On Fri, 2008-10-31 at 18:04 +0100, Hanssens Bart wrote:
d) Add this version info in the meta info of the document itself
Pro: very easy and quick to do
Pro: you don't have to deal with other tools / hacks
Con: this would actually make atomic tests a little bit dirtier
(probably nitpicking, you could remove the data using a script
when you really want to remove every bit not part of the test)
d1) Using ODF's metadata functionality
d2) Adding elements in a separate namespace
Suggestions ? Remarks ? :-) Personally, I would opt for d1 or d2,
but thats's just my two cents...
On first glance (d) seems to be the right way to go but since we expect
can expect a really large number of files, running say all tests for
version 1.0 would require that _all_ files are accessed, the metadata
retrieved from them etc.
In this situation a mechanism that works on the file level e.g. (a) or
(b) would make it easier to select the files of interest.
Andreas J. Guelzow <ague...@math.concordia.ab.ca>
Andreas J. Guelzow