\documentclass[12pt,a4paper]{article} %\newif\ifpdf\ifx\pdfoutput\undefined\pdffalse\else\pdftrue\fi %\usepackage{german} \usepackage{times} %\usepackage{mathptm} %\usepackage[T1]{fontenc} %\usepackage{covington} %\usepackage{multicol} %\usepackage{fancyhdr} %\usepackage{graphicx} \usepackage{url} \usepackage{alltt} \usepackage{rotating} \usepackage{array} \usepackage{maketitel} \usepackage{serto} \usepackage{assyr} \usepackage{longtable} \usepackage{chicago2} %\ifpdf \usepackage[pdftex,colorlinks=true, urlcolor=urlcol, % URL: http:// citecolor=bibcol, % bibTeX linkcolor=linkcol, % sections, footnotes, ... pagecolor=linkcol, % links to other pages filecolor=filecol, % URL: file:/ pdftitle={Serto - A font for Syriac (Aramaic)}, pdfauthor={Johannes Heinecke}, pdfsubject={Oriental alphabets, Semitic languages},pdfkeywords={},pagebackref, pdfpagemode=None,bookmarksopen=true]{hyperref} \usepackage{color} \definecolor{linkcol}{rgb}{0.75,0,0} \definecolor{bibcol}{rgb}{0,0.5,0} \definecolor{urlcol}{rgb}{0,0,0.75} \definecolor{filecol}{rgb}{0,0,0.75} %\else %\fi %\addtolength{\topmargin}{-15mm} %\addtolength{\textheight}{30mm} %\addtolength{\textwidth}{0mm} %\addtolength{\oddsidemargin}{0mm} %\addtolength{\evensidemargin}{0mm} % EDIT sertodoc.ptex % serto.py sertodoc.ptex > sertodoc.tex % pdflatex sertodoc.tex \usepackage{parskip} %\let\S=S % for verbatim %\let\T=T % for verbatim \def\st{Ser\d to} \def\texxet{{\tt TeX-{}-XeT}} \def\pdflatex{{\tt pdflatex}} \def\SertoVersion{1.0} \newcommand{\showS}[1]{{\tt #1}} \newcommand{\showT}[1]{{\tt <{}T>#1}} \newcommand{\showST}[1]{{\tt #1}} \newcommand{\showC}[1]{{\tt #1}} \newcommand{\showCT}[1]{{\tt #1}} \newcommand{\SE}{{\tt }} \newcommand{\eSE}{{\tt }} \newcommand{\TR}{{\tt }} \newcommand{\eTR}{{\tt }} \newcommand{\ST}{{\tt }} \newcommand{\eST}{{\tt }} \font\logo=logo10 scaled \magstep1 \def\MF{{\logo METAFONT}} \begin{document} \title{ser.t=a' \emph{\st} -- a font for Syriac (Aramaic)} \author{Johannes Heinecke} \date{Version \SertoVersion, 30\textsuperscript{th} March 2013} \maketitle \tableofcontents \section{Introduction} \emph{\st} is a form of the alphabet used for Aramaic (a western semitic language) which has been spoken in the Near East since at least 1100 BC. More precisely, \st is used for Syriac which is the variant of Aramaic spoken since the second century AD.\footnote{See \citeNP{ungnad1932}, \citeNP{brockelmann1960}, \citeNP{costaz1986} or \citeNP{noeldeke1986} (English translation \citeNP{noeldeke2001}) for further information on Syriac.} Syriac used other alphabets as well, notably \emph{Es\d trangelo}, which is not (yet) contained in this package. Since \st{} is as the Arabic alphabet a syllabic script, vowels are marked by diacritic marks above (or under) the consonantic letters. Modern forms of Aramaic still use either \st{}, the Chaldean alphabet or Es\d trangelo. Since Syriac split up in two main dialects in the fifth century AD, two differing systems of vowel-marking were established: Whereas the western dialect (Edessean) used Greek letters as vowel symbols (these are the only ones provided by this package for the time being), the eastern dialects uses dots to indicate the vowels (Chaldean vowels). This package enables you to typeset words or paragraphs in \st{} using a preprocessor which chooses the correct letter form depending on context. In order to typeset paragraphs the use of a recent version of \pdflatex\ is needed, which can handle the right-to-left typesetting. For older version of \LaTeX, the preprocessor must be used with the option {\tt -o} (see section \ref{prep}). This package also included an adapted version of a Chaldean font (thanks to Tony Khoshaba, who put this font to the Web). %qa.t.tel qa.tel % ^sapiyr biy^s l=ak . \section{The alphabets} Every letter in \st{} (and some letters in the Chaldean alphabet) has several forms, depending on its position in the word: An initial, medial or final form. Since some letters do not connect to the following letter, there are isolated forms as well (i.e. a letter which is not connected to the right nor the left). The coding column in the following table refers to the preprocessor described below (section \ref{prep}). If you do not want to use the preprocessor, please refer to the encoding table in section \ref{enc}. \subsection{Consonants}\label{consonants} %\begin{tabular}{| >{\rule[-2mm]{0pt}{9mm}}c|c|c|c| r >{\it}l | >{\tt}c|} \begin{longtable}{| >{\rule[-2mm]{0pt}{9mm}}c|c|c|c| c | r l | >{\it}c | >{\tt}c | } \hline \multicolumn{4}{|c|}{ser\d to forms} & & \multicolumn{2}{c|}{name} & \rm translit. & \rm coding \\ \cline{1-4} %\hline \begin{turn}{90}isolated\end{turn} & \begin{turn}{90}final\end{turn} & \begin{turn}{90}medial\end{turn} & \begin{turn}{90}initial\end{turn} & \begin{turn}{90}Chaldean\end{turn} & & & & \endhead \hline ' & -' & & & ' & '=alap+ & '=alap+ & \raise.2ex\hbox{$\rhook$} & '\\ \hline b & -b & -b- & b- & b & beyt+ & beyt+ & b\footnotemark & b \\ & & & & v\footnotemark & & & v & v \\ \hline g & -g & -g- & g- &g & g=amal & g=amal & g & g\\ & & & & .g & & & \d g & g \\ & & & & j & & & j & j \\ \hline d & -d & & &d & d=alat+ & d=alat+ & d & d\\ \hline h & -h & & &h & he' & he' & h & h\\ \hline w & -w & & &w & waw & waw & w & w \\ \hline z & -z & & &z & zayn & zayn & z & z \\ \hline .h & -.h & -.h- & .h- &.h & .heyt+ & .heyt+ & \d h & .h\\ \hline .t & -.t & -.t- & .t- &.t & .teyt+ & .teyt+ & \d t & .t\\ \hline y & -y & -y- & y- &y & y=ad+ & y=ad+ & y & y \\ \hline k & -k & -k- & k- &k & k=ap+ & k=ap+ & k & k\\ & & & & _k & & & \b h & \verb._k. \\ & & & & ^k & & & \v c & \verb.^k. \\ \hline l & -l & -l- & l- & l & l=amad+ & l=amad+ & l & l\\ \hline m & -m & -m- & m- & m & miym & miym & m & m\\ \hline n & -n & -n- & n- &n & nuwn & nuwn & n & n\\ \hline s & -s & -s- & s- &s & semkat+ & semkat+ & s & s\\ \hline ` & -` & -`- & `- &` & `e' & `e' & \raise.2ex\hbox{$\lhook$} & ` \\ \hline p & -p & -p- & p- &p & pe' & pe' & p & p \\ & & & & f & & & f & f \\ \hline .s & -.s & & &.s & .s=ade' & .s=ade' & \d s & .s\\ \hline q & -q & -q- & q- &q & q=ap+ & q=ap+ & q & q \\ \hline %\newpage %\hline r & -r & -r- & r- &r & riy^s & riy^s & r & r \\ %\hline R & -R & -R- & R- &R & & & & R\\ \hline ^s & -^s & -^s- & ^s- &^s & ^siyn & ^siyn & \v s & \verb.^s.\\ \hline t & -t & & &t & taw & taw & t & t\\ & & & & _t & & & \b{t} & \verb._t. \\ \hline \end{longtable} %\end{tabular} \addtocounter{footnote}{-1} \footnotetext{The \emph{be\=ga\b dke\=pa\b t} are not yet always processed. In general, the doubling of the consonant creates a qu^s^s=ay=a' in the syriac text and does not change the transliteration. On the other hand, a consonant followed by {\tt+} will receive a ruwkk=ak+=a' and in the transliteration \emph{bgdkft} will appear as \emph{\b b\=g\b d\b h\=p\b t}.