Changes in version 1.9.1                        

    o   Move example files into package to avoid spurious check errors in windows

                        Changes in version 1.7.1                        

    o   Imports generics from ProtGenerics

    o   Export base classes

                        Changes in version 1.5.3                        

    o   Use "location" attribute instead of "name" to create the
	spectrumFile/databaseFile column in the flattened data.frame.

                        Changes in version 1.3.6                        

    o   Huge restructuring in documentation

    o   Fixed a bug that would throw an error when an empty mzID object was flattened

    o   Restructuring of unit tests so they conform with 'new' guidelines and gets
	called during check

                        Changes in version 1.3.5                        

    o   Bug fixes

                        Changes in version 1.3.4                        

    o   Bug fixes

                        Changes in version 1.3.3                        

    o   Added method removeDecoy, to remove decoy specific information from an mzID or
	mzIDCollection object thus decreasing its size.

    o   Added formal getter methods to extract specific information from mzID and
	mzIDCollection objects

    o   All attribute names are now parsed as-is despite inconsitency between schema
	versions. Controlling compatability is now done using the safeName parameter 
	in flatten and getter methods

    o   Substantial cleanup in documentation

                        Changes in version 1.3.2                        

    o   Fixed bug due to R's change in type.convert() that would result in precise 
	numbers being reported as strings

                        Changes in version 1.3.1                        

    o   Fixed bug that would cause error on files with no modifications in the search
	parameters

                         Changes in version 1.2                         

    o   Release on Bioconductor

                        Changes in version 1.1.6                        

    o   Added documentation

                        Changes in version 1.1.5                        

    o   Now computes mzR compatible acquisitionNum for the scans (thanks to Sebastian
	Gibb)

    o   Checks for existence of local files (thanks to Sebastian Gibb)

    o   XML moved from depend to import

                        Changes in version 1.1.4                        

    o   Now computes mzR compatible acquisitionNum for the scans (thanks to Sebastion
	Gibb)

    o   Various bug fixes

                        Changes in version 1.1.3                        

    o   Introducing the mzIDCollection class to handle multiple mzID objects

                        Changes in version 1.1.2                        

    o   Added the possibility to create the different 'sub'classes of mzID directly
	from a file without first having an internal xml representation and namespace

                        Changes in version 1.1.1                        

    o   Fixed bug where multiple names in the modification rules would crash the
	parsing. Multiple names gets collapsed with '/'

                        Changes in version 1.0.0                        

    o   Accepted in Bioconductor

                       Changes in version 0.99.3                        

    o   Fix for issue #6. It now correctly parses files with empty AnalysisData
	nodes as empty mzID objects.

    o   Full support for mzIdentML v1.0 and v1.1 

                       Changes in version 0.99.2                        

    o   Accepted in Bio 2.13

                        Changes in version 0.9.1                        

    o   Added unit test using testthat for low level functions

    o   Added BiocViews DataImport, MassSpectrometry, Proteomics

                        Changes in version 0.9.0                        

    o   Submitted to Bioconductor

                        Changes in version 0.3.1                        

    o   Comment unused functions <2013-07-05 Fri>

    o   Minor typos in vignette  <2013-07-05 Fri>

                        Changes in version 0.3.0                        

NEW FEATURES AND FUNCTIONS

    o   Support for mzIdentML v1.0 (thanks Laurent Gatto)

                        Changes in version 0.2.1                        

NEW FEATURES AND FUNCTIONS

    o   Version counter now in BioConductor style

    o   Namespace now extracted from file instead of hardcoded in the constructor

    o   mzID constructor now checks the mzIdentML version before parsing (only 1.1.0
	supported atm)

                        Changes in version 0.1-1                        

NEW FEATURES AND FUNCTIONS

    o   Added function `flatten` to create tabular representation of results

                        Changes in version 0.0-2                        

NEW FEATURES AND FUNCTIONS

    o   The package is now fully documented

    o   created helper functions: `countChildren`, `attrExtract` and
	`attrExtractNameValuePair`

    o   Added NEWS file

    o   Added README.md file

    o   The parser now succesfully imports all test files on the HUPO mzIdentML page

PERFORMANCE

    o   Improvements in the mzIDpsm constructor makes it ~2-3x faster depending on the
	file size (thanks to `countChildren`, `attrExtract` and
	`attrExtractNameValuePair`)

                        Changes in version 0.0-1                        

NEW FEATURES AND FUNCTIONS

    o   First build of the package

    o   Introduction of the classes: `mzID`, `mzIDdatabase`, `mzIDevidence`,
	`mzIDparameters`, `mzIDpeptides` and `mzIDpsm` with related constructors