[DDI-SRG] Naming conventions for packaging

[Joachim Wackerow]

Here is what we discussed already for the CR2 release:

A ZIP file should unpack into a subfolder with the same name. Below two 
directories exist, one for the schemas and one for the HTML documentation.

The ZIP file respectively the subfolder should follow this naming rule:

The angle brackets mark the variable parts in the name, they are not 
part of the name.
The square brackets mark optional parts.

DDI_<MAJOR_VERSION>_<MINOR_VERSION>[_CR<NUMBER>]_<ISO_DATE>_Documentation_XMLSchema

If a release has only schemas then (but that shouldn't happen):

DDI_<MAJOR_VERSION>_<MINOR_VERSION>[_CR<NUMBER>]_<ISO_DATE>_XMLSchema

This way it is always clear what it is independently in which folder the 
people unpack the files. This is especially important when dealing with 
different versions.

For example:

DDI_3_0_CR3_2008-02-12_Documentation_XMLSchema.zip

This should unpack into:

DDI_3_0_CR3_2008-02-12_Documentation_XMLSchema
   Documentation
     Modules
       module HTML documentation files ...
     DDI_3_0_CR3.html
   XMLSchema
     XHTML
       XHTML files ...
     archive.xsd
     comparative.xsd
     conceptualcomponent.xsd
     datacollection.xsd
     dataset.xsd
     dcelements.xsd
     ddi-xhtml11-model-1.xsd
     ddi-xhtml11-modules-1.xsd
     ddi-xhtml11.xsd
     ddiprofile.xsd
     DDI_CR3_Notes.txt
     group.xsd
     inline_ncube_recordlayout.xsd
     instance.xsd
     logicalproduct.xsd
     ncube_recordlayout.xsd
     physicaldataproduct.xsd
     physicalinstance.xsd
     proprietary_recordlayout.xsd
     reusable.xsd
     simpledc20021212.xsd
     studyunit.xsd
     tabular_ncube_recordlayout.xsd
     xml.xsd

These notes should perhaps available at the TIC web site.

Achim