} \stepcounter{footnote} \footnotetext{Modern Aramaic dialects using the Chaldean alphabet have diacritic symbols (dots and tildes) which can be typeset directly.} \subsection{Vowels}\label{vowels} The package allows to typeset the greek vowels or Chaldean vowels symbols. To have the vowel symbol written in inversed form under the consonant, user upper case input. \begin{quote} {\bf ATTENTION:} I had to change the coding of zq=ap+=a' (_=a) from {\tt o} to \verb.=a. in order to accomodate the new letter _o (coding {\tt o}). In texts encoded up to version 0.4 {\tt o} has to be replaced by \verb.=a. Sorry for any inconveniances. \end{quote} \begin{tabular}[t]{| >{\rule[-3mm]{0pt}{11mm}}c | c | r l | c | >{\tt}c |} \hline Greek & Chaldean & \multicolumn{2}{c|}{name} & \rm transliteration & \rm coding \\ %form & form & & & & \\ \hline _a & _:a & pt+=a.h=a' & pt+=a.h=a' & a & a \\ \hline _e & _:e & rb+=a.s=a' & rb+=a.s=a' & e & e \\ \hline _i & _:i & .hb+=a.s=a' & .hb+=a.s=a' & i & i \\ \hline _=a & _:=a & zq=ap+=a' & zq=ap+=a' & =a & =a \\ \hline _u & _:u & `.s=a.s=a' & `.s=a.s=a' & u & u \\ \hline _o & _:o & & & =a & o \\ \hline _P & & sy=a--Pme' & sy=a--Pme' & & P \\ \hline \end{tabular} ~~~~~~ \begin{tabular}[t]{| >{\rule[-4mm]{0pt}{11mm}}c | >{\tt}c |} \hline Greek & \rm coding \\ %form & \\ \hline _A & A \\ \hline _E & E \\ \hline _I & I \\ \hline _=A & =A \\ \hline _U & U \\ \hline \end{tabular} Note: The .hb+=a.s=a' and `.s=a.s=a' of the Eastern or Chaldean vowels do in general occur together with a \emph{mater lectionis}: l:iy, l:ow (or l:iy, l:ow) The sy=a--Pme' is processed as a vowel sign, even if it is not so from a linguistic point of view. Its coding \showS{P} is chose because of its plural signification. If you do not want it over a letter, put it over a word stretch: The Aramaic at the beginning of this paragraph has been typset in the following: \showS{sy=a--Pme'}. To avoid that the simple preprocessor does not mess about with vowels in ligatures (notably \emph{L\=ama\b d-\=alaf} and \emph{\=Alaf-l\=ama\b d} the vowel must be set after both consonants of the ligature: \showS{l=a'} yields (incorrect) l=a' but \showS{l'=a} yields the vocalized ligature l'=a \showS{'Al=Ah=a'} yields (incorrect) 'Al=Ah=a' but \showS{'la=Ah=a'} yields the vocalized ligature 'la=Ah=a' The Chaldean letters do not have this ligature. Instead, a taw-=al=af ligature is provided: sipr=ayuwt'a The default vowels are the greek-based vowels. In order to get Chaldean vowels, it suffices to add {\tt:} in front of the vowel in coding. Thus you can set the most famous Aramaic phrase in all Syriac alphabets in either vowel system: \showS{eliy eliy lm=an=a' s=ab=akt=aniy}\\ eliy eliy lm=an=a' s=ab=akt=aniy \showS{:el:iy :el:iy lm:=an:=a' s:=ab:=akt:=an:iy}\\ :el:iy :el:iy lm:=an:=a' s:=ab:=akt:=an:iy \showC{eliy eliy lm=an=a' s=ab=akt=aniy}\\ eliy eliy lm=an=a' s=ab=akt=aniy \showC{:el:iy :el:iy lm:=an:=a' s:=ab:=akt:=an:iy}\\ \hfill :el:iy ':el:iy lm:=an:=a' s:=ab:=akt:=an:iy \subsection{Transliteration and long vowels} As mention in section \ref{prep} below, the preprocessor can produce a transliteration as well. The transliteration can be defined in the table used by the preprocessor {\tt serto.font} and {\tt assyr.font}. The current definition uses the transcription as shown in the tables in sections \ref{consonants} and \ref{vowels}, with the exception of long vowels. In words using a vowel symbol together with a \emph{mater lectionis}, the transliteration