Hi Don, Arved
i would be greatfull to work with someone on a related problem, who  
better overlooks FOP then i do at the moment .
right now i have some questions:

1. i do not fully understand about the numerical and symbolic values for 
the keep-* stuff.
the standard says (in "4.8 Keeps and Breaks") that there is any interger 
value allowed which specify the "strength", whereas higher number means 
more strength is imposed on the keep condition. in other words the 
bigger the values is the more the fo-processor should try to keep the 
stuff together, right?   but there are in addition to symbolic values
"auto" (which means no keep condition) and "always" (which means " keep 
always ") .
on the other hand in FOP   "always" is mapped to "5" in fop and "auto" 
is mapped to "7".
which would mean only the values 5,6 and 7 are meaningfull  and moreover 
it looks as if the always should actually at least be greater then "auto".
could you help me reducing my confusion?

2. i'm unsure about the "within-*" (e.g. "within-page") property 
refinement.
if for example, i have a block with
keep-with-next.within-page="always" but
keep-with-next.within-column="auto"
does this mean that i should make a page break before that block if 
 normaly the pagebreak would occure after it. but i should not start new 
column before that block instead let occure the column break after this 
block as if no keep-with-next property would have been set, right?

3. i don't yet understand how the mechanics of the FONode.rollback() 
works. it seems it has something to do with undoing layout operations in 
order to redoing this part of the  layout with eg. different column or 
page breaks (e.g. if a keep condition occured that could not be 
fulfilled). am i on the right trak here? do you any more details on 
that, or do i need just to look more deeply in to the source?

Bodo

 

Don Wellington wrote:

>Hi Arved-
>
>I submitted patches against 20.1 to do an effective
>keep-with-next on list-item-label and list-item-body,
>except I assumed an implied keep-with-next on
>list-item-label.  I would be glad to work with bodo on
>this time permitting.  It would probably be best, so
>that there are not two different pieces of code
>floating around doing nearly the same thing.
>
>Don
>
>--- Arved Sandstrom <[EMAIL PROTECTED]> wrote:
>
>>Hi, Bodo
>>
>>I'd be very happy to help. I'm badly sidelined due
>>to work pressure, but 
>>there is always some time I can pry loose...I'd
>>rather spend that time 
>>getting new developers up to speed.
>>
>>I'll review the source this evening and give you a
>>synopsis of what we are 
>>facing here.
>>
>>Regards,
>>Arved Sandstrom
>>
>>At 12:18 PM 8/27/01 +0200, [EMAIL PROTECTED] wrote:
>>
>>>hi,
>>>i tried to use the keep-with-next property on a
>>>
>>multi column page with 
>>
>>>some  text blocks that  normally fits on one page
>>>
>>with 4 columns.
>>
>>>my text consists of this sequence of text blocks:
>>>
>>every 2nd block is 
>>
>>>meant to be a headline and  the the following block
>>>
>>is the paragraph 
>>
>>>belonging to that headline.
>>>therefore i set the keep-with-next attribute on the
>>>
>>headline blocks but 
>>
>>>still the column break occurred between the
>>>
>>headline and the paragraph.
>>
>>>looking in to the code i found that  in 
>>>org.apache.fop.fo.flow.Block.layout(Area) the line:
>>>this.keepWithNext =
>>>                          
>>>
>>this.properties.get("keep-with-next").getEnum();
>>
>>>does not have any effect , since the getEnum()
>>>
>>method always returns (0) 
>>
>>>because the
>>>org.apache.fop.fo.KeepProperty(Keep) does not
>>>
>>overwrite the getEnum() 
>>
>>>method.
>>>
>>>so i patched this so that this.keepWithNext got the
>>>
>>value="5" (just to 
>>
>>>see what will happen then ) but immediately  i run
>>>
>>in to another error:
>>
>>>the headline appeared on the bottom of the old
>>>
>>column, a wrong page 
>>
>>>break occurred (instead of a column break)
>>>and the same headline appeared on the top of the
>>>
>>new page.
>>
>>>if nobody else is currently working on this
>>>
>>problems i would be willing 
>>
>>>to do this, if i (as a someone who just started to
>>>
>>try to understand the 
>>
>>>fop source code) get some assistance from the fop
>>>
>>gurus.
>>
>>>would you appreciate this?
>>>
>>>bodo teichmann
>>>
>>>
>>>
>>>
>>---------------------------------------------------------------------
>>
>>>To unsubscribe, e-mail:
>>>
>>[EMAIL PROTECTED]
>>
>>>For additional commands, email:
>>>
>>[EMAIL PROTECTED]
>>
>>>
>>Fairly Senior Software Type
>>e-plicity (http://www.e-plicity.com)
>>Wireless * B2B * J2EE * XML --- Halifax, Nova Scotia
>>
>>
>>
>---------------------------------------------------------------------
>
>>To unsubscribe, e-mail:
>>[EMAIL PROTECTED]
>>For additional commands, email:
>>[EMAIL PROTECTED]
>>
>
>
>__________________________________________________
>Do You Yahoo!?
>Make international calls for as low as $.04/minute with Yahoo! Messenger
>http://phonecard.yahoo.com/
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: [EMAIL PROTECTED]
>For additional commands, email: [EMAIL PROTECTED]
>
>



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]

Reply via email to