesn't have:
for example, a method in my extension to Frame when the object I'm dealing
with is really a Frame.
If searchButton is supposed to reference an extension to Button that has a
setImageURL method but actually references a Button (extension) that
doesn't have this method, this is what you'll get.
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: request for a sendmail java script
On Thu, 22 Oct 1998, Matt Zagni wrote: > > Hi, > > Please could someone mail me an example of a sendmail java class file > and .java file or even a url where examples are located. IBM of course: http://www.alphaworks.ibm.com/ Look for the smtp packackage. Includes smtp (talks to sendmail) and a gui to write letters and a demo application. While there, get jikes (a faster java compiler) and jikes debugger. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Does Sun/JavaSoft support Java on Linux?
On Fri, 23 Oct 1998, Steve Byrne wrote: > John Summerfield writes: > > On Fri, 23 Oct 1998, Baron Roberts wrote: > > > > > > > > Hi All, > > > > > > I just read the java-linux FAQ but need to get a > > > bit more clarification. As of today, does Sun/JavaSoft > > > officially support Java and its APIs on Linux? I > > > > No > > John is speaking from his knowledge, which is understandably > not correct, as we haven't advertized the fact until now that > Sun has licensed the JDK 1.2 pre-release sources to the porting > team and that we are starting to work with the 1.2 port. Being a > full licensee (just like IBM or that other large company) means that > we have full access to the JDK sources, updated on a regular basis, > as well as the Java Compatibility Kit, so that we can validate that > the Linux port is 100% Java Compatible. I maintain my earlier statement: Sun does not support the API on Linux, just as Sun does not support in on OS/2. Anyone with a problem specific to the Linux implementation should NOT report their problems to Sun. An example of a problem Sun would not want to hear about: "My program dumps core when..." otoh if you try to extend StreamTokenizer and find something in ints implementaion is spastic, tell Sun. Of course, the porting team has special access. It's fair to say that Sun welcomes Java on Linux: it's politically to Sun's advantage to jave bog-standard Java on as many platforms as possible. I imagine Sun sees commercial advantage too: the more java apps there are, the more software there is that runs on Sun boxes. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: CoffeeShop(java IDE)
On Sat, 24 Oct 1998, Ryuji Yokoyama wrote: > Hello All! > > Is anybody using CoffeeShop? please tell me con and pro about it. Is it > worth to spend $85? Costs $0 to try. Only you can decide what it's worth to you. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
A bug in appletviewer
Extrace the attached java applet into a new directory, say ~/javabug. Call
the file "bug.java."
Compile it as you'd compile any java applet, maybe
javac bug.java
Run the applet thus:
appletviewer bug.java
Click your various mouse buttons on the frame window and take not of its
behaviour.
Change directory and run it again: maybe
cd ..
appletviewer javabug/bug.java
Note that this time there's a warning message in the top of the window
warning the user that this is an applet window. You will also note that
some of the text the applet attempts to display is obscured by this
message.
This behaviour is different from that I observe with
Appletviewer on NT 4.0 Sp1 JDK 1.1.4
Netscape 4.something on NT
Netscape 4.06 on Linux
Appletviewer on OS/2 JDK 1.1.4
Netscape 2.02 with Java 1.1.4 on OS/2
and in pictures in my book Just Java, Sun Microsystems.
In all these cases, the warning message is at the bottom of the window: in
principle the user can, if something seems obscured, resize the window to see
what's hidden.
I have
[summer@possum public_html]$ java -version
java version "1.1.6"
I don't seem to have the downloaded file to tell me more precisely.
However, the timestamp on java is "Aug 29 19:07."
Cheer
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
//
import java.applet.*;
import java.awt.*;
import java.awt.event.*;
public class bug extends Applet {
demoMouse F;
public void init() { F = new demoMouse("TestFrame"); }
}
class demoMouse extends Frame implements MouseListener {
String s,s2,s3;
public void init() {
s="test";
s2="";
s3="";
addMouseListener( this);
setBounds(100,300,900,150);
setBackground(Color.white);
show();
}
public demoMouse() {super();}
public demoMouse(String p) {
super(p);
init();
}
public void mouseExited(java.awt.event.MouseEvent e) {;}
// public void mousePressed(java.awt.event.MouseEvent e) {;}
public void mouseReleased(java.awt.event.MouseEvent e) {;}
public void mouseEntered(java.awt.event.MouseEvent e) {;}
public void mouseClicked(java.awt.event.MouseEvent e) {;}
public void mousePressed(MouseEvent e) {
s="test";
s2="";
if ((e.getModifiers() & InputEvent.BUTTON1_MASK) != 0)
s= "button1 pressed" ;
if ((e.getModifiers() & InputEvent.BUTTON2_MASK) != 0)
s="button2 pressed" ;
if ((e.getModifiers() & InputEvent.BUTTON3_MASK) != 0)
s="button3 pressed" ;
if ((e.getModifiers() & InputEvent.ALT_MASK) != 0)
s2="alt pressed" ;
if ((e.getModifiers() & InputEvent.META_MASK) != 0)
s2="meta pressed";
if (e.isPopupTrigger())
s3="Pop the menu";
else s3 = e.toString();
repaint();
} // end method
public void paint(Graphics g)
{
super.paint(g);
g.drawString(s,5,40);
g.drawString(s2,5,80);
g.drawString(s3,5,120);
}
}
Re: A bug in appletviewer
On 27 Oct 1998, Juergen Kreileder wrote:
John> public void paint(Graphics g)
> John> {
> John> super.paint(g);
> John> g.drawString(s,5,40);
> John> g.drawString(s2,5,80);
> John> g.drawString(s3,5,120);
> John> }
> John> }
>
> When working with frames you have to honor the frame's insets.
> The insets include the frames border size, the title bar size,
> the menu bar size, the warning bar size.
> So it's no bug, your code is not 100% portable ;-)
>
> Have you tried to use something like:
>
> Insets insets = getInsets();
> int top = insets.top;
> int left = insets.left;
> super.paint(g);
> g.drawString(s,5 + left ,40 + top);
> g.drawString(s2,5 + left ,80 + top);
> g.drawString(s3,5 + left ,120 + top);
Then let's blame Peter van der Linden whose book, Just Java, doesn't
mention insets in the index nor, so far as I can see scanning chapters, in
any code examples.
I have seen insets mentioned in the JDK documentation, but only in the
context of a ScrollPane.
Seems I've some code to change. Sob.
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: A bug in appletviewer
On Thu, 29 Oct 1998, John Summerfield wrote: > > I have seen insets mentioned in the JDK documentation, but only in the > context of a ScrollPane. > > Seems I've some code to change. Sob. > I'll assume that Insets fixes the problem - it was, in any event, secondary to something else that was bugging me; MouseEvent.isPopuptrigger() always returns false on OS/2 & NT - that's why my sample code was doing mousey things. One thing about this Insets caper bugs me though: I have a Frame: its background is white. I position a window (MenuBar, grey) at the top using the value returned by Insets. On OS/2 it looks fine: on Linux there's a 1-2 pixel white line above the MenuBar. I'll hide the problem for now by changing my colours. btw I have noticed the Insets returned by OS/2 are different. I've also noticed that Insets returns all zeroes if the window's not visible - another spastic idea. Perhaps I'll bug Sun with that one though. Thanks for the help. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Alpha port?
