openjava-support

[openjava-support] FASD project: Online survey launched

[<ben...@id...>]

Dear Open Source developer

I am doing a research project on "Fun and Software Development" in which I kindly invite you to participate.
You will find the online survey under http://fasd.ethz.ch/qsf/. The questionnaire consists of 53 questions and you will need about 15 minutes to complete it.

With the FASD project (Fun and Software Development) we want to define the motivational significance of fun when software developers decide to engage in Open Source projects. What is special about our research project is that a similar survey is planned with software developers in commercial firms. This procedure allows the immediate comparison between the involved individuals and the conditions of production of these two development models. Thus we hope to obtain substantial new insights to the phenomenon of Open Source Development.


With many thanks for your participation,
Benno Luthiger


PS:
The results of the survey will be published under http://www.isu.unizh.ch/fuehrung/blprojects/FASD/.
We have set up the mailing list fa...@we... for this study. Please see http://fasd.ethz.ch/qsf/mailinglist_en.html for registration to this mailing list.

_______________________________________________________________________

Benno Luthiger
Swiss Federal Institute of Technology Zurich
8092 Zurich

Mail: benno.luthiger(at)id.ethz.ch
_______________________________________________________________________

[openjava-support] More OpenJava compiler issues

[Michael J. <msj...@uw...>]

While I did finally see the CVS update, I haven't yet tested to see if 
non-public classes compile. I'm sorry - I'll take a look at this a bit later 
on.

In any case, got another interesting bug, again in Compiler.java. The javac 
method, when dealing with generated classes, reuses the File objects in the 
'files' array when building the list of files to compile (from added_cu). 
These file objects of course are completely wrong and don't match the new 
compilation units, but the error didn't manifest itself until I generated 
more classes out than I was feeding in, at which point you get an 
ArrayOutOfBounds exception.

Cheers,
Michael Jarrett
University of Waterloo

[openjava-support] OJ_Unknown

[Michael J. <msj...@uw...>]

One question: why OJ_Unknown for non-public classes?

I've been trying to compile code with a large number of package-access 
classes, and each one generates an OJ_UnknownX.java and OJ_UnknownXOJMI.java.

Each one of these internally declares a class of the original name. The OJMI 
files won't compile at all since they declare a public class with a name that 
does not match the file name.

This seems to be generated by openjava.ojc.Compiler in getMainClassName. This 
actually causes a variety of bugs, since "nonpubclassid" does not get reset 
when generating Java files in "outputToFile" causing the OJMI file and the 
Java source to have different IDs, and not reset again in the "javac" phase, 
meaning the compiler can't find any of these OJ_Unknown files anyways.

The above error can be fixed by placing the line "nonpubclassid=0;" at the 
beginning of outputToFile() and javac(), but it still doesn't fix the fact 
that the OJMI classes won't be compiled by Java due to the classname-filename 
conflict. I can't really offer a fix to this since I don't really understand 
why you use OJ_Unknown in the first place.

Any help appreciated.
I am now using OpenJava_1.1 branch from CVS from Sourceforge.

Michael Jarrett
University of Waterloo

Re: [openjava-support] Overzealous attempts to resolve types?

[Michiaki T. <TA...@jp...>]

Michael,

You are on the sane way.  Unfortunately, such error suppression is not yet 
supported in OpenJava.
We should be able to do this by overriding OJClass.resolveError() or 
something.
Thanks for your suggestion and I am going to provide it in OpenJava later 
(not in this week, sorry).

Mich
-
Michiaki (Mich) Tatsubori, PhD.  mi...@ac...
IBM Tokyo Research Laboratory




Michael Jarrett <msj...@uw...>
Sent by: ope...@li...
2003/11/10 07:18
 
        To:     ope...@li...
        cc: 
        Subject:        [openjava-support] Overzealous attempts to resolve 
types?

 

Hey all. Rather new to OpenJava, hope someone can give me some advice on 
this.
(OpenJava Compiler Version 1.1 build 20030618)

I am trying to write a metaclass to reproduce a method for a number of 
primitive types. Unfortunately, in the case of method parameters, the 
types 
are attempted to be resolved a bit too early, producing error messages 
both 
in initializing the parse tree and during my translateDefinition. For the 
latter, it appears for some strange reason that OJClass.removeMethod 
requires 
the ability to resolve the method parameter's types? None of this is a 
problem for the return type, which I suppose gets recognized as a keyword, 
or 
is not checked as often?

In the end, it still generates the output correctly, but the errors are 
annoying. The errors are shown below.

Is there a sane way in general to translate code where members/fields/
parameters/types are missing or invalid until after translateDefinition?

Cheers,
Michael Jarrett
University of Waterloo


Generating parse tree.
..done.

