Re: Thanks :)
** ** WARNING: WinProxy has detected a virus in file attached to this e-mail message! The attachment has been automatically removed to protect your network. WinProxy Administrator: [EMAIL PROTECTED] 06/09/04 11:03:53 WinProxy (Version 5.1 R1d (5.0.50.10)) - http://www.Ositis.com/ Antivirus Vendor: Panda Software Scan Engine Version: 2.10.1.6_3.1.5.211 Pattern File Version: 3.78749 (Timestamp: 2004/06/08 11:33:36) Machine name: STAG-DOWNLOAD Machine IP address: 61.95.203.145 Server: 209.237.227.199 Client: 192.168.100.102 Protocol: SMTP Virus: "W32/Bagle.N.worm" found! Attachment: MoreInfo.pif ** ** - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks!
Your file is attached. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks :)
<>- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks for the note on JNI and class loading in the release notes
Hi Jonathan,
we had the same problem and fixed it in the way described now in the Tomcat
documentation.
Probably one additional remark should be added to the documentation :
If you place the Java code loading the native library outside of the web
application (for example in $CATALINA_HOME/common/lib) it is loaded only
once and the problem is solved.
This sollution has some implication : The classes containing the native code
are loaded by a classloader which has no knowledge about any class which
resides in \Web-inf\lib. You will get an exception if you try to instanciate
a class which resides in the \Web-inf\lib directory from your native code!
You will also get an exception if you try to import a class which resides in
the \Web-inf\lib\ directory from your java code in $CATALINA_HOME/common/lib
since the two classes are loaded by different classloaders.
This will not work (ClassA in $CATALINA_HOME/common/lib ansd ClassB in
\Web-inf\lib\ ) :
ClassA :
import ClassB;
public native static void doSomething(ClassB obj);
ClassB :
import ClassA
public static void main(String[] args) {
ClassA.doSomething(this);
}
Now it works again :
ClassA :
public native static void doSomething(Object obj);
ClassB :
import ClassA
public static void main(String[] args) {
ClassA.doSomething((Object)this);
}
Is this description correct? How do you handle this problem? Is there a more
elegant sollution ?
Regards,
Jochen
P.S: Sorry for posting this message to tomcat-user. Mea culpa.
-
Tomcat 4.0 and JNI Based Applications:
-
Applications that require native libraries must ensure that the libraries
have
been loaded prior to use. Typically, this is done with a call like:
static {
System.loadLibrary("path-to-library-file");
}
in some class. However, the application must also ensure that the library
is
not loaded more than once. If the above code were placed in a class inside
the web application (i.e. under /WEB-INF/classes or /WEB-INF/lib), and the
application were reloaded, the loadLibrary() call would be attempted a
second
time.
To avoid this problem, place classes that load native libraries outside of
the
web application, and ensure that the loadLibrary() call is executed only
once
during the lifetime of a particular JVM.
- Original Message -
From: "Jonathan Eric Miller" <[EMAIL PROTECTED]>
To: "Tomcat Developer List" <[EMAIL PROTECTED]>
Sent: Saturday, September 15, 2001 5:59 AM
Subject: Thanks for the note on JNI and class loading in the release notes
> I'm guessing that Craig is the one that added the section about JNI and
> class loading in the RC1 release notes. I just wanted to say that I
> appreciate that you documented this.
>
> I also noticed that you fixed a problem that I noticed with the Base64
> encoder where it had trailing zeroes.