[ Please CC me your reply! ] Hello, I just got a DEC Alpha 21164 box. I'm running Debian on it. I went to the ports page but the port from "Uncle George" is not accessible (permission denied error). Can anybody point me to where I might find the Alpha version, appropriate diffs to compile it, etc, etc. I would really like to get it working on Alpha but can't seem to find the Alpha version anywhere! Thanks, John -- John Goerzen Linux, Unix consulting & programming [EMAIL PROTECTED] | Developer, Debian GNU/Linux (Free powerful OS upgrade) www.debian.org | + Visit the Air Capital Linux Users Group on the web at http://www.aclug.org
Uncle George's Alpha Port
Hi, The archives don't show e-mail addresses or else I'd write to Uncle George directly. Anyway, I understand there is some problem with your Web provider. I'd encourage you to upload your port of JDK to Alpha to sunsite or someplace like that. I believe there is significant interest in it, and it's a shame to have a situation where nobody can use the port to such an excellent machine because your web provider is being nasty. Sunsite has many, many mirrors and once it's on there, this ought to eliminate all future problems. Is there any other way to get it from you? Thanks, John Goerzen
Re: Announce: JDK117 for alpha
(BTW, I tried replying personally to you, but it bounced.) George, I just want to say a BIG thanks for your work on this. This has eliminated one of the things that I was lacking on Alpha, and was sorely missed. No longer! Just a quick question: Does it use green threads or native threads? If not native threads, can we expect such support in the future? Thanks, John On Wed, Nov 11, 1998 at 10:21:05AM -0500, Uncle George wrote: > Yes folks ( I think ) it ready for prime time. There was a binary > version v1, but there was an awt problem that showed up in the ide > "supermojo" ( i have no connection, AND have only a demo model ), and > has been superceeded by the binary files version v2. The classes file is > still version v1. -- John Goerzen Linux, Unix consulting & programming [EMAIL PROTECTED] | Developer, Debian GNU/Linux (Free powerful OS upgrade) www.debian.org | + Visit the Air Capital Linux Users Group on the web at http://www.aclug.org
RE: JDK licensing (was Re: A Scenerio)
On Tue, 10 Nov 1998, Mark H. Wilkinson wrote: > > Could someone explain the differences between a reference implementation and > > a production. > > I suspect that the reference and production JDK releases from Sun are > pretty closely related because the JDK reference implementation _is_ > actually pretty high quality in most areas. When a new JDK 1.1 reference > release is posted the Solaris production release is typically pretty > close behind, presumably because the reference->production patches don't > change much between minor releases. I will speculate here: Buggy code doesn't demonstrate well how things should work: if the spec says "This class works this way" and the reference implementation doesn't do it reliably, it's not a lot of good as a reference. Clearly Sun needs to get the reference implementation out the door AND Sun doesn't want to be sued if it breaks, So, they say, "Don't bet your business on this - it could fail." If the reference implementation IS of good quality, it makes porting (including the work done by the Solaris folk) easier and so promotes the use of java. When java comes from the Solaris people (or from IBM for OS/2 etc and so forth) it's going to be used by people who ARE betting their business on it and the quality perception is different. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Low level disk access
On Wed, 11 Nov 1998 [EMAIL PROTECTED] wrote: > While this is not directly a Java question does anybody know of any > library functions etc which will provide low level disk access to > harddisks. i.e sector access etc. > > I am looking at writing a Java class for providing direct access to a > dedicated harddisk so that I can avoid some of the overhead of Java and > normal filesystem based disk access. On Linux, a disk is a file like any other. Use /dev/hda for the device, /dev/hda1 for the partition. Be aware if you give youself write access and stuff it up, the disk content is effectively destroyed and you'll need to repartition & reformat. However, if the overhead's a problem, Java's not the language to use: it's full of overheads. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Retrieving a fully qualified hostname under NT
On Thu, 12 Nov 1998, Mark Hofmann wrote: > Hi all, > > I've posted that question a few days ago to java.lang.programmer > but got no response :-( > But, since this problem is really bugging me, and there might be some > people in this list that run there progs under NT and Linux, just as me, > I hope to find a solution here. > Anyway, here is the problem, easy and simple > > when I use > String localhostname=(InetAddress.getLocalHost()).getHostName(); > under Linux I get the fully qualified hostname, under NT just the local > hostname, eg. host instead of host.my.domain.com > I've seen divergent behaviours too and never been quite sure. I suggest that, having got your host name you then get its IP address and then do a reverse lookup to get the fully-qualified name associated with that IP address. Of course, this might not always give the desired result: if an IP address is associated with more than one hostname (a valid confguration) you'll need to decide for yourself how important that is. Here's what might happen here: [summer@emu JAVA]$ nslookup proxy Server: localhost Address: 127.0.0.1 Name:emu.os2.ami.com.au Addresses: 192.168.1.2, 192.168.0.2 Aliases: proxy.os2.ami.com.au [summer@emu JAVA]$ nslookup 192.168.1.2 Server: localhost Address: 127.0.0.1 Name:emu.os2.ami.com.au Address: 192.168.1.2 Well, actually that machine's "hostname" command returns emu.os2.ami.com.au, but it illustrates the point. Be aware the domain names might not match either: I could configure a domain charlie.ibm.com and map it to the same IP addresses I use for os2.ami.com.au. I imagine I can provide both reverse mappings too: that could really confuse things. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: [Java-Linux] about this list ...
On Thu, 12 Nov 1998, olivier Dulac wrote: > > Hi, > > Sorry for this isn't a java-related message, but it a few suggestions > about the list itself, and people using it... > > 1) include [Java-linux] in your Subject:, for every mails to this list > >*please* put a header in front of every email for the mailing list >(exemple: "Subject: [Java-linux]". Been through this on another list. Consensus there was that it's a rotten idea. Plenty other headers to use to filter. Causes problems with many mailers: folk can't see the really interesting part of the subject which is NEVER [the name of this mailing list inserted here] I suggest sorting on any of these: >From [EMAIL PROTECTED] Fri Nov 13 23:00:37 1998 Return-Path: <[EMAIL PROTECTED]> X-From_: [EMAIL PROTECTED] Fri Nov 13 06:38:07 1998 Resent-From: [EMAIL PROTECTED] X-Mailing-List: <[EMAIL PROTECTED]> archive/latest/3476 X-Loop: [EMAIL PROTECTED] Resent-Sender: [EMAIL PROTECTED] > (And i hope i won't be flamed, these are only my opinions after all... ;^) I don't flame people. However, if I see a stoopid idea I say so, and your first qualifies. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Mouse Clicks
On Sat, 14 Nov 1998, Karthik Vishwanath wrote: > Hi all, > On receiving a mousePressed, how does one determine which mouse > button is the one clicked? > > ps: i am sorry if this is a purely java based question, but then i do not > know where else i can post it. > See javap java.awt.event.MouseEvent javap java.awt.event.InputEvent Use getModifiers() in conjunction with the info the above commands provide. Read the online docs for more. Be aware that isPopupTrigger() is active on mousepress on Unix, mouserelease on WIN* and OS/2. This is consistent with the host environment, but a great way to trick programmers into writing non-portable code. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Linux JDK Bug -
On Sat, 14 Nov 1998, Jim Arlet wrote: > Hello, > > Sorry to have to bother you with support questions, but I am an experienced > Java developer on Solaris, and I have been trying to get your JDK port to > run on my Linux machine, without luck. I have tried both the glibc and > libc5 versions of the JDK, and both have been unsuccessful, yet with > different results. > > My machine is configured as follows: > > Linux version: Red Hat 5.0 (Kernel 2.0.32) > Platform: Intel (P90) > JDK:jdk1_1_6-v2-glibc_tar.gz (first try) > jdk_1_1_7-v1a-libc5-x86_tar.gz (second try) > Path: my_install_dir/bin > > I assume I have glibc, because when I enter the "ls -l /lib/libc.so.*" > command is see the following: You certainly have glibc. I suggest you get ftp://sunsite.unc.edu/pub/Linux/distributions/redhat/contrib/hurricane/i386/jdk-sn-1.1.6-1.2glibc.i386.rpm which should install without any drama. If you have a fast connexion, you could even do this: rpm -i ftp://sunsite.unc.edu/pub/Linux/distributions/redhat/contrib/hurricane/i386/jdk-sn-1.1.6-1.2glibc.i386.rpm If you want online docs, your solaris docs are fine. ps Quite likely you have libc as well: I understand it gets installed as a compatibility measure as glibc isn't binary-compatible with libc. This command will tell you: rpm -q libc glibc and maybe return libc-5.3.12-24 glibc-2.0.7-19 Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Basic installation. Please Help.