shows the transcription instead, for instance .heyt+ transliterates as .heyt+ and not \emph{\d hey\b t} and q=aye'm: \begin{tabular}{| >{\rule[-4mm]{0pt}{11mm}}r | r | c | >{\tt}c |} \hline Greek & Chaldean & \rm transliteration & \rm coding \\ \hline --=a' & --:=a' & _=a' & =a' \\ \hline --ey & --:ey & _ey & ey \\ \hline --e' & --:e' & _e' & e' \\ \hline --iy & --:iy & _iy & iy \\ \hline --uw & --:uw & _uw & uw \\ \hline \end{tabular} \subsection{Punctuation and paragraph marks} \begin{tabular}{| >{\rule[-4mm]{0pt}{11mm}}c | >{\tt}c |} \hline form & \rm coding \\ \hline . & . \\ \hline : & : \\ \hline .X. & .X. \\ \hline .:. & .:. \\ \hline \end{tabular} ~~~~~~~ \begin{tabular}{| >{\rule[-4mm]{0pt}{11mm}}c | >{\tt}c |} \hline form & \rm coding \\ \hline :: & :: \\ \hline :- & :- \\ \hline .~. & \verb:.~.: \\ \hline ./. & ./. \\ \hline \end{tabular} \subsection{Unicode} This package has a limited Unicode support in that texts encoded in UTF8 can be directly typeset. In order to activate the UTF8 interpretation, either use UTF8 for your whole document by declaring \verb+\usepackage[utf8]+\verb+{inputenc}+ in the preamble of you document, or just put \verb+%\usepackage[utf8]+\verb+{inputenc}+ somewhere at the beginning of your document. \subsection{The encoding}\label{enc} \def\cell#1{\char'#1} \def\cellrow#1{ & \cell{#10} & \cell{#11} & \cell{#12} & \cell{#13} & \cell{#14} & \cell{#15} & \cell{#16} & \cell{#17} \\ \hline} \def\reihe#1{\it '#10 \cellrow{#1}} \def\xcell#1{\char"#1} \def\xcellrow#1{ & \xcell{#10} & \xcell{#11} & \xcell{#12} & \xcell{#13} & \xcell{#14} & \xcell{#15} & \xcell{#16} & \xcell{#17} & \xcell{#18} & \xcell{#19} & \xcell{#1A} & \xcell{#1B} & \xcell{#1C} & \xcell{#1D} & \xcell{#1E} & \xcell{#1F} \\ \hline} \def\xreihe#1{\it "#1n \xcellrow{#1}} The following table shows the internal encoding of the defined letters of \st\ and the Chaldean variant. \newif\ifhextabelle \hextabelletrue %\pdfoutput\undefined\pdffalse\else\pdftrue\fi \ifhextabelle {\serto \begin{tabular}{ r || c | c | c | c | c | c | c | c | c | c | c | c | c | c | c | c | } & \it 0 & \it 1 & \it 2 & \it 3 & \it 4 & \it 5 & \it 6 & \it 7 & \it 8 & \it 9 & \it A & \it B & \it C & \it D & \it E & \it F \\ \hline\hline \xreihe{0} \xreihe{1} \xreihe{2} \xreihe{3} \xreihe{4} \xreihe{5} \xreihe{6} \xreihe{7} \xreihe{8} \xreihe{9} \xreihe{A} %\xreihe{B} %\xreihe{C} %\xreihe{D} %\xreihe{E} %\xreihe{F} \end{tabular}} {\assyr \begin{tabular}{ r || c | c | c | c | c | c | c | c | c | c | c | c | c | c | c | c | } & \it 0 & \it 1 & \it 2 & \it 3 & \it 4 & \it 5 & \it 6 & \it 7 & \it 8 & \it 9 & \it A & \it B & \it C & \it D & \it E & \it F \\ \hline\hline \xreihe{0} \xreihe{1} \xreihe{2} \xreihe{3} \xreihe{4} \xreihe{5} \xreihe{6} \xreihe{7} \xreihe{8} \xreihe{9} \xreihe{A} %\xreihe{B} %\xreihe{C} %\xreihe{D} %\xreihe{E} %\xreihe{F} \end{tabular}} \else {\serto \begin{tabular}{ r || c | c | c | c | c | c | c | c | } & \it 0 & \it 1 & \it 2 & \it 3 & \it 4 & \it 5 & \it 6 & \it 7 \\ \hline\hline \reihe{00} \reihe{01} \reihe{02}\reihe{03} \reihe{04} \reihe{05}%\reihe{06} \reihe{07} \reihe{10}\reihe{11} \reihe{12}\reihe{13}%\reihe{14} \reihe{15}\reihe{16} \reihe{17} \reihe{20} \reihe{21} \reihe{22} \reihe{23} \reihe{24} \reihe{25} \end{tabular}} {\assyr \begin{tabular}{ r || c | c | c | c | c | c | c | c | } & \it 0 & \it 1 & \it 2 & \it 3 & \it 4 & \it 5 & \it 6 & \it 7 \\ \hline\hline \reihe{00} \reihe{01} \reihe{02} \reihe{03} \reihe{04} \reihe{05} %\reihe{06} \reihe{07} \reihe{10} \reihe{11} \reihe{12} \reihe{13} \reihe{14} \reihe{15} \reihe{16} \reihe{17} %\reihe{20} %\reihe{21} %\reihe{22} %\reihe{23} %\reihe{24} %\reihe{25} \end{tabular}} \fi \urlstyle{same} \section{The preprocessor {\tt serto.py}}\label{prep} Typesetting of texts is still not yet possible with standard \LaTeX, since the right-to-left typesetting (as for instance as in Arab\TeX) has not yet been implemented. For the time being I propose a preprocessor (written in {\sc Python}\footnote{Every version from 2.4 onwards up to version 2.7 should do; Python 3.* however does not work {\sc Python} is an interpreted object oriented programming language available for many operating systems (\url{http://www.python.org}).}) and \pdflatex. I'm well aware that {\tt serto.py} is not part of the most beautiful pieces of software code, on the contrary, it's rather spaghetti code. Many things could have been in a more intelligent way, but it works, which is the most important thing. If you find the time to improve it please share your changes with me! \subsection{Using the preprocessor} The preprocessor is called with the \LaTeX-file as argument: \verb| serto.py [-o] ppfilename.tex > filename.tex| The resulting \LaTeX-file can be \LaTeX ed as usually. Please make sure to have the \verb.\usepackage{serto}. included in your preamble. The option {\tt -o} is necessary if you use an older version of \LaTeX\ which is not capable to typeset texts from the right to the left (\texxet extension). The {\tt -o} -option tells the preprocessor to inverse the letters on its own. In order to typeset whole paragraphs \pdflatex\ is the better solution. Usually it comes with every modern \TeX-distribution. At least on Ubuntu 12.04 (texlive package), the standardly installed \pdflatex\ behaves correctly. The preprocessor recognizes two types of commands. Within a single line you can put Syriac words between \SE\ and \eSE: For example \showS{ser.t=a'} becomes ser.t=a'. \ST\ and \eST\ generate the enclosed part in \st\ and generates a transliteration as well (\showST{mdiyt=a'} becomes mdiyt=A' ``city''), whereas \TR\ and \eTR\ can be used for parts only need in transliterated form (\showT{ser.t=a'} becomes ser.t=a'). Since in transliteration a ``neutral vowel'' is needed, which does not appear in \st, the code \verb.@. can be used: \showST{\^{}s\symbol{64}m=a`} produces ^s@m=a` For multiple lines, start a block using {\tt }\footnote{Using {\tt } or {\tt } implies the using of the {\tt -e} option of the preprocessor {\tt serto.py}. If you do not use the {\tt -e} option the resulting \LaTeX-file is likely to be syntactically incorrect.} in a line on its own. This block can be closed by {\tt }. %To have a block in \st\ and in transliterated form, use {\tt %}. If you need transliterated Syriac, use {\tt } and {\tt }. The commands {\tt }/{\tt } and {\tt }/{\tt } do not work properly with the {\tt -o} option of the preprocessor and and older \LaTeX. If you add \TeX-commands in these blocks, a right-to-left typesetting version of \LaTeX\ is obligatory. For the time being the preprocessor tries to set the hard sign qu^s^s=ay=a' %\emph{qu\v s\v s\=oy\=o} on top of a consonant if the consonant is doubled in the input: \showS{q.tl} yields q.tl but \showS{q.t.tl} yields q.t.tl In cases where you need a qu^s^s=ay=a' %\emph{qu\v s\v s\=oy\=o} without wanting to double the consonant, a {\tt *} can be used after the letter to typeset a dot above a letter: \showS{h*=anon} produces h*=anon and \showS{\^{}sl=amk+on} results in ^sl=amk+on %wadd*=ame' To avoid a qu^s^s=ay=a' (when you need to adjacent identical consontants, either use a vowel on the first, use the stretching symbol: \showS{maml'e} yields maml'e\\ {\tt m\verb.