The Unicode Consortium Discussion Forum (CLOSED)

In 3.11 Normalization Forms under Combining Classes one can read :

"D106 Typographic interaction: Graphical application of one nonspacing mark in a position relative to a grapheme base that is already occupied by another nonspacing mark, so that some rendering adjustment must be done (such as default stacking or side-byside placement) to avoid illegible overprinting or crashing of glyphs.

The assignment of combining class values for Unicode characters was originally done with the goal in mind of defining distinct numeric values for each group of nonspacing marks that would typographically interact. Thus all generic nonspacing marks above are given the value \p{ccc=230}, while all generic nonspacing marks below are given the value \p{ccc=220}. Smaller numbers of nonspacing marks that tend to sit on one “shoulder” or another of a grapheme base, or that may actually be attached to the grapheme base itself when applied, have their own combining classes."

I'm having some difficulty understanding what was written before and after the word Thus above, as marks above and below don't interact typographically, as defined in D106. Also, from Figure 2.24 of Chapter 2, I got the impression that, when two non-spacing diacritical marks interact typographically they should have the same combining class value.

I'm also having some difficulty understanding the next paragraph :

"When assigned this way, canonical ordering assures that, in general, alternate sequences of combining characters that typographically interact will not be canonically equivalent, whereas alternate sequences of combining characters that do not typographically interact will be canonically equivalent."

Speaking of Figure 2.24, I must say that there is an error there. The ccc for the character U+0328 should be 202 instead of 220.

The text is fine, note that each group of characters in the example consists of characters that

1. interact typographically
2. have a common ccc value

Does that help you in reading it?

Please use the reporting form (under Consortium / Contact Us) on the home page to report the alleged mistake in the figure. That's the only way this will get to the attention of someone who might be in a position to do something about it.

I think it's clear now. But could you clarify the next paragraph :

"When assigned this way, canonical ordering assures that, in general, alternate sequences of combining characters that typographically interact will not be canonically equivalent, whereas alternate sequences of combining characters that do not typographically interact will be canonically equivalent."

Thanks for your input.

The use of the word "alternate" is not helpful - it's intended as short hand for "sequences of the same characters, but in an alternate order", but when read in isolation makes you think you of sequences that alternate between two characters, like 101010.

I definitely didn't follow you on this. But I think I'm starting to see the light. Please correct me if I'm wrong : given two characters x and y, x != y, with the same ccc, by definition, <x, y> is not canonically equivalent to <y, x>, and vice-versa, given two characters x and y, x != y, with different ccc's, by definition, <x, y> is always canonically equivalent to <y, x>.

I'm assuming above that all ccc's are not null.

PS: Is there a way, similar to MS Word, to insert a unicode character on this editor, by simply typing the Unicode hexadecimal code followed by Alt + x ? In other words, how should I proceed to insert a Unicode character here ?

Many thanks for your attention.

I should have written: the term "alternate sequence" is not defined - it's intended as short hand for "sequences of the same combining characters, but in an alternate order"


Yes, that's correct.



That must be so, because characters with ccc=0 are never reordered, so even if one of them has ccc=0, any other sequence, with an alternate ordering, would be distinct (not canonically equivalent).



The only ways that I know are to either set your keyboard so you can type the character you want, for example ß from the German keyboard, or cut&paste a character from another application. I personally use Unibook (http://unicode.org/unibook/). When you view a code chart, you can click on the character and copy it with Ctrl-C. (Windows only).

asmus

I want to thank you for your valuable input

See http://en.wikipedia.org/wiki/Unicode_input for some suggestions.

BobH

That was a very good link.

Thanks