On Tue, 8 Feb 2005, Andries Brouwer wrote:

> Thanks! However, in fact that is what I do. In order not to have
> to type the utf8 for every symbol, I use the SERA (System for
> Ethiopic Representation in ASCII) abbreviations in my text,
> and an OTP file to convert them to the right symbols.
>
> What I see is the following:
>
>   \documentclass{article}
>   \begin{document}
>   \char'0
>   \end{document}
>
> with latex gives a single Gamma. Also with lambda.
>
> Adding just the line \usepackage{omega} gives the error message
>   "Package babel Error: You haven't defined the language english yet."
>
> Also adding the line \usepackage[english]{babel} causes the mentioned
>   ! Text line contains an invalid character.
>   OCP stack 1.0 entry 0:^^@
> (Here it is clear, but in general it is annoying that no line number
> in given in the error messages.)
>
> Your suggestion
>
>   \documentclass{article}
>   \usepackage[english]{babel}
>   \usepackage{omega}
>   \begin{document}
>   {\clearocplists\char'0}
>   \end{document}
>
> does not give error messages, but the page remains empty, apart
> from the page number. (Are things different for you?)


I didn't test, but definitely because the current font doesn't
have any glyphs at position zero.


> Your \clearocplists does help a little:
> With \char'043 instead of \char'0 lambda gave
>   ! You can't use `macro parameter character #' in horizontal mode.
>   OCP stack 1.0 entry 0:#
> without it, and a # on the page with \clearocplists.
>
> Things look like lambda parses the input twice, where \charN is
> interpreted on the first pass and replaced by the appropriate
> character, where then that character is parsed again.


Well, it's kind of a bug in Omega that is passing `#' down the
stack, but what happens is, like you said, the stream of
characters is passed through all the current OCP list members, so
a \char'043 is not a font position anymore, it is processed by
OCPs first.


> Andries

BTW, it's sad that the Omega project is not supported anymore.  I
just tried to find its old web pages with Google and failed.


--behdad
http://behdad.org/

--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/

Reply via email to