^.ml'} yields m^ml'\\ \showC{maml'e} yields maml'e\\ {\tt m\verb.^.ml'} yields m^ml'\\ To get the soft sign ruwkk=ak+=a' %\emph{rukk\=o\b h\=o} a {\tt +} must \emph{follow} the letter: \showS{'ab+d=a'} yields 'ab+d=a' An \emph{\=alaf} is autmatically prefixed before an initial vowel: \showS{etqa.tel} and \showS{'etqa.tel} both yield etqa.tel Sometimes the letter \emph{r\=\i\v s} is written with two points. To achieve this, use {\tt R} instead of {\tt r} in the input: \showS{\^{}sapiyRe'} yields ^sapiyRe' Silent consonants have a bar mb+a.tl=an=a' %\emph{mva\d tl\=on\=o} (\emph{linea occultans}) under the line which is produced by {\tt =} just before the consonant (attention {\tt =a}, however, yields =a: \showS{'an=tt} yields 'an=tt In order to have the \emph{linea occultans} on top of the letter, use {\tt ==}: \showS{h==wiyt} yields h==wiyt There is no automatic stretching yet, but the \verb.--. can be used to insert a ``manual stretch'': \showS{napiyqt=a'} becomes napiyqt=a' but \showS{na-{}-piyq-{}-t=a'} is printed as na--piyq--t=a' This works also for the Chaldean letters \showC{n:ap:iyqt:=a'} becomes n:ap:iyqt:=a' but \showC{n:a-{}-p:iyq-{}-t:=a'} is printed as n:a--p:iyq--t:=a' If you do not use the preprocessor, you can activate \st\ by the command \verb.\serto.. In this case you have to choose the correct letters yourself, and use the commands \verb.\upperserto{.\emph{vowelnumber}\verb.}{.\emph{letter}\verb.}. or \verb.\lowerserto{.\emph{vowelnumber}\verb.}{.\emph{letter}\verb.}. to set vowels. Please see the encoding table in section \ref{enc} for the correct vowel numbers. In order to get bold letters, you can use \verb.\sertob. with or without preprocessor (see section \ref{prep} for more information on the preprocessor. \section{The format of the {\tt*.font} files} These files are necessary to tell the preprocessor where (in the font) a certain letter is found, and whether it has different forms. The format is straight forward, with, however, a few idiosynchrasies. In general there are two sections, the first (starting with a line \verb.#FONT.) indicates which letter has which form in which position and a second (\verb.#TRANS.) to define the transcription. The first part consists of lines like the following\\ \verb.b beth 66+124 66 66 66+124 1.\\ which reads, coding \verb.b. is for the letter \emph{beth}, its isolated form is character 66 followed by character 124, its initial and medial form is character 66, it's final form is character 66 followed by character 124 and the next letter (if any) must take it's medial form. However, if one of the four last columns has a value of $-1$, no form is provided in the font. Values from 0 to 15 are reserved for accents/vowels above the line, values from 16 to 31 are reserved for vowels under the line. The lines \verb. ~ blank 32 32 32 32 0.\\ \verb. Q shadda 6 6 6 6 2.\\ \verb. -- stretch 45 45 45 45 1.\\ must not be deleted. The coding for digits (starting with 0 up to 9 in the first column cannot have the \verb.