Re: problems with tables

[Bh <bh...@hotmail.com>]

Any ideas when the auto layout is going to be supported? Really waiting for
this feature (damn, i wish i had some time to help these ppl out)


----- Original Message -----
From: "Victor Mote" <vi...@outfitr.com>
To: <fo...@xml.apache.org>
Sent: Friday, October 18, 2002 9:24 AM
Subject: RE: problems with tables


Andreas Kroop wrote:

> i have a problem with tables (fop 0.20.4)

...

>
> message:
> [WARNING] table-layout=auto is not supported, using fixed!

The default value of the "table-layout" property is "auto". However, FOP
does not yet support that feature. To get the WARNING message to go away,
add the 'table-layout="fixed"' attribute to your table element.

Victor Mote

Re: problems with tables

["J.Pietschmann" <j3...@yahoo.de>]

Alex McLintock wrote:
>> Any ideas when the auto layout is going to be supported? Really 
>> waiting for
>> this feature (damn, i wish i had some time to help these ppl out)
> 
> I'm sure we could find someone to implement it if you want to pay for 
> the enhancement.

I wouldn't hold my breath. Automatic table layout is one of the more
tricky things to implement, for example it is not compatible with
page streaming and/or needs backtracking. While it looks similar
to column rebalancing, and probably is similar, column rebalancing
is confined to one page, while tables might span multiple pages (I
still remember the guy who produced a 3000 page table - ouch!)

J.Pietschmann

Re: problems with tables

[Alex McLintock <al...@OWAL.co.uk>]

At 15:40 18/10/02, you wrote:
>Any ideas when the auto layout is going to be supported? Really waiting for
>this feature (damn, i wish i had some time to help these ppl out)


I'm sure we could find someone to implement it if you want to pay for the 
enhancement.

Alex





Openweb Analysts Ltd, London.
Software For Complex Websites http://www.OWAL.co.uk/
Open Source Software Companies please register here 
http://www.OWAL.co.uk/oss_support/

RE: problems with tables

["Matthew L. Avizinis" <ml...@gleim.com>]

I wasn't implying FOP is poor in concept -- in fact, I think it's excellent; that's why we're using it.  I was just stating that many organizations use products like from Quark or Adobe for electronic publishing and haven't found a purpose for a program like FOP just yet....that is, they haven't seen the light.  :)

> *** disagree.... for the record, the concept of fop is just gr8. 
> i am using
> fop to transform my xml documents in to 8 different formats (some still to
> be done) including xml, pdf, csv, txt, mif, svg, ps, pcl, tiff 
> and rtf (rtf,
> tiff, svg, mif to be done) I dont know of anyother product which provides
> these many options. agreed fop is clunky to use but works gr8. 
> i'll stop now
> :)
> 

Re: problems with tables

[Bh <bh...@hotmail.com>]

I disagree.  If one were to look at the proposed development timelines of
other (for instance) Apache projects, such as Xalan, one would see that they
are striving to make reasonable timelines.

*** agreed xalan is awesome

This is, imho, probably due to having larger numbers of active developers.
Also, because projects like Xalan have wider commercial application right
now than FOP.  Using XSLfo is not a widely accepted publishing paradigm.
We're moving to it right now for paper book, software, and online products,
but I believe that is not the anywhere near the norm yet, since there are
many other good electronic publishing methods "out there".  As more (and
richer) companies find a need for FOP (and support the Open Source concept)
they will be more willing to allow use of company resources to support this
project.

*** disagree.... for the record, the concept of fop is just gr8. i am using
fop to transform my xml documents in to 8 different formats (some still to
be done) including xml, pdf, csv, txt, mif, svg, ps, pcl, tiff and rtf (rtf,
tiff, svg, mif to be done) I dont know of anyother product which provides
these many options. agreed fop is clunky to use but works gr8. i'll stop now
:)

Me, I'm not a java developer, so I'm not much help with this project other
than to report problems I've encountered and help with the occasional
question.  I've got my hands so heaping full at work developing the XSL
transforms to generate the fo for all the books (content people can be quite
particular in how they want to present things) we've got and online stuff,
and a 10-month-old to chase after at home in my off time that I don't have
much time anyhow-- I've submitted _one_ FOP extension function for writing
page numbers to an alternate file and that's about all I'm up for.

Am I rambling?  I'll stop now....