On Sun, 15 Nov 1998, Kenny Freeman wrote: > Hi, you have to add the path to the jar files like (for example): > > export > CLASSPATH=$CLASSPATH:/usr/local/swing/swing.jar:/home/freeman/jars/net.jar:etc > > that should work a little better for you (once you've added the path to > each jar you want to use). I do hope someone comes up with a better idea than classpath RSN: mine runs to well over 2K. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
BDK
Some time ago I downloaded BDK from Sun: it's a little while ago, but I
seem to recall three choices:win32, Solaris, and "run anywhere."
Run Anywhere makes sense to me (I run linux, OS/2 and NT) and i thought I
had that version.
What I really have is a file called BDKJul98.bin and which includes a shell
script for Solaris (was I really so stupid?) (a really tricky shell script)
that picks bits out of the rest of the file, including JRE for Solaris. It
then uses the jre to run the installer.
Havin thought about it for a while I hit on the idea of extracting the
script and fixing it: it sets the path, uses absolute paths to where
programs such as tail, dd etc live on Solaris (not the same as on Linux)
and using my Linux jre to run the installer.
After a couple of false starts I got into the installer, but then i saw
1 A blank licence (easy to agree with that)
2 blank "important information" (that's cool, can't be any).
3 Telling the installer where to install is decidedly tricky: I can't read
where it thinks it's installing.
I do see a few (quite a few) of these:
Warning:
Name: textfield
Class: XmTextField
Character '/' not supported in font. Discarded.
Warning:
Name: textfield
Class: XmTextField
Character 'u' not supported in font. Discarded.
Warning:
Name: textfield
Class: XmTextField
Character '/' not supported in font. Discarded.
Cute. What do I do now? While you folk are laughing yourselves silly at my
stupidity, I'm going back to have another look at sun just in case.
If sun's not the best place to get BDK, then where?
Oh yes, I'll attach my fixed script just in case someone finds it useful:
hasn't done me any good yet.
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
#!/bin/sh
ARCHSIZE=1625742
JRESIZE=2586013
ARCHANDJRESIZE=4211755
BDK=/u/summer/BDKJul98.bin
#d17
echo "InstallAnywhere is preparing to install..."
PPWD=`pwd`
if [ -x /tmp ] ; then
INSTBASE=/tmp/
else
INSTBASE=./
fi
cd $INSTBASE
mkdir install.dir.$$
cd install.dir.$$
mkdir Solaris
cd Solaris
mkdir resource
cd resource
cd "$PPWD"
jreTCname="$INSTBASE"jre115req.tar.Z
jreTname="$INSTBASE"jre115req.tar
jrejavahome="$INSTBASE"jre1.1.5
JRESIZEPARAM=-"$JRESIZE"c
blah=`tail $JRESIZEPARAM "$BDK" > $jreTCname` || \
{
echo "Can't run tail; cannot install."
exit 1
}
if [ -x /usr/bin/cksum ] ; then
cksum=`/usr/bin/cksum $jreTCname`
index=1
for s in $cksum
do
case $index in
1) thesum=$s;
index=2;
;;
2) thesize=$s;
index=3;
;;
esac
done
if expr $thesize != $JRESIZE ; then
echo ""
echo "The download file appears to be corrupted. Please try"
echo "to download again, and make sure you are using 'binary' mode."
echo ""
echo "Please do not attempt to install this archive file."
exit 2
fi
fi
cd "$INSTBASE"install.dir.$$/Solaris/resource
blah=`/usr/bin/uncompress $jreTCname` || \
{
echo "Can't run uncompress; cannot install."
exit 3
}
blah=`tar xvpf $jreTname` || \
{
echo "Can't run tar; cannot install."
exit 4
}
rm -f $jreTCname $jreTname ;
cd "$INSTBASE"install.dir.$$
mkdir InstallerData
cd "$PPWD"
archname="$INSTBASE"install.dir.$$/InstallerData/Installer+JRE.zip
ARCHANDJRESIZEPARAM=-"$ARCHANDJRESIZE"c
blah=`tail $ARCHANDJRESIZEPARAM "$BDK" > $archname` || \
{
echo "Can't run tail; cannot install."
exit 5
}
cd "$INSTBASE"install.dir.$$/InstallerData
blah=`/bin/dd if=Installer+JRE.zip of=Installer.zip bs=$ARCHSIZE count=1 ` || \
{
echo "Can't run dd; cannot install."
exit 6
}
rm Installer+JRE.zip
cd "$INSTBASE"install.dir.$$
CLSPTH=InstallerData:InstallerData/Installer.zip:Solaris/resource/jre/lib/rt.jar
jre -classpath $CLSPTH com.zerog.ia.installer.Main
cd $INSTBASE
rm -rf install.dir.$$
exit 0
Re: Basic installation. Please Help.
On Tue, 17 Nov 1998, Pavel Tolkachev wrote: > > IMO, the cross-platform capabilities of java are seriously hurt by all the hoops > > that end-users have to jump through (setting up classpaths, .sh/.bat files, etc) > > just to get someone elses application to run. Admittedly, java installshield and > > the like do do a good job of setting this stuff up, but a more general (and built > > in) solution is required. > > > > [ bryce ] > According the Sun (unfortunately I do not remember URL) the 'preferred' > way to specify classpath is do not set CLASSPATH environment variable at > all (namely because it is not portable) but to use -cp and -classpath > for jre start (apparently -classspath is for 'everywhere used' packages > like Swing, probably JGL and jpython) and -cp for the application > classpath itself and other its specific packages. Anyway you will > usually have non-portable scripts to start applications... probably > concatenating command line from other environment variables not > deprecated by Sun like SWING_RELEASE_PATH :). For development (where we > need to use java instead of jre) all should be put in -classpath, I > beleive. My 2K CLASSPATH is a problem in itself: I did a bit of creative scripting to do it, and had do do the same thing on OS/2 (the OS/2 commndline can't be that long before substitutions) Worse, though, is the time Java spends searching through all the directories, jars and zips (and those zips causes some confusion to the script - some contain source, not classes). If I developed in C++ on OS/2 using OS/2's Visual Age C++, I would have tools to repackage and redistribute necessary DLLs. If I use all the bits of java I have here (and it's all free), and I create something roughly equivalent to hotjava using the POP3, SMTP & FTP bts I got from IBM, Swing 1.1, collections 1.1 (SUN) then my distribution's getting cumbersome and may not perform well either on some platforms, at least during startup. Worse yet, an applet version might be truly appalling, at least to users on the wrong side of a modem. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: BDK
On Wed, 18 Nov 1998, John Summerfield wrote: > What I really have is a file called BDKJul98.bin and which includes a shell Laugh on. I now see a file called BDKJul98.zip which is smaller and (presumambly) really will run anywhere. Now, if someone can stand up, salute their national flag and say "It's Sun's fault," we can all laugh them. I suppose it won't happen. Now. where did I put the Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Keyboard input problems
Greetings - I am trying to evaluate a large Java package from Object International called Together/J. It comes up, analyzes classes, and generates documentation both with the jdk1.1.7v1a interpreter and with the tya compiler. With the compiler, it's performance is tolerable, except that it can't really handle keyboard input in most input contexts. If you type r..e..a..l..l..y s..l..o..w, it works OK, but as soon as you type at a normal pace, the characters get scrambled badly -- you don't lose characters, but they are out of order. It's very repeatable. I'm writing to java-linux about this because the same app on NT works fine. About the same performance, but keyboard input works. Is this a known problem? Environment: RH 5.0, kernel 2.0.34, glibc 2.0.7, Dell Pentium2/266, 128 meg RAM. Cheers - John Collins University of Minnesota
Re: Keyboard input problems [and TYA...]
