» Home Page
Nota sugli Accenti

Gli accenti e la lingua italiana

Oltre a non ricordare la differenza tra accento acuto e grave, accade che non si sappia come scrivere le lettere accentate (maiuscole o minuscole) con tastiere che non le riportano, specialmente quelle maiuscole. Succede quindi che si preferisca sostituire alle lettere accentate il digramma lettera+apice, magari anche per usare solo i caratteri ASCII (American Standard Code for Information Interchange) a 7 bit. Questo era accettabile prima della diffusione degli standard a 8 bit ma evitabile da un po' di anni, con un piccolo sforzo.

  1. Una pagina dettagliata su questo problema: Accademia della Crusca (Copia Cache)
  2. Una pagina sulle regole di ortografia: Accademia della Crusca (Copia Cache)
  3. Indicazioni generali sull'uso di accento grafico e apostrofo: Accademia della Crusca (Copia Cache)

LaTeX/TeX

In TeX e LaTeX gli accenti sono facili da inserire: per \`e ==> , \'e ==> , \`E ==> (dopo la compilazione). Purtroppo i correttori ortografici non sempre considerano perch\'e equivalente a perch, anche con dizionari italiani. pi facile scrivere e leggere un testo cos, piuttosto che uno cos\`\i (ci vuole \i per evitare il puntino della i): basta usare caratteri Western (ISO-8859-1), detti anche LATIN-1 invece che l'insieme ASCII standard. Bisogna per ricordarsi di richiamare il pacchetto latex opportuno:

\usepackage[latin1]{inputenc}

Per selezionare latin1 per VI (per esempio)

vim/gvim:

let &termencoding = &encoding 
set encoding=latin1
set fileencodings=latin1

anche:

LANG=en_US.ISO-8859-1

I possibili metodi di uso della tastiera sono i seguenti.

Dead keys

(copiato da Wikipedia) Dead keys are commonly used to generate letters with accents (diacritics), because that way one does not need one key for each possible combination of letter and accent, but only one key for each accent (the dead key) plus the usual letter keys.

For example, if a keyboard has a dead key "", the French character e accent aigu () can be generated by pressing first "", then "e".

This functionality resembles the way diacritics are put on letters in a typewriter. In such machines, when you type a diacritical mark, such as "", or "^", the carriage won't move unless you press space or type a letter below the mark, just like a dead key. Clearly, this has no restrictions, so you could accentuate an "f" for example.

In Microsoft Word and related programs, dead keys are typed using the Ctrl key with the punctuation mark that looks most like the accent.

Su XFree86 si possono definire i dead_keys con un comando tipo:
``keycode  20 = minus underscore dead_acute question dead_grave questiondown``
(per xmodmap)
oppure caricando una tabella dei caratteri diversa:
``setxkbmap us_intl``

Compose

(copiato da Wikipedia) On some computer systems, a compose key is a key which is designated to signal the software to interpret the next keystrokes as a combination in order to produce a character not found on the keyboard.

For example, typing compose, then a, then e may produce the AE ligature, . Typing compose, then e, then ' can yield an e with an acute accent, .

The compose key is known as "Multi_key" in the XFree86 environment.

Per averlo su XFree86 un modo : ``setxkbmap -option compose:rwin``

Particularly on modern systems which support customisable compose sequences and Unicode, the table shown is not complete. Given the vast number of sequences permissable and the vast number of characters desirable in Unicode, a complete table would be incredibly long.

Microsoft Windows does not use a compose key, but characters not found on the keyboard can be generated by holding down alt, and then typing the ASCII or unicode key code corresponding to the desired character.

SCIM/UIM

Per una vera internazionalizzazione (linux): Unicode (UTF-8) e i metodi SCIM/UIM: A mini guide to SCIM/UIM. Purtroppo il LaTeX/TeX non ancora immediatamente utilizzabile in Unicode, anche se una buona parte dei caratteri grosso modo funziona. Si sta aspettando l'estensione a 16 bit Lambda/Omega:

(copiato da http://omega.cse.unsw.edu.au/omega/omega.jsp?userId=370)

Omega is an extension of TeX that is aimed primarily at improving TeX's multilingual abilities.

