%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % File: montex.tex % Author: Oliver Corff % Date: January 31st, 1999 % Version: \VersionRelease % Copyright: Ulaanbaatar, Beijing, Berlin % % Description: MonTeX -- Mongolian for LaTeX2e % Implementation Level \ImplementationLevel % System Documentation % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % ----------------- identification ends here ------------------- % \documentclass[11pt,a4paper]{article} %\usepackage[T3,LMS,LMC,T1]{fontenc} % <- Uncomment these lines \usepackage[latin1]{mls} % if you use the TIPA %\usepackage[noenc]{tipa} % <- package \usepackage{longtable} % % If emtex goofs with (printer) memory overflow % when attempting to print this document then % set the following number to "1", recompile and % increase the number step by step until all % examples are printed. The maximum value is 6. \newcounter{FontSamples} \setcounter{FontSamples}{6} % <--- Modify this number! % Make sure your installation includes ec sources % for Dunhill and Funny typefaces otherwise LaTeX % will complain about font files not found. % It is safe to set the FontSamples value to 4 on % emtex systems. % % You obtain a nicer `gamma' by uncommenting the following line % but if you do so make sure the TIPA package gets included. % You should then uncomment line no. 18 of this file. % %\renewcommand{\g}{\textipa{\textgamma}} %% %% The following commands and command groups are only used %% for abbreviating the typing work of the documentation. %% \newcommand\exa{\nopagebreak \begin{flushleft}\smallskip \nopagebreak \begin{minipage}[t]{6cm}\sloppy} \newcommand\exb{\end{minipage}\kern 1cm\begin{minipage}[t]{8cm}\sloppy } \newcommand\exc{\end{minipage}\kern -3cm \smallskip\end{flushleft}} %\renewcommand{\MonTeX}{Mon\TeX\ \textsf{Mongolian for \LaTeX}} \newcommand{\Soyombo}{\textsc{Soyombo}} \newcommand{\Zanabazar}{\textsc{Zanabazar}} \newcommand{\XD}{\textsc{X\"awt\"a\"a D\"orw\"oljin}} \newcommand{\LMC}{\textsf{LMC}} \newcommand{\LMS}{\textsf{LMS}} \newcommand{\LMT}{\textsf{LMT}} \newcommand{\Tone}{\textsf{T1}} \newcommand{\Ttwo}{\textsf{T2}} \newcommand{\TW}{{\mnr S\"UXBAT}} \newcommand{\WT}{{\rnm S\"UXBAT}} \newcommand{\Tw}{{\mnr S\"uxbat}} \newcommand{\Wt}{{\rnm S\"uxbat}} \newcommand{\ff}[1]{\fontfamily{#1}} \newcommand{\fs}[1]{\fontseries{#1}} \newcommand{\fsh}[1]{\fontshape{#1}} \newcommand{\tttt}[1]{\texttt{\char92#1}} \newcommand{\CaptionsList}[1]{% \begin{center} \begin{tabular}{l|l|l} Command & English & #1 \\ \hline \tttt{prefacename} & Preface & \prefacename\\ \tttt{refname} & References & \refname\\ \tttt{abstractname} & Abstract & \abstractname\\ \tttt{bibname} & Bibliography & \bibname\\ \tttt{chaptername} & Chapter & \chaptername\\ \tttt{appendixname} & Appendix & \appendixname\\ \tttt{contentsname} & Contents & \contentsname\\ \tttt{listfigurename} & List of Figures& \listfigurename\\ \tttt{listtablename} & List of Tables& \listtablename\\ \tttt{indexname} & Index & \indexname\\ \tttt{figurename} & Figure & \figurename\\ \tttt{tablename} & Table & \tablename\\ \tttt{partname} & Part & \partname\\ \tttt{enclname} & encl & \enclname\\ \tttt{ccname} & cc & \ccname\\ \tttt{headtoname} & To & \headtoname\\ \tttt{pagename} & Page & \pagename\\ \tttt{seename} & see & \seename\\ \tttt{alsoname} & see also & \alsoname\\ \end{tabular} \captionsenglish \caption{Captions in #1% \label{#1captions}} \end{center}% } %% %% Meaningful text of document montex.tex begins here %% \title{% \ifthenelse{\value{PostScriptAvailable}=1} {\mbosoo{mongGol}\mbosoo{bicig tai}\mbosoo{lateh}\\[0.75cm] }{% \textit{mong\g ol bicig-täi lateh}\protect\footnote{% Instead of the transliterated text, the same text should appear in vertical Mongolian writing. Since \MonTeX\ did not find PostScript support on your system, it decided to print this text horizontally and in transliteration. It is suggested that you update your \LaTeX\ system by adding PostScript support (see section~\ref{PostScriptSupport}).}\\} \xalx{Kirill "Us"agt"a"i La\TeX}\\ \MonTeX\\ Mongolian for \LaTeXe\\ Implementation Level \ImplementationLevel\\ v.~\VersionRelease} \author{Oliver Corff\protect\thanks{The author currently receives a DFG (Deutsche Forschungsgemeinschaft) postdoctoral grant (Habilitationsstipendium) and expresses his sincere gratitude.} \and Dorjpalam Dorj\protect\thanks{Institute for Language and Literature, Mongolian Academy of Sciences, Ulaanbaatar. D.~Dorj's work on this software was partially funded by a DAAD scholarship for which he expresses is heartfelt gratitude.}} \date{\VersionDate} \begin{document} \maketitle \begin{abstract} \MonTeX\ is now released in Implementation Level \ImplementationLevel\ implying that there is not only Cyrillic Support for Modern Mongolian (Xalx dialect), Buryat and Russian; this version includes rudimentary support for Mongolian Script (also known as Uighur Script) embedded in Latin and Cyrillic texts as it is used in dictionaries, philological work and official documents. The present release (\VersionRelease) is still very early; expect bugs and irregular behaviour. Especially the Mongolian Script input method cannot be but called crude. \end{abstract} \clearpage \tableofcontents \listoftables \clearpage \section{\MonTeX\ Features} \MonTeX\ is a package which provides several Mongolian scripts to the \LaTeXe\ user community. At present, it features Cyrillic typefaces for modern Mongolian as it is used in Mongolia and Buryatia. The package can be used for writing Mongolian text documents as well as for writing single Mongolian words embedded in other languages. It is also possible to include short portions of classical Mongolian in Latin and Cyrillic texts. Instead of delaying the release of the whole package any further, it was decided to release every module to the user community as soon as it is truly usable. Thus, we start with Cyrillic Mongolian, i.~e. Xalx and Buryat, and Classical Mongolian. In addition, rudimentary support (mainly for captions and the \verb'\today' command) for Russian is available as Russian is a language frequently used in Mongolia and Buryatia. The implementation levels are assigned as follows: \begin{center} \begin{tabular}{rl} Implementation Level & Features \\\hline I & Modern Mongolian in Cyrillic Script and Buryat\\ II & I and Mongolian script in LR mode horizontal\\ IIa & II and Mongolian script portions in LR mode vertical\\ III & II and Mongolian script text in horizontal RL mode\\ IV & III and Mongolian script text printed vertically \\ \end{tabular} \end{center} Implementation level I is good for producing documents in Modern Mongolian; implementation level II adds support for words and lines of Classical Mongolian embedded in other (Cyrillic and Latin) scripts which is essentially useful for dictionaries etc.; implementation level IIa allows single words to be placed in vertical capsules; implementation level III allows the composition of purely Classical Mongolian documents while implementation level IV finally allows the combination of both scripts in freely assignable quantities and locations within the document. Mongolian linguistic culture provides a perplexing richness of writing systems of varying regional, historical and socio-political importance. Developing a Mongolian system which covers \emph{all} documented writing systems is tantamount to writing a Mongolian Babel system and cannot be done in a truly elegant manner with respect to the current \LaTeXe\ limitations. Full ISO 10646/Unicode character set support is one desideratum, the free switching of all sorts of writing directions in the document is another cornerstone of success. The word \emph{Mongolian} is actually an umbrella term for several languages rather than the precise name of a single language. Things become more complicated when names of ethnic groups, languages and writing systems are mixed. \begin{description} \item [Xalx] or Khalkha is the name of the Mongolian nationality residing in Mongolia proper. Their dialect forms the basis of Mongolian written with Cyrillic letters. Throughout this text, \emph{Modern Mongolian} is used as a synonym. \item [Buryat] is the name of the Mongolian nationality residing in Buryatia, north of Mongolia, east of Lake Baikal, being a part of the Russian Federation. The Buryat call themselves \emph{Buryaad} while Xalx Mongolians call them \emph{Buriad}. The English name follows the Russian orthography. Linguistically, Xalx and Buryat Mongol are fairly close languages; Buryat has a slightly different sound system in which the phoneme /s/ partially shifted to /h/; the modern Buryat Cyrillic alphabet (virtually