Bryce McKinlay wrote: > > Hi, > > I had this problem as well - it appears to be specific to Together/J on the > Blackdown JDK when using green threads. If you set THREADS_FLAG=native then > the problem goes away. The problem does not appear with the Open Group JDK > (which also uses native threads). > > On the subject of native threads, they do benchmark a little slower due to > increased overhead, but in most cases application > performance/responsiveness feels faster - I find that text inputs and > editors in particular are snappier. > > ... OK. I loaded and installed the native threads package. Now Together runs much slower than it did before I installed TYA, but the keyboard input problem is fixed. I guess I haven't found any combination that makes it really usable. Has anyone else gotten Together working to their satisfaction on a reasonably large project (160 classes, 8 packages)? John Collins University of Minnesota
Re: Java programming env (newbie question)
On Mon, 23 Nov 1998, Matt Welsh wrote: > > You probably have Kaffe installed, which is a free Java Virtual Machine > replacement. > the command rpm -qf `which javac` will confirm. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Tried to free bogus memory
this is an applet and the problem occurs in appletviewer This does NOT happen in netscape (Java 1.1.5). [summer@possum os2]$ ./r os2.Ftpenquiry$1[,0,0,10x280,invalid] free: double-freed at 83F8748 free: double-freed at 83F87A0 free: double-freed at 83F9900 free: double-freed at 83F9958 Tried to free bogus memory 83ffdd8, ignored Tried to free bogus memory 83ffe30, ignored Tried to free bogus memory 83ffe88, ignored Tried to free bogus memory 83ffee0, ignored free: double-freed at 8401BD8 free: double-freed at 8401C30 README.linux for JDK 1.1.6, Version 4 08/28/98 This is the Blackdown Java-Linux port of JDK 1.1.6 to Linux. It extends Steve Byrne's JDK 1.1.6 Version 2 port. RedHat 5.0 glibc If you need source say so now: it's latish at night and I wanna sleep. I'll have your reply if you're reasonably prompt in the morning (and if all goes well, jdk117). If I don't hear in time I'll asome you think t might be fixed and I'll ignore it, at least till I have jdk117 in place. I can give it a try on OS/2 if needed: it's got IBM's latest 116 (and I'm trying to get 117 for that too. I might be sleeping, but my computer will be hard at it.) Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: How to Write MakeFile ?
On Wed, 25 Nov 1998 [EMAIL PROTECTED] wrote: > Heres a simple Makefile I use. > > BIN = /home/jarvi/java/gps/bin/ > JAVAC = javac > JFLAGS= -d /home/jarvi/java/gps/bin -O > JFILES := $(patsubst %.java,$(BIN)%.class,$(wildcard *.java)) > SUBDIRS = almanac misc route trackpoint waypoint rxtx > > .SUFFIXES:.class .java > > $(BIN)%.class: %.java > $(JAVAC) $(JFLAGS) $< > > all:$(JFILES) $(SOFILES) > @set -e; for i in $(SUBDIRS); do $(MAKE) -C $$i; done I use a script, typically in which ever directory's most likely to be convenient. The script will typically have the command ... better, here's one: #!/bin/bash jikes -d ~/Classes MyStocks.java \ Records/*.java \ Reports/*.java \ Reports/*/*.java \ || exit java Stocks.MyStocks which will probably complete before javac's compiled anything. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: CLASSPATH confusion
On Tue, 24 Nov 1998 [EMAIL PROTECTED] wrote:
> Having it "fixed" in 1.2 won't help us, as Jikes is designed to run with
> all but most ancient JDK's. You supply the classes.zip (or jar, or
> whatever), we provide the bytecode. Sigh.
I imagine you'll think of a switch such as +1.2 to give it 1.2 behaviour.
If you decide to use a jikesrc file, can we please start a nw conventiion?
my home dir's getting litterd with .*rc files. Let's start putting them in
~/.etc - if the directory exists, use it. If it doesn't and there's no
existing config file, make it.
Now my contribution to the CLASSPATH problem. I drop all my java stuff into
/usr/java. I thought of scanning from there for *.jar and *.zip but
discovered most of the zips are source, so gave scanning for them.
I decided doing it every time's not so bright either as it's not especially
quick and doesn't change all that often, even though I'm a bit of a
software junkie. Besiddes, I don't generally want demo jars in my
classpath.. So I scan as needed (with the find command) to create a list of
jars.
In /etc/profile.d I have
drwxr-xr-x 2 root root 1024 Oct 21 20:37 .
drwxr-xr-x 22 root root 3072 Nov 25 15:20 ..
-rw-r--r-- 1 root root 2499 Oct 21 20:34 java
-rwxr-xr-x 1 root root 485 May 17 1998 jdk.csh
-rwxr-xr-x 1 root root 775 Oct 21 20:37 jdk.sh
"java" contains a list of my jars, created with a find command then editted
to remvoe the obvious junk.
Here's how I use it:
[summer@possum profile.d]$ cat jdk.sh | grep -v ^#
if [ -z "${JAVA_HOME}" ] ; then
export JAVA_HOME=/usr/lib/jdk-1.1.6
fi
if [ -z "${JDK_HOME}" ] ; then
export JDK_HOME=/usr/lib/jdk-1.1.6
fi
SW=`cat /etc/profile.d/java`
SW=`echo ${SW} | sed -e "s/ /:/g"`
MYPATH="${JDK_HOME}/lib/classes.zip:${SW}"
export CLASSPATH="${MYPATH}"
MYPATH="${JDK_HOME}/bin/"
echo ":${PATH}:"| grep "${MYPATH}" >/dev/null ||\
PATH="${PATH}:${MYPATH}"
export PATH
For those who haven't twigged, in a convoluted sequence of events, bash
gets to run /etc/profile.d/*.sh (and csh runs /etc/profile.d/*.csh) during
login processing. This is on RedHat - it may work a little differently on
other distributions.
I have to do it a little differently on OS/2 which can't cope with the
commandline size needed to set all these jars at once.
Here's the find command to locate the jars. The list will need some
tweaking to weed out demos and any other junk.
find /usr/java -name *.jar
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: Using "setSize" with a Frame.
On Fri, 27 Nov 1998, Karthik Vishwanath wrote:
> Hi all,
> when i create an application as :
> public class Myapp extends Frame
> {
> public Myapp()
> {
> super("MyApp");
> setSize(100,100);
> show();
> }
> }
>
> and run it, the setSize does not resize the frame.
>
> Can anyone tell me what must i do inorder to resize a frame?
>
> -Karthik.
Karthik
I did expect to see more sensible replies than I have thus far.
I expanded your code to what you SHOULD have posted - with this example,
the whole program isn't excesive.
This works here:
import java.awt.*;
class Myapp extends Frame
{
public Myapp()
{
super("MyApp");
setSize(1100,100);
show();
}
public static void main(String [] s)
{
Myapp m = new Myapp();
// m.setSize(1100,100);
// m.show();
}
}
I've not read through the docs, but setSize(100,100); does not resize here
either.
btw Arguably the setsize/show are better done in the main() method: this
help give you the flexibiliy of using the app in a different context and
perhaps setting its frame to another size.
