Many warnings from InMemoryDAO while reading DIP data

[GoogleCodeExporter]

When parsing DIP db (human only, PSI-MI 2.5, 
ftp://dip.doe-mbi.ucla.edu/2014/mif25/Hsapi20140117.mif25.gz) 
I am getting hundreds of warnings from 
psidev.psi.mi.xml.dao.inMemory.InMemoryDAO class
(mostly about ExperimentDescription or Participant, and id - either 0, 1, or 
2....):

2014-03-24 12:53:06,308 WARN  p.p.m.x.d.i.InMemoryDAO - Attempt to overwrite an 
existing ExperimentDescription with id: 0
2014-03-24 12:53:06,308 WARN  p.p.m.x.d.i.InMemoryDAO - The new 
ExperimentDescription was different
2014-03-24 12:53:06,308 WARN  p.p.m.x.d.i.InMemoryDAO - Original : 
ExperimentDescription{id=0, names=null, 
bibref=Bibref{xref=Xref{primaryRef=DbReference{attributes=null, db='pubmed', 
dbAc='MI:0446', id='7739537', refType='identity', refTypeAc='MI:0356', 
secondary='null', version='null'}, secondaryRef=null}, attributes=null}, 
xref=Xref{primaryRef=DbReference{attributes=null, db='dip', dbAc='MI:0465', 
id='DIP-698X', refType='identity', refTypeAc='MI:0356', secondary='null', 
version='null'}, secondaryRef=null}, hostOrganisms=null, 
interactionDetectionMethod=CvType{names=Names{shortLabel='null', 
fullName='coimmunoprecipitation', aliases=null}, 
xref=Xref{primaryRef=DbReference{attributes=null, db='psi-mi', dbAc='MI:0488', 
id='MI:0019', refType='identity', refTypeAc='MI:0356', secondary='null', 
version='null'}, secondaryRef=null}}, participantIdentificationMethod=null, 
featureDetectionMethod=null, confidences=null, 
attributes=[Attribute{value='MI:0218', name='dip:evidence-for-ac', 
nameAc='null'}, Attribute{value='physical interaction', 
name='dip:evidence-for-name', nameAc='null'}, Attribute{value='dip:0002', 
name='dip:evidence-scale-ac', nameAc='null'}, Attribute{value='small scale', 
name='dip:evidence-scale-name', nameAc='null'}, Attribute{value='', 
name='dip:evidence-processing-status-ac', nameAc='null'}, Attribute{value='', 
name='dip:evidence-processing-status-name', nameAc='null'}]}

Are these norm, i.e., the parser successfully recovers from those errors; can 
we depend on the data/model after it's read or something to be fixed?

Original issue reported on code.google.com by rod...@gmail.com on 24 Mar 2014 at 9:55