Initializing parse tree.
VariableBinder.bindName() openjava.mop.OJClassNotFoundException: typename 
: 
typename
ClosedEnvironment
class object table : {}
binding table : {}
parent env : ClassEnvironment
class : blah
member classes : []
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : GlobalEnvironment
class object table : {char=char, long=long, 
ca.uwaterloo.ece750.meta.PrimitiveTemplate=class ca.uw
aterloo.ece750.meta.PrimitiveTemplate, float=float, short=short, 
void=void, 
int=int, blah=class bl
ah, java.lang.Object=class java.lang.Object, <type>null=<type>null, 
double=double, java.lang.Strin
g=class java.lang.String, boolean=boolean, byte=byte}

..done.

Translating callee side
getTypename templated!
OJClass.forNameAnyway() failed for : typename 
openjava.mop.OJClassNotFoundException: typename
ClassEnvironment
class : blah
member classes : []
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : GlobalEnvironment
class object table : {char=char, long=long, 
ca.uwaterloo.ece750.meta.PrimitiveTemplate=class ca.uw
aterloo.ece750.meta.PrimitiveTemplate, float=float, short=short, 
void=void, 
java.lang.System=class
 java.lang.System, int=int, blah=class blah, java.lang.Object=class 
java.lang.Object, java.lang.St
ring[]=class java.lang.String[], <type>null=<type>null, double=double, 
java.lang.String=class java
.lang.String, boolean=boolean, byte=byte}




-------------------------------------------------------
This SF.Net email sponsored by: ApacheCon 2003,
16-19 November in Las Vegas. Learn firsthand the latest
developments in Apache, PHP, Perl, XML, Java, MySQL,
WebDAV, and more! http://www.apachecon.com/
_______________________________________________
openjava-support mailing list
ope...@li...
https://lists.sourceforge.net/lists/listinfo/openjava-support

[openjava-support] Overzealous attempts to resolve types?

[Michael J. <msj...@uw...>]

Hey all. Rather new to OpenJava, hope someone can give me some advice on this.
(OpenJava Compiler Version 1.1 build 20030618)

I am trying to write a metaclass to reproduce a method for a number of 
primitive types. Unfortunately, in the case of method parameters, the types 
are attempted to be resolved a bit too early, producing error messages both 
in initializing the parse tree and during my translateDefinition. For the 
latter, it appears for some strange reason that OJClass.removeMethod requires 
the ability to resolve the method parameter's types? None of this is a 
problem for the return type, which I suppose gets recognized as a keyword, or 
is not checked as often?

In the end, it still generates the output correctly, but the errors are 
annoying. The errors are shown below.

Is there a sane way in general to translate code where members/fields/
parameters/types are missing or invalid until after translateDefinition?

Cheers,
Michael Jarrett
University of Waterloo


Generating parse tree.
..done.

Initializing parse tree.
VariableBinder.bindName() openjava.mop.OJClassNotFoundException: typename : 
typename
ClosedEnvironment
class object table : {}
binding table : {}
parent env : ClassEnvironment
class : blah
member classes : []
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : GlobalEnvironment
class object table : {char=char, long=long, 
ca.uwaterloo.ece750.meta.PrimitiveTemplate=class ca.uw
aterloo.ece750.meta.PrimitiveTemplate, float=float, short=short, void=void, 
int=int, blah=class bl
ah, java.lang.Object=class java.lang.Object, <type>null=<type>null, 
double=double, java.lang.Strin
g=class java.lang.String, boolean=boolean, byte=byte}

..done.

Translating callee side
getTypename templated!
OJClass.forNameAnyway() failed for : typename 
openjava.mop.OJClassNotFoundException: typename
ClassEnvironment
class : blah
member classes : []
parent env : FileEnvironment
        package : null
        main class : blah
        local classes :
        imported classes :
        imported packages : java.lang ca.uwaterloo.ece750.meta
        local class table : {}
parent env : GlobalEnvironment
class object table : {char=char, long=long, 
ca.uwaterloo.ece750.meta.PrimitiveTemplate=class ca.uw
aterloo.ece750.meta.PrimitiveTemplate, float=float, short=short, void=void, 
java.lang.System=class
 java.lang.System, int=int, blah=class blah, java.lang.Object=class 
java.lang.Object, java.lang.St
ring[]=class java.lang.String[], <type>null=<type>null, double=double, 
java.lang.String=class java
.lang.String, boolean=boolean, byte=byte}

[openjava-support] Test

[Michiaki T. <TA...@jp...>]

This is a test.
-
Michiaki (Mich) Tatsubori, PhD.  mi...@ac...
IBM Tokyo Research Laboratory