RE: Re[2]: Xerces 1 use in XMLC

Hello Richard,

I tested on linux with jdk 1.4.1 and jikes 1.5.1 . Everything works
fine.
When I try without jikes (I ran 'ant clean', then moved away the jikes
binary and typed 'ant tests') I get:

------------------------------------------------------------------------
--------
[...]
lib:
    [mkdir] Created dir:
/home/enrico/Projects/xmlc/libraries/xml/modules/xerces/build/classes/ME
TA-INF/services
     [copy] Copying 2 files to
/home/enrico/Projects/xmlc/libraries/xml/modules/xerces/build/classes/ME
TA-INF/services
      [jar] Building jar:
/home/enrico/Projects/xmlc/release/lib/xerces.jar
init:
compile:
    [mkdir] Created dir:
/home/enrico/Projects/xmlc/libraries/xml/modules/jtidy/
build/classes
    [javac] Compiling 45 source files to
/home/enrico/Projects/xmlc/libraries/xml/modules/jtidy/build/classes
lib:
      [jar] Building jar:
/home/enrico/Projects/xmlc/release/lib/jtidy.jar
init:
compile:
    [mkdir] Created dir:
/home/enrico/Projects/xmlc/xmlc/modules/xmlc/build/clas
ses
    [javac] Compiling 268 source files to
/home/enrico/Projects/xmlc/xmlc/modules/xmlc/build/classes

BUILD FAILED
file:/home/enrico/Projects/xmlc/xmlc/modules/xmlc/build.xml:135: Error
running jikes compiler

------------------------------------------------------------------------
-----------

I think something is hardcoded in the tests to look for jikes to compile
them... and since jikes is not there the build fails before it gets to
run them....

If instead I try to run after all unit tests are compiled (and after
moving away the jikes binary) everything looks OK.

I will try the same exact tests on Windows, just to verify everything is
good.....


Enrico.




> -----Original Message-----
> From: Richard Kunze [mailto:kunze@xxxxxxxxx] 
> Sent: Thursday, October 24, 2002 4:51 AM
> To: xmlc@xxxxxxxxxxx
> Subject: Re: Re[2]: Xmlc: Xerces 1 use in XMLC
> 
> 
> Hi Enrico,
> 
> I just fixed some small encoding bugs that surfaced with JDK 
> 1.4.1 and updated 
> the test cases. I've run the test suite with all combinations 
> of JDK 1.4.1, 
> JDK 1.3.1_04, javac and jikes 1.15. Environment is Linux 
> (Kernel 2.4.19, 
> Distribution is Debian 3.0 with a number of packages upgraded to the 
> "unstable" branch), ant version is 1.5.1
> 
> Results are as follows:
> 
> JDK 1.4.1 + jikes 1.15 and JDK 1.4.1 + javac: All tests pass OK.
> 
> JDK 1.3.1_04 + jikes 1.15 and JDK 1.3.1_04 + javac: 
> 
> With the tidy parser, test 3 from the "EncodingTests" suite 
> fails because the 
> generated output differs from the expected result. This test 
> tries to work 
> with a file in the GB2312 (simplified chinese) encoding, and 
> it looks like 
> there were changes in how Java interprets this encoding 
> between JDK 1.3.1 and 
> JDK 1.4.1. As I don't read chinese, I've no idea which 
> interpretation is 
> correct (or even if the test case file really *is* encoded in 
> GB2312). For 
> now, I simply assume the newer version got it right, so I've 
> checked the 
> results from the JDK 1.4.1 run into CVS as expected results. 
> 
> All tests that use the swing HTML parser fail because the 
> output differs from 
> the expected results - seems like SUN changed the swing 
> parser between 1.3.1 
> and 1.4.1. Here, I assume as well that the newer version is 
> correct, so I 
> checked the results from the 1.4.1 run into CVS as expected results.
> 
> All other tests pass on both JDK 1.3.1 and JDK 1.4.1
> 
> Can you pelase update from CVS and verify my results?
> 
> Thanks,
> 
> Richard
> -- 
> Richard Kunze 
> 
> [ t]ivano Software, Bahnhofstr. 18, 63263 Neu-Isenburg
> Tel.: +49 6102 80 99 07 - 0, Fax.: +49 6102 80 99 07 - 1 
http://www.tivano.de, kunze@xxxxxxxxx 


_______________________________________________
XMLC mailing list
XMLC@xxxxxxxxxxx http://www.enhydra.org/mailman/listinfo.cgi/xmlc