identical with the Cyrillic alphabet used for writing Modern Mongolian) has one additional letter (H/h, \xalx{H/h}) for marking the difference to /s/. \item [Bicig] (literally \emph{script} in Mongolian) denotes text written in the traditional Mongolian script which is also referred to as Uighur. \end{description} Xalx Mongolian, or Modern Colloquial Mongolian, is about as different from the form written in Classical script as modern English in phonetical spelling (assume it be written in Shavian letters) from the highly historical orthography of Standard English. Beyond these differences, Mongolian written in Classical Script usually preserves a substantial amount of historical grammatical features which make it look a bit like Elizabethan English. \subsection{Requirements and Limitations} In order to run \MonTeX\ a recent version of \LaTeXe\ is necessary. \MonTeX\ relies on the NFSS font selection mechanism and the ligature capacities of Metafont. This package has not been tested under \LaTeX2.09 and will most certainly not function satisfactorily under that environment. Depending on the implementation level, further software support becomes necessary since not all features can be realized smoothly in \LaTeXe\ alone. \begin{center} \begin{tabular}{r|l} Implementation Level & Requirements \\\hline I & A working \LaTeXe\ system\\ II & A working \LaTeXe\ system\\ IIa & like II, plus PostScript support\\ III & like II, plus functional \texttt{TeX--XeT} system\\ IV & like III, plus PostScript support \\ \end{tabular} \end{center} The e\TeX\ (available for DOS and UNIX based computers alike) system provides full Right-to-Left writing support; for obtaining the \verb"\LR" and \verb"\RL" primitives it is also possible to build \TeX\ from the \texttt{web2c} (or teTeX) sources, replacing \texttt{tex.web} with \verb'tex--xet.web' and \texttt{tex.ch} with \verb'tex--xet.ch' unless you have a new version of teTeX which includes the \texttt{etex} and \texttt{elatex} commands already. Post\-Script support is a standard feature of most UNIX installations and supplied with most of the available \TeX\ for Windows9x distributions.% \footnote{The authors use Linux (Red Hat 4.2 and 5.1) systems for daily work; on the same hardware, PostScript under Windows\emph{xx} is significantly slower than under Linux; this holds true for document and font compilation as well.} A word of warning is necessary here. \MonTeX\ is not a small, convenient system which can be used without any effort. Much like its very foundation \LaTeX\ it requires some willingness to study a few (and indeed simple) rules; occasionally one or the other old habit has to be overcome. The reward is text typeset in excellent quality so that scholarly achievement no longer disappears in badly typeset documents. \subsection{PostScript support\label{PostScriptSupport}} PostScript is used for creating vertical capsules of text within horizontal text for \MonTeX\ implementation levels IIa as well as for complete pages with implementation level IV. This requires the presence of the \texttt{rotating} package for \LaTeX\ which itself relies on the \texttt{graphics} package. The \texttt{rotating} and \texttt{graphics} packages come with teTeX but do not come with emtex. They can be found at CTAN. Besides the above-mentioned packages it is necessary that the generated \texttt{.dvi} files can be processed further, e.\,g. by \texttt{dvips} which generates a PostScript file out of a \texttt{.dvi} file. If there is no PostScript printer at your site, PostScript emulation is necessary which is usually provided by GhostScript and GhostView. Implementations of these systems are available for a number of operating systems and can also be found at CTAN. Linux, a free UNIX system, comes which GhostScript, and the winemtex distribution of \LaTeX\ includes GhostScript as well. A sample command sequence to produce and preview a document can be found in illustration~\ref{RunExampleTeTeX}. \begin{figure} \begin{verbatim} $ latex montex.tex # Compile document $ dvips montex.dvi # Create PostScript out of DVI $ gv montex.ps # Preview document (allows printing) \end{verbatim} \caption{A sample compilation and preview cycle for a document requiring PostScript support}~\label{RunExampleTeTeX} \end{figure} Without PostScript support, only implementation level II can be realized (instead of IIa): Mongolian script can be printed horizontally but not vertically. Illustration~\ref{RunExampleEmtex} shows the commands for emtex (in fact, there is no difference to the usual command sequence). \begin{figure} \begin{verbatim} C:\text> latex2e montex.tex # Compile document C:\text> v montex.dvi # Preview document \end{verbatim} \caption{A sample compilation and preview cycle for a document without PostScript support}~\label{RunExampleEmtex} \end{figure} \subsection{Acknowledgements} The authors wish to thank the creators of \TeX\ and \LaTeX\ as well as the designers of the existing fonts for their generosity of providing the world with such inspiring pieces of software. The packages from which pieces of code originated by inspiration or blunt copy are far too numerous; the Russian captions were taken from the file \texttt{russian.sty} (as were the English captions), most of the Cyrillic letters were produced with fonts by Nana Glonti and Alexander Samarin; additional letters were taken from J.~Knappen's font files. Special thanks go to David Carlisle who offered the solution for a serious problem with the ligature mechanism in \TeX. Among the friends and colleagues in Mongolia and Germany who offered information, support and encouragement the authors wish to name B.~Nerguy, Urgamal, M.~Balk, Q.~\"Anxzayaa and K.~Maezono (without implication of any particular order or precedence). They contributed test data as well as their ideas for encoding, font shapes, user interfaces, and, last but not least, were patient alpha testers who helped the authors with numerous problem reports. Most of the improvements between version 0.1 and the present version are not actually improvements; they are simply eliminations of partially awful bugs as well as ugly hacks (rather than \emph{code}) and aim to make this package simply usable (if not useful). \subsection{Sources of Code and Inspiration} Some Cyrillic packages have been available for a few years. All Cyrillic packages available for \TeX/\LaTeX\ stem from one of two lines of ancestry: \begin{itemize} \item Fonts developed at the University of Washington \item Fonts by Nana Glonti and Alexander Samarin \end{itemize} The two lines differ substantially in scope of characters and printing quality. The University of Washington series in \textsf{OT2} encoding has a broad support for East European languages, but the praise for printing quality is given to the characters designed by Nana Glonti and Alexander Samarin. The Glonti/Samarin line of characters has undergone numerous minor modifications, not so much in glyph shapes but basically in determining encoding slots. Fortunately, the fonts are set up in a way that allows for convenient redefining of individual code positions. Neither original line of Cyrillic characters offers the additional characters necessary for writing non-Slavic languages like Mongolian; already in the beginning of the 1990s, J\"org Knappen filled the gap and designed additional letters which were intended to be used with Bashkir, a Turkic language. In fact, most of the letter forms employed there can also be used in other non-Slavic languages used throughout Central Asia since these letters are not specific to Bashkir. Some of J.~Knappen's letter forms (accidentally mostly those which are not necessary for writing Modern Mongolian) do need some refinement, and are then immediately suitable for a range of languages including Kasakh, Tuvinian etc. In the present stage of the system, only those letters used in Mongolian and Buryat are incorporated from J.