| |
| |
| |
|
Page: 1 2
Comments:
<0> hello all
<0> JNI and Oracle.. anyone ?
<1> specifics?
<0> is it possible to call shared library function through java + JNI within Oracle stored procedure ?
<0> I have no problem in using java stored procedure witnin oracle, and to call function from shared libraries from java using JNI,
<0> but not when combining both.
<0> when I try to load the java cl*** file to the oracle using loadsql command, I get the following error :
<1> i don't understand, perhaps too late or so
<0> ORA-29521: referenced name Link could not be found
<0> the "Link" string need to point to the libLink.so file
<2> dwork are you around?
<3> Is there a way that I can echo the raw data (in hex format) in a HTTP request to System.out or the log?
<3> I've got a J2ME midlet which is sending UTF-8 strings (hebrew characters) in a POST to an ASP server.
<3> The ASP server belongs to the customer and I think something's going wrong at his end, so I want to make a Servlet to prove that the data is being sent (although i believe he's not reading it properly)
<4> hi is this chat room about the java programming language?
<5> yes it is
<4> how can i copy an array of objects by VALUE (not by reference)?
<4> It's not letting me use <array>.clone()
<5> u want to clone every object in the array?
<1> Thorion: possibly related: http://www.google.com/search?complete=1&hl=en&q=deep+cloning+in+java&btnG=Google+Search
<4> hmmm that gives me the idea to copy the array into a vector then into another array, but i would think there a more common sense way... thankyou for the link
<4> srry yes aliveuser the entire array
<5> well i think clackwell have helped u
<5> u was gonna tell u to try to clone every object in the array and make a new one
<5> not cloning the whole array
<4> yeah i did that and it still p***ed the objects as references
<4> ah... no, i didn't do clone()... thanks
<4> hmmm no Object.clone() method in J2ME MIDP 1.0
<5> so override clone and do ur own clone for ur objects
<3> Thorion: what about creating your own copy method?
<4> CrosswordUnit[] copy = new CrosswordUnit[original.length];
<4> for (int i=0;i<original.length;i++) {
<4> copy[i]=original[i];}
<3> that's an aarray, not an object
<3> copy.var = original.var; ...
<4> yes but i don't really want to copy each variable by hand, although if there's no other option...
<6> hey
<6> i have a problem
<7> state it
<6> hey
<6> im portuguese so im not very good on english
<6> but i have a database
<6> and i want to insert something there
<6> when i click on button insert it insert on database
<6> but i cant see in the program
<6> but if i restart the program i can see
<6> anyone know ho i do to see without restart?
<4> repaint()?
<7> when you've inserted the data, do you also refresh your current dataset in your application?
<7> your insert seems to be working, but you must make sure you're viewing a fresh set of data
<6> how i do that??
<6> the comand??
<7> i'm not used to jdbc at all, so that's up to you to find out
<7> how do you fetch the data when you run the program?
<7> repeat that process after you've inserted the new data
<4> Well it looks as if I'll have to write my own clone() method and p*** the variables... there is a MIDP method System.arrayCopy() but that still p***es by reference :( Thanks for your help guys
<7> doesn't every object inherit the clone() method by itself?
<6> like sql = slect * table??
<7> ||EU||: exactly
<6> i do that
<6> but it still dont work
<4> no <object>.clone() method in MIDP 1.0
<6> if i close the aplication and open the new data apeer
<7> Thorion: ah. MIDP. my bad then
<6> but if not i dont appear
<7> ||EU||: you must output the new data to the screen again
<6> i do that
<6> sql = select * table
<7> ||EU||: surely there must be a method that fetches your data rows?
<7> just modifying a variable without using it won't do anything
<6> yeah
<6> like refreshdata?
<7> sounds very likely :)
<6> i do that
<8> is it possible to do chroot() in Java within the program
<8> so that when i exit the program im in the same root env i was before i ran the program..
<9> hi
<9> if i persist an interned String, will it be duplicated?
<3> Are form elements in HTTP POST messages included in the URL or the body of the message?
<10> neither
<10> request Headers
<3> Why does every example I've seen tell me to do: "hc.openDataOutputStream().write()"?
<3> I want to send UTF-8 characters, by the way.
<3> hc.setRequestProperty( "Content-Encoding", "UTF-8" );
<3> what do I need to do so that the Servlet can do "request.getParameter("MyName")"?
<3> (on a J2ME client)
<11> hello i have this error why ?<identifier> expected
<11> System.out.println(getName()+" est termine");
<12> pastebin your code
<11> ? here
<12> pastebin.com
<11> k thx
<13> morning
<14> hi all
<14> any Java gurus online who can help me out?
