Re: [alreq] Shortcomings of Characters Table

I would like to summarize the discussion regarding shortcomings of 
character tables to make them actionable (either on ALReq side or 
CLDR's). For those of you who are inclined regarding the positioning 
of percent sign, please do open another issue and discuss the matter 
there. Thank you.

Here are the summary of the issues and actions needed to be taken:

1. Add double angle-quotes, U+00AB and U+00BB (« and ») to CLDR Arabic
 exemplar set for punctuation;
_1.1. Question:_ is that if there is any common usage of the single 
angle-quotes, U+2039 and U+203A (‹ and ›) for nested quotations or 
alike?
2. Re: U+066A (ARABIC PERCENT SIGN), U+066B (ARABIC DECIMAL 
SEPARATOR), 
U+066C (ARABIC THOUSANDS SEPARATOR), ALReq tables need to include the 
data provided in CLDR [Number 
Symbols](http://unicode.org/reports/tr35/tr35-numbers.html#Number_Symbols)
 to resolve the discrepancies for Arabic. No action needed on the data
 for CLDR;
3. Re: U+00D7 (MULTIPLICATION SIGN), ALReq tables need to include the 
CLDR Number Symbols data (see item 2);
4. Re: U+00F7 (DIVISION SIGN), CLDR doesn’t have a dedicated data item
 (according to UTS#35) to record that for each locale. ALReq included 
this in the charts for Persian based on ISIRI keyboard layout 
standards. Need to be discussed in CLDR TC to see if a data item to 
record such symbol might be possible. Otherwise, discuss in ALReq on 
how to proceed including it;
5. Re: U+0640 (ARABIC TATWEEL), CLDR data doesn’t include it at the 
moment for either Arabic or Persian. ALReq tables included it in 
charts for Persian based on ISIRI standards (see item 4). Propose to 
CLDR to add U+0640 to auxiliary exemplar sets for both Persian and 
Arabic;
6. Re: U+2026 (HORIZONTAL ELLIPSIS), CLDR data includes it in exemplar
 punctuation for Persian, but not for Arabic. Propose to CLDR to add 
U+2026 to Arabic punctuation exemplar as it’s already in use for 
ellipsis patterns;
7. Re: U+2212 (MINUS SIGN), Arabic data from CLDR under Number Symbols
 shows that Arabic is using U+002D (HYPHEN-MINUS) both for `latn` and 
`arab` number systems.    ALReq need to include this code-point in the
 tables;
_7.1. Question:_ Is CLDR data in this regard acceptable and denotes 
common usage?

@mostafah, I think we need to revise the CLDR lookup in the script 
generating the charts to go down in the numbers > symbols sections of 
the data (or get it from ICU4J).

@khaledhosny, @ntounsi, @mmh-docufy-de; please review and comment on 
my summary. After we gathered the comments, there would be an update 
on the ticket that @brawer kindly filed on our behalf.

Thanks to all involved in this discussion.

-- 
GitHub Notification of comment by shervinafshar
Please view or discuss this issue at 
https://github.com/w3c/alreq/issues/49#issuecomment-269246952 using 
your GitHub account

Received on Monday, 26 December 2016 22:48:54 UTC