+. in the position definition The transcription definition defines for each coding symbol (defined in the font-section) a valid \TeX-string to be used if transcription is needed, e.g. \verb- .t \d{t}- i.e. the coding \verb-.t- will be representend by \emph{\d{t}} in transcribed portions. \section{Two examples} Some of the following can be found in {\tt example.ptex} which comes with this package. %If you do not have \elatex, you cannot set texts in \st\ but only %single words. The following input was used to generate the output below: \begin{quote}\small \begin{verbatim} \documentclass[12pt,a4paper]{article} usepackage{serto} \begin{document} \end{verbatim} \begin{alltt} men qadiy^se' ho' tetqada^s kmo' dat+basb@suwn pagdo' \end{alltt} \begin{verbatim} \end{document} \end{verbatim} \end{quote} \begin{quote} men qadiy^se' h=a' tetqada^s km=a' dat+basb@suwn pagd=a' \end{quote} %In order to set more than just words %\elatex\ (needed for the \texxet\ extension) must be used. %\elatex\ properly sets text from the right to the left. \begin{quote} wk=ad .hzaw kuwmd'e da.h.tIyt=A': medem da`bad na.siy.h=a': lAbkuwh=y= wa'sruwh=y= wa'ytIywh=y= qd=am malk=a'. we'^stA`iyw lmalk=a' me----dem da`bad {\sertob l'Al=Ahayhwn}. m=alk=a' deyn 'ema----r. l'=a hw=a' me^stAwd=ay=aw 'e^stEwdiyt lIy dtEdba.h l'Ap=aluw: na.siy.h=a' 'emar. liy 'la=Ahe' 'ayleyn dma't`eyn. l'=a hw=a' 'la=Ahe' 'Enuwn. qareb 'Enuwn lIy lh=ark=a'. \sertob wl'=A ^suwbh=adhuwn dakiys.ty=ane' l'=a ^s=abeq '=n=a' .had menhuwn dl'=a mdaqeq '=n=a' `adm=a' lram^s=a'. \end{quote} The preceding text was set typeset with the following input: \begin{quote}\small \begin{verbatim} \documentclass[12pt,a4paper]{article} \usepackage{serto} \begin{document} \end{verbatim} \verb.. \begin{verbatim} wk=ad .hzaw kuwmd'e da.h.tIyt=A' : medem da`bad na.siy.h=a': lAbkuwh=y= wa'sruwh=y= wa'ytIywh=y= qd=am malk=a'. we'^stA`iyw lmalk=a' me----dem da`bad {\sertob l'Al=Ahayhwn}. m=alk=a' deyn 'emar. l'=a hw=a' me^stAwd=ay=aw 'e^stEwdiyt lIy dtEdba.h l'Ap=aluw: na.siy.h=a' 'emar. liy 'la=Ahe' 'ayleyn dma't`eyn. l'=a hw=a' 'la=Ahe' 'Enuwn. qareb 'Enuwn lIy lh=ark=a'. \sertob wl'=A ^suwbh=adhuwn dakiys.ty=ane' l'=a ^s=abeq '=n=a' .had menhuwn dl'=a mdaqeq '=n=a' `adm=a' lram^s=a'. \end{verbatim} \verb.. \verb.\end{document}. \end{quote} Replacing {\tt } by {\tt } results in this \begin{quote} wk=ad .hzaw kuwmd'e da.h.tIyt=A': medem da`bad na.siy.h=a': lAbkuwh=y= wa'sruwh=y= wa'ytIywh=y= qd=am malk=a'. we'^stA`iyw lmalk=a' me----dem da`bad {\assyrb l'Al=Ahayhwn}. m=alk=a' deyn 'ema----r. l'=a hw=a' me^stAwd=ay=aw 'e^stEwdiyt lIy dtEdba.h l'Ap=aluw: na.siy.h=a' 'emar. liy 'la=Ahe' 'ayleyn dma't`eyn. l'=a hw=a' 'la=Ahe' 'Enuwn. qareb 'Enuwn lIy lh=ark=a'. \assyrb wl'=A ^suwbh=adhuwn dakiys.ty=ane' l'=a ^s=abeq '=n=a' .had menhuwn dl'=a mdaqeq '=n=a' `adm=a' lram^s=a'. \end{quote} The file was preprocessed using {\tt serto.py}. \section{Things still missing} The todo-list is long. I try to add features etc. as soon as possible. Please tell me items you would like to have, but which are not yet on this list. Any volunteers are welcome! \begin{itemize} \item proper typsetting of texts (without preprocessor, maybe in the Arab\TeX\ package) \item proper treatment of \emph{matr\=es lectionis} (long vowels) %\item treatment of the sy=ame' %\emph{sey\=om\=e} \item proper treatment of the silent bar mb+a.tl=an=a' %\emph{mva\d tl\=on\=o} \item proper treatment of \emph{be\=ga\b dke\=pa\b t} with hard sign