>
> Thanks, Jon
>
>
Re: Thanks for the make links and a rant
On Mon, 13 Nov 2000, Roy Wilson wrote: > Nick, > > I agree that your example is simple. Even I can understand and create > such "toy" makefile. What I am complaining about is getting Not a toy. Really works. By the way, Matthew, my boss, sent this earlier but it somehow never made it to the list: MATTHEW Hi Roy! > > If someone can provide a link to something like "Make for Dummies" I'd > appreciate it. > You betcha, and fyi-- this format is VERY similar to the make file Nick Bauman, myself and others use at our place of employment. (Some might say better!?) At the very least, it's well documented. http://geosoft.no/javamake.html enjoy. -Matthew PS>I've been resisting the temptation to get involved on the ant rant thread, but this link might suggest I'm a Make fan. PPS> I'm not. MATTHEW > evil/nasty/nested makefiles that don't work (unlike the MAKE_CONF > experience you've had) and having to understand all [exaggeration] of it > to find the presumably few line(s) that must be changed. If you don't know make, this will happen. Same with Ant. =) > I assume you're referring to Nash the game theorist. Back before WWII and > after, von Neumann (a founder of the discipline of computing) applied > functional analysis as developed by Banach to problems of mathematical > physics and economics. I didn't know that Nash used functional analysis > in game theory. Thanks for the opportunity to rant :-). Well, Nash is purported to have solved the embedding theorem and was working on a general theorem on turbulance before he lost his mind (maybe he was writing Makefiles when he snapped. I rememeber it was either Makefiles or Fermat's Last Theorem, I don't recall exactly, gah!) > Roy > > >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< > > On 11/13/00, 5:58:34 PM, Nick Bauman <[EMAIL PROTECTED]> wrote regarding > Re: Thanks for the make links and a rant: > > > > > > The makefile I used to compile Tomcat is thus: > > > -8<-- > > # Keep track of the package name > > PACKAGE=org.apache > > > # Keep track of the package version > > VERSION=3_2 > > > # where the maketools be > > MAKE_CONF=./maketools > > > include $(MAKE_CONF)/config.mk > > include $(MAKE_CONF)/rules.mk > > -8<-- > > > Can it be any simpler? > > > Now I know that there is a lot of magic in the MAKE_CONF, but then you > > don't have to worry about that. It just works. > > > On Mon, 13 Nov 2000, Roy Wilson wrote: > > > > Nick, > > > > > > I have a copy of the FSF make manual: As our president used to say "I > > > recur to my former statement [about make documentation]." :-) I'll have > > > to check out the O'Reilly reference. See my rant below. > > > > > > Roy > > > > > > -- > > Nicolaus Bauman > > Software Engineer > > Simplexity Systems > > > > > - > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > -- Nicolaus Bauman Software Engineer Simplexity Systems - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks for the make links and a rant
Nick, I agree that your example is simple. Even I can understand and create such "toy" makefile. What I am complaining about is getting evil/nasty/nested makefiles that don't work (unlike the MAKE_CONF experience you've had) and having to understand all [exaggeration] of it to find the presumably few line(s) that must be changed. I assume you're referring to Nash the game theorist. Back before WWII and after, von Neumann (a founder of the discipline of computing) applied functional analysis as developed by Banach to problems of mathematical physics and economics. I didn't know that Nash used functional analysis in game theory. Thanks for the opportunity to rant :-). Roy >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< On 11/13/00, 5:58:34 PM, Nick Bauman <[EMAIL PROTECTED]> wrote regarding Re: Thanks for the make links and a rant: > The makefile I used to compile Tomcat is thus: > -8<-- > # Keep track of the package name > PACKAGE=org.apache > # Keep track of the package version > VERSION=3_2 > # where the maketools be > MAKE_CONF=./maketools > include $(MAKE_CONF)/config.mk > include $(MAKE_CONF)/rules.mk > -8<-- > Can it be any simpler? > Now I know that there is a lot of magic in the MAKE_CONF, but then you > don't have to worry about that. It just works. > On Mon, 13 Nov 2000, Roy Wilson wrote: > > Nick, > > > > I have a copy of the FSF make manual: As our president used to say "I > > recur to my former statement [about make documentation]." :-) I'll have > > to check out the O'Reilly reference. See my rant below. > > > > Roy > > > -- > Nicolaus Bauman > Software Engineer > Simplexity Systems > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks for the make links and a rant
Roy, Having just read a biography of John Forbes Nash, I have but a wisp of understanding of what a Banach space is. However, you echo what I already said. Make is unlike any other tool in the unix world. I think you could write a white paper on this, seriously; like, entitled "where the hell did Make come from?" because it's actually an example of how diverse a genius (or sadist, depending on your prespective. I prefer the former) RMS is. It resembles, at times, treatise on lingusitics. But I digress. The makefile I used to compile Tomcat is thus: -8<-- # Keep track of the package name PACKAGE=org.apache # Keep track of the package version VERSION=3_2 # where the maketools be MAKE_CONF=./maketools include $(MAKE_CONF)/config.mk include $(MAKE_CONF)/rules.mk -8<-- Can it be any simpler? Now I know that there is a lot of magic in the MAKE_CONF, but then you don't have to worry about that. It just works. On Mon, 13 Nov 2000, Roy Wilson wrote: > Nick, > > I have a copy of the FSF make manual: As our president used to say "I > recur to my former statement [about make documentation]." :-) I'll have > to check out the O'Reilly reference. See my rant below. > > Roy > -- Nicolaus Bauman Software Engineer Simplexity Systems - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
