It is the UCA that defines the concept of ignorable characters. And
the DUCET ("Default" UCET) that makes characters ignorable by default,
but this can still be tailored in specific collations).Where did you see another confusive expression in TUS defining "default ignorable" differently as being intended for rendering purposes ? If there's such a place it should be changed (and TUS should not standardize what renderers should do, they have great flexibility as long as this does not disrupt the text semantics 'too much", including not rendering all possible semantic differences but rendering them in contexts where confusion is not possible for readers in most used locales and with common presentations used in that language and script pair). Many "zero-width" characters can in fact have a visible advance width (e.g. the circumflex accent or macron above do have an inherent width which may influence the advance width of a lowercase i below it, even if in most common case they don't cause any additional advance width above capitals). In other cases, that will in fact be kerned if necessary (and if possible) into the characters on both sides, provided they do not collide with them. The acute accent in Greek will also be typically written on the left side of capitals, causing them to move to the left, as of the acute accent had a non-zero width and was a normal character... 2013/3/18 Richard Wordingham <[email protected]>: > On Mon, 18 Mar 2013 08:32:03 +0100 > Philippe Verdy <[email protected]> wrote: > >> The "Default ignorable" property has nothing to do with rendering or >> being zero-width, it's just a matter of collation (comparing strings >> for similarity, for plain-text searches, or sorting them), it does not >> necesarily mean that the character is zero-width (that's a rendering >> property). > > "Default ignorable" is all about display and has nothing to do with > collation. Perhaps you are confusing it with "completely ignorable". > Completely ignorable characters are not completely ignorable; they can > disrupt contractions. > > Richard. >
