Re: Closure of tabular-data-model issues

hi Mati,

the document in question moved to CR back in mid-July, so i think it 
would possibly be best to raise new issues in github for these comments 
(and point to them from the tracker).


On 20/09/2015 17:14, Matitiahu Allouche wrote:
> Sorry for coming back to the party so late in the process.
>
> I have reviewed issues 477, 478 and 479, which are related to bidi, and I think that the handling of bidi is good.
> Two remarks:
>
> 1. About inheritance of the text direction annotation by column or cells, it might be worth specifying that if the parent's annotation is "auto", what is inherited is "auto" and not the resolved value of the table direction (for columns) or text direction (for cells) of the parent (which is either "ltr" or "rtl").
>
> 2. I have stated that the hierarchy of inheritance should be table > column > cell, and JeniT agreed on the principle in his comment dated July 2. However, section 4.5 "Cells" says:
> [quote]
> text direction — which direction the text within the cell should be displayed, as described in section 6.5.1 Bidirectional Tables; the value of this annotation may be derived from the table direction annotation on the table, if the textDirection property is set to inherit (the default).
> [end of quote]
> As far as I understand, this means that there is no possibility to make all cells in a certain column default to text direction = ltr and the cells in another column default to text direction = rtl. This is a very frequent use case.
> This can be cured by changing the text above to:
> [changed text]
> text direction — which direction the text within the cell should be displayed, as described in section 6.5.1 Bidirectional Tables; the value of this annotation may be derived from the text direction of the containing column if the textDirection property is set to inherit (the default), which itself may be derived from the table direction annotation on the table (see section 4.3 "Columns").
> [end of changed text]
>
> Another remark: it seems to me that nothing has been done on account of issue 463 (language metadata elsewhere than for columns), not even rejecting the suggestion to allow a language declaration at the table level and at the cell level (I think that it is difficult to justify having a language declaration at the row level).

iirc, it was possible to set the language in this way using inherited 
properties - it's just that that wasn't terribly clear from the original 
text, which JeniT worked on as a result of the issue we raised.

i think there are a number of things that are difficult to understand if 
you just read the spec, because of the way that things are scattered 
around the document. Hopefully, when people pull together tutorials for 
real users, they will make this easier.

anyway, i'll close the issues we currently have in our tracker for now.

ri

Received on Wednesday, 23 September 2015 16:15:57 UTC