[OFFICE-2950] 19.364 number:transliteration-format - "...which number characters to use." ??? For what? (I know the answer, I think, but we don't say here.) - OASIS Technical Committees Issue Tracker

Details

Type: Bug
Status: Applied
Priority: Major
Resolution: Fixed
Affects Version/s: ODF 1.2 CD 05
Fix Version/s: ODF 1.2 CD 06
Component/s: Locale
Labels:
None

Proposal:

Hide

The number:transliteration-format attribute specifies a set of number characters to be used for the representation of digits in a text.

The set of number characters is identified by the value of the number:transliteration-format attribute listing the decimal "DIGIT ONE" character with numeric value 1 as listed in the Unicode Character Database file UnicodeData.txt with value 'Nd' (Numeric decimal digit) in the General_Category/Numeric_Type property field 6 and value '1' in the Numeric_Value fields 7 and 8, respectively as listed in DerivedNumericValues.txt for a set of numbers.

(Question: How to determine the extent of the block of digit characters? For ODF-Next - Remember that we need to support Ideographic Numerical Values - Unicode 4.6)

If the number:transliteration-format attribute is absent or has no value, the ASCII representation of Latin-Indic digits is used. The number:transliteration-country, number:transliteration-language, and, number:transliteration-style attributes are ignored.

Note: Together the transliteration-format, number:transliteration-country, number:transliteration-language, and, number:transliteration-style attributes enable the complex conditions for the display of digits.

(Question: But we don't say how they interact. A question for ODF-Next as well.)

(Comment: BTW, this isn't transliteration. This is display of digit in what is alleged to be native characters. Hard to see how that is transliteration.)

Show
The number:transliteration-format attribute specifies a set of number characters to be used for the representation of digits in a text. The set of number characters is identified by the value of the number:transliteration-format attribute listing the decimal "DIGIT ONE" character with numeric value 1 as listed in the Unicode Character Database file UnicodeData.txt with value 'Nd' (Numeric decimal digit) in the General_Category/Numeric_Type property field 6 and value '1' in the Numeric_Value fields 7 and 8, respectively as listed in DerivedNumericValues.txt for a set of numbers. (Question: How to determine the extent of the block of digit characters? For ODF-Next - Remember that we need to support Ideographic Numerical Values - Unicode 4.6) If the number:transliteration-format attribute is absent or has no value, the ASCII representation of Latin-Indic digits is used. The number:transliteration-country, number:transliteration-language, and, number:transliteration-style attributes are ignored. Note: Together the transliteration-format, number:transliteration-country, number:transliteration-language, and, number:transliteration-style attributes enable the complex conditions for the display of digits. (Question: But we don't say how they interact. A question for ODF-Next as well.) (Comment: BTW, this isn't transliteration. This is display of digit in what is alleged to be native characters. Hard to see how that is transliteration.)
Resolution:

Hide

See EIke's listing of digits to include.

Show
See EIke's listing of digits to include.

Description

19.364 number:transliteration-format - "...which number characters to use." ??? For what? (I know the answer, I think, but we don't say here.)

Now reads: "The number:transliteration-format attribute specifies which number characters to use."

This is a case where I would concede the recasting of the text lost information but some critical bits were missing even in the 1.0 text.

Which read:

*****
14.7.10

The various number:transliteration-* attributes specify the native number system of the
style to display the number using, for example, CJK number characters. The notation is inspired
by the W3C XSLT 2.0 draft, see §12.3 of [XSLT2]. However, to be able to fully distinguish
between all possible native number systems additional attributes are needed in combination. For
example, Korean uses 11 different systems where the digits are not always different but short and
long and formal and informal forms exist.
*****
Then,

*****
Transliteration Format

The number:transliteration-format attribute specifies which number characters to use.
The value of the attribute is the digit "1" expressed as a native number.
If no format is specified the default ASCII representation of Arabic digits is used, other
transliteration attributes present in this case are ignored.
*******

Do you see the missing part?

We never say that the value of the digit "1" expressed as a native number identifies a set of digits in a particular language in the Unicode standard.

That is implied but not very well.

Plus in its current structure, we need to identify the number:transliteration-* attributes that are to be ignored.

Possibly a note on usage of the other attributes?

19.364 number:transliteration-format - "...which number characters to use." ??? For what? (I know the answer, I think, but we don't say here.)

Details

Description

Attachments

Activity

People

Dates