~Knappen's files. \begin{sloppypar} After discussing the typeface issues with Mongolian specialists, the Glonti/Samarin letter forms were chosen for their superior appearance in volume text. The fonts had to be renamed; failing to do so would have resulted in unpermissible ambiguity. \end{sloppypar} One feature of the traditional Cyrillic font packages for \TeX\ (besides their lacking support for non-Slavic languages) is the intimate relationship between input encoding and output encoding. The first step in building Mongolian support was to separate these two spheres as numerous Mongolian encodings exist which should all be supported by the Mongolian package. A new encoding was then defined (\LMC\ --- Local Mongolian Cyrillic) which is a close approximation of a transliteration based on Latin1 encoding, notably with front vowels \"a, \"o, \"u (\xalx{"a, "o, "u}) and \"i\ (\xalx{"i}) in matching positions. The encoding is completely detached from the existing Cyrillic codepages of which there are too many; in addition it should be possible to produce Mongolian documents in 7-bit environments so as to assure maximum document portability. An additional ligature table for Metafont was then supplied which takes care of most of the two-letter combinations necessary for entering Cyrillic since the Cyrillic alphabet has more letters (36 in the present version) than the Latin alphabet which prohibited any 1:1-mapping scheme. The used transliteration is very closely modelled after the MLS system yet provides enough transparency for accepting alternative spellings in some cases. \subsection{Output Encodings} Several output encodings are defined for \MonTeX: \begin{description} \item [\LMC] Local Mongolian Cyrillic: This encoding was defined in order to avoid collisions with existing Cyrillic encodings for \TeX\ and \LaTeX. \item [\LMS] Local Mongolian Script: The system's generic encoding for the Mongolian script which is still given in glyph encoding. The language is represented by a Latin transliteration the letters of which are essentially treated as future Mongolian canonical code positions. Once Mongolian Unicode will be available, the Latin transliteration can be seamlessly replaced by Mongolian canonical characters. \item [\LMT] Local Mongolian-Tibetan: This encoding is reserved to ensure access to the characters in the future \Zanabazar\ package: \Soyombo\footnote{% It is possible to use the Soyombo package available since 1996 as long as \Zanabazar\ is not available.} and \XD. It is designed to comprise Tibetan as well, and Sirlin's Tibetan fonts can be directly used with this encoding. \end{description} \subsection{\MonTeX\ and recent \TeX\ trends} At present (early 1999, that is) a new Cyrillic encoding for \TeX/\LaTeX\ is being developed. The encoding is known as \Ttwo; \Ttwo\ comes in several varieties depending on which of the numerous extended Cyrillic characters for non-Slavic languages (Altaic, Turkic, Caucasian, etc.) is needed. The \Ttwo\ encoding comes with its own set of Cyrillic fonts maintained by Vladimir Volovich and Werner Lemberg. As soon as the \Ttwo\ encoding is stable and becomes a standard part of the \TeX\ distributions \MonTeX\ will switch to \Ttwo. At the moment the private encoding \LMC\ is favoured over \Ttwo\ just because it could be tested; future implementations of \MonTeX\ will provide a smooth transition for the user: documents developed with older versions of \MonTeX\ will be upward compatible. The present version of \MonTeX\ understands the Cyrillic character names as given in LH (which is closer to Unicode than the original design). The \texttt{babel} package will also be supported in due course; at the moment, \texttt{babel} support is lacking mainly due to font encoding questions. At present, \MonTeX\ is \emph{not} built with \texttt{babel} compatibility in mind. It must be seen as a stand-alone extension similar to \texttt{german.sty} or the \textsf{CJK} package. The future belongs to 16-bit character sets; the first \TeX\ development supporting larger character sets is $\Omega$mega of which experimental versions exist. One of the great features of $\Omega$mega is the capability to process canonical input encodings in order to generate glyph variants for document presentation. These so-called translation processes are far more powerful than anything Metafont can offer via ligatures, and they are the only feasible way to avoid external preprocessors or internal retransliteration engines coded in \TeX\ needed to process Mongolian script.\footnote{% The retransliteration engine provided with \MonTeX\ has a rather `combined' approach; basic letter forms are selected in the retransliteration section while typical ligatures are composed with the ligature tables of Metafont. The authors express their sincerest gratitude to David Carlisle who contributed the missing link between characters in the output list and \TeX/Metafont's ligature mechanism.} Prof.~Lagally's Arab\TeX\ is the only \LaTeX\ package known to the authors where an extensive retransliteration engine is realized as pure \TeX\ code; it is an impressive piece of work defying any simple-minded imitation. So far, $\Omega$mega translation processes exist for Tibetan and Arabic (paragons of complex relations between original script and any attempted romanization). \section{Installation} The installation procedure depends on the nature of the actual \TeX\ system. The directory tree of teTeX is different from the emtex tree; hence the source archive \texttt{montex}\textit{nn}\texttt{.zip} features the following subdirectories the contents of which has to be placed into appropriate branches of the \TeX\ installation: \begin{itemize} \item \texttt{mfinput} holds the complete Metafont sources for the Mongolian fonts. The suggested path for emtex users is \verb"\emtex\mfinput\"; for teTeX users \verb"$TEXMF/fonts/source/public/" is a suitable choice. \item \texttt{tfm} holds all necessary font metrics files. The suggested path for emtex users is \verb"\emtex\tfm\"; for teTeX users \verb"$TEXMF/fonts/tfm/public/" is a suitable choice. \item \texttt{texinput} holds all style files, font encoding definitions etc. which are read by \LaTeXe. The suggested path for emtex users is \verb"\emtex\texinput\montex"; for teTeX users \verb"$TEXMF/tex/latex/montex/" is a suitable choice. \item \texttt{doc} contains the documentation (the document which you are reading right now). It can be placed in \verb"\emtex\doc\montex" (for emtex users) or \verb"$TEXMF/doc/latex/montex" (for teTeX users). \end{itemize} It may become necessary to rehash the directory database of the \TeX\ system. When in doubt, consult your system administrator or local \TeX\ guru\footnote{\TeX\ wizard, that is. Personal communication with David Kastrup.}. On teTeX systems, the command \texttt{texhash} will perform this service. If you want to use \MonTeX\ Implementation Level IIa or IV it is necessary to ensure PostScript support both on the \LaTeX\ side (\texttt{graphics} and \texttt{rotating} packages, \texttt{dvips} command) and the system side (PostScript printer or emulation must be installed). \subsection{Hyphenation Patterns} \MonTeX\ provides hyphenation rules for Modern Mongolian (Xalx). Hyphenation patterns for English are activated with English as selected language; hyphenation patterns for Russian exist at CTAN but they are unfortunately not suited for \MonTeX\ withour prior work. Hyphenation patterns for Buryat have not been developed yet. Due to the very nature of \TeX, hyphenation patterns for a given language cannot be loaded at run-time but must be compiled into a so-called format file which gets loaded by \TeX\ whenever the command \texttt{latex} is executed. A format file is usually created when a new \TeX\LaTeXe\ system is installed, but creating a new format can be done at any later time again. A special variant of \TeX\ called \texttt{initex} is used for this purpose. The procedure sounds more intimidating than it actually is. Since there are many different types of \TeX\ installations, the procedure is somewhat system-dependent. There is detailed on-line documentation available for performing this task, either in form of a text file for emtex, or in form of a FAQ file which can be displayed using the command \texttt{texconfig faq} on teTeX systems. \section{User Commands} \subsection{General Settings} In order to access the commands of \MonTeX\ the package must be loaded in the document preamble by saying \begin{verbatim} \usepackage[,]{mls} \end{verbatim} The options include choices for the basic document language and input encodings. \subsubsection{Document Language} The document language can be set with one of \ifthenelse{\value{BuryatAvailable}=1}{\texttt{buryat}, }{} \verb"english", \verb"russian" or \verb"xalx" like in \begin{verbatim} \usepackage[xalx]{mls} \end{verbatim} which issues all captions and the date in Modern Mongolian \ifthenelse{\value{BuryatAvailable}=1} {(for \texttt{buryat} see table~\ref{Buryatcaptions},}{} for \verb'english' the default English captions of \LaTeXe\ are used, for \verb'russian' see table~\ref{Russiancaptions}, for \verb'xalx' see table~\ref{Xalxcaptions}). The option \verb"english", at least as a \verb"\usepackage" option, is essentially a do-nothing: it sets captions to English (which is the