I use FOP 0.20.5 and XEP 4.
When producing tables the footnotes are normally well known in advance
and most tables have sufficient rows that there will always be enough
for all footnotes to be printed at the bottom of the page.
I agree the behaviour is not clearly defined by the spec. and obviously
how the processor treats table headers.
If it processes them once and saves the output for reproducing on
subsequent pages then
the treatment of footnotes may be restricted to the first page.
But I reasoned that - if subsequent pages could have different layout
then any formatter would have to process the headers for each page
and would be behaving very strangely if they 'decided' that some output
wasn't to be
reproduced on each page.
Eliot Kimber wrote:
>Mike Trotman wrote:
>>If your tables break over pages and you have a table-header (which
>>therefore repeats at the top of each page)
>>then you could put your table footnote in a real footnote in a cell in
>>the table-header section.
>>Then - whenever the table-header repeats - you get the footnote at the
>>bottom of the table.o
>>This does work - because I use it for exactly the effect you are talking
>While FO implementation are you using? I tried this with XSL Formatter
>3.2 and I only got a footnote on the first page of the table but XEP 4
>repeated it on each page.
>In reviewing the FO 1.1 CR spec I don't see anything that clearly
>specifies what the implication of having out-of-line objects in the
>table header are.
>I think it's reasonable to expect that a footnote specified in the table
>header be repeated on each page on which the header is repeated but I
>don't see anything in the spec that either demands it or disallows it.
>One potential danger with approach, although probably not a likely
>problem, is that implementations are not obligated to place a footnote
>on the same page as the callout--they may choose to force them to a
>following page if there's not enough room on the callout's page.
Message Scanned by ClamAV on datalucid.com