Funding for Float support in FOP?

["Aaron W. Hsu" <ar...@sacrideo.us>]

It's nice that I've been able to work around the lack of floats in FOP
for my sigplan article stylesheet, but it still irks me quite a bit. I
don't have the time an inclination to fix the problem myself, given the
lack of time and expertise in this area. However, I at least want to
look into whether it might not be possible to support getting float
support into FOP somehow. Thus, the following:

Given appropriate amounts of funding, is there someone with the
appropriate expertise willing to implement float support and the like in
FOP? If so, what is the funding amount that would be necessary? 

I'm looking for honest answers here. I'm not sure I can raise this level
of funding, but at the very least it would be good to know how much it
would cost. I have enough people somewhat interested in this that it
would make sense to make an effort if possible. 

-- 
Aaron W. Hsu | arcfide@sacrideo.us | http://www.sacrideo.us
לֵ֤ב חֲכָמִים֙ בְּבֵ֣ית אֵ֔בֶל וְלֵ֥ב כְּסִילִ֖ים בְּבֵ֥ית שִׂמְחָֽה׃


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Re: Retrieve-Marker retrieve-position = all?

[Luis Bernardo <lm...@gmail.com>]

I don't think a general solution is possible within XSL-FO. If the 
number of entries (titles) per page is fixed (except for the last page) 
then it is simple. If the number of entries per page is bounded, and 
there is enough real estate in the page to place the resolved markers 
then it can be done with some manipulation of intermediate format (IF).

The approach would be for every marker to register the current value and 
the next few ones. Say, you do not expect more than 5 entries per page. 
Then for every title you would use two markers. So marker 
"this-and-next" for title10 would be "title10, title11, title12, 
title13, title14", and marker "this" would be "title10". Then you would 
retrieve the first "this-and-next" marker in the page, and the last 
"this" marker in the page. The resolved values in IF would then look 
like "title10, title11, title12, title13, title14, title13", where the 
last entry if from the last "this" marker in the page. Then you would 
process the IF to reduce it to "title10, title11, title12, title13".

On 2/14/14, 6:45 PM, Matthias Reischenbacher wrote:
> Hi,
>
> I'd like to display multiple markers with the same name on the page 
> header. Conceptually those markers correspond to chapter titles and 
> all chapter titles, visible on a single page, should be printed. Up 
> until now I've used the retrieve-position attribute for showing the 
> first or last chapter title. But this time I'd need the first, last 
> and all in between on the current page. Is this even possible with 
> XSL-FO/FOP? If not, is it achievable by manipulating the fop 
> intermediate structure?
>
> Best regards,
> Matthias
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
> For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org
>


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Retrieve-Marker retrieve-position = all?

[Matthias Reischenbacher <ma...@gmx.at>]

Hi,

I'd like to display multiple markers with the same name on the page 
header. Conceptually those markers correspond to chapter titles and all 
chapter titles, visible on a single page, should be printed. Up until 
now I've used the retrieve-position attribute for showing the first or 
last chapter title. But this time I'd need the first, last and all in 
between on the current page. Is this even possible with XSL-FO/FOP? If 
not, is it achievable by manipulating the fop intermediate structure?

Best regards,
Matthias

---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Re: Funding for Float support in FOP?

["Aaron W. Hsu" <ar...@sacrideo.us>]

Vincent Hennebert <vh...@gmail.com> writes:

> On top of this, an implementation of the standard fo:float element
> wouldn’t allow you to achieve the Sigplan layout you mentioned in this
> other thread:
> http://mail-archives.apache.org/mod_mbox/xmlgraphics-fop-users/201402.mbox/%3C87ob2mkv58.fsf%40sacrideo.us%3E

Thanks. I'm aware that floats won't help me there. That was made clear
to me from the outset, though I do wish they would. However, floats seem
helpful in some other elements of handling DocBook documents. 

> All that said, implementing before-floats should be doable in the
> current layout engine without too much effort. There’s a branch in
> Subversion with the start of an implementation that would need to be
> brought up-to-date, debugged, completed and cleaned up. I can’t say much
> more without studying the problem in more details, but we are talking in
> terms of weeks rather than months or years.

Thanks for this information. I think it would be nice if FOP could get
an extension to XSL-FO that would allow it to put "headers" or "footers"
on specific pages and specific columns. I do not know if that is a
feature generally desirable or not, but I know it would be nice for at
least me. :-) On the other hand, I have a working workaround, so I'm not
worried about it right now.

-- 
Aaron W. Hsu | arcfide@sacrideo.us | http://www.sacrideo.us
Please support my work: https://www.gittip.com/arcfide/

לֵ֤ב חֲכָמִים֙ בְּבֵ֣ית אֵ֔בֶל וְלֵ֥ב כְּסִילִ֖ים בְּבֵ֥ית שִׂמְחָֽה׃


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Re: Funding for Float support in FOP?

