You are viewing a plain text version of this content. The canonical link for it is here.
Posted to fop-dev@xmlgraphics.apache.org by Andreas L Delmelle <a_...@pandora.be> on 2006/06/03 01:36:53 UTC
Re: Bug in whitespace handling with linefeed-treatment="ignore"?
On Apr 27, 2006, at 09:58, Jeremias Maerki wrote:
Sorry for the extra-late reply, still catching up...
> However, I'm not sure it this is right, but
> white-space-treatment is handled "before any linefeed-treatment
> handling
> is considered" (XSL-FO 1.0, 7.15.8). I think the afterLineFeed
> variable
> is currently not properly set. It also looks like our testcases don't
> cover this particular constellation. What adds to my uncertainty is
> that
> XEP does not behave as I would expect in this case (it seems to remove
> the tab but not the spaces).
I think you are quite correct here. Even more, I think this should be
extended to all occurrences of the linefeed character, meaning all
the way at the beginning of the loop --linefeed converted to a
space-- and I'd even move it out of the switch-block... if there is a
linefeed character, no matter what happens to it, the afterLinefeed
condition should always be true.
There could be nasty side-effects, I'm not sure, so I'll have to play
with it a bit more before making those changes.
Cheers,
Andreas
Re: Bug in whitespace handling with linefeed-treatment="ignore"?
Posted by Andreas L Delmelle <a_...@pandora.be>.
On Jun 3, 2006, at 11:42, Manuel Mall wrote:
>
> I thought that was fixed in
> http://svn.apache.org/viewcvs?rev=397556&view=rev
Yep, missed that one.
Thanks for the quick pointer.
Later,
Andreas
Re: Bug in whitespace handling with linefeed-treatment="ignore"?
Posted by Manuel Mall <mm...@arcus.com.au>.
On Saturday 03 June 2006 17:35, Andreas L Delmelle wrote:
> On Jun 3, 2006, at 05:14, Manuel Mall wrote:
> > <snip />
> > I don't think anything is wrong here. The definition of
> > linefeed-treatment was changed in XSL 1.1 (its all part of
> > refinement now) and I believe we behave correctly with respect to
> > 1.1.
>
> Ah yes, I always forget the changes to these properties from 1.0 to
> 1.1...
>
> Still, if, in the original example, the tab characters aren't removed
> but spaces are, then I'm inclined to think that Jeremias' suspicion
> is correct, and there is *something* not entirely right. :/
>
I thought that was fixed in
http://svn.apache.org/viewcvs?rev=397556&view=rev
>
> Later,
>
> Andreas
Manuel
Re: Bug in whitespace handling with linefeed-treatment="ignore"?
Posted by Andreas L Delmelle <a_...@pandora.be>.
On Jun 3, 2006, at 05:14, Manuel Mall wrote:
> <snip />
> I don't think anything is wrong here. The definition of
> linefeed-treatment was changed in XSL 1.1 (its all part of refinement
> now) and I believe we behave correctly with respect to 1.1.
Ah yes, I always forget the changes to these properties from 1.0 to
1.1...
Still, if, in the original example, the tab characters aren't removed
but spaces are, then I'm inclined to think that Jeremias' suspicion
is correct, and there is *something* not entirely right. :/
Later,
Andreas
Re: Bug in whitespace handling with linefeed-treatment="ignore"?
Posted by Manuel Mall <mm...@arcus.com.au>.
On Saturday 03 June 2006 07:36, Andreas L Delmelle wrote:
> On Apr 27, 2006, at 09:58, Jeremias Maerki wrote:
>
> Sorry for the extra-late reply, still catching up...
>
> > However, I'm not sure it this is right, but
> > white-space-treatment is handled "before any linefeed-treatment
> > handling
> > is considered" (XSL-FO 1.0, 7.15.8). I think the afterLineFeed
> > variable
> > is currently not properly set. It also looks like our testcases
> > don't cover this particular constellation. What adds to my
> > uncertainty is that
> > XEP does not behave as I would expect in this case (it seems to
> > remove the tab but not the spaces).
>
> I think you are quite correct here. Even more, I think this should be
> extended to all occurrences of the linefeed character, meaning all
> the way at the beginning of the loop --linefeed converted to a
> space-- and I'd even move it out of the switch-block... if there is a
> linefeed character, no matter what happens to it, the afterLinefeed
> condition should always be true.
>
> There could be nasty side-effects, I'm not sure, so I'll have to play
> with it a bit more before making those changes.
>
Andreas,
I don't think anything is wrong here. The definition of
linefeed-treatment was changed in XSL 1.1 (its all part of refinement
now) and I believe we behave correctly with respect to 1.1.
>
> Cheers,
>
> Andreas
Manuel