default of this package anyway). \ifthenelse{\value{BuryatAvailable}=1} {\begin{table}[h] {\captionsburyat \CaptionsList{Buryat}} \end{table}}{} \begin{table}[h] {\captionsrussian \CaptionsList{Russian}} \end{table} \begin{table}[h] {\captionsxalx \CaptionsList{Xalx}} \end{table} The date form follows \TeX\ conventions and is thus a mixture of numbers and words. Thus for \verb"\today" (\today) we have% \footnote{The actual date at compilation time is used for the examples}:\label{today} \begin{center} \begin{tabular}{ll} \textbf{Buryat} &\BuryatToday\\\label{BuryatToday} \textbf{Xalx} &\XalxToday\\\label{XalxToday} \textbf{Russian} &\RussianToday\\\label{RussianToday} \end{tabular} \end{center} The language specifiers \ifthenelse{\value{BuryatAvailable}=1}{\texttt{buryat},}{} \verb"english", \verb"russian" and \verb"xalx" can also be used anywhere in the document as arguments to the \verb"\selectlanguage" command. Instead of stating an argument to \verb"\usepackage{mls}" it is possible to say in your document \begin{quote} \verb"\selectlanguange{xalx}" \end{quote} which would set captions to Xalx Mongolian. \subsubsection{Input Encodings} \MonTeX\ is flexible enough to deal with several kinds of input encodings including code pages with Cyrillic letters. In an environment based on Latin1 encoding (ISO 8859-1, ANSI or Windows) no input encoding conversion is necessary. Users working on IBM compatible DOS platforms can specify the option \verb"mls" in the preamble: \begin{verbatim} \usepackage[mls]{mls} \end{verbatim} The MLS codepage is compatible with the IBM 437 codepage as far as the front vowels are concerned but features additional Cyrillic letters. Other available input encodings support either front vowels or Cyrillic letters but usually not both. Basically it is possible to use \MonTeX\ without anything else but the plain 7-bit ASCII encoding since internal and external mechanisms are available which can render transliterated texts into Mongolian. Using Latin1 encoding is however much more elegant as the Mongolian front vowels can be entered directly. If a local environment supports Cyrillic and Script codepages then texts can be composed using these codepages. Table \ref{inputenc} shows which codepages are supported. Those codepage names which are followed by a `(+)' are supplied by \MonTeX\ whereas the other codepage declarations are recognized and passed through to the system. The column ``Front Vowels'' indicates whether the vowels \"o and \"u (and their Mongolian counterparts \xalx{"o, "u}) are available in that particular codepage. \begin{table} \begin{center} \begin{tabular}{l|c|c|c|c|c|c|c|c|c|c} Encoding Option&\multicolumn{7}{|c|}{Transliteration Symbols}% &Cyrillic&\multicolumn{2}{|c}{Front Vowels}\\\hline \rule{0mm}{2.25ex}% &\"A/\"a&\"O/\"o&\"U/\"u&\"E&\"e&\"I&\"i& &\xalx{"O/"o}&\xalx{"U/"u}\\ \hline \texttt{mls} (+) &yes &yes &yes &no &yes&no &yes&yes &yes &yes \\ \texttt{ncc} (+) &no &no &no &no &no &no &no &yes &yes &yes \\ \texttt{mos} (+) &no &no &no &no &no &no &no &yes &yes &yes \\ \texttt{mnk} (+) &no &no &no &no &no &no &no &yes &yes &yes \\ \texttt{dbk} (+) &no &no &no &no &no &no &no &yes &yes &yes \\ \texttt{ctt} (+) &no &no &no &no &no &no &no &yes &yes &yes \\ \texttt{ibmrus}(+)&no &no &no &no &no &no &no &yes &no &no \\ \texttt{koi} (+) &no &no &no &no &no &no &no &yes &no &no \\ \texttt{850} &yes &yes &yes &yes&yes&yes&yes&no &no &no \\ \texttt{852} &yes &yes &yes &yes&yes&no &no &no &no &no \\ \texttt{mac} &yes &yes &yes &yes&yes&yes&yes&no &no &no \\ \texttt{atari} &yes &yes &yes &yes&yes&yes&yes&no &no &no \\ \texttt{roman8} &yes &yes &yes &yes&yes&yes&yes&no &no &no \\ \end{tabular} \caption{Available Characters in Input Encodings\label{inputenc}} \end{center} \end{table} \subsection{Cyrillic Text -- \xalx{Kirill "us"ag}} \subsubsection{Cyrillic Text in Transliteration (LMC) Mode} \MonTeX\ provides two basic modes of operation: in \begin{itemize}\label{SetDocumentEncoding} \item Transliteration Mode (intimately linked to the LMC encoding) all incoming text is regarded as transliterated Cyrillic. This allows users to compose Cyrillic documents on pure ASCII machines. In contrast, the \item Immediate Mode does nothing and waits for explicit Cyrillic characters in the input in order to generate Cyrillic output. \end{itemize} Two commands are used to switch between these modes: \begin{quote} \begin{verbatim} \SetDocumentEncodingLMC \SetDocumentEncodingNeutral \end{verbatim} \end{quote} The first command switches to Transliteration Mode, the second command deactivates the transliteration and thus, by definition, activates Immediate Mode. In the LMC encoding, most Cyrillic characters are mapped directly to a single Latin character but for some characters there is a text command which became necessary since there are more Cyrillic than Latin characters. For convenience, a few ligatures were defined, too. Details are given in table~\ref{cyralpha}. \input{cyralpha} Front vowels can be entered directly using the encoding slot of a valid and active input encoding, or they can be expressed via an abbreviated \verb'"'\emph{v} notation where \emph{v} stands for any desired vowel. In the \LMC\ encoding used by \MonTeX, \verb'"' is not an active character; selecting the proper letter is done by ligature statements in the Metafont sources. Some letters can be entered with or without a preceding \verb"\", like \cyryu\ and \cyrya. Both \verb"\yu" and \verb"yu" will produce a \cyryu. While \verb"yu" is interpreted as a ligature, \verb"\yu" allows for the character \cyryu\ to be combined with accents. Accents are not commonly used in Mongolian since there are precise rules for word stress. This feature is taken from the \textsf{OT2} encoding and is included mainly for the sake of completeness, convenience and compatibility\footnote{The magic triple-C!}. Here now a sample of Mongolian text: \exa {\mnr<> x"am"a"an aldarshsan, Z"u"un xyazgaaryg toxinuulax sa"id N.~Dugarjaw ardyn xuw\i sgalyn b"u"ur "ax"an "ue"as xamgi"in "agz"agt"a"i am\i\ d"u"is"an alband tomilogdox c"ar"ag da"iny olon quxal daalgawryg xiq"a"ang"u"il"an biel"u"ulj yawsan t"u"uxt"a"i x"un.} \exb \begin{verbatim} {\mnr<> x"am"a"an aldarshsan, Z"u"un xyazgaaryg toxinuulax sa"id N.~Dugarjaw ardyn xuw\i sgalyn b"u"ur "ax"an "ue"as xamgi"in "agz"agt"a"i am\i\ d"u"is"an alband tomilogdox c"ar"ag da"iny olon quxal daalgawryg xiq"a"ang"u"il"an biel"u"ulj yawsan t"u"uxt"a"i x"un.} \end{verbatim} \exc In order to make the document you are reading at the moment truly portable, the somewhat more clumsy \verb'"'\emph{v} notation was used in this example; if your environments supports an 8-bit codepage (what it usually does), all front vowels can be entered as \texttt{\"a}, \texttt{\"o} and \texttt{\"u} etc. using the slots of those vowels in the particular active codepage.\footnote{% Looking at the source code of this document the astute reader will discover that all front vowels are indeed produced using the \tttt{ "a} (etc.) notation; thus the document source can be viewn and manipulated on any 7-bit ASCII platform; it can also safely be transmitted via e-mail.} \subsubsection{Entering Cyrillic Text in Immediate Mode} For freely combining Latin and Cyrillic characters without using any explicite commands it is necessary that the codepage in use supports some Cyrillic encoding. It should be noted, however that these documents are not easily portable between different platforms anymore since they need recoding; some of the Cyrillic codepages are defective in one or the other way thus individual characters can get lost. \begin{sloppypar} The user simply specifies the desired input encoding as a \verb:\usepackage[:\emph{}\verb:]{mls}: option, and \MonTeX\ takes care of the rest. It is a feature and not a bug that input encoding and document language are chosen independently. It is well possible that a user working on a computer with default Mongolian codepage wants to create a document in Russian, English or any other language yet wants to include Mongolian fragments in her text without explicitely issuing any command. \end{sloppypar} In case a need arises for switching from Transliteration Mode to Immediate Mode the command \verb:\SetDocumentEncodingNeutral: can be issued anywhere in the preamble or the document itself; like \verb:\SetDocumentEncodingLMC: it affects the Cyrillic transliteration only and leaves the document language in its chosen state. \subsubsection{Entering Cyrillic Characters by Name} Outside the Cyrillic environments, individual Cyrillic characters can be entered by using the commands beginning with \verb"\cyr"\textit{x} from the two right columns of table \ref{cyralpha} where \textit{x} stands for the letter name. This command works in any encoding. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% %% Temporarily disabled. Do we really need it? %% %% \subsubsection{Choosing between MLS and LH Names} %% %% The naming conventions used in the MLS (Mongolian Language Support) %% system and LMC (Local Mongolian Cyrillic) character set were %% developed independently from the extended Cyrillic %% character set definition which will form the forthcoming \LaTeXe\ %% standard Cyrillic character set. Unfortunately there are differences %% in certain glyph names which are sometimes ambiguous for \MonTeX\ %% and the LH world. %% %% In order to provide compatibility with the future standard, \MonTeX\ %% provides a switch which allows setting the desired character naming %% convention: \verb|\CyrLHStyle|. With the setting \verb|CyrLHStyletrue| %% the LH names are activated, with \verb|CyrLHStylefalse| the %% established MLS names as given in the right-side columns of %% table~\ref{cyralpha} (``Generic Command'') are selected.\footnote{% %% Once it is possible to use standard \Ttwo\ fonts with %% \MonTeX\ this switch will probably be set automagically.} %% This is necessary for backward %% compatibility so that existing documents should not be %% re-edited. The character names affected are given in %% table~\ref{cyrlhstyle}. %% %% \begin{table} %% \begin{center} %% \begin{tabular}{l|cccccccccccccc} %% \tttt{CyrLHStylefalse}&w &ye &j &ij &oe &ue &x\\ %% \tttt{CyrLHStyletrue} &v &e &zh &ishrt &otld &y &h\\ %% \hline %% \tttt{CyrLHStylefalse}&h &q &qh &xat &zln &e \\ %% \tttt{CyrLHStyletrue} &hsha &ch &chsh &hrd &sft &e \\ %% \end{tabular} %% \caption{\MonTeX\ Style and LH Style Character Names}\label{cyrlhstyle} %% \end{center} %% \end{table} \subsubsection{Entering Special Characters} A few special characters are available, notably the guillemots frequently used for quoting text, the currency symbol and the ordinal number symbol. See table~\ref{SpecialCharacters}. \begin{table} \begin{center} \begin{tabular}{cll} Symbol & Symbolic Command & Alternative\\ \mnr\lgu& \verb|\lgu| & \verb|<<|\\ \mnr\rgu& \verb|\rgu| & \verb|>>|\\ \No & \verb|\No| & \\ \Togrog & \verb|\Togrog| & \\ \togrog & \verb|\togrog| & \\ \end{tabular} \caption{\MonTeX\ Special Characters\label{SpecialCharacters}} \end{center} \end{table} The command producing the guillemots (\verb|\lgu|, \verb|\rgu|) \emph{only} works in a Cyrillic environment --- it is not a generic command. There are actually two versions of the \verb|\togrog| command. While \verb|\Togrog| produces a sans serif \Togrog\ (considered standard) with any font selected it is also possible to print serif (\MyTogrog), italic (\textit{\MyTogrog}) and typewriter (\texttt{\MyTogrog}) versions of this symbol.\footnote{The currency symbol is not limited to these three typefaces; all typefaces can be selected.} For achieving this result the commands \verb|\MyTogrog| and \verb|\mytogrog| are available. Unlike the standard command they simply pick the current font style of the surrounding letters for the currency symbol. \subsubsection{Running Text with Embedded Words in Different Encodings} Independently of the document language it is possible to produce portions of Cyrillic text within Latin text and vice versa. The two commands \verb"\mnr" and \verb"\rnm" switch from ordinary Latin text to transliterated Cyrillic text and back to Latin text. The command stands for \emph{m}ongolian \emph{n}ew \emph{r}omanization and its reversal (which can, by accident, also be read as \emph{r}eturn to \emph{n}or\emph{m}al). They can be used as stream commands or for initializing groups:\label{mnrnm} \exa \mnr mongol x"al ba \rnm english text with a {\mnr mongol} word inserted \exb \begin{verbatim} \mnr mongol x"al ba \rnm english text with a {\mnr mongol} word inserted \end{verbatim} \exc For enhanced convenience, portions of text can also be encapsulated into the commands \verb"\xalx{...}" for Cyrillic text and \verb"\lat{...}" for neutral (i.~e. Latin) texts.\label{capsules} The commands \verb"\mnr", \verb"\rnm", \verb"\xalx{...}" and \verb"\lat{...}" do \emph{not} switch the default encoding; this shows up when a construct like \verb:\lat{\verb|article|}: is placed in Transliteration Mode; the result will be \texttt{\xalx{article}} rather than \texttt{article}; in order to generate the desired form, the mode switching commands must be used. \subsubsection{Font Selection Commands} The Cyrillic fonts are set up in a manner which allows for seamless switching between Roman and Cyrillic typefaces. The font switching commands used for modifying typefaces (by \verb"\text..") are completely transparent to the encoding; no precaution whatsoever has to be taken. Most of the typefaces supplied with the traditional \textsf{OT1} encoding are also available for \MonTeX; Dunhill and Funny Roman are included.\footnote{% A complete overview of the NFSS classification of the Computer Modern fonts can be found in The \LaTeX\ Companion, by Michel Goossens, Frank Mittelbach and Alexander Samarin, Addison-Wesley 1994, p.~181.} \MonTeX\ offers the following font families: \begin{center} \begin{tabular}{l|l} \verb'\fontfamily{...}' parameter & Family Description \\ \hline cmr & Computer Modern Serif\\ cmss & Computer Modern Sans Serif\\ cmtt & computer Modern Typewriter\\ cmvtt& Computer Modern Variable Width Typewriter\\ cmfr & Computer Modern Funny\\ cmfib& Computer Modern Fibonacci\\ cmdh & Computer Modern Dunhill\\ cmssq& Computer Modern Sans Serif Quotation Style 8pt\\ \end{tabular} \end{center} The word ``Roman'' was avoided since in \MonTeX\ these families also cover matching typefaces in Cyrillic script. The first three families have support for combinations of different weights and shapes (e.\,g. bold and italic) whereas the other series usually only offer an italic variant. The Sans Serif Quotation Style 8pt typeface is not by default installed in standard \LaTeX\ distributions hence it cannot be guaranteed that switching to and from Cyrillic letters maintains the typeface. The fonts (upright and slanted) can be accessed via the \verb|\fontfamily{cmssq}| command but are not shown in table~\ref{typeface}. See table~\ref{typeface} for a therefore incomplete list of available typeface examples. \begin{table} \begin{center} \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \hline \multicolumn{2}{|c|}{Family and Command Example}& \multicolumn{2}{c|}{Typeface Examples} \\ \hline \multicolumn{2}{|c|}{Computer Modern Serif}&& \\ cmr & (default) & {\Tw}& {\Wt}\\ & \tttt{ textbf\{...\}}& \textbf{\Tw}&\textbf{\Wt}\\ & \tttt{ textsl\{...\}}& \textsl{\Tw}&\textsl{\Wt}\\ & \tttt{ textsc\{...\}}& \textsc{\Tw}&\textsc{\Wt}\\ & \tttt{ textit\{...\}}& \textit{\Tw}&\textit{\Wt}\\ & \tttt{ fontseries\{bx\}\char92 textit\{...\}}& \fs{bx}\textit{\Tw}&\fs{bx}\textit{\Wt}\\ \hline \end{tabular} \ifthenelse{\value{FontSamples}>1}{% \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \multicolumn{2}{|c|}{Computer Modern Typewriter}&& \\ cmtt & \tttt{texttt\{...}\}& \texttt{\Tw}&\texttt{\Wt}\\ & \tttt{texttt\{\char92textit\{...\}\}}& \texttt{\textit{\Tw}}&\texttt{\textit{\Wt}}\\ & \tttt{texttt\{\char92textsl\{...\}\}}& \texttt{\textsl{\Tw}}&\texttt{\textsl{\Wt}}\\ & \tttt{texttt\{\char92textsc\{...\}\}}& \texttt{\textsc{\Tw}}&\texttt{\textsc{\Wt}}\\ \hline \end{tabular}}{\rule{0mm}{4ex}Computer Modern Typewriter example suspended\dots} \ifthenelse{\value{FontSamples}>2}{% \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \multicolumn{2}{|c|}{Computer Modern Variable Width Typewriter}&& \\ cmvtt & \tttt{ fontfamily\{cmvtt\}...}& \ff{cmvtt}\Tw&\ff{cmvtt}\Wt\\ & \tttt{ fontfamily\{cmvtt\}\char92 textit\{...\}}& \ff{cmvtt}\textit{\Tw}&\ff{cmvtt}\textit{\Wt}\\ \hline \end{tabular}}{\rule{0mm}{4ex}Computer Modern Variable Width Typewriter example suspended\dots} \ifthenelse{\value{FontSamples}>3}{% \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \multicolumn{2}{|c|}{Computer Modern Sans}&& \\ cmss & \tttt{ textsf\{...\}} & \textsf{\Tw}&\textsf{\Wt}\\ & \tttt{ textsf\{\char92 textsl\{...\}\}}& \textsf{\textsl{\Tw}}&\textsf{\textsl{\Wt}}\\ & \tttt{ textsf\{\char92 fontseries\{bx\}...\}}& \textsf{\fs{bx}\Tw}&\textsf{\fs{bx}\Wt}\\ & \tttt{ textsf\{\char92 fontseries\{sbc\}...\}}& \textsf{\fs{sbc}\Tw}&\textsf{\fs{sbc}\Wt}\\ \hline \end{tabular}}{\rule{0mm}{4ex}Computer Modern Sans example suspended\dots} \ifthenelse{\value{FontSamples}>4}{% \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \multicolumn{2}{|c|}{Computer Modern Funny}&& \\ cmfr & \tttt{ fontfamily\{cmfr\}...