2013/10/20 Jukka K. Korpela <[email protected]> > 2013-10-20 11:47, Jukka K. Korpela wrote: > > What you could do in a web page is to put U+00A0 U+25CC in one element >> and U+0E31 in another and position the elements in the same place, set >> to have the same width and to be horizontally centered. >> > > Oops. I meant U+25CC and U+00A0 U+0E31.
Good point, all combining marks should work OK when they follow U+0020 (legacy), U+00A0 (prefered), or U+25CC, these three charcters being valid bases in all scripts Additionally the "ideographic" space could be added to the list in an ideographic font, or probably all space characters Or the joining sharada in an arabic font (provided that the base character and the combining mark resolve with the same direction under Bidi, otherwise they will be logically separated in distinct spans; the three characters above have neutral or inherited directions, they should be part of the same directional span under Bidi). However I'm not sure that you can sfely ssume that these bases followed by multiple combiing marks from scripts with distinct strong RTL and LTR directions should be treated as a single rendered cluster. E.g. U+00A0 + <Arabic or Hebrew combining mark> + <Indic combining mark> (caveat: they are single cluster with the **default** grapheme cluster boundaries and for the normalisation purpose which may reorder these sequences)
