OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: RE: [office] Use of <text:s> versus &nbsp;

Yes, I understand all that.  I was only talking about the way of preventing collapse of a run of spaces, not making the entire run non-breaking.

So "x&nbsp;&nbsp; y" (with the space at the right) should do the same thing that

   "x <text:s text:c='2'/>y" does.

(The problem with &nbsp; is that, when editing, the user can end up with "x&nbsp;&nbsp;y" which has an undesirable effect.)

My main point was that the ODF specification 
favors "x <text:s text:c='2'/>y" 
and does not favor "x<text:s text:c='3'/>y" as a practice.
The form "x<text:s text:c='2'/> y" is not considered at all.

It's also not specified anywhere in the ODF specification what the effect of fixed runs of spaces is when some but not all of the spaces are past a line-break position and whether a (soft [page]) break can occur in the midst of the run.  (The next question becomes, what happens with the spaces beyond the line break.)

 - Dennis


I notice that in OpenOffice.org and legacy implementations, "X <text:s text:c='19'/>Y" when moved one character beyond the right margin, will drop the Y to the beginning of the next line and no spaces precede the Y on the new line until the X is also moved to that line.  Then the 20 spaces reappear.   The same thing happens when the string is pushed pass the right margin of the last line on a page.
It is also the case that, when removing characters on the line where the X still appears, when the Y is brought back to the same line, there will be 20 spaces between the X and the Y.

I also notice that OpenOffice.org-legacy implementations *do* put line breaks in the midst of runs of non-breaking spaces, and this will lead to spaces appearing on a new (page) line in front of the Y character.  So, while insertion of non-breaking spaces is supported, they do not render properly!

-----Original Message-----
From: Jos van den Oever [mailto:jos@vandenoever.info] 
Sent: Monday, July 22, 2013 11:44 PM
To: office@lists.oasis-open.org; dennis.hamilton@acm.org
Cc: 'Andreas J Guelzow'
Subject: Re: [office] Use of <text:s> versus &nbsp;

On Tuesday 23 July 2013 00:14:52 AM Dennis E. Hamilton wrote:
> My comment was about where producers tend to put the <text:s> and what the specification recommends, versus the use of &nbsp; directly in the Unicode by other systems.   (This is not unrelated to the conversation about BiDi).  I understand what the Unicode code point for &nbsp; is.  One similarity is that <text:s> and &nbsp; render as spaces and they are not collapsed.

The difference betweeen <text:s/> and &nbsp; is that &nbsp; is a non-breaking space. "Hello&nbsp;world!" and "Hello<text:s/>world!" can be layed out differently. "Hello&#20;world;" and "Hello<text:s/>world!" render the same.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]