Open GoogleCodeExporter opened 9 years ago
Modularization of jmztab is good idea to cover different requirements of users.
We will think about it in the coming version.
You would like to separate small molecules part from jmztab, however, I thought
your modularization way not support it. Because of, there exists many
relationships among different sections in mzTab file. In jmztab project, we
create a separate module
(https://code.google.com/p/mztab/source/browse/#svn%2Fjmztab%2Ftrunk%2Fsrc%2Fmai
n%2Fjava%2Fuk%2Fac%2Febi%2Fpride%2Fjmztab%2Fmodel) to maintain five sections
(metadata, protein, peptide, PSM, small molecule) as a whole target.
In the next stage, we will think about separating parse, validate and gui/cli
utils from jmztab core module, and create a couple of dependences projects.
Hopefully new solution will archive your requirement.
I looked your work in git repository, it looked very nice. We will reference
your modularization way and hopefully keep communication in the future.
Original comment by xuqing...@gmail.com
on 22 Feb 2014 at 2:38
I changed the groupId in the github project to net.sf.maltcms.mztab for now to
avoid any artifact clashes between your main-line and the fork. I will be happy
to help with a reintegration in the future. I am not planning to add any major
changes, so that should be manageable.
The modularization seemed necessary for Maltcms due to the large number of
dependencies that were pulled in transitively through the ion-generator
dependency. The reorganization helped me to understand the interdependencies
within jmztab much better. Maybe the (transitive) dependencies that are not
required for jmztab's functionality could be removed in the future?
Original comment by nils.hof...@web.de
on 4 Mar 2014 at 7:55
Original issue reported on code.google.com by
nils.hof...@web.de
on 21 Feb 2014 at 4:48