<15> RoyME: Who knows. Perhaps if you asked a more precis question, someone might have a chance at answering it.
<1> hi wlf
<14> Heh, ok, I'll have a go
<14> I'm creating a application, and I found out that using Integer>>1 is faster then Integer/2...why is that? Doesn't Java optimise that sort of thing?
<1> RoyME: you cannot do SomeInteger/2.
<1> so now we have to see a test case, because we cannot rely on you properly testing it.
<14> Just int i = 10 -> i/2 is slower then i>>1
<1> RoyME: analyse the generated bytecode if you like.
<14> Err..how do I analyse that..? (feeling nooby)
<1> RoyME: learn java bytecode, perhaps read the jvm spec, get a java decompiler or use javap to decompile.
<14> I have the DJ Java Decompiler and it regenerates the same code as I put in, IE. i >>= 1; and i /= 2;
<16> That's because they are different, obviously.
<16> Adding the extra if-power-of-two check to the / operator would be more expensive than just letting the user optimize himself.
<1> RoyME: i don't understand. does it or does it not generate the same bytecode?
<16> It doesn't, if I understood correctly.
<16> Turns it back into the original code for whichever case
<14> Ah dwork, that makes sense.. that probably is because it doesn't have a compiler that could filter it out?
<14> Doing it runtime woulnd't make much sense no :)
<1> RoyME: perhaps contact sun and ask them why (after you have verified) it doesn't "optimize" that.
<16> It doesn't make any sense that it would
<1> maybe their stance is/was that these are trivial cases which are taken care of by the JIT anyway.
<14> Well, if it would be replaced while creating the cl*** file it would be faster in runtime
<1> which brings us to the question of how you benchmarked it. you need to know what you are doing when benchmarking. it is not as easy as taking the time that an operation takes.
<16> But the parser only sees intvar opdiv inttype, it doesn't know it's a 2 and therefore cheaper to bit shift
<16> Nor it should, because if it was to check it to see which way to go, it'd be gruesomely expensive
<16> It's a lot better to cache the whole block in the JIT and if it's "static", just reuse it
<14> In the case the "/2" (or 4/8/16/32 etc) part is static it makes a bit sense right?
<16> It makes sense in the JIT step, not the bytecode generation.
<14> Yeah..
<14> It just wondered me why it still makes so much difference while running it in a simple test-benchmark program
<1> RoyME: show us your benchmark, then we'll tell you why it is broken.
<14> Although that speed in my case is neglectable
<14> Heh ok:
<1> pastebin.com or so
<14> Its in pastebin.com now (RoyvanRIjn)
<1> even better perhaps: study "Watch your hotspot compiler go" (or so) at faq.
<1> you can post the url at pastebin.com
<14> http://pastebin.com/579932
<1> anything that cuts down on effort of other people is good
<1> RoyME: that code doesn't take JIT into account, does it? i don't know if the jit compiles code blocks inside a method. i ***ume whole methods only. so you benchmarked some raw performance, which is at least a relevant piece of information that you may not want to omit when stating such results.
<1> see the faq article, it details on that i think.
<14> You suggest putting the calculations in methodes, testing the difference?
<1> RoyME: i suggest read the article in the faq (for the third time), so that you know what you are doing instead of doing the next 1 billionth pointless benchmark which says nothing of relevance, because it is broken.
<1> while at it, also try using the server jit/hotspot, you may see noticable differences in comparison to the workstation jit/hotspot, and fewer differences between the results of your operation.
<1> microbenchmarking is ok, but people who do them on java typically go about it in such a naive manner, without knowing about JIT, etc. that their findings are useless.
<1> also relevant on the matter of microbenchmarking: GAMES Evaluating Java for Game Development: http://www.rolemaker.dk/articles/evaljava/
<1> ignore the topic of that article, the author points out how many benchmarks he looked at are fundamentally flawed, that is the interesting bit.
<14> That last link gave me a 404.. :(
<1> really? ah, shame. it was lengthy but good.
<1> not 404 for me, but "The account is closed"
<14> Ah yes, me too
<14> And it isn't that bad, I'm working as a J2ME game programmer now, so might even learn a couple of more good thing
<1> ah :)
<17> hello!
<17> I'm working with ReportViewerBean of CrystalReports, anyone used it?
<1> RoyME: google still has it, the author moved it to some subscription site it seems.
Return to
#java
or
Go to some related
logs:
#linux
durentu
#AllNiteCafe
#chatzone
#linux
#linuxhelp
#linux
#MissKitten
#mirc
#MissKitten
|
|