Re: [css3-writing-modes] i18n-ISSUE-279: text-combine-horizontal & full/half-width characters

I think my lack of understanding here stems from a misapprehension on my 
part. For some reason I thought that only non-full-width characters or 
those that could be convered to non-full-width were affected by 
text-combine-upright:all.

Koji helped me understand that the all value will try to put anything 
and everything inside the element it is applied to on one horizontal 
line, and that consequently you have to apply that styling carefully and 
sparingly, which is somewhat different from the 'digit' approach.

I think the text is probably ok.

RI


On 27/03/2014 04:28, John Daggett wrote:
>
> Richard Ishida wrote:
>
>> I still find this a little confusing in the CR version of the spec.
>> I think the phrase:
>>
>> " when the combined text consists of more than one character, then any
>> full-width characters must first be converted to their non-full-width
>> equivalents"
>>
>> suggests, with the example that follows, and that uses
>> text-combine-upright: digits 2;, that ordinary full-width characters
>> in the content are affected by the digits keyword. Whereas, i
>> suspect that what is meant is
>>
>> " when the combined text consists of more than one character, then any
>> properties that would display non-full-width characters as full-width
>> characters must be disabled by text-combine-upright"
>
> You're confusing the conditions for applying a combination with how
> the combination is performed. The definition of the 'digits' value
> implies that value only affects a "maximal sequence of consecutive
> ASCII digits". It has no affect on full-width digits. If the 'all'
> value is used instead, then a sequence of full-width digits would be
> first converted to half-width digits before a combined form is laid
> out.
>
> The reason for this is that full-width digit glyphs are not the ideal
> form to use if compression is required. If you search the www-style
> archives for "text-combine-horizontal half-width", you'll find many
> posts discussing this last fall, along with examples and details
> regarding this.
>
> Regards,
>
> John Daggett
>
>

Received on Monday, 31 March 2014 18:35:21 UTC