   Matthew L. Avizinis <ma...@gleim.com>
Gleim Publications, Inc.
   4201 NW 95th Blvd.
 Gainesville, FL 32606
(352)-375-0772
      www.gleim.com <http://www.gleim.com>

=======================================================================
computing (kum' pyoot ing)
1. n the art of calculating how much time you wasted and money you spent in
a doomed attempt to master a machine with a mind of it's own. --from
Computing: A HACKER'S DICTIONARY


> -----Original Message-----
> From: Victor Mote [mailto:vic@outfitr.com]
> Sent: Friday, October 18, 2002 11:28 AM
> To: fop-user@xml.apache.org
> Subject: RE: problems with tables
>
>
> Bh wrote:
>
> > Any ideas when the auto layout is going to be supported? Really
> > waiting for
> > this feature
>
> If one were predicting such a thing in a commercial environment,
> with relatively predictable resources, one would expect to be off
> by a factor of at least 3. To predict such a thing on an
> open-source environment is not possible.
>
> > i wish i had some time to help these ppl out)
>
> Well, if you don't have time, perhaps your employer could fund
> the salary of a developer who could. There are probably many
> companies out there that could benefit greatly by investing in
> FOP that way instead of putting a greater number of dollars into
> a commercial product that they have no control over. Another
> alternative would be to investigate the commercial
> implementations. I don't know whether any support this feature or
> not, but even if they do not, they could probably also be
> persuaded by dollars.
>
> Victor Mote
>
>

Re: problems with tables

[Gour <go...@mail.inet.hr>]

Oleg Tkachenko (olegt@multiconn.com) wrote:

Hi!

> Hey, that's a horse of a different color! Xalan's maiden name is 
> lotusxsl and the code was donated to apache by blue giant, which still 
> actively supports it, look at 
> http://xml.apache.org/xalan-j/readme.html#developer-list - 10 of 13 
> developers have emails at ibm.com domain, you know, fop doesn't have 
> such support, so we are on our own.

I'm new on this list as well in the world of XML, XSLT, FO .. moving from LyX &
LaTeX into DocBook.

As it was stated n previous posts on this thread, FO needs some time to become
recognized as a method for generating quality output.

Dave's book is on the way from Amazon to my home, and I believe that soon I'd 
able to get more into FO and, at least, give some feedback to FOP developers.

For now I only can say: THANK YOU and congratulations for coping with this very
important project providing the community with such a useful tool. (I'm helping
one non-profit educational institution and preparing course for another 
non-profit organization and it's great to have the choice of free open-source
tools like FOP).

Although I'm not Java programmer, I hope I will be able to somehow help your
development.

Sincerely,
Gour

Re: problems with tables

[Oleg Tkachenko <ol...@multiconn.com>]

Matthew L. Avizinis wrote:
 > I disagree.  If one were to look at the proposed development
 > timelines of other (for instance) Apache projects, such as Xalan, one
 > would see that they are striving to make reasonable timelines.
Hey, that's a horse of a different color! Xalan's maiden name is 
lotusxsl and the code was donated to apache by blue giant, which still 
actively supports it, look at 
http://xml.apache.org/xalan-j/readme.html#developer-list - 10 of 13 
developers have emails at ibm.com domain, you know, fop doesn't have 
such support, so we are on our own.

-- 
Oleg Tkachenko
eXperanto team
Multiconn International, Israel

RE: problems with tables

[Victor Mote <vi...@outfitr.com>]

Matthew L. Avizinis wrote:

> have larger and more active teams of developers than FOP.  Thus, 
> their websites are more uptodate, their documentation is clearer 
> and more thorough, and they publish to-do lists, i.e. things that 
> people currently think are important enough to work on.  They 
> just seem to be more publicly open about what's going on.  With 

This is a valid point, and there are at least two initiatives under way to try to improve this situation. The cost of jumping in to FOP development is probably higher than it needs to be. There are a handful of dedicated (and I think pretty brilliant) developers who know the code well enough to work with it, and we are trying to build the infrastructure to attract more & bring them up to speed quickly.

> shouldn't find their way into the documentation.  Sure I know the 
> line, "your welcome to contribute" to fix problems like this but 
> the _process_ set up for doing this is just too time consuming. 

Well, except for the time lag difference between committers and non-committers, the process is the same for everyone. So, if it is too time-consuming for you, it seems a bit unreasonable to suggest that someone else should be doing it more and harder.

> get FOP working.  I don't think it's reasonable to always give 
> the off-hand remark to just do it themselves if they think it's 
> important, or pour money out to someone (maybe they're operating 
> on a shoestring budget and can't afford to).  At the least I 

This sounds like 1) I have no money, 2) I have no time, 3) I have no skills, 4) I demand that someone solve my problem for me anyway, and please don't be slow about it. I hate to mention it, but perhaps the developers are operating on a shoestring budget as well. I don't think it is reasonable to DEMAND of volunteers that they work harder or longer than they are. There was a very clever television commercial in the U.S. a few years ago (I think from EDS) that depicted an airplane being built while it was flying. That is really not a bad analogy for FOP. It is good enough to be useful to a lot of people, but there still is not enough time to serve coffee to passengers. If the passengers are willing and able to pick up a rivet gun, please do, but at the very least, please put the whips down.

