"(in 6429)" in allkeys.txt

Whistler, Ken ken.whistler at sap.com
Tue Mar 11 13:57:23 CDT 2014

Per asked:

> In the DUCET file allkeys.txt,
> http://www.unicode.org/Public/UCA/latest/allkeys.txt ,
> there is "(in 6429)" as a comment for some characters.
> I first didn't understand why, but then I realized those are control
> characters that are part of ISO/EIC 6429.
> Why is that pointed out explicitly in that context?

1. To make it clear that those are not actually Unicode
character names, but names for control functions associated
with ISO 6429. (Note that this practices dates back a long
time now in the DUCET data files -- it predates the addition of
the ISO 6429 control function names as formal name
aliases in NameAliases.txt in the UCD.)

2. Because the same "names" which appear in the comments
in allkeys.txt for UCA also appear in comments for the CTT
in ISO 14651 (which is generated with the same tool). And
the "(in 6429)" notes were added there to forestall people
asking questions about these "names" that aren't "names".

3. And the reason they *continue* to appear in the comments
in both allkeys.txt and in the CTT for ISO 14651 is to preclude
people asking questions about why they would be removed. ;-)

> The reason I'm asking is that I was looking at the proposed new version
> of this file, and was thinking about suggesting a short note in the
> comments in the beginning of the file.

My personal preference, rather than larding up the header
of a machine-generated file with more commentary, would be
a suggestion for further clarification in the text of UTS #10, if
necessary. After all, the allkeys.txt header already points to
UTS #10 for more information -- which anyone needs to understand
and use the data file, anyway.


More information about the Unicode mailing list