Sometimes it's a good idea to use getToolkit() to find the underlying
screen size and making your window some proportion of this so it is about
the same physical size regardless of screen resolution.
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: Using "setSize" with a Frame.
On 29 Nov 1998, Juergen Kreileder wrote: > It opens a frame with the size 1100x100. You and I should read more closely. 1100x100 works correctly here. 100x100 doesn't. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Using "setSize" with a Frame.
On 29 Nov 1998, Juergen Kreileder wrote: > Ok, the code does exactly what it's supposed to do on my machine: > It opens a frame with the size 1100x100. > > What window managers do you guys use? What JDK version? fvwm2 -cmd FvwmM4 -debug /etc/X11/AnotherLevel/fvwm2rc.m4 XFree86-3.3.2-5 RH 5.0++ JDK116 (117 has arrived here and will be installed RSN) Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Stupid PopupMenu question (retry)
On Sat, 28 Nov 1998, Thomas Okken wrote:
> I'm trying to make a PopupMenu appear in a Component (a subclass of
> Canvas, to be precise).
> I create the menu in my Component's constructor:
>
> mypopup = new PopupMenu();
> mypopup.add("Blah");
> // etc...
> mypopup.addActionListener(this);
>
> In my processMouseEvent method, I check for mouse clicks in a certain
> place, and then I try to pop up the menu by saying
>
> mypopup.show(this, X, Y);
>
> The result is this:
>
> Exception occurred during event dispatching:
> java.lang.NullPointerException: parent is null
> at java.awt.PopupMenu.show(PopupMenu.java:92)
> at MyComponent.processMouseEvent(MyComponent.java:358)
> at java.awt.Component.processEvent(Component.java:2249)
> at java.awt.Component.dispatchEventImpl(Component.java:1781)
> at java.awt.Component.dispatchEvent(Component.java:1708)
> at java.awt.EventDispatchThread.run(EventDispatchThread.java:81)
>
> The documentation for PopupMenu.show() mentions a "parent", but I can't
> figure out how to
> set it.
add(mypopup);
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: Using "setSize" with a Frame.
On Sun, 29 Nov 1998, Pavel Tolkachev wrote: > > Also be sure that your window manager ALLOWS so small Frames at all. > Otherwise, peer can prevent your setSize() from doing its work. If I use setSize(100,100); I get a frame that's smaller than that: something like 90x72 I think: I've deleted the source and can't be bothered retyping it. However, by changing setSize(100,100); to setSize(1100,100); and nothing else, the program works as expected. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: FUCKING MAIL LIST
Is there any particular reason that you ignore the instructions sent when you subscribed, fail to make use of the information available to you at the website for these lists, and further proceed to make a royal fool of yourself while endeavoring to inappropriately insult the list and others on it? On Sat, Nov 28, 1998 at 05:28:28PM -0600, apersil wrote: > TELL ME HOW TO REMOVE MYSELF FROM THIS FUCKING MAIL LIST! -- John Goerzen Linux, Unix consulting & programming [EMAIL PROTECTED] | Developer, Debian GNU/Linux (Free powerful OS upgrade) www.debian.org | + Visit the Air Capital Linux Users Group on the web at http://www.aclug.org
Re: jdk1.2
On Mon, 30 Nov 1998, Troy Wu wrote: > On Mon, 30 Nov 1998, Singleton, Terry wrote: > > Is there any sense of when the JDK will be ported once 1.2 is released. I am > > Perhaps the best place to track JDK-1.2 development is at the Sun site > since they have offered to port JDK-1.2 to Linux. I read the article at Sun rather carefully. It does not lead me to believe that Sun is doing thw writing, but rather that the folk here at Blackdown are given similar standing to other third-party porters such as IBM. I've also seen a report to that effect here. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Idiot Trap!!!
On Mon, 30 Nov 1998 [EMAIL PROTECTED] wrote: > Just an idea. > > It appears that an automated mailing list has become the ideal tool to > trap idiots. Any idiot can somehow manage to subscribe themselves, but, > even though they are given unsubscribe information during the I'm on several mailing lists, and subscribe/unsubscribe methods vary. I can't remember how to get off any particular list. Most list managers include unsubscribe info at the foot of every message: on some lists, it's as simple as echo unsubscribe | mail > hours ago ([EMAIL PROTECTED]), are bound to get struck on the mailing list. If you emailed [EMAIL PROTECTED], you will have found this idiot had an altogether different agenda. Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: JDK install
On Wed, 2 Dec 1998, Jaimini Ram wrote: > > hi, > > I downloaded the jdk for linux from blackdown.org and installed it. > But when i say 'java' in the jdk bin directory, it gives the message : > > ./../bin/i586/green_threads/java: can't load library 'libXpm.so.4 > > X windows is NOT installed, so is there any way to make it run without > requiring x libs ? [summer@possum jserv0.9.11]$ locate README | grep jdk then see Java Virtual Machine variations --- There are three versions of the Java interpreter that are shipped with this release. The default one depends on having X libraries present on your system, and has Motif statically linked in (so it does not require Motif to be installed on your machine). One other version has no dependency on X, and -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: The bug is still there (and problems with the 1.1.7-v1a) (Re:A bug in the parse me
On Tue, 1 Dec 1998, Kontorotsui wrote: > > On 26-Nov-98 Juergen Kreileder wrote: > > [About parse method] > > >Get 1.1.7-v1a and try again, 1.1.7 has some bugfixes for the parse methods. > > I got the 1.1.7-v1a and the bug is still there. The blanks are not skipped, > while in the docs it is clearly written that they should be. > > >The parse-methods are 100% pure Java, so if 1.1.5 Linux and 1.2 Solaris > >show different behavior it doesn't have to be a Linux bug, but it means > >that there have been some changes to the java code between 1.1.5 and 1.2. > > Yes, probably it's not a problem with the Linux port, but is a fact that > the behaviour is different from what I read in the docs. > > Should I do a bug report to the Java developers? That is a good idea. It would be sensible first to verify it in an applet in Netscape (which does not use the Blackdown port). > > > Anyway, here is a real Linux port JDK 1.1.7-v1a problem. > > I get this message: > > /usr/local/java/bin/../bin/checkVersions: /tmp/ldd.out.889: Permission denied This is more usually a permissions problem, though it's odd that it happens with root. Can you, from the commandline, execute the command ls >/tmp/Clancy.was.here Does this show anything odd? ls -dl /tmp For reference, I get this: [summer@emu summer]$ ls -dl /tmp -d drwxrwxrwt 8 root root 3072 Dec 3 07:06 /tmp You can bypass this by running export JDK_NO_VERS_CHECK=don\'tbother or the csh equivalent. > > every time I run java and javac. The number seems to be the PID and this > happens with my standard user and the superuser. > Beside the annoying message, anything works fine. > -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: nested classes, packages, and javac depend
On Fri, 4 Dec 1998, Steve Cohen wrote: > I find that I'm unable to get rid of the following errors when I do a > javac depend of my whole project: > > error: File /usr/local/java/xyz/Scheduler.java does not contain nested > class xyz.Scheduler. 1 as expected. Please adjust the class path so that > the file does not appear in the package xyz. > > > What the hell is this trying to tell me? > Of course Scheduler.java doesn't contain the class xyz.Scheduler 1. > It contains xyz.Scheduler$1 > > I've looked and looked. There appears to be nothing wrong with my code. > > Is this a known bug in javac in jdk 1.1.6? spend a few minutes getting jikes from http://www.alphaworks.ibm.com/, but wait until about Monday there. The current version is 0.40 and you do not want that one. 0.41 is promised for Monday: I have it here and it seems fine. Whether javac is broken or not, in such cases it's good to get a second opinion. Besides, jikes is about ten times faster: I kid you not. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Servlets ??
