Found another inconsistency between Unicode and the TM/TMW docs. I've sent e-mail to Tony Duff asking who's right, but I'm putting this in the errata under the assumption that even if Unicode is wrong, Unicode's wrong view will somehow rule the day.

Also, TMW->EWTS now generates \uF021-\uF0FF or \u0F00-\u0FFF escapes when appropriate.  A few TMW glyphs still give errors.

Also, there's now a test to be sure that TM<->TMW and TMW->EWTS won't break in the future (except for the one glyph in TMW that isn't in TM, that one isn't tested).  The baselines have not been hand-verified, but changes will be detected.
This commit is contained in:
dchandler 2003-11-24 05:49:15 +00:00
parent 9a247f5932
commit 5d053b41fe

View file

@ -84,10 +84,10 @@ M~~~8,91~~~~~~~0F7E
<~95,5~~9,95~~~~~~~0F3A
// gug.rtags.gyas:
>~96,5~~9,96~~~~~~~0F3B
// yar.tshes.rtags:
}~134,5~~10,39~~~~~~~0F3E
// mar.tshes.rtags
{~135,5~~10,40~~~~~~~0F3F
// {m/y/DLC?}ar.tshes.rtags: DLC NOW UNICODE?
{~134,5~~10,39~~~~~~~0F3F
// {y/m/DLC?}ar.tshes.rtags: DLC NOW UNICODE?
}~135,5~~10,40~~~~~~~0F3E
// dzud.rtags.me.long.can:
\u0F13~94,5~~9,92~~~~~~~0F13
// hard tsheg:
@ -715,30 +715,6 @@ a+r+y~145,4~~8,65~1,109~8,121~1,123~1,125~8,107~8,114~f68,fb2,fb1
<?ToWylie?>
M~238,1~~8,90~~~~~~~0F7E
// You can't convert these half-numbers to Unicode, but there exist
// half-measures: use U+0F20-U+0F29 and formatted text (such as RTF or
// HTML)
\tmw8053~50,5~~9,53~~~~~~~none
\tmw8054~51,5~~9,54~~~~~~~none
\tmw8055~52,5~~9,55~~~~~~~none
\tmw8056~53,5~~9,56~~~~~~~none
\tmw8057~54,5~~9,57~~~~~~~none
\tmw8058~55,5~~9,58~~~~~~~none
\tmw8059~56,5~~9,59~~~~~~~none
\tmw8060~57,5~~9,60~~~~~~~none
\tmw8061~58,5~~9,61~~~~~~~none
\tmw8062~59,5~~9,62~~~~~~~none
\tmw8063~60,5~~9,63~~~~~~~none
\tmw8064~61,5~~9,64~~~~~~~none
\tmw8065~62,5~~9,65~~~~~~~none
\tmw8066~63,5~~9,66~~~~~~~none
\tmw8067~64,5~~9,67~~~~~~~none
\tmw8068~65,5~~9,68~~~~~~~none
\tmw8069~66,5~~9,69~~~~~~~none
\tmw8070~67,5~~9,70~~~~~~~none
\tmw8071~68,5~~9,71~~~~~~~none
\tmw8072~69,5~~9,72~~~~~~~none
~~~3,45~~~~~~~0F0B
~~~4,45~~~~~~~0F0B
@ -954,10 +930,6 @@ y~175,4~~8,70~~~~~~~0FB1
// ra.btags:
r~176,4~~8,71~~~~~~~0FB2
// DLC FIXME: remove the <?Ignore?> tag and you get TMW->Wylie
// mappings for all these.
<?Ignore?>
// DLC EWTS doesn't give mappings for these half numerals...
\u0F33~70,5~~9,73~~~~~~~0F33
\u0F2A~71,5~~9,74~~~~~~~0F2A
@ -986,11 +958,6 @@ r~176,4~~8,71~~~~~~~0FB2
// kuruka:
\uF035~92,5~~9,90~~~~~~~none
// yungs.drung (reversed):
\tmw8097~97,5~~9,97~~~~~~~5350
// yungs.drung (standard):
\tmw8098~98,5~~9,98~~~~~~~534D
// mchan rtags trailing:
\uF038~99,5~~9,99~~~~~~~none
// mchan rtags leading:
@ -1048,11 +1015,6 @@ r~176,4~~8,71~~~~~~~0FB2
// sdong.tshugs:
\u0F19~133,5~~10,38~~~~~~~0F19
// utsama ka:
\tmw2059~57,2~~3,59~1,109~4,120~1,125~1,123~4,106~4,113~f88,f90
// utsama kha:
\tmw2060~58,2~~3,60~1,109~4,120~1,125~1,123~4,106~4,113~f88,f91
// tza.'phru:
^~174,4~~8,67~~~~~~~0F39
// reversed tza.'phru: (non-reversed is in Unicode, but not this)