qu^s^s=ay=a' %\emph{qu\v s\v s\=oy\=o} and soft sign ruwkk=ak+=a' %\emph{rukk\=o\b h\=o} \item automatic transcription mode \item interpunction \item proper dealing with ligatures %\item Chaldean vowels %\item unicode input \item 'es.trangel=a' %and Chaldean glyphs \item numbers %\item support for typical encodings and for unicode (UTF-8) either via %%the preprocessor or via the \verb:inputenc.sty: package. %\item creating Type 1 fonts from the \MF-sources \item support for existing Syriac fonts \end{itemize} \section{Installation} The easiest way to install the fonts and the preprocessor is by installing the debian package (this includes only the {\tt pfb}, {\tt tfm} and {\tt afm} files for the fonts), the needed styles and the preprocessor, but not the \MF{} sources: {\verb+ +\tt sudo dpkg -i serto-\SertoVersion.deb} If you are not on a Debian or Ubuntu plateform, you need to install manually from the {\tt .tgz} file: \subsection{Using Metafont sources} Put the {\tt *.mf} files into a subdirectory {\tt serto} of your metafont branch in your {\tt texmf}-directory. For example using the texlive distribution under Linux, you should put them into {\tt\url{/usr/local/share/texmf/fonts/source/serto/}}. Do not forget to call {\tt texhash} in order to make the tex software find the newly installed fonts. \subsection{Using vector fonts} Using vector fonts depends a little from the \TeX-installation used, the following is tested for Ubuntu 12.04 and 10.04, it will probably work on Debian platforms as well, or other platforms using the texlive installation. \begin{itemize} \item copy {\tt syriac.map} to {\tt /usr/local/share/texmf/fonts/map/dvips/config/} \item copy {\tt *.afm} to {\tt /usr/local/share/texmf/fonts/afm/syriac/serto/} \item copy {\tt *.pfb} to {\tt /usr/local/share/texmf/fonts/type1/syriac/} \item add {\tt Map syriac.map} to {\tt /etc/texmf/updmap.d/10local.cfg} \item run {\tt sudo update-updmap} \item run {\tt sudo updmap-sys} \end{itemize} \subsection{Other files} The Stylefile etc. {\tt *.sty}, {\tt *.fd} go into a directory for stylefiles, e.g. \url{/usr/local/share/texmf/tex/latex/serto/}. The preprocessor {\tt serto.py} and the encoding file {\tt serto.font} somewhere where it can be found (e.g {\tt \url{/usr/local/bin}}). They must reside in the same directory unless you specify in the environment variable {\tt SERTOFONTDIR} the directory containing {\tt serto.font} and {\tt assyr.font}. Possibly you have to adjust the first line of the preprocessor \verb.#!/usr/bin/python. if your python interpreter is somewhere else. \section{License} This Material is subjec to the LaTeX Project Public Li­cense 1.3 (\url{http://ctan.org/license/lppl1.3}). \section{Changelog} \begin{itemize} \item Version 1.0 \begin{itemize} \item adding a character for the \emph{linea occultans} above the letter \item {\tt SERTOFONTDIR} enviromnment variable to specify the directory of {\tt *.font} files \item some UTF8 support \end{itemize} \item Version 0.7 \begin{itemize} \item Chaldean vowels \item Integration of the Chaldean font provided by Tony Khoshaba \item Major adjustements to the {\tt serto.py} preprocessor \end{itemize} \item Version 0.2, 0.3 and 0.4 \begin{itemize} \item can't remember, didn't keep track of changelog those days\dots \end{itemize} \item Version 0.1 \begin{itemize} \item Initial version \end{itemize} \end{itemize} \bibliographystyle{engl} \bibliography{buch} \end{document}