Uniocde protocol or URNs?

From: Brett Zamir (brettz9@yahoo.com)
Date: Wed Jan 27 2010 - 21:29:29 CST

Next message: spir: "algorithm implementation: coping with code properties"

Previous message: John H. Jenkins: "Re: What is the Source of "kMandarin" in Unihan.txt?"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Mail actions: [ respond to this message ] [ mail a new topic ]

Hello all,

In the latest version (2.6.2) of my Firefox add-on, Unicode Input
Tool/Converter (at https://addons.mozilla.org/en-US/firefox/addon/5235
-- further localizers welcome!), I've added support for my own custom
protocol.

This protocol lets you use, within regular HTML links, such as the
following (in place of regular HTTP links):

     x-unicode:?find;char=a
     x-unicode:?searchName;string=Latin%20letter
     x-unicode:?searchkDefinition;string=house

Clicking on such a link will trigger the opening of the Firefox
extension so that you can view the characters in context, either within
their Unicode range (the first example), or among all the other
characters that match the criterion (the second and third examples).

You can see some examples at
http://brett-zamir.me/tests/unicodeProtocolExamples.html , though you
must first install the latest version of the extension (2.6.2) for this
to work.

I was wondering whether the Unicode Consortium were interested in:

   1. standardizing on a protocol such as that above (removing the need
      for the custom "x-" prefix)? and/or
   2. adopting a URN namespace to support specification of Unicode
      characters and names or possibly ranges and categories as well

The protocol would allow for specification of specific behaviors in
addition to targeting specific characters and ranges (e.g., it might
indicate which fields to display), while a URN would be confined to
identifying a particular letter, range, or category. (In hindsight, I
probably should have split up search and Name or kDefinition into
key-value pairs rather than a single identifier.)

Incidentally, I've filed a bug with Mozilla (at
https://bugzilla.mozilla.org/show_bug.cgi?id=539889 ) to suggest better
support for URNs and protocol handlers so that multiple extensions could
offer themselves as choices to handle these. Feel free to vote on the
bug if you are in favor of it.

Any interest in making a standard protocol or URN?

best wishes,
Brett Zamir

Next message: spir: "algorithm implementation: coping with code properties"
Previous message: John H. Jenkins: "Re: What is the Source of "kMandarin" in Unihan.txt?"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Mail actions: [ respond to this message ] [ mail a new topic ]

This archive was generated by hypermail 2.1.5 : Wed Jan 27 2010 - 21:36:17 CST