https://issues.apache.org/bugzilla/show_bug.cgi?id=51710
--- Comment #3 from Antti Koskimäki <antti.koskim...@joinex.com> 2011-08-23 20:52:18 UTC --- I stared the xml for a while and then noticed, that only thing that doesn't make sense is "ref" attribute of formula. The declared shared formula is always correct relative to the containing cell. The ref-attribute on the other hand is almost random, and does not in any way reflect to the formula copying process. What it seems to define is boundaries where the shared instance is in use. When I count the references, it appears that there's familiar magic numbers involved :=) 176 si="0" 255 si="1" 255 si="2" 255 si="3" 64 si="4" 255 si="5" 255 si="6" 64 si="7" .. So, it seems to me that A) one formula "instance" is shared maximum 255 times B) distribution of these shared instances is somewhat random (mystical optimization?) C) ref-attribute really does only define the "boundaries", nothing else I also peeked the POI code, and found out that there's calculations on ss-level. It raised a question if HSSF-heritage is misleading the XSSF-implementation here ? P.S. document that made me understand this shared-thing: http://www.codeproject.com/KB/XML/ooxml_is_defective.aspx (see chapter 3) -- Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@poi.apache.org For additional commands, e-mail: dev-h...@poi.apache.org