}& \ff{cmfr}\Tw&\ff{cmfr}\Wt\\ \hline \end{tabular}}{\rule{0mm}{4ex}Computer Modern Funny example suspended\dots% For printing the Modern Funny example the user is required to edit this file (\texttt{montex.tex}) at line 26 and set the value of \texttt{fontSamples} to ``5''.} \ifthenelse{\value{FontSamples}>5}{% \begin{tabular}{|p{0.75cm}p{7cm}|p{1.5cm}p{1.5cm}|} \multicolumn{2}{|c|}{Computer Modern Dunhill}&& \\ cmdh & \tttt{ fontfamily\{cmdh\}...}& \ff{cmdh}\TW&\ff{cmdh}\WT\\ \hline \end{tabular}}{\rule{0mm}{4ex}Computer Modern Dunhill example suspended\dots% For printing the Modern Dunhill example the user is required to edit this file (\texttt{montex.tex}) at line 26 and set the value of \texttt{fontSamples} to ``6''.} \caption{Typefaces are consistent for both Cyrillic and Latin.\label{typeface}} \end{center} \end{table} Besides these transparent commands for scalable fonts \MonTeX\ also offers two inch-high variants of bold Computer Modern Sans typefaces for Latin and Cyrillic: \verb|\cminch| and \verb|kminch|. These commands bypass the NFSS font setup and should only be used for book titles etc. The command sequence \verb|{\cminch AB} {\kminch AB}| produces the following output:\label{cminchkminch} \vspace{5mm} \begin{center} {\cminch AB} {\kminch AB} \end{center} \vspace{5mm} \subsubsection{Shorthands for embedding words in a different typeface}\label{typefacecapsules} Sometimes it may be necessary to give short portions of text not only in a different encoding (for which the \verb:\lat{...}: and \verb:\mnr{...}: commands are useful) but it may also be necessary to switch the typeface temporarily. Usually capsules using \verb'\text'\emph{xx} do the work if only the typeface is concerned, and building nested commands like \verb'\textsf{\lat{...}}' is cumbersome if these changes have to be applied very often. \MonTeX\ provides an abbreviated style following the rule \begin{quote} \texttt{[k|l]}\emph{two letter font style code}\verb'{...}' \end{quote} where the font style code is one of \verb'rm', \verb'bf', \verb'it', \verb'sl', \verb'sf', \verb'sc' and \verb'tt', like \verb'\ksl{...}', \verb'\lsc{...}', etc. \subsubsection{Shorthands for writing transliterated texts} \MonTeX\ provides shortcuts for writing certain accented symbols used in conventional transliterating of Mongolian by haceks, the nasal and the gamma. These shortcuts are essentially mnemonics replacing the somewhat more tedious accent notation (see table~\ref{shortcuts}). \begin{table} \begin{center}\begin{tabular}{ll|ll} %\hline Letter & Input & Letter & Input \\ & & &\\ \hline & & &\\ \ch & \verb"\ch" & \Ch & \verb"\Ch" \\ \jh & \verb"\jh" & \Jh & \verb"\Jh" \\ \sh & \verb"\sh" & \Sh & \verb"\Sh" \\ \zh & \verb"\zh" & \Zh & \verb"\Zh" \\ \ng & \verb"\ng" & \Ng & \verb"\Ng" \\ \g & \verb"\g" & & \\ %\hline \end{tabular}\end{center} \caption{Shortcuts for Transliteration Symbols\label{shortcuts}} \end{table} It must be observed that these commands are by default dependent on the environment they are used in. \verb"\Sh" yields a \Sh\ when used in a Latin environment but results in a \mnr\Sh\rnm\ when used in a Cyrillic context\footnote{The authors wish to thank J.~Knappen for resolving one instability in the original code for these letters.}: \exa \emph{\Sh agdar} and \emph{\Ch adraa} are transliterations for {\mnr\Sh agdar} and {\mnr\Ch adraa}. \exb \begin{verbatim} \emph{\Sh agdar} and \emph{\Ch adraa} are transliterations for {\mnr\Sh agdar} and {\mnr\Ch adraa}. \end{verbatim} \exc All accented characters which are contained in the \textsf{T1} encoding or can be generated out of these via accents can be produced. This comes in conveniently for transliterating Oirat texts which needs vowels with double diacritics, like \rule{0mm}{2.5ex}\={\"a} which can be entered as any combination of two nested accent commands (like \verb:\={\"a}:) or one accent command and a vowel with diacritics (provided an 8-bit input codepage is available). \subsubsection{Numbering by Cyrillic letters} Analogous to the \verb:\Alpha: command which provides an alphabetical counter in English, \MonTeX\ features counters for Buryat, Modern Mongolian, and Russian. \begin{description} \item [Buryat]\label{Uzeg} The counter for Buryat is invoked with \verb"\Uzeg{"\emph{n}\verb|}| or \verb"\uzeg{"\emph{n}\verb|}| and is valid for $1\le n\le32$. \def\NR #1{$^{#1}$\Uzeg{#1}/\uzeg{#1}} \begin{tabular}{cccccc} \NR{1} &\NR{2} &\NR{3} &\NR{4} & \NR{5}&\NR{6} \\ \NR{7} &\NR{8} &\NR{9} &\NR{10}&\NR{11}&\NR{12}\\ \NR{13}&\NR{14}&\NR{15}&\NR{16}&\NR{17}&\NR{18}\\ \NR{19}&\NR{20}&\NR{21}&\NR{22}&\NR{23}&\NR{24}\\ \NR{25}&\NR{26}&\NR{27}&\NR{28}&\NR{29}&\NR{30}\\ \NR{31}&\NR{32}\\ \end{tabular} \item [Modern Mongolian]\label{Useg} The counter for Modern Mongolian is invoked with \verb"\Useg{"\emph{n}\verb|}| or \verb"\useg{"\emph{n}\verb|}| and is valid for $1\le n\le31$. \def\NR #1{$^{#1}$\Useg{#1}/\useg{#1}} \begin{tabular}{cccccc} \NR{1} &\NR{2} &\NR{3} &\NR{4} & \NR{5}&\NR{6} \\ \NR{7} &\NR{8} &\NR{9} &\NR{10}&\NR{11}&\NR{12}\\ \NR{13}&\NR{14}&\NR{15}&\NR{16}&\NR{17}&\NR{18}\\ \NR{19}&\NR{20}&\NR{21}&\NR{22}&\NR{23}&\NR{24}\\ \NR{25}&\NR{26}&\NR{27}&\NR{28}&\NR{29}&\NR{30}\\ \NR{31}\\ \end{tabular} \item [Russian]\label{Asbuk} The counter for Russian is invoked with \verb"\Asbuk{"\emph{n}\verb|}| or \verb"\asbuk{"\emph{n}\verb|}| and is valid for $1\le n\le28$. \def\NR #1{$^{#1}$\Asbuk{#1}/\asbuk{#1}} \begin{tabular}{cccccc} \NR{1} &\NR{2} &\NR{3} &\NR{4} & \NR{5}&\NR{6} \\ \NR{7} &\NR{8} &\NR{9} &\NR{10}&\NR{11}&\NR{12}\\ \NR{13}&\NR{14}&\NR{15}&\NR{16}&\NR{17}&\NR{18}\\ \NR{19}&\NR{20}&\NR{21}&\NR{22}&\NR{23}&\NR{24}\\ \NR{25}&\NR{26}&\NR{27}&\NR{28}\\ \end{tabular} \end{description} \newpage \subsection{Mongolian Script Support -- \bcg{mongGol bicig}}\label{bcg} Beginning with Implementation Level II \MonTeX\ provides Mongolian Script support. Mongolian Script can be used in running text like \marginpar{\mbosoo{mongGol}\mbosoo{bicig} \vspace{2.54mm} \raggedright\small without PostScript support Mongolian text enclosed in vertical capsules will be printed \emph{horizontally}!} this: \bcg{mongGol bicig}, and if your system supports PostScript, it can even be encapsulated in vertical capsules containing one word or a short series of words, like in the keyword entry of dictionaries.\footnote{Famous dictionaries with a mixture of vertical and horizontal printing are I.~J.~Schmidt's Mongolian-Russian-German dictionary (1835) and F.~Lessing's Mongolian-English dictionary (1960).} See chapter \ref{VerticalCapsules} for the necessary commands. Vertical capsules like these \mbosoo{mongGol} \mbosoo{bicig} (\textit{mong\g ol bicig}) can appear anywhere in your text; line spacing etc. will adapt automatically. It is also possible to create margin notes with Mongolian Script. \subsubsection{Mongolian Script Fundamentals} Mongolian Script, or \emph{bicig}, is a writing with an intriguing and complex relationship between the canonical letters of the alphabet and their presentations in context. Virtually any canonical letter can assume several shapes. As a rule of thumb, there are three or four basic shapes: the letter in isolated form, the letter in initial, medial and final position of a word. Only a few letters stay the same, and in rare cases there are up to ten possibilities for representing a single letter. On the other hand, some letters share the same shape in different contexts; one so-called \emph{glyph} can represent more than one letter, sometimes three or four different letters. Decomposing the writing system and using glyphs as the atoms of writing is a feasible way of writing Mongolian script. \MonTeX\ provides a transliteration engine but can also use the glyph input method developed for the MLS system. A complete guide to the principles of glyph analysis can be found in the MLS Report by one of the authors.\footnote{Oliver Corff: MLS Report. UNU/IIST Report No.~8, Macau 1993} In \MonTeX, Mongolian script can be entered in three ways, either by writing transliterated Mongolian, by an approximated symbol for every glyph or by generic name. \MonTeX\ features a rather simple-minded Transliteration Mode for Mongolian Script; every input which consists only of letters, digits and punctuation marks is accepted, \TeX\ and \LaTeX\ commands make the system derail. \subsubsection{Entering Mongolian Script in Transliteration Mode} \begin{sloppypar} In Transliteration Mode (activated with the commands \verb|\bcg{...}| or \verb|\mbosoo{...