On Nov 15, 2005, at 10:03, Jeremias Maerki wrote:

<snip />
The fix is probably to extract handleWhitespace
from Block into a separate class and call it from Block and Marker.

In this respect: I still wonder whether it wouldn't be more convenient to split up the whitespace handling, and deal with the inlines separately. Currently, InlineCharIterator needs to generate boundary characters to indicate start- or end-inline. If we would deal with the whitespace of the inlines at inline-level itself, it should become far more straightforward to apply the 'special' rules (no removal of the first/last space of the inline, or before it).

On top of that, it does away with the need to chain together all FOText instances of a whole block (thus making that ugly static 'lastFOTextProcessed' obsolete?)

Extracting handleWhitespace() into a separate class would, in any case, be A Good Thing.

My 2 cents.

Cheers,

Andreas

Reply via email to