> think it would be better to have a section of the website for 
> "feature requests" and whether or not anyone is working on them.  
> This is, imho, more respectful of the user.  It would also allow 
> a prominent location for all of those folks looking for some way 
> to contribute a list of things that need to be done.  Bugzilla?  
> I don't know, but it seems about as easy as using cvs without a 
> WinGUI (like TortoiseCVS) -- ouch it hurts and takes too long.

Again, I agree. One project in the back of my mind that seems useful is to research Bugzilla, and find out how to build URLs that will query it in a more convenient way. I am pretty sure that this can be done, and it wouldn't take a developer to figure out how. Documenting these URLs in the xml-docs so that they are available to all wouldn't require a developer either. I really think we need someone to volunteer to cull through the postings on this list each day & update the FAQs -- that wouldn't require a developer either. There is more, but we all grow weary ... 'Nuff said.

Victor Mote

RE: problems with tables

["Matthew L. Avizinis" <ml...@gleim.com>]

I don't want to get in semantics here about what reasonable is.  No, Xalan team hasn't precisely promised anything.  Perhaps I should have been more precise in stating my opinion.  All I was trying to say is that some of the other Apache projects seem to have larger and more active teams of developers than FOP.  Thus, their websites are more uptodate, their documentation is clearer and more thorough, and they publish to-do lists, i.e. things that people currently think are important enough to work on.  They just seem to be more publicly open about what's going on.  With FOP I know that it is being "redeveloped", but that's about it-- if I'm not really a java person, but really really like FOP, why should I have to check out the cvs source, or subscribe to the fop-dev list (I do, but that's beside the point) to see how things are progressing?  Why not just every now and then, like every couple weeks or once a month, let _everyone_ know, i.e. via the website, where the project stands?
There's no doubt, afaics, that FOP releases less frequently than some other Apache projects.  Sure, they've been around longer, but judging from their webpages, they seem to be better organized-- for example their examples fit their release.  I was just looking at widowsandorphans.fo and, as an example (supposed to help a newbie) is somewhat less than desireable -- table-layout="fixed" is not defined, the intended borders do not render because the border-style is not defined, and the table-columns + border is too wide to fit in the allotted space.  This results in error messages popping up that might tend to confuse rather than help new users.  Files like this just shouldn't find their way into the documentation.  Sure I know the line, "your welcome to contribute" to fix problems like this but the _process_ set up for doing this is just too time consuming. (But I don't know what would be better.) I fixed the file but what do I do with it?
End user's are always going to have feature requests, but they are not always going to be java developers.  Just look at this user list, some folks can barely put together an XSLT file and get FOP working.  I don't think it's reasonable to always give the off-hand remark to just do it themselves if they think it's important, or pour money out to someone (maybe they're operating on a shoestring budget and can't afford to).  At the least I think it would be better to have a section of the website for "feature requests" and whether or not anyone is working on them.  This is, imho, more respectful of the user.  It would also allow a prominent location for all of those folks looking for some way to contribute a list of things that need to be done.  Bugzilla?  I don't know, but it seems about as easy as using cvs without a WinGUI (like TortoiseCVS) -- ouch it hurts and takes too long.
OK, now that I've probably stuck my foot in my mouth (and shoved it in quite a ways) I invite any constructive responses.

   Matthew L. Avizinis <ma...@gleim.com> 
Gleim Publications, Inc.
   4201 NW 95th Blvd.
 Gainesville, FL 32606
(352)-375-0772
      www.gleim.com <http://www.gleim.com> 

=======================================================================
computing (kum' pyoot ing)
1. n the art of calculating how much time you wasted and money you spent in a doomed attempt to master a machine with a mind of it's own. --from computing: A HACKER'S DICTIONARY



> 
> There is a lot I don't know, and perhaps I have just missed 
> something, but... Have the Xalan developers actually promised or 
> estimated dates for features along the way? If not, then who is 
> defining "reasonable"? "Reasonable timeline" does not imply 
> "deadline", even an estimated one. I certainly think FOP proceeds 
> on a "reasonable timeline", especially considering the quantity 
> of resources.
> 
> If someone wants to step forward and state a "reasonable 
> timeline" for completion of table-layout="auto" or any other 
> feature, then we'll expect to see a contribution of code or 
> resources that can be turned into code to accomplish that. 
> Otherwise, the answer is that it will get done when it is 
> important enough to someone to get to it.
> 
> Victor Mote
> 

RE: problems with tables

[Victor Mote <vi...@outfitr.com>]

Matthew L. Avizinis wrote:

> I disagree.  If one were to look at the proposed development 
> timelines of other (for instance) Apache projects, such as Xalan, 
> one would see that they are striving to make reasonable timelines.

