[XeTeX] bug in xecyr

[Ross Moore]

On 17/05/2009, at 3:05 PM, Ross Moore wrote:

> LaTeX usage seems to be controlled by  inputenc.sty  where relevant
> coding is:

Sorry, that should be  fontenc.sty .

>
>
> \def\update at uclc@with at cyrillic{%
>  \expandafter\def\expandafter\@uclclist\expandafter
>   {\@uclclist
>   \cyra\CYRA\cyrabhch\CYRABHCH\cyrabhchdsc\CYRABHCHDSC\cyrabhdze
>   \CYRABHDZE\cyrabhha\CYRABHHA\cyrae\CYRAE\cyrb\CYRB\cyrbyus
>   \CYRBYUS\cyrc\CYRC\cyrch\CYRCH\cyrchldsc\CYRCHLDSC\cyrchrdsc
>   \CYRCHRDSC\cyrchvcrs\CYRCHVCRS\cyrd\CYRD\cyrdelta\CYRDELTA
>   \cyrdje\CYRDJE\cyrdze\CYRDZE\cyrdzhe\CYRDZHE\cyre\CYRE\cyreps
>   \CYREPS\cyrerev\CYREREV\cyrery\CYRERY\cyrf\CYRF\cyrfita
>   \CYRFITA\cyrg\CYRG\cyrgdsc\CYRGDSC\cyrgdschcrs\CYRGDSCHCRS
>   \cyrghcrs\CYRGHCRS\cyrghk\CYRGHK\cyrgup\CYRGUP\cyrh\CYRH
>   \cyrhdsc\CYRHDSC\cyrhhcrs\CYRHHCRS\cyrhhk\CYRHHK\cyrhrdsn
>   \CYRHRDSN\cyri\CYRI\cyrie\CYRIE\cyrii\CYRII\cyrishrt\CYRISHRT
>   \cyrishrtdsc\CYRISHRTDSC\cyrizh\CYRIZH\cyrje\CYRJE\cyrk\CYRK
>   \cyrkbeak\CYRKBEAK\cyrkdsc\CYRKDSC\cyrkhcrs\CYRKHCRS\cyrkhk
>   \CYRKHK\cyrkvcrs\CYRKVCRS\cyrl\CYRL\cyrldsc\CYRLDSC\cyrlhk
>   \CYRLHK\cyrlje\CYRLJE\cyrm\CYRM\cyrmdsc\CYRMDSC\cyrmhk\CYRMHK
>   \cyrn\CYRN\cyrndsc\CYRNDSC\cyrng\CYRNG\cyrnhk\CYRNHK\cyrnje
>   \CYRNJE\cyrnlhk\CYRNLHK\cyro\CYRO\cyrotld\CYROTLD\cyrp\CYRP
>   \cyrphk\CYRPHK\cyrq\CYRQ\cyrr\CYRR\cyrrdsc\CYRRDSC\cyrrhk
>   \CYRRHK\cyrrtick\CYRRTICK\cyrs\CYRS\cyrsacrs\CYRSACRS
>   \cyrschwa\CYRSCHWA\cyrsdsc\CYRSDSC\cyrsemisftsn\CYRSEMISFTSN
>   \cyrsftsn\CYRSFTSN\cyrsh\CYRSH\cyrshch\CYRSHCH\cyrshha\CYRSHHA
>   \cyrt\CYRT\cyrtdsc\CYRTDSC\cyrtetse\CYRTETSE\cyrtshe\CYRTSHE
>   \cyru\CYRU\cyrushrt\CYRUSHRT\cyrv\CYRV\cyrw\CYRW\cyry\CYRY
>   \cyrya\CYRYA\cyryat\CYRYAT\cyryhcrs\CYRYHCRS\cyryi\CYRYI\cyryo
>   \CYRYO\cyryu\CYRYU\cyrz\CYRZ\cyrzdsc\CYRZDSC\cyrzh\CYRZH
>   \cyrzhdsc\CYRZHDSC}%
>  \let\update at uclc@with at cyrillic\relax
> }
> \DeclareOption*{%
>    \let\encodingdefault\CurrentOption
>    \edef\reserved at f{%
>      \lowercase{\def\noexpand\reserved at f{\CurrentOption enc.def}}}%
>    \reserved at f
>    \InputIfFileExists\reserved at f
>         {}{\PackageError{fontenc}%
>          {Encoding file `\reserved at f' not found.%
>           \MessageBreak
>            You might have misspelt the name of the encoding}%
>          {Necessary code for this encoding was not
>           loaded.\MessageBreak
>           Thus calling the encoding later on will
>           produce further error messages.}}%
>   \let\reserved at f\relax
>   \expandafter\in@\expandafter{\CurrentOption}%
>                               {T2A,T2B,T2C,X2,LCY,OT2}%
>   \ifin@
>      \expandafter\in@\expandafter\cyra\expandafter
>                               {\@uclclist}%
>      \ifin@
>      \else
>        \update at uclc@with at cyrillic
>      \fi
>   \fi
> }
> \ProcessOptions*
>
>
> Also relevant are files in  .../texmf/tex/latex/cyrillic
>
> viz.
> [GlenMorangie:tex/latex/cyrillic] rossmoor% grep CYRYI *.def
> cp1251.def:\DeclareInputText{175}{\CYRYI}
> cp855.def:\DeclareInputText{141}{\CYRYI}
> cp866.def:\DeclareInputText{244}{\CYRYI}
> cp866nav.def:\DeclareInputText{248}{\CYRYI}
> iso88595.def:\DeclareInputText{167}{\CYRYI}
> isoir111.def:\DeclareInputText{183}{\CYRYI}
> koi8-ru.def:\DeclareInputText{183}{\CYRYI}
> koi8-u.def:\DeclareInputText{183}{\CYRYI}
> lcyenc.def:\DeclareTextSymbol{\CYRYI}{LCY}{248}
> maccyr.def:\DeclareInputText{186}{\CYRYI}
> macukr.def:\DeclareInputText{186}{\CYRYI}
> t2aenc.def:\DeclareTextSymbol{\CYRYI}{T2A}{136}
> x2enc.def:\DeclareTextSymbol{\CYRYI}{X2}{136}
>
>
> That gives at least 9 different encoding positions in non-Unicode
> fonts.

there are 3 output encodings, and 10 input encodings ...

> Are they all accessing a glyph for
>  a.  the same character ?
>  b.  the *correct* character;  i.e.  Ux0407  ?
> or
>  c.  the incorrectly named character;  Ux0406 ?

My own testing shows that  \CYRYI  gives  Ї  (with double-dot above,
hence  Ux0407 )  and that  \CYRII  gives  І  (with no dots above,
hence Ux0406).

This seems to confirm the reported error.


> Would  Evgenie Medvedev  or  A. Shipunov , or anyone else who has
> had lots of experience with these encodings, please comment on this,
> so that a possible error is not propagated any further; nor any new
> error created by acting in haste.


Cheers,

	Ross

------------------------------------------------------------------------
Ross Moore                                       ross at maths.mq.edu.au
Mathematics Department                           office: E7A-419
Macquarie University                             tel: +61 (0)2 9850 8955
Sydney, Australia  2109                          fax: +61 (0)2 9850 8114
------------------------------------------------------------------------