}|)\label{mbosoo} Mongolian text portions can be entered using a transliteration which is a rough approximation to the MLS system. The major difference is that only pure Latin alphabetical symbols can be used for virtually all letters. Front vowels are either entered via the traditional vowels with diacritics (\emph{\"a, \"o, \"u}) or can be entered with capitalized versions of the normal vowels. Capitalized letters have to be used for entering {\g} and {\sh} which are entered as \textit{G} and \textit{S}. Special variants for certain letters can be selected with Form Variant Selectors. Examples are given in section~\ref{bicig-examples}.\footnote{% The suggested solution has the advantage that it can be used on computers featuring codepages without umlaut symbols as most of the Cyrillic code pages are `defective' in this point.} \end{sloppypar} The available Mongolian characters (\emph{ca\g an tolu\g ai}) are shown in table~\ref{bcgcagan}.\footnote{The alphabetical arrangement follows large that given on p.~17 of N.~Poppe's \textit{Grammar of Written Mongolian}, Wiesbaden 1954, 1964, 1974 (third printing). Letters not given there are appended to Poppe's list.} \input{bcgcagan} \subsubsection{Entering Mongolian Script in Immediate Mode} For freely combining Mongolian Script with other characters without using any explicite commands it is necessary that the codepage in use supports Mongolian Script glyphs; currently this is the MLS codepage. The MLS input encoding is specified like \verb:\usepackage[mls]{mls}:. As with Cyrillic codepages it should be noted that these documents are not easily portable between different platforms anymore since they cannot be recoded at ease. See table~\ref{bcgalpha} for a list of available symbols. \subsubsection{Entering Mongolian Script by Symbol or Name} Without MLS codepage support, Mongolian words can also be entered using the \verb"\glyphbcg{...}" command in running text. Within these groups, Mongolian Script glyphs are entered in the form of approximated symbols; sometimes these symbols reflect the underlying canonical letter, sometimes functional equivalents (for punctuation marks etc.) are chosen; sometimes there is no evident relation between glyph and input symbol simply because a free slot within the ASCII range $\le127$ was chosen. Please consult table~\ref{bcgalpha} for a list of available glyphs and their input equivalents. Without any preparations on the side of the text environment it is possible to enter individual Mongolian glyphs by name in a way similar for that of entering Cyrillic characters; the Mongolian glyph names can be found in table~\ref{bcgalpha}. Thus, \verb|\shilbe| produces a \shilbe. A number in the MLS column indicates the encoding position of the MLS codepage; a missing number in this column indicates that the glyph is part of extended \MonTeX\ glyph set without being part of the original MLS. %\newpage \input{bcgalpha} \subsubsection{Entering Mongolian Script Special Symbols} For the correct operation of retransliterating systems processing Mongolian script additional symbols are needed. These include Form Variant Selectors (\textsf{FVS}), the Vowel Separator, and other symbols like the Mongolian Positional Indicator. As can be seen from its usage in table~\ref{bcgcagan}, entering \verb|*ng| tells the system to consider this \emph{ng} to be in non-initial position.\footnote{Unfortunately, though it is now commonly agreed in the scientific community that these symbols are needed, their definition is still in a state of flux, and thus the symbols given here are presented on a preliminary basis.} Besides these symbols, table~\ref{LMS-SpecialSymbols} includes also some useful punctuation marks etc.\ as they are used in Mongolian Script. \begin{longtable}{|c|l|l|} \hline\endhead \hline\endfirsthead \hline\endfoot \hline\endlastfoot Symbol & Name & Input \\ \hline \bosoo{\glyphbcg{!}} & Exclamation Mark & \verb|!| \\ \bosoo{\glyphbcg{?}} & Question Mark & \verb|?| \\ \bosoo{\glyphbcg{!?}} & Exclamation Question Mark& \verb|!?| \\ \bosoo{\glyphbcg{?!}} & Question Exclamation Mark& \verb|?!| \\ \bosoo{\glyphbcg{*}} & Mong. Positional Indicator& \verb|*| \\ \bosoo{\glyphbcg{\char32}} & Mongolian Space & \verb*|-| \\ \bosoo{\glyphbcg{(}} & Opening Bracket & \verb|(| \\ \bosoo{\glyphbcg{)}} & Closing Bracket & \verb|)| \\ \bosoo{\glyphbcg{<}} & Opening Angle Bracket & \verb|<| \\ \bosoo{\glyphbcg{>}} & Closing Angle Bracket & \verb|>| \\ \bosoo{\glyphbcg{<<}} & Opening Guillemot & \verb|<<| \\ \bosoo{\glyphbcg{>>}} & Closing Guillemot & \verb|>>| \\ % \bosoo{\glyphbcg{\{}} & Opening Parenthesis & \verb|{| \\ % \bosoo{\glyphbcg{\}}} & Closing Parenthesis & \verb|}| \\ \bosoo{\glyphbcg{'}} & Form Variant Selector 1& \verb|'| \\ \bosoo{\glyphbcg{"}} & Form Variant Selector 2& \verb|"| \\ \bosoo{\glyphbcg{\char43}}& Mong. Vowel Separator & \verb|=| \\ \bosoo{\glyphbcg{|}} & Mongolian Nuruu & \verb'|' \\ \bosoo{\glyphbcg{.}} & Period & \verb|.| \\ \bosoo{\glyphbcg{,}} & Comma & \verb|,| \\ \bosoo{\glyphbcg{:}} & Colon & \verb|:| \\ \bosoo{\glyphbcg{;}} & D\"orw\"oljin & \verb|;| \\ \bosoo{\glyphbcg{..}} & Ellipsis & \verb|..| \\ \bosoo{\glyphbcg{0}} & Digit zero & \verb|0| \\ \bosoo{\glyphbcg{1}} & Digit one & \verb|1| \\ \bosoo{\glyphbcg{2}} & Digit two & \verb|2| \\ \bosoo{\glyphbcg{3}} & Digit three & \verb|3| \\ \bosoo{\glyphbcg{4}} & Digit four & \verb|4| \\ \bosoo{\glyphbcg{5}} & Digit five & \verb|5| \\ \bosoo{\glyphbcg{6}} & Digit six & \verb|6| \\ \bosoo{\glyphbcg{7}} & Digit seven & \verb|7| \\ \bosoo{\glyphbcg{8}} & Digit eight & \verb|8| \\ \bosoo{\glyphbcg{9}} & Digit nine & \verb|9| \\ \hline \caption{Mongolian Script Special Symbols and Punctuation Marks}\label{LMS-SpecialSymbols} \end{longtable} \subsubsection{Displaying Transliterations} For huge word lists and similar material it is convenient to enter the transliteration only once and use it as input both for the Mongolian retransliteration engine and the presentation of the transliteration. A construct like \exa \vspace{8mm} \newcommand{\Keyword}[1]{#1 \bcg{#1}} \Keyword{anda} / \emph{Looks nice.} \Keyword{SaGdur} / \emph{Not as nice.} \exb \begin{verbatim} \newcommand{\Keyword}[1]{#1 \bcg{#1}} \Keyword{anda} / \emph{Looks nice.} \Keyword{SaGdur} / \emph{Not as nice.} \end{verbatim} \exc is helpful as long as no capitalized single-letter entity is used. Capitalized entities look less pleasing in conventional texts; for these purposes, the command \verb|\PrettyMLS| is provided which takes input with single-letters entities and converts it to a more traditional representation.\label{PrettyMLS} \exa \vspace{5mm} \newcommand{\Keyword}[1]{% \PrettyMLS{#1} \bcg{#1}} \Keyword{anda} / \emph{Good.}\par \Keyword{SaGdur} / \emph{Good again.} \exb \begin{verbatim} \newcommand{\Keyword}[1]{% \PrettyMLS{#1} \bcg{#1}} \Keyword{anda} / \emph{Good.}\par \Keyword{SaGdur} / \emph{Good again.} \end{verbatim} \exc Two additional flags, \verb|\ShowSpecialMLStrue| and \verb|\ShowSpecialMLSfalse|, can be used to activate canonical identifiers instead of the conventional notation for the special characters of table~\ref{LMS-SpecialSymbols}.\label{ShowSpecialMLS} \exa \PrettyMLS{SaGdur blam=a} \vspace{8mm} \ShowSpecialMLStrue \PrettyMLS{SaGdur blam=a} \vspace{8mm} \ShowSpecialMLSfalse \PrettyMLS{SaGdur blam=a} \exb \begin{verbatim} \PrettyMLS{SaGdur blam=a} \ShowSpecialMLStrue \PrettyMLS{SaGdur blam=a} \ShowSpecialMLSfalse \PrettyMLS{SaGdur blam=a} \end{verbatim} \exc The complete set of characters covered by \verb|\PrettyMLS| is shown in table~\ref{PrettyMLS-Table}. \input pretty.tex \subsubsection{Text in Vertical Capsules\label{VerticalCapsules}}\label{bosoo} With PostScript support, it is possible to create vertical capsules containing short portions of text (typically one or two words); the commands \verb"\bosoo{...}" does not change the current encoding and can be used for non-Mongolian words whereas \verb"\mbosoo{...}" sets the encoding temporarily to \LMS. Accepting the same input as \verb"\bcg{...}", writing \verb"\mbosoo" is simply a shortcut for writing \verb"\bosoo{\bcg{...}}" and facilitates the printing of individual Mongolian words in vertical appearance. \subsection{Mongolian Script Input Examples}\label{bicig-examples} \textbf{Nota bene}: The following examples are all given in horizontal mode. It is visible that grammatical endings are always separated by \verb|-| whereas final vowels are separated from the stem by \verb|=|. Form Variant Selector are used for differentiating between various forms of the same letter in similar environments; \verb|Ed'| \bcg{Ed'} and \verb|sayid| \bcg{sayid} will show you the difference. Without the Form Variant Selector, *\verb|Ed| and \verb|on| will be the same: \bcg{on}. \exa \raggedright \bcg{mongGol bicig. ulaGanbaGatur xota bol mongGol ulus-un nEyislEl. Ed'-Un zasaG, sayid. EngkEbayar. 