>From this point forward, I was looking for your estimate of when table-layout="auto" would be implemented. I never did see it   :-)

There is a lot I don't know, and perhaps I have just missed something, but... Have the Xalan developers actually promised or estimated dates for features along the way? If not, then who is defining "reasonable"? "Reasonable timeline" does not imply "deadline", even an estimated one. I certainly think FOP proceeds on a "reasonable timeline", especially considering the quantity of resources.

If someone wants to step forward and state a "reasonable timeline" for completion of table-layout="auto" or any other feature, then we'll expect to see a contribution of code or resources that can be turned into code to accomplish that. Otherwise, the answer is that it will get done when it is important enough to someone to get to it.

Victor Mote

RE: problems with tables

["Matthew L. Avizinis" <ml...@gleim.com>]

I disagree.  If one were to look at the proposed development timelines of other (for instance) Apache projects, such as Xalan, one would see that they are striving to make reasonable timelines.

This is, imho, probably due to having larger numbers of active developers.  Also, because projects like Xalan have wider commercial application right now than FOP.  Using XSLfo is not a widely accepted publishing paradigm.  We're moving to it right now for paper book, software, and online products, but I believe that is not the anywhere near the norm yet, since there are many other good electronic publishing methods "out there".  As more (and richer) companies find a need for FOP (and support the Open Source concept) they will be more willing to allow use of company resources to support this project.

Me, I'm not a java developer, so I'm not much help with this project other than to report problems I've encountered and help with the occasional question.  I've got my hands so heaping full at work developing the XSL transforms to generate the fo for all the books (content people can be quite particular in how they want to present things) we've got and online stuff, and a 10-month-old to chase after at home in my off time that I don't have much time anyhow-- I've submitted _one_ FOP extension function for writing page numbers to an alternate file and that's about all I'm up for.

Am I rambling?  I'll stop now....

   Matthew L. Avizinis <ma...@gleim.com> 
Gleim Publications, Inc.
   4201 NW 95th Blvd.
 Gainesville, FL 32606
(352)-375-0772
      www.gleim.com <http://www.gleim.com> 

=======================================================================
computing (kum' pyoot ing)
1. n the art of calculating how much time you wasted and money you spent in a doomed attempt to master a machine with a mind of it's own. --from Computing: A HACKER'S DICTIONARY


> -----Original Message-----
> From: Victor Mote [mailto:vic@outfitr.com]
> Sent: Friday, October 18, 2002 11:28 AM
> To: fop-user@xml.apache.org
> Subject: RE: problems with tables
> 
> 
> Bh wrote:
> 
> > Any ideas when the auto layout is going to be supported? Really 
> > waiting for
> > this feature
> 
> If one were predicting such a thing in a commercial environment, 
> with relatively predictable resources, one would expect to be off 
> by a factor of at least 3. To predict such a thing on an 
> open-source environment is not possible.
> 
> > i wish i had some time to help these ppl out)
> 
> Well, if you don't have time, perhaps your employer could fund 
> the salary of a developer who could. There are probably many 
> companies out there that could benefit greatly by investing in 
> FOP that way instead of putting a greater number of dollars into 
> a commercial product that they have no control over. Another 
> alternative would be to investigate the commercial 
> implementations. I don't know whether any support this feature or 
> not, but even if they do not, they could probably also be 
> persuaded by dollars. 
> 
> Victor Mote
> 
> 

RE: problems with tables

[Victor Mote <vi...@outfitr.com>]

Bh wrote:

> Any ideas when the auto layout is going to be supported? Really 
> waiting for
> this feature

If one were predicting such a thing in a commercial environment, with relatively predictable resources, one would expect to be off by a factor of at least 3. To predict such a thing on an open-source environment is not possible.

> i wish i had some time to help these ppl out)

Well, if you don't have time, perhaps your employer could fund the salary of a developer who could. There are probably many companies out there that could benefit greatly by investing in FOP that way instead of putting a greater number of dollars into a commercial product that they have no control over. Another alternative would be to investigate the commercial implementations. I don't know whether any support this feature or not, but even if they do not, they could probably also be persuaded by dollars. 

Victor Mote