On Fri, 4 Dec 1998, Markus Fritz wrote:
> If you want to run servlets with your apache-webserver, you need
> a servlet-engine like JServ (java.apache.org) or JRun
> (www.livesoftware.com).
> Both come with JSDK (JavaServletDevelopmentKit from SUN) included.
>
the JSDK with jserv is oldish. You can get a newer one from Sun.
I found the instructions with jserv hopeless (and the list of email
addresses to report problems/fixes no better). However, I have it running
(but no servlets no run yet). I'l attaching a couple of scripts I used to
build jserv and apache 1.3.3.
Note the scripts are as I used them and run on my computer with bas 2.0. If
you're using bash 1.4 or some other script you may need to make some
changes.
--
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
#!/bin/bash
# BuildJserv
export CLASSPATH=${JDK_HOME}/lib/classes.zip
AVER=1.3.3
JVER=0.9.11
Apache=${1:-~/updates/apache_${AVER}.tar.gz}
Jserv=${2:-/redhat/incoming/jserv${JVER}.tar.gz}
cd
[ -d apache ] || mkdir apache
cd apache
CD=`pwd`
[ -d ${CD}/apache_${AVER} ] && rm -rf ${CD}/apache_${AVER}
[ -d ${CD}/jserv${JVER} ] && rm -rf ${CD}/jserv${JVER}
echo untarring ${Apache}
tar -zxf ${Apache} || exit 4
echo untarring ${Jserv}
tar -zxf ${Jserv} || exit 4
CONF=`find ${CD} -name Configuration`
echo editting ${CONF}
##ed ${CONF} <
#!/bin/sh
#cd /usr/src/redhat/SOURCES/apache_1.3.1
./configure \
--prefix=/usr \
--sysconfdir=/etc/httpd/conf \
--localstatedir=/var \
--enable-suexec \
--suexec-caller=nobody \
--suexec-userdir=cgi-bin \
--suexec-uidmin=500 \
--suexec-gidmin=500 \
--suexec-safepath="/bin:/usr/bin" \
--logfiledir=/var/log/httpd \
--enable-module=most \
--enable-shared=max \
$@
make
Re: The bug is still there (and problems with the 1.1.7-v1a) (Re
On Thu, 3 Dec 1998, Kontorotsui wrote: > > On 03-Dec-98 John Summerfield wrote: > >> [About parse method] > >> Should I do a bug report to the Java developers? > > > >That is a good idea. > >It would be sensible first to verify it in an applet in Netscape (which > >does not use the Blackdown port). > > I verified that the bug is there also in a Solaris 1.2 JDK, which should have > been released directly by JavaSoft. Should this be enough as a proof that it's > not a problem with the Blackdown port? Sounds heaps to me. Can't blame Blackdown for bugs in software they've not even seen. Mention to Sun the environments where you've produced the problem. > > >You can bypass this by running > >export JDK_NO_VERS_CHECK=don\'tbother > >or the csh equivalent. > > It works, thank you very much. > > Just another question, please, can I miss something important by asking to the > JDK not to bother with the version check? > If there is a way, I'd like to help you (or who takes care of the checkVersions > script) to fix this little bug. I think the script is just making sure your not mixing different releases of java. If you know you're not, then you know the test isn't needed. Mind you, anyone who's been programming more than a few days has seen someone who "knows" something proven wrong. However, as it's working you probably don't have a problem. go win fame and infamy by telling Sun about their problem. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: nested classes, packages, and javac depend
On Fri, 4 Dec 1998, David Warnock wrote: > Steve, > > > I find that I'm unable to get rid of the following errors when I do > a > > javac depend of my whole project: > > I have read soimewhere sometime that javac -depend is not reliable. > > Can I suggest you download the JIKES compiler from IBM. It's free, it > runs an order of magnitude faster than JAVAC and so far as I can see > the -depend works. You may want to use the +E compiler switch to get > errors in the same/similar format to JAVAC. > The address is http://www.alphaworks.ibm.com/formula/Jikes/ (it's > available as a link from the blackdown site). but wait til after the weekend. The current version, 0.40 has a problem with the -d switch (no worries if you don't use it). 0.41 is supposed to be in place Mon. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: jikes vs. javac
On Fri, 4 Dec 1998, Axel Zeilner wrote: > yes jikes is an impressing alternative for development of large java projects > but beware of jikes bytecode running on symantec jit compilers. > > the bytecode is according to IBM obeying the specifications of sun but doesn't > run on windows symantec JIT VMs. we got multiple exceptions and system > crashes with our applications. this 'works' on sun jdk 1.1.6 for windows and > netscape 4.05 (builtin VM). the only workaround is to disable the jit by > renaming the symjit.dll and compilation with javac (see towerJ for getting > performance with sun javac). What does David say about this? -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: oracle
On Fri, 4 Dec 1998, Jinpeng Xie wrote: > Hi, > > I got news about ORACLE server for linux. Does Oracle finish > porting ORACEL SERVER to Linux? When is it available to us? Go visit oracle & see. There's quite a lot to download: [summer@emu summer]$ ls oracle/ total 142919 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:16 805ship.1 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:18 805ship.2 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:18 805ship.3 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:19 805ship.4 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:19 805ship.5 -rw-r--r-- 1 summer summer 20971520 Oct 10 06:20 805ship.6 -rw-r--r-- 1 summer summer 19214240 Oct 10 06:21 805ship.7 -rw-rw-r-- 1 summer summer 726035 Nov 14 17:22 oas_install.pdf -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: [FYI] Java violates US patent?
On Mon, 7 Dec 1998, Christopher Hinds wrote: > So What? Ever heard of a spreadsheet product called Visicalc. Visicalc > ran > on CPM OS( Late 70s early 80s). The company( the name excapes me ) that > developed Visicalc held the patent on the spreadsheet type GUI's also. > Then along came Lotus 1.2.3 , MuliPlan , and even MS Excel. > I have not seen any of the products removed from the market by patent > infringement violations. Vision Software Inc I think. Was bought by Lotus while Lotus was suing Borland and others re L&F of 123. > > Ernst de Haan wrote: > > > Found a link to this article on JavaLobby: > > > > http://www.pbs.org/cringely/pulpit/pulpit19981203.html > > > > Could turn out bad. Very bad. For those who've not seen it, this seems to be most relevant. On November 17th, little Eolas came into effective control of U.S. patent number 5,838,906 for an invention described as "a distributed hypermedia method for automatically invoking external application providing interaction and display of embedded objects within a hypermedia document." Say that fast three times. The patent is held by The University of California in the name of inventors Michael Doyle, David Martin and Cheong Ang. Doyle is the CEO of Eolas, which is the exclusive licensee of the patent. I've been using OS/2 for some years, since before the WWW existed. OS/2 manuals are distributed as hypertext documents. One of the capabilities of the viewer is the ability to launch external programs which could be the program being described, or perhaps an audio supplement or other special viewer. As others have remarked, the idea of executable tokenised code in one form or another is not new. I used Software AG's Natural on IBM-compatible mainframes in the early 80s and Natural has a compilation phase which creates tokenised object code. Point & shoot was in use to some extent then too: no mice, but we had light pens to click on onscreen selectable fields not very different from the hypertext links we use now. ISPF was around in te late 80s too, and it did do hypertext and run supplementary application code that could be scripted (MVS command procedures and the like) or coded in compiled languages such as PL/1, COBOL and assembler. So the general idea is not new, but HTML is no more than an implementaion of the ideas embodied in ISPF dialogues. Syntactically, they're about as similar as C and COBOL. Arguably the client-server notion's covered with Natural, an early fourth-generation programming language. It provides a simplified programming environment for access to databases which need not be on the same computer. Natural runs under the control of a TP monitor (equates somewhat with web browser). Screen layout is ordinarily done from Natural as one can with Java. If anyone ever bothered to run a Natural program in ISPF (entirely possible in the late 80s) then they've done pretty much everything mentioned in Cringle's article. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: [FYI] Java violates US patent?
