Log of the #nice channel on irc.freenode.net

Using timezone: Central European Time
<CIA-3>03arjanb * 10Nice/src/bossa/syntax/ (4 files): Converted ImportedConstructor.01:28
* leblin joins02:58
<arjanb>hello03:00
<leblin>hey there.. is daniel around?03:01
<arjanb>he is here but might not be awake because it's 3 am03:03
could i help you?03:04
<leblin>do you know if he received my emails?03:06
<arjanb>no idea03:07
have you got any reply?03:13
<leblin>no, but i get so much junk mail i have problems sorting through it.03:14
<cm_>arjanb: How does option type for primitives work, eg ?int ?
<arjanb>cm_: ?int types are always boxed atm03:15
<cm_>ok
Is it possible to do it non-boxed in the jvm?03:17
<leblin>will it be difficult to fix the protected member access problem?
<arjanb>leblin: it might be a too restrictive check somewhere from the time most methods weren't compiled inside classes but i'm not sure03:22
cm: it might be possible but you need to represent the null case somehow03:24
<cm_>heh, I was referring to that delicate case - implicitly :-)03:26
<leblin>i will have to check it out. the eclipse api uses protected members extensively. it needs to be fixed if any improvements to the plugin are done. i will see if i can find the problem
<cm_>arjanb: Why not "unboxing" of values. Would that be possible?03:30
<arjanb>the bytecode generator does boxing/unboxing when it is neccessary but is not smart in avoid it03:35
it would be possible to make it smarter but that's not easy i think03:36
<cm_>Is the 1.0 release scheduled?03:38
<arjanb>there is no schedule
leblin: the last check in bytecode.Access.legal() doesn't make sense to me03:44
does it make a difference removing?: && receiver.isSubtype(c)03:46
hmm probably not03:49
* fcb joins04:08
<leblin>i dont believe that check is necessary imo
the gnu.bytecode package has little documentation04:29
arjan are you still here?04:44
* fcb leaves05:34
* ChanServ leaves06:13
* leblin leaves
* arjanb leaves
* ChanServ joins06:14
* leblin joins
* arjanb joins
* ChanServ leaves06:15
* leblin leaves
* arjanb leaves
* ChanServ joins
* leblin joins
* arjanb joins
* leblin leaves07:40
* bonniot leaves10:10
* cm_ leaves10:36
* CIA-3 leaves11:32
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins11:33
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves11:34
* CIA-3 joins
* CIA-3 leaves
* CIA-3 joins11:38
* CIA-3 leaves
* CIA-10 joins11:39
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves
* CIA-3 joins11:43
* CIA-3 leaves
* CIA-10 joins11:44
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves
* CIA-3 joins11:48
* CIA-3 leaves11:49
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves11:50
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves11:51
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves11:52
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves11:53
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins11:54
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins11:55
* CIA-3 leaves
* CIA-3 joins11:59
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves12:00
* CIA-3 joins
* CIA-3 leaves12:01
* CIA-3 joins12:04
* CIA-3 leaves12:06
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves12:07
* CIA-10 joins
* CIA-10 leaves12:08
* CIA-3 joins
* CIA-3 leaves12:13
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins
* CIA-3 leaves12:14
* CIA-3 joins12:18
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins12:19
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins12:20
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins12:21
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves
* CIA-3 joins12:22
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves12:23
* CIA-3 joins
* CIA-3 leaves
* CIA-10 joins
* CIA-10 leaves13:38
* CIA-4 joins13:39
* leblin joins17:33
* CIA-4 leaves17:39
* CIA-10 joins
<arjanb>hi17:40
<leblin>hello. thanks for applying my patches.17:42
i'll get the style correctly on any newer ones i make
* CIA-10 leaves17:44
* CIA-4 joins
<arjanb>are you working on other patches now?17:47
<leblin>not yet. i just woke up :)
the reason we cannot access protected methods of java classes is because the methods are compiled to the fun class for the package17:48
<arjanb>most methods are compiled in the class itself but maybe a copy in the fun class too17:54
<leblin>in all my test cases, the method always gets called from the fun class17:56
<arjanb>the general solution is to create redirect methods in the class for all used protected methods17:57
<leblin>i was going to propose that idea.17:58
modifying the access check is not good enough. a security manager can catch the illegal access18:09
<arjanb>indeed18:13
leblin: if you had no response than you could send the classloading patch to me18:46
* CIA-4 leaves18:54
* CIA-7 joins18:55
<leblin>ok i will. i will also add a check for unsupported class version files before doing so.18:56
<arjanb>is there any problem with class versions?18:57
<leblin>yes, i have nicec using java 1.4, i must compile my java files using 1.4 as well and not 1.5 or an exception will be thrown when i try to load it.18:59
<arjanb>what exception?19:01
<leblin>java.lang.UnsupportedClassVersionError19:07
<arjanb>thrown by nicec?19:08
<leblin>it is thrown by the class loader however it is not caught anywhere in the compiler19:09
<arjanb>i'm failing to understand why that exception is thrown..19:14
<leblin>i have java version 1.4 and 1.5 installed. nicec is using java 1.4 on my system. the error is thrown if i try to load a 1.5 class file19:19
where do i send the patch?19:22
<arjanb>to my sourceforge mail address19:24
<leblin>ok. do we use whitespace or tabs for indentation in nice?19:25
<arjanb>i use spaces19:27
* CIA-7 leaves19:33
<leblin>i have sent the email19:41
<arjanb>got it, i will apply after finishing my current changes19:45
* CIA-3 joins19:56
<leblin>brb20:10
* leblin leaves
* leblin joins20:15
<CIA-3>03arjanb * 10Nice/src/bossa/syntax/ (5 files): Avoid misleading error messages with named arguments.20:40
03arjanb * 10Nice/src/nice/tools/code/TypeImport.java: Handling of classloading exceptions. (patch by Luc Perrin)21:01
<leblin>i just got an exception in the parser21:48
<arjanb>what is it?21:49
<leblin>try compiling the following21:51
void check( ) {
?foo = null;
}
<arjanb>never found such problems with javacc before but i had the same a few days ago :(21:53
maybe it knows that i want to replace the parser...21:54
<leblin>do you know which parser you want to use?
<arjanb>antlr 3 but unfortunately it will take some time before it's released21:57
<CIA-3>03arjanb * 10Nice/src/bossa/syntax/if.nice: Improved pretty printing of if and ?:.22:46
* leblin leaves23:55

Generated by Sualtam