While http://tools.oasis-open.org/issues/browse/OFFICE-1243 http://tools.oasis-open.org/issues/browse/OFFICE-1246 http://tools.oasis-open.org/issues/browse/OFFICE-1370 http://tools.oasis-open.org/issues/browse/OFFICE-1380 http://tools.oasis-open.org/issues/browse/OFFICE-1470 http://tools.oasis-open.org/issues/browse/OFFICE-1371 http://tools.oasis-open.org/issues/browse/OFFICE-1373 and http://tools.oasis-open.org/issues/browse/OFFICE-1379 are primarily concerned with some implementaion not performing the white space handling described in the standard and the misunderstanding that white space collapse or removal implies that white space cannot be correctly represented, this issue relates to some inconsistence I believe to have run into:
The general text in 1.5 states:
"White-space characters occurring in elements with element content only are ignored. "
"For any other element, white-spaces are preserved by default. Elements may declare specific rules
for the handling of white-space characters. 5.1.1"
I am wondering why there is a difference in white space handling between the <text> element with a reasonable if not perfect mechanism (the mechanism still doesn't allow pretty printing of the document) and the <number:text> element in which white space suddenly matters. Especially in <number:text> whitespace is probably occurring rarely so that a requirement to use <text:s> and a complete whitespace collapse would be quite useful!