On Tue, 8 Dec 1998, Alex Pozgaj wrote: > > Just a minor nit-pick: not every implementation did that. > > Let me guess... you too are comming from the Sinclair world, aren't you? miscrosoft (mbasic) did. cbasic did. Natural & Adacom did it in 1980. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Killing Applets in browsers
On Sun, 13 Dec 1998, Leo Cyr wrote: > If you use HotJava (current is 1.1.5) for some reason ;-) it automatically > re-downloads the classes/jars specified in the Applet tag. I'm not sure why > -- I guess it is realted to the browser's cache, but caching of html is > still on... It just works... > > Christopher Hinds wrote: > > > Since the browser caches HTML pages in a history list of some type, > > every > > you time leave and return to the HTML page, the applet will be activated > > ( start() ) and deactivated ( stop() ) respectively. Therefore you can > > try clearing the cached pages which should force the embedded JVM > > to call the destroy() on the applet. Or you can uses JavaScript to > > clear the browser cached history > > list if the browser supports JavaScript. > > > > Cheers > > Chris > > > > Thor Erik Karlsen wrote: > > > > > This is a proplem that has been irritating me since the day I started > > > developing applets(not that long ago). Is there a WORKING way an > > > applet can destroy itself and tell the browser that its dead? I always > > > have to start the browser again and reload the applet during > > > development. Destroy() dont work. What im looking for is some code > > > within the applet that solves the problem once for all. Thor Erik > > > Karlsen > > > Catalyst ONE AS > > > [EMAIL PROTECTED] afaik the applet can't do it, but the user CAN. With netscape, hold the shift key down while clicking "reload." Or use the appletviewer. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Reading (All) Environment Variables in W95/NT
On Mon, 14 Dec 1998 [EMAIL PROTECTED] wrote:
The stuff you outline here depends on your shell: even on Unix it won't
work in all cases without some changes.
Solutions on NT:
1 Get bash and use that to run your java app. Then you can use your
backticks and other funnies. I think you can get it at www.cygnus.com. I
was looking at bash source code a while back and it looks intended to
compile for OS/2. Maybe the standard version will compile for NT with the
right compiler (probably gcc), maybe not. Look for gcc at cygnus too.
2 Use the set command:
set http://database/cgi-bin/
MyCharts.y=698
MyCharts.x=1216
watchlist=http://database/cgi-bin/list
[summer@possum etc]$
> I am getting some requests from NT users on an application that I wrote on
> a Java/Linux environment. I have a Java application that runs fine on Unix
> and I am trying to get to run perfectly on NT. I have a run time shell
> script that passes the entire environment to the JVM using a system
> property.
>
> java -Denv=`env` xenon.xsql.editor.Xsql
>
> The env produces the multiline environment variables list that we know and
> love. On running the application the program looks for the value of system
> property called `env'. Hence the JVM gets the entire environment. I have
> source code that breaks down the multiline value and cuts out each variable
> definition like `HOME=/home/peterp'. The definition is further split into a
> name and value (`HOME' and `/home/peterp') and then place stored in the
> system properties as renamed property `env.HOME' and `/home/peterp'. In
> short all the environment variable are prepended with `env.'. Using this
> hand-me-down old deprecated source code from an unknown URL, I can write
> code like this:
>
> String imagesDir=System.getProperty("env.IMAGES_DIR");
> String searchPath=System.getProperty("env.JAVAFILESEARCHPATH");
>
>
> Of course this does not work with Windows NT (or 95/98). I am not an expert
> on NT. I do not think there is an equivalent for `env' under NT.
> Is there such thing ?
>
> Of course I could write a work around that involve just passing a limited
> number of properties to JVM in a NT batch command `.BAT' file.
>
> java -DIMAGES_DIR=%IMAGES_DIR% ... xenon.sql.editor.Xsql
>
> Admittedly the full final code would look terrible ugly. I am clueless if
> the DOS? W95/NT Shell allows long character lines. Can you even do `back
> subtitution' using the W95/NT shell is also quite another thing as well?
> Also when I tried extending my `PATH' under W95 I got `out of environment
Read the docs on command.com - you need to expand the environment space.
Commandline lenght is acutely limited and you can't expand it.
> space' or something?
>
> However do any other java-linuxers know of a better solution or API?
>
> Any help receives my grateful kudos.
>
> Pete
>
--
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: Linux Red Hat ethernet driver
On Mon, 14 Dec 1998, Dan Finkelstein wrote: > Hi- > > Sorry if this is the wrong list, but... It is. > > I'm installing Red Hat Linux 5.2 on a Compaq. The ethernet driver, tlan.o, > for this machine is copied to the file system. But Red Hat's install > program doesn't list it amongst its drivers from which to select. I have > the file, but don't know how to "install" it manually. Its name is tlan.o > and is in the directory with all the other ethernet drivers. > > This is probably really easy, but I'm stumped, > Dan > -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: java rpms (jdk, tya, kaffe, documnetations)
On Tue, 15 Dec 1998, Levente Farkas wrote: > hi, > I'm just uploading a few java releated rpm to incoming.redhat.com > and in-rhcn.redhat.com. All of this try to conform to rhcn and updates. > > last week updated tutorial IMHO one of the best java doc. > java-tutorial-1.3-1.noarch.rpm > java-tutorial-1.3-1.src.rpm Sun expressly forbids this. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Reading (All) Environment Variables in W95/NT
On Tue, 15 Dec 1998 [EMAIL PROTECTED] wrote: > Ok assuming you can do 3 __where__ is the standard place for the property files > per user per machine? I think it is best to write a software installer program > to do this task. It would write the property file to `user.home' say and the > main application would know where to look for it and load that file and make it > a environment settings default. > There is a convention on unix that user's config stuff ins written in ~/ with the result my home directory's become littered with dozens of .*rc files. I've settled on putting mine in ~/etc/ You can find the home directory from the system properties and build a valid directory name for the runtime environment using other information there. I don't know what MACs impose, but on OS/2 it's reasonable to assume long file names (HPFS is optional, but lots of other software requires it possibly including java: I have OS/2 plus java, but as I have no FAT partitions the question didn't come up. NT's NTFS also allows long names: neither's case sensitive though so AB.DATA = ab.datA. Don't know the rules for WIN9x, but I assume that it's compatible in this respect with NT. > -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: Request For OS Sys Props [was Re: Reading (All) EnvVars ]
On Wed, 16 Dec 1998 [EMAIL PROTECTED] wrote:
> Right then.
>
> I notice that the GIMP is very in that it creates a `~/.gimp subdirectory'
> to place multiple configurations. So does the the CDE on my Solaris box in
> `~/.dt'. On non-Unix machine the `dot' directory does not make a lot of
> sense. Is there a way to find if the JVM is running on a unix box or not?
>
>
> If not I can do it the hard way.
> can you mail me your UNIX & non UNIX settings:
>
> System.out.println( "os.name= "+ System.getProperty( "os.name"));
> System.out.println( "os.arch= "+ System.getProperty( "os.arch"));
> System.out.println( "os.version = "+ System.getProperty( "os.version"));
>
> Somebody must have already have done this, if not I publish the results in
> a web table in 1999!
