[Biojava-dev] build failure strangeness: bogus errors

[Matthew Pocock]

A number of people have reported this. It has generaly gone away after 
doing an "ant clean package"

Matthew

Keith James wrote:
> I've got build failures on a fresh checkout since the projection
> changes (identical to Sylvian's):
> 
> compile:
>     [javac] Compiling 1069 source files to /nfs/team81/kdj/dev/biojava-live/ant-build
> /classes/main
>     [javac] /nfs/team81/kdj/dev/biojava-live/ant-build/src/main/org/biojava/bio/seq/i
> mpl/RevCompSequence.java:47: reference to ProjectedFeatureHolder is ambiguous, both c
> lass org.biojava.bio.seq.projection.ProjectedFeatureHolder in org.biojava.bio.seq.pro
> jection and class org.biojava.bio.seq.ProjectedFeatureHolder in org.biojava.bio.seq m
> atch
>     [javac]     private ProjectedFeatureHolder pfh;
>     [javac]             ^
>     [javac] /nfs/team81/kdj/dev/biojava-live/ant-build/src/main/org/biojava/bio/seq/i
> mpl/RevCompSequence.java:63: reference to ProjectedFeatureHolder is ambiguous, both c
> lass org.biojava.bio.seq.projection.ProjectedFeatureHolder in org.biojava.bio.seq.pro
> jection and class org.biojava.bio.seq.ProjectedFeatureHolder in org.biojava.bio.seq m
> atch
>     [javac]         pfh = new ProjectedFeatureHolder(new TranslateFlipContext(seq,thi
> s,seq.length()+1,true));
>     [javac]                   ^
>     [javac] /nfs/team81/kdj/dev/biojava-live/ant-build/src/main/org/biojava/bio/seq/S
> equenceTools.java:90: cannot resolve symbol
>     [javac] symbol  : constructor SubSequence (org.biojava.bio.seq.Sequence,int,int,j
> ava.lang.String)
>     [javac] location: class org.biojava.bio.seq.SubSequence
>     [javac]     return new SubSequence(seq, start, end, name);
>     [javac]            ^
>     [javac] 3 errors
> 
> Looking at the code, I don't think this should be happening. If I
> change the imports on RevCompSequence so that I explictly import the
> required classes from seq.projection before I do an import on demand
> from seq then RevCompSequence will compile.
> 
> The SubSequence error is also bogus because this public constructor
> does exist. If I do explicit imports again this also goes away.
> 
> I'm not sure what the underling cause of this is - any ideas?
> 
> maul[kdj]~/dev/biojava-live: uname -a
> OSF1 maul V5.1 2650 alpha
> 
> maul[kdj]~/dev/biojava-live: java -version
> java version "1.4.1"
> Java(TM) 2 Runtime Environment, Standard Edition
> Fast VM (build 1.4.1-1, native threads, mixed mode, 02/26/2003-14:52)
> 


-- 
BioJava Consulting LTD - Support and training for BioJava
http://www.biojava.co.uk