87736 - sil Greek/Hebrew/Apparatus Fonts does not work

Status: RESOLVED WONTFIX

(Core :: Internationalization, defect)

Description

[Frank Tang]

reproduce procedure
1. visit http://www.ccel.org/m/machen/greek/htm/i.htm
2. visit http://www.sil.org/computing/fonts/silgreek/ to download and install 
the SIL Greek font package.
3. visit visit http://www.ccel.org/m/machen/greek/htm/i.htm
expect result-
The greek should shown as IE.
Actual result-
It does not shown correctly.

Comment 1

[Frank Tang]

I find this bug when I try to print that public domain book for my wife since 
she is learning NT Greek right now. I also received a mail for SIL.org recently 
about this issue. 

Two issue-
1. we don't know that font, that font is a symbol encoded font. We need to hack 
fontEncoding.propertires to teach it.
2. we don't round trip windows-1252 to/from Unicode for the following code point
0x81, 0x8d, 0x8f, 0x90, 0x9d.

Comment 2

[Frank Tang]

reassign to ftang

Comment 3

[Frank Tang]

Attachment: make the gfx know these fonts.

Comment 4

[Frank Tang]

Attachment: make the windows-1252 round trip.

Comment 5

[Frank Tang]

rbs- can you review this one?
I think this will also impact 72265. Code point , 0x8d, 0x8f, 0x90, 0x9d won't 
work in 72265 without the 2nd patch here. 

Comment 6

[Frank Tang]

sorry, it should be 77265, not 72265

Comment 7

[rbs]

I don't mind supporting more Symbol fonts. Nevertheless, I am cc:ing some folks 
that I know may raise eyebrows :-)

r=rbs

Comment 8

[Jean-Marc Desperrier]

Ftang,

People who continue to use proprietary symbol fonts encoding to encode ancient
greek are in error !

There is support for ancient greek in unicode, many unicode fonts around support
ancient greek.
For interoperability, ancient greek should be displayed in UTF-8.

ccel.org would be much more inspired to switch to UTF-8. 
SIL should stop diffusing that font, or at least warn people to stop creating
new document that use it. 
Some peoples that had created custom ancient greek fonts have already taken this
step.

I can give list of ressources to download ad-hoc unicode font, and ancient greek
pages that use unicode, but most are in french.
But I searched, and found rapidly some in english, see :
http://www.ziplink.net/~ptrourke/fonts.htm
http://members.aol.com/AtticGreek/
http://www.users.dircon.co.uk/~hancock/antioch.htm
http://titus.uni-frankfurt.de/unicode/samples/grbeisp.htm

The only trouble for ancient greek in unicode is the lack of support on Macintosh.
If Mozilla were the browser application able to do that, it would be great, and
I believe it would not take so much work.

Comment 9

[Frank Tang]

too risky to change 1252 converter right now. 
mark it as moz1.0 for now. 

Comment 10

[Frank Tang]

 Jean-Marc Desperrier :
agree with you, but we probably also maintains IE compatibility.

Comment 11

[Hixie (not reading bugmail)]

Could we possibly limit this to quirks mode? The more we support these fonts
the more we are ruining the web and the more we are making a mockery of our
pledge to support standards.

Comment 12

[Frank Tang]

>Could we possibly limit this to quirks mode? The more we support these fonts
>the more we are ruining the web and the more we are making a mockery of our
>pledge to support standards.
I do not agree with you. Mozilla is NO LONGER the 300 lb gorillia. IE IS. People
will use these font regardless we support it or not. The more we do, more user
will use our product. If we don't do this, those user simply won't use Mozilla
and use IE or Netscape 4.x. Your statement is right if Netscape+Mozilla get 90%
of market share today. Unfortunately, it is not now. 

Comment 13

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

This not only makes a mockery of our standards compliance, but it makes a
mockery of any attempts for us to have a font selection algorithm that properly
handles mixed international text and author- or user-suggested fonts.

Comment 14

[Frank Tang]

file seperate bug 116143 for the windows-1252 round trip problem. I think we 
need to fix bug 11643 even we may not want to fix this bug. 

Comment 15

[Frank Tang]

push off to future for now.

Comment 16

[Adam Hauner]

-> to default owner (rather than ftang's WONTFIX)

Comment 17

[Simon Montagu :smontagu]

This is soooo WONTFIX.

Comment 18

[Simon Montagu :smontagu]

*** Bug 362761 has been marked as a duplicate of this bug. ***