At 02:33 PM 12/30/01 -0500, Tex Texin wrote:
>It is a bit inconsistent and therefore confusing.
>
>I searched for "bidirectional" which immediately pointed me at the
>general punctuation pages in a pdf file.
>Searching for "bidrectional" in that file turns up empty.
This is one of the few cases of an index entry that has no corresponding
line in the nameslist file. Usually the index entry is derived directly
from the character names and aliases, or the text of the block names and
sub headers. That's the reason you couldn't find "bidirectional" in the pdf
file. The subheaderin this case is just "Formatting characters" and that's
not very specific.
>If you search
>for left-to-right, right-to-left, override, or embed, there you do get
>to the characters. However a saving grace is that when you are first
>pointed at the general punctuation file, the character code 202A is
>mentioned, so if you notice that you can go right to the character
>range.
I'll make sure that is clearly worded in the instructions.
>Maybe the initial index needs to be more comprehensive. It is usually a
>difficult task for any large book to get right. However, tracking the
>web queries might help improve it over time...
The problem you encountered was one where the index is already more
comprehensive and detailed than the nameslist. ;-)
One could monkey with the nameslist, adding the subheader for the
bidirectional controls, but then we would pick up a number of one-character
ranges with subheaders, which becomes awkward in itself.
A./
PS: I'm in the process of updating the HTML files for the index to match
the contents of the Index-3.2.0dnn.txt file in the BETA directory. That
file covers the new 3.2 character names etc. but does not pick up new or
revised aliases and subheaders in the existing repertoire...
This archive was generated by hypermail 2.1.2 : Sun Dec 30 2001 - 23:40:10 EST