[Vincent Hennebert <vh...@gmail.com>]

On 13/02/14 18:29, Aaron W. Hsu wrote:
> Chris Bowditch <bo...@hotmail.com> writes:
>
>> Assuming the prototype can be finished within say another 12 months,
>> it would then take several further man years to port the rest of FOP's
>> functionality onto the prototype. We estimate the implementation cost
>> would be in the range of several hundred thousand GBP :-o
>>
>> This is why fo:float has not yet been implemented. Sorry to be the
>> bearer of bad news.
>
> Thanks for the information. It's not bad news for me at all, but more
> just a data point. I wanted to see an actual estimate of just how
> difficult it is to add float support to FOP, and your estimate is quite
> helpful. I haven't read the specification myself, so I have no intuition
> about how crazy floats are, and it doesn't surprise me that a feature
> like this could take years to implement.

On top of this, an implementation of the standard fo:float element
wouldn’t allow you to achieve the Sigplan layout you mentioned in this
other thread:
http://mail-archives.apache.org/mod_mbox/xmlgraphics-fop-users/201402.mbox/%3C87ob2mkv58.fsf%40sacrideo.us%3E

You can’t use side-float because you don’t know in which paragraph to
put the anchor such that it would end up at the bottom of the first
column. You can’t use before-float because it would flow up to the top
of the page. Neither can you use a footnote, because it would take up
the whole width of the page, so it would push both columns up. Plus in
some cases it might be deferred to the next page.

So to my knowledge your requirement is not achievable in plain XSL-FO.

All that said, implementing before-floats should be doable in the
current layout engine without too much effort. There’s a branch in
Subversion with the start of an implementation that would need to be
brought up-to-date, debugged, completed and cleaned up. I can’t say much
more without studying the problem in more details, but we are talking in
terms of weeks rather than months or years.

Implementing side-floats is IMO not achievable without undertaking
a major redesign of the layout engine.


HTH,
Vincent

---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Re: Funding for Float support in FOP?

["Aaron W. Hsu" <ar...@sacrideo.us>]

Chris Bowditch <bo...@hotmail.com> writes:

> Assuming the prototype can be finished within say another 12 months,
> it would then take several further man years to port the rest of FOP's
> functionality onto the prototype. We estimate the implementation cost
> would be in the range of several hundred thousand GBP :-o
>
> This is why fo:float has not yet been implemented. Sorry to be the
> bearer of bad news.

Thanks for the information. It's not bad news for me at all, but more
just a data point. I wanted to see an actual estimate of just how
difficult it is to add float support to FOP, and your estimate is quite
helpful. I haven't read the specification myself, so I have no intuition
about how crazy floats are, and it doesn't surprise me that a feature
like this could take years to implement.

-- 
Aaron W. Hsu | arcfide@sacrideo.us | http://www.sacrideo.us
לֵ֤ב חֲכָמִים֙ בְּבֵ֣ית אֵ֔בֶל וְלֵ֥ב כְּסִילִ֖ים בְּבֵ֥ית שִׂמְחָֽה׃


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org

Re: Funding for Float support in FOP?

[Chris Bowditch <bo...@hotmail.com>]

Hi Aaron,

Implementation of fo:float is one of the most trickest algorithms within 
the XSL-FO specification. Vincent H worked on a prototype layout engine 
that could accomodate fo:floats in 2008, but after 12 months we had to 
abandon it as it was still some distance from being finished. Thats just 
the prototype, and excluded all the output generation and many other 
features in FOP.

Assuming the prototype can be finished within say another 12 months, it 
would then take several further man years to port the rest of FOP's 
functionality onto the prototype. We estimate the implementation cost 
would be in the range of several hundred thousand GBP :-o

This is why fo:float has not yet been implemented. Sorry to be the 
bearer of bad news.

Of course, I'm happy to learn if anyone has any idea how to implement a 
solution using the existing Knuth Layout Algorithm.

Chris

On 11/02/2014 03:47, Aaron W. Hsu wrote:
> It's nice that I've been able to work around the lack of floats in FOP
> for my sigplan article stylesheet, but it still irks me quite a bit. I
> don't have the time an inclination to fix the problem myself, given the
> lack of time and expertise in this area. However, I at least want to
> look into whether it might not be possible to support getting float
> support into FOP somehow. Thus, the following:
>
> Given appropriate amounts of funding, is there someone with the
> appropriate expertise willing to implement float support and the like in
> FOP? If so, what is the funding amount that would be necessary?
>
> I'm looking for honest answers here. I'm not sure I can raise this level
> of funding, but at the very least it would be good to know how much it
> would cost. I have enough people somewhat interested in this that it
> would make sense to make an effort if possible.
>


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org