>
> Peter Pilgrim
>
If you run this (snitched from the book justJava)
// Show how to read and write property files
import java.io.*;
import java.util.*;
public class y {
public static void main(String args[]) {
try {
//String p = System.getProperty("os.version"); // get one property
//System.out.println("ver="+p);
ByteArrayOutputStream bos = new ByteArrayOutputStream();
Properties P = new Properties(System.getProperties());
System.getProperties().save(bos,"basic props");
DataOutputStream ds = new DataOutputStream(bos);
ds.writeBytes("awt.image.incrementaldraw=true\n");
ds.writeBytes("awt.image.redrawrate=50\n");
ds.close();
byte[] tmp = bos.toByteArray();
ByteArrayInputStream bis = new ByteArrayInputStream(tmp);
P.load(bis);
System.setProperties(P);
System.getProperties().list(System.out); // list all properties
} catch (Exception e) {e.printStackTrace();}
}
}
you will get a list of all properties. Amongst them you will find file and
path separators: use these and don't care what you're running on. By using
these properties, youa void the possibility of being caught out with a
system that you've not heard of.
--
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: Java 1.2 (Java 2) functionality
On Sun, 20 Dec 1998, Nathan Meyers wrote: > Mark Murphy wrote: > > > > Will there be a port for Linux having the swing components and other > > Java 1.2 functionality ? You can download swing NOW from Sun for JDK1.1. > > There's also some publicity at Sun's Java site > (http://java.sun.com/pr/1998/11/pr981102-01.html) describing Sun's > support for a Linux JDK 1.2. Of course, there's no clue what that > "support" entails -- writing fan mail, perhaps :-). It seems clear enough to me: Sun works WITH ... (Sun does not do the work) Sun does validate the port. Sun does supply the source code and (resumably) technical assistance. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
Re: An IDE for C and JAVA
On Mon, Dec 21, 1998 at 10:06:10PM -0500, Kirk Hutchinson wrote: > First of all, XEmacs is not an IDE. It's a code editor - that's it. You obviously know little about it then. It has built-in and (virtually) seamless interfaces to compilers, debuggers, and interpreters. > It's really too bad that more IDEs are not available for Linux. What do these give you that XEmacs doesn't? -- John Goerzen Linux, Unix consulting & programming [EMAIL PROTECTED] | Developer, Debian GNU/Linux (Free powerful OS upgrade) www.debian.org | + Visit the Air Capital Linux Users Group on the web at http://www.aclug.org
Re: An IDE for C and JAVA
On Mon, 21 Dec 1998, John Goerzen wrote: > On Mon, Dec 21, 1998 at 10:06:10PM -0500, Kirk Hutchinson wrote: > > > First of all, XEmacs is not an IDE. It's a code editor - that's it. > > You obviously know little about it then. It has built-in and (virtually) > seamless interfaces to compilers, debuggers, and interpreters. > > > It's really too bad that more IDEs are not available for Linux. > > What do these give you that XEmacs doesn't? I have used IBM's Visual Age for Java on OS/2. VAJ understands beans. I can gt beans from any source & import them into VAJ. Once I've done that. I can use those beans as components to build on in my applications. VAJ can edit the properties of a bean as I add it to my application code. Just as I can edit the properties of a Frame or an Applet or a Panel, I can edit the properties of, oh let's say, an ftp bean I obtain separately. I've begun readong a book on Java beans: It seams that whereas an IDE such as Visual Age for C++ would be editing some reporesentation of a window, frame etc, VAJ is editing my application program itself. It gets the info to do this from the Beaninfo classes acompanying the beans (and generates Beaninfo classes for my application). That's what a good java IDE can give, and I don't really imagine XEmacs can come close. It's also why VAJ (and I imagine its competitors) need a strong computer. no puny P100s with 32 Mb. I have a PII 233 64 Mb and would not wish to use anything much less. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
java.lang.ClassCastException
This code
Frame parent = null;
...
parent = (Frame) getParent();
Compiles and runs on Linux. However, on OS/2, it gets a ClassCastException
if in fact the parent is an applet.
I've programmed my way round it:
if (p instanceof Frame)
{
parent = (Frame) getParent();
System.out.println("p(" + parent + ")" );
if (parent instanceof Frame)
parent.setTitle( "some title text " );
};
(I know it's untidy: I was pressed at the time).
I don't know which implementation is at fault: if there's someone from IBM
sees this, please pass it on.
I'll be away for up to a week: I don't plan to read all the old
contributions to this list on my return. if anyone needs more info, please
mail me direct and do NOT cc the list (else my mail filter will catch it).
--
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.
Re: An IDE for C and JAVA
On Tue, Dec 22, 1998 at 11:11:08AM +0100, Artur Biesiadowski wrote: > John Goerzen wrote: > > I personally do not use highly integrated IDEs and it seems that you do > not also, but do not think that Xemacs is an IDE - it is just very smart > editor. What it lacks ? > > Graphical composition of GUI components, including positioaning them, I'm sorry, but you can certainly have IDEs without this. I remember using Borland C++ 2.0 under DOS, which had what everyone would call an IDE. Before that, I used Turbo Pascal 5.5 under DOS, which also had an IDE -- and everyone called it that. XEmacs has far more features than either of those, both IDE-wise and otherwise, so I cannot understand how you can claim that something that does more than other IDEs is not an IDE. > disvovering their properties by beans mechanism, creating handlers for > them by simple clicking, editing their properties with instant effect on > the screen. So what you're wanting is a Java visual development environment. Incidentally, I have yet to see one of those that I like, for Java or otherwise. They often work by laying out components at certain pixel locations, which is even worse in Java than elsewhere, because fonts and sizes of widgets can vary tremendously between systems. Also, none of them that I've seen will do Swing, which I use, so that rules them all out. Back to XEmacs. > is really pitty one. I'm talking about something with ability to check > variables by moving mouse over them , graphically displaying and > followingg instance variables (as ddd does) etc. I've never used jdb, but I must say that for a debugger, I prefer gdb. Nothing else, except perhaps the XEmacs interface to it, is really acceptable. I find that limitations of a GUI get to be cumbersome in many cases.
Re: An IDE for C and JAVA
On Tue, 22 Dec 1998, John Goerzen wrote: > On Tue, Dec 22, 1998 at 11:11:08AM +0100, Artur Biesiadowski wrote: > > > John Goerzen wrote: > > > > I personally do not use highly integrated IDEs and it seems that you do > > not also, but do not think that Xemacs is an IDE - it is just very smart > > editor. What it lacks ? > > > > Graphical composition of GUI components, including positioaning them, > > I'm sorry, but you can certainly have IDEs without this. I remember using > Borland C++ 2.0 under DOS, which had what everyone would call an IDE. > Before that, I used Turbo Pascal 5.5 under DOS, which also had an IDE -- and > everyone called it that. XEmacs has far more features than either of those, > both IDE-wise and otherwise, so I cannot understand how you can claim that > something that does more than other IDEs is not an IDE. What was state of the art with DOS doesn't compare with the modern scene, eigher in function or in machine requirements. > Incidentally, I have yet to see one of those that I like, for Java or > otherwise. They often work by laying out components at certain pixel > locations, which is even worse in Java than elsewhere, because fonts and > sizes of widgets can vary tremendously between systems. Also, none of them > that I've seen will do Swing, which I use, so that rules them all out. Back > to XEmacs. IBM's Visual Age for Java includes Swing. The entry version is free and available for OS/2 and NT. I've heard rumours of a Linux version coming. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