@ -1062,18 +1024,6 @@ r~176,4~~8,71~~~~~~~0FB2
// half a.chen:
\u0F01~179,4~~8,73~~~~~~~0F01
// ITHI secret sign:
//
// f68,fa0,f80,f72 comes close, but fa0 might be larger. The OpenType
// font will probably have had this very glyph in mind if that works!
// DLC FIXME: No way this is Unicode f68,fa0,f80,f72, though: tha is
// small, and gi-gu plus reversed gi-gu together is ill-defined.
//
// DLC FIXME: WHAT'S THE EWTS?
//
// ITHI secret sign:
\tmw7074~180,4~~8,74~~~~~~~none
// Terton's mark:
\uF023~181,4~~8,75~~~~~~~none
// Terton's mark:
@ -1092,10 +1042,6 @@ r~176,4~~8,71~~~~~~~0FB2
// Mark:
\uF030~187,4~~8,82~~~~~~~none
// DLC see chinese unicode:
// Chinese letter:
\tmw9060~155,5~~10,60~~~~~~~none
// dril.bu:
\u0FC4~190,5~~10,62~~~~~~~0FC4
// rdo.rje:
@ -1115,10 +1061,6 @@ r~176,4~~8,71~~~~~~~0FB2
// nor.bu.bzhi.khyil:
\u0FCC~198,5~~10,70~~~~~~~0FCC
// bindu + datse + thigle: DLC NOW 0f82 or not? 91,5 is a candidate too
// DLC NOW: \uF03A or not? Awaiting Than's reply...
\tmw7095~242,1~~8,95~~~~~~~none
// zhu.yig.mgo.rgyan:
\u0F0A~33,5~~9,33~~~~~~~0F0A
// bka'.shog.mgo.rgyan:
@ -1150,3 +1092,63 @@ r~176,4~~8,71~~~~~~~0FB2
\uF036~93,5~~9,91~~~~~~~none
// nyi.zla:
\u0F82~91,5~~9,89~~~~~~~0F82
// DLC FIXME: remove the <?Ignore?> tag and you get TMW->Wylie
// mappings for all these.
<?Ignore?>
// bindu + datse + thigle: DLC NOW 0f82 or not? 91,5 is a candidate too
// DLC NOW: \uF03A or not? Awaiting Than's reply...
\tmw7095~242,1~~8,95~~~~~~~none
// DLC see chinese unicode:
// Chinese letter:
\tmw9060~155,5~~10,60~~~~~~~none
// ITHI secret sign:
//
// f68,fa0,f80,f72 comes close, but fa0 might be larger. The OpenType
// font will probably have had this very glyph in mind if that works!
// DLC FIXME: No way this is Unicode f68,fa0,f80,f72, though: tha is
// small, and gi-gu plus reversed gi-gu together is ill-defined.
//
// DLC FIXME: WHAT'S THE EWTS?
//
// ITHI secret sign:
\tmw7074~180,4~~8,74~~~~~~~none
// You can't convert these half-numbers to Unicode, but there exist
// half-measures: use U+0F20-U+0F29 and formatted text (such as RTF or
// HTML)
\tmw8053~50,5~~9,53~~~~~~~none
\tmw8054~51,5~~9,54~~~~~~~none
\tmw8055~52,5~~9,55~~~~~~~none
\tmw8056~53,5~~9,56~~~~~~~none
\tmw8057~54,5~~9,57~~~~~~~none
\tmw8058~55,5~~9,58~~~~~~~none
\tmw8059~56,5~~9,59~~~~~~~none
\tmw8060~57,5~~9,60~~~~~~~none
\tmw8061~58,5~~9,61~~~~~~~none
\tmw8062~59,5~~9,62~~~~~~~none
\tmw8063~60,5~~9,63~~~~~~~none
\tmw8064~61,5~~9,64~~~~~~~none
\tmw8065~62,5~~9,65~~~~~~~none
\tmw8066~63,5~~9,66~~~~~~~none
\tmw8067~64,5~~9,67~~~~~~~none
\tmw8068~65,5~~9,68~~~~~~~none
\tmw8069~66,5~~9,69~~~~~~~none
\tmw8070~67,5~~9,70~~~~~~~none
\tmw8071~68,5~~9,71~~~~~~~none
\tmw8072~69,5~~9,72~~~~~~~none
// yungs.drung (reversed):
\tmw8097~97,5~~9,97~~~~~~~5350
// yungs.drung (standard):
\tmw8098~98,5~~9,98~~~~~~~534D
// utsama ka:
\tmw2059~57,2~~3,59~1,109~4,120~1,125~1,123~4,106~4,113~f88,f90
// utsama kha:
\tmw2060~58,2~~3,60~1,109~4,120~1,125~1,123~4,106~4,113~f88,f91