287166 - Georgian list numbering is wrong

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect, P1)

Description

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Our Georgian list numbering is wrong.  See:
http://lists.w3.org/Archives/Public/www-international/2005JanMar/0157
http://lists.w3.org/Archives/Public/www-international/2005JanMar/0158

Steps to reproduce:
 1. load http://dbaron.org/css/test/2005/georgian-lists

Actual results: first list matches, second one doesn't
Expected results: second list matches, first one doesn't

I'm cc:ing the two people who have volunteered for Georgian localization.  I'd
appreciate if they could confirm that our current behavior is incorrect and that
we should make this change.

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: patch

Comment 2

[Jungshik Shin]

Comment on attachment 178211 [details] [diff] [review]
patch

r=jshin

Comment 3

[Aiet Kolkhi]

I confirm. The Georgian numbering should indeed be lowercase.

Georgian uppercase range in Unicode is mainly resevred for old Georgisn script, 
that was used centuries ago.

Comment 4

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 178211 [details] [diff] [review]
patch

sr=bzbarsky

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Fix checked in to trunk, 2005-03-22 15:27 -0800.

Comment 6

[Gia Shervashidze]

There is no lowercase/uppercase in Georgian at all, but different Alphabets.

Character-Number correspondence scheme for Georgian is based on old Asomtavruli.
 [Unfortunately] contemporary 10E3 is missed - see http://www.gia.ge/Alphabets.html
So, sequences from 10B3(400) in source and correspondingly in patch aren't valid.
Agree with John Cowan regarding Mkhedruli - that was usual practice for medieval
Georgian script (mainly for dates).  Patch applied.

P.S. For new localization voluteer - if there are no remarks/corrections NO
confirmations needed nor for earlier, nor for this, nor for my future patches in
any lists!

Comment 7

[Gia Shervashidze]

Attachment: Patch2

Comment 8

[Gia Shervashidze]

Sorry...

Comment 9

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

According to the URL in comment 6, 50 and 60 are also backwards.  Should that be
corrected as well?

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Actually, that URL disagrees about 50 and 60 between the characters given and
the Hex values given.

Comment 11

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

So, I think the following errors exist:
 * http://www.gia.ge/Alphabets.html swaps the Hex values for 50 and 60 but gets
the characters correct (I'm guessing it's that way because of the Obsolete
notation and because it's a lot easier to check the characters if you know the
language)
 * Opera gets 400 wrong (uses U+10E3 instead of U+10F3) but otherwise agrees
with http://www.gia.ge/Alphabets.html

Could you (and others?) just confirm that those two errors are as-stated (which
is what your patch does, I think) and not the other way around?

Comment 12

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: testcase

Comment 13

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I committed your patch, but I'd still appreciate answers to the questions in
comment 11 so that we can improve css3-lists.

Comment 14

[Gia Shervashidze]

Right, HEX values missed between 50 & 60.  i check table.
U+10E3 instead of U+10F3 is "free interpretation" by Opera, but it could accepted.
Patch must be OK (You can use U+10E3 instead of U+10F3 for full compatibility
with Opera)
As i mentioned initial character-number correspondence was wrong in source - so
Your new attachment must based on right sequence of old Georgian letters.

Comment 15

[Aiet Kolkhi]

(In reply to comment #6)
> P.S. For new localization voluteer - if there are no remarks/corrections NO
> confirmations needed nor for earlier, nor for this, nor for my future patches
> in any lists!

The issue Reporter asked for confirmation.

Comment 16

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Do any of the others here have opinions on which character is better to use for
400?  Compatibility with Opera is probably only a minor concern if we should
worry about it at all.

Comment 17

[Gia Shervashidze]

(In reply to comment #15)
> nor for earlier, nor for this, nor for !!!my!!! future patches in any lists!

Comment 18

[Aiet Kolkhi]

I think we should leave it the correct way and I will have the issue reported 
and corrected to Opera Software.