https://issues.apache.org/bugzilla/show_bug.cgi?id=46905





--- Comment #30 from Andreas L. Delmelle <[email protected]>  2009-06-26 
01:25:44 PST ---
Created an attachment (id=23886)
 --> (https://issues.apache.org/bugzilla/attachment.cgi?id=23886)
sample PDF result showing a remaining issue

Further testing and digging revealed that the last attached FO/PDF sample
contains a fluke... 
If I lengthen the nested block-4a so that it causes a page-level overflow, none
of the breaks in the last column will be considered/registered as feasible
(page-)breaks. As a consequence, the next restart is triggered from the last
column-break. An additional node is inserted to fill the last column, and the
remainder is pushed to the next page, while it should actually just overflow
the column/page. Since the last part is also run through the column-balancing
algorithm, I end up with suboptimal break-decisions...

For the rest, in the meantime, I have made a few more modifications to the last
patch, so am now down to four failing testcases (five, when including the
above). Will post an updated patch later today.

-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

Reply via email to