1999 on. 40 tOgOrOg 20 mOnggU. bUri. blam=a badm=a. utasu: 00976,1,321654} \exb \begin{verbatim} \raggedright \bcg{mongGol bicig. ulaGanbaGatur xota bol mongGol ulus-un nEyislEl. Ed'-Un zasaG, sayid. EngkEbayar. 1999 on. 40 tOgOrOg 20 mOnggU. bUri. blam=a badm=a. utasu: 00976,1,321654} \end{verbatim} \exc \section{External Support Software} With \MonTeX, it is still possible to process documents generated with the MLS software package. The MLS converter produces Cyrillic and Mongolian Script texts out of transliterations using the MLS codepage. Documents encoded in MLS can be directly processed, no further conversion is necessary. \section{Shortcomings, Desiderata, Bugs and Error Reporting} \subsection{Hyphenation Patterns} The Mongolian hyphenation patterns delivered with \MonTeX\ are still under development, so please expect occasional hyphenation errors. It must be also noted that for proper hyphenation of critical words \sh\ should be entered as \verb|\sh|, not as \verb|sh| since the first is a character command processed by \LaTeXe\ while the latter is a ligature processed by Metafont. If a wrong hyphenation is spotted please check first whether the word in question contains ligature statements (\verb|sh|, \verb|ya|, \verb|yu| etc.) which should then replaced by the proper character commands. It is usually sufficient to add a leading back slash and include the entity in braces: \verb|{\ya}| is as good as \verb*|\ya |. Hyphenation patterns for Russian exist but are still to be re-encoded in LMC encoding; Buryat hyphenation rules are still to be defined. \subsection{Retransliteration Problems} Apart from being incomplete as far as some arcane writing variants are concerned, the retransliteration engine provided with \MonTeX\ has two serious shortcomings. Firstly, the input can only consist of letters, punctuation marks and numbers. Any \TeX\ or \LaTeX\ command (including \verb|\"a| for \emph{\"a} and friends) makes the retransliteration engine fail. Secondly, for large quantities of text, the retransliteration system is agonizingly slow. \subsection{Missing Caption Definitions} The translated captions provided with \MonTeX\ are not completely translated at the moment. Notably \tttt{ccname} and \tttt{headtoname} are missing in Mongolian and Buryat, mainly due to grammatical reasons. This will be fixed in later versions. \subsection{Page Headers and Text Encodings} In rare cases it is possible that a \verb|\section| text which is supposed to be typeset in Cyrillic letters is output in Latin letters. This happens if the text on that very page contains encoding selection commands which happen to fall near the page boundary. The only remedy is to enclose the argument text in an additional \verb|{\mnr ...}| (or vice versa for Latin) command. \subsection{The \texttt{kminch} Font} The Cyrillic typefaces of \MonTeX\ are completed by inch-high sans serif capital letters good for book titles etc. Unfortunately, it cannot be used orthogonally with the other fonts in \Tone\ (Latin characters) and \LMC\ (Cyrillic characters) encoding since its definition is based on \TeX\ primitives rather than \LaTeXe's NFSS font selection scheme. \subsection{The \emph{gamma} and its Shape} The original \TeX\ environment only provides an italic gamma which is intended for the math mode. This gamma is used for transliterating Mongolian; it does not perfectly blend into roman text. The \texttt{tipa} provides a gamma which goes better with roman text. If \texttt{tipa} is installed on your system then uncomment lines 18 and 42 of this file. \begin{quote} \begin{verbatim} \documentclass[11pt,a4paper]{article} %\usepackage[T3,LMS,LMC,T1]{fontenc} % <- Uncomment these lines \usepackage[latin1]{mls} % if you use the TIPA %\usepackage[noenc]{tipa} % <- package. Check line 42 [...] % % You obtain a nicer `gamma' by uncommenting the following line % but if you do so make sure the TIPA package gets included. % You should then uncomment line no. 18 of this file. % %\renewcommand{\g}{\textipa{\textgamma}} \end{verbatim} \end{quote} \subsection{Error Message \texttt{option clash}} The interaction of \MonTeX\ with other language support packages for \LaTeXe\ has not been thoroughly tested and problems may well arise from using conflicting commands, character presentations etc. One known constellation causing concern is the \texttt{tipa} package. This package calls \texttt{fontenc} with its own options producing an option clash. In this case follow the instructions given in the \texttt{tipa} manual and start the \texttt{tipa} package with the \texttt{noenc} option. The following excerpt from a document preamble is known to work: \begin{quote} \begin{verbatim} \usepackage[T3,LMS,LMC,T1]{fontenc} \usepackage[latin1]{mls} \usepackage[noenc]{tipa} \end{verbatim} \end{quote} \subsection{Error Message \tttt{selectlanguage}} There seem to be differences in the nature of installed \LaTeXe\ platforms; emtex shows a behaviour different from teTeX with regard to pre-loaded language options. On teTeX systems, no error message concerning the redefinition of the \texttt{selectlanguage} command occurs, on emtex systems such a message may occur if no other language support packages are loaded. This error message can be safely ignored but the author hopes to find a solution later. \subsection{Printer Memory Overflow} Depending on the printing system it may happen that a Printer Memory Overflow message is generated when attempting to print this text. So far, this happened only on emtex systems running on plain \textsf{DOS}. This is an exceptional situation caused by the very high number of fonts used for this document. The error message never occurred on systems with PostScript postprocessing of the DVI file. It is very simple and straightforward to reduce the number of typeface examples of this document. Near the beginning of the source file of this very text, the lines \begin{quote} \begin{verbatim} % If emtex goofs with (printer) memory overflow % when attempting to print this document then % set the following number to "1", recompile and % increase the number step by step until all % examples are printed. The maximum value is 6. \newcounter{FontSamples} \setcounter{FontSamples}{6} % <--- Modify this number! \end{verbatim} \end{quote} can be found. It is now possible to increase the number of printed typeface samples step by step until either memory saturation is reached or the system manages to print all fonts. In addition, it should be noted that printing this documentation for the first time may take some time until all fonts are computed. \subsection{Error Reports} Time is a most precious resource and one of the main reasons why the authors decided \emph{not} to support other environments than \LaTeXe. If \MonTeX\ does not work for you because you use a \LaTeX 209 installation, or expect to find a working system for plain \TeX\ support, then the author cannot assist you beyond the advice to install the most recent version of \LaTeXe. If you find a bug or think a feature is missing which you'd like to see included then your comments are most welcome. One of the authors can be reached by e-mail (\texttt{corff@zedat.fu-berlin.de}), and available updates will appear in Infosystem Mongolei (\texttt{http://userpage.fu-berlin.de/\~{}corff}). Please check the MLS directory for available releases and patches. \vspace{1cm} \hfill\parbox{6cm}{\kit{Migj"ad Janra"isig Burxny m"almi"i n"a"as"an o"in "olzi"it"a"i "od"or biqiw.}} \newpage \section{Commands in Alphabetical Order} All user level commands available in \MonTeX\ are given here in alphabetical order. Every entry in the following list has up to seven sections which are only present if necessary: \begin{description} \item [Synopsis] shows the usage of the command; \item [Function] states its purpose and function; \item [Limitations] in functional range, allowed input etc. are stated here; \item [Comments] and additional information about purpose and nature of the command; \item [Related commands] in the command reference; \item [See page] of the main text; \item [Example] shows a typical application. If several related commands have the same usage and command syntax, then only one example is given which is typically found at the first place a command is mentioned. One example is the command for numbering by letters: The commands \verb|\Asbuk|, \verb|\Useg| and \verb|Uzeg| are similar, and an example is only given under \verb|\Asbuk|. \end{description} \input{comref.tex} \end{document}