RE: [1.2T-LC] text-align and text-anchor with direction

Sorry for the lateness of this comment.  It just dawned on me that there is no text afaict indicating that in SVG Tiny automatically changes the result of text-align or text-anchor when direction="rtl" is specified on a block element or its ancestor (unlike CSS).  Is this intentional?

The textArea element has a text-align property whose start and end values are dependent on the writing system being used. The text says "For right to left horizontal (Hebrew, Arabic, etc): start is right and end is left ", but it's not clear to me what indicates to an implementation that we are dealing with right to left horizontal text.  (I suspect there may be some legacy-derived reliance on the setting of writing-mode lingering in the text here.) Shouldn't an implementation align text to the right in textArea by default if direction is set to rtl on that textArea element or inherited from higher up the hierarchy (as it would in CSS)?  

For the text element 'alignment' is achieved using the text-anchor property.  Judging from the expected result of the test text-align-202-t start means put the right edge of the rendered text at the initial cursor position if the first character in the text block is an Arabic or Hebrew character, although that seems at odds with the text in 10.8.1 about start that says "The rendered characters are aligned such that the start of the resulting rendered text is at the initial current text position. For Latin or Arabic, which is usually rendered horizontally, this is comparable to left alignment. "  Isn't it comparable to left alignment for left-to-right text, but right alignment for right-to-left?  

I think that if the text element or its ancestor has a direction property set to rtl, this should make the text appear to the left of the initial cursor position too, and that that should be clarified in the text.


Richard Ishida
Internationalization Lead
W3C (World Wide Web Consortium)

> -----Original Message-----
> From: Doug Schepers []
> Sent: 11 October 2008 04:02
> To:
> Cc:;
> Subject: Re: [1.2T-LC] i18n comment 9: Geometric middle of the text string
> (ISSUE-2111)
> Hi, I18N-
> wrote (on 10/10/08 3:44 PM):
> > Comment from the i18n review of:
> >
> >
> > Comment 9 At
> > Editorial/substantive: E Tracked by: RI
> >
> > Location in reviewed document: 10.8.1
> > []
> >
> > Comment: "middle The rendered characters are aligned such that the
> > geometric middle of the text string is at the current text position."
> >
> > It could be made clearer that this refers to the geometric middle of
> > the resulting rendered text, rather than the text string, since 'text
> > string' commonly refers to the characters in memory. This behaviour
> > has to take into account combining characters, conjunct forms, and
> > character widths in the rendered text.
> Fixed, by replacing those three uses of the text string "text string"
> with the text string "resulting rendered text".  Please let us know if
> this satisfies your comment.
> Regards-
> -Doug

Received on Thursday, 30 October 2008 18:09:44 UTC