When the TeX program was originally developed in the mid seventies by Donald Knuth it was mainly aimed at typeseting mathematical texts in the English language. Since then TeX has made inroads in broader and broader areas of scientific, literary and other scholarly activities in many countries all over the world. In 1991, Knuth froze TeX, mainly in the interest of stability. However, he allows the TeX code to be used as the basis for further developments, so long as the resulting system is distributed under a different name.

In Omega all characters and pointers into data-structures are 31-bit wide, instead of 8-bit, thereby eliminating many of the trivial limitations of TeX. Omega also allows multiple input and output character sets, and uses programmable filters to translate from one encoding to another, to perform contextual analysis, etc. Internally, Omega uses the universal Unicode/ISO-10646 character set. Omega also includes support for multiple writing directions.

These improvements not only make it a lot easier for TeX users to cope with multiple or complex languages, like Arabic, Indic, Khmer, Chinese, Japanese or Korean, in one document, but also form the basis for future developments in other areas, such as native color support and hypertext features.

The LaTeX format adapted to the special features of Omega is called Lambda.

The Omega 2.0 distribution is being prepared in time for the next TUG meeting, Xanthi, Greece, August 30 to September 3, 2004.

LaTeX and Unicode

(Copiato da stephan.walter.name)

It is possible to use languages with non-roman scripts like Japanese in LaTeX, without using language-specific packages like CJK (http://cjk.ffii.org/) or ArabTeX (http://www.informatik.uni-stuttgart.de/ifi/bs/research/arab_e.html).

You'll need three things: latex-ucs, ttfucs and a UTF-8 capable text editor. These are described here: latex-ucs

latex-ucs is a package by Dominique Unruh that enables UTF-8 as an input encoding. Download it as tgz (http://www.unruh.de/DniQ/latex/unicode/), deb (http://packages.debian.org/cgi-bin/search_packages.pl?searchon=names&version=all&exact=1&keywords=latex-ucs) or rpm (http://rpmfind.net//linux/RPM/suse/9.0/i386/suse/noarch/latex-ucs-20030605-22.noarch.html).

Once you installed it, use it as follows:

\usepackage{ucs}
\usepackage[utf8x]{inputenc}

Now you can write your LaTeX documents in UTF-8. ttfucs

While latex-ucs enables you to write in any script system contained in the Unicode standard, you may not have the correct fonts installed to actually display (and print) the glyphs. The easiest way to solve this is to install a TrueType font containing (almost) all Unicode characters. There are a number of free-of-charge fonts:

Segni diacritici per il greco antico

Un problema in pi si ha se si vogliono scrivere (in LaTeX, ottenendo un risultato ottimo) testi in greco antico; oltre che le lettere, il problema sono i segni diacritici: spirito aspro e dolce, accento acuto, grave e circonflesso, dieresi (medioevale) e iota sottoscritto. Un modo quello di inserire il greco con il package babel con l'opzione polutonikogreek:

\RequirePackage[italian,polutonikogreek]{babel}

Il testo greco pu essere scritto con i caratteri latini (senza Unicode), come argomento del comando \textgreek:

\textgreek{>Arqim'hdhs}

oppure dell'environment polutonikogreek (in cui si vedono gli esempi di tutti i segni sopra elencati):

\foreignlanguage{polutonikogreek}{
\begin{greektext}
E>is'i tines >en >epip'edw| kamp'ulai gramma`i peperasm'evai,
a<`i t~wn t`a p'erata >epizeugnuous~wn a>ut~wn e>ujei~wn >'htoi
<'olai >ep`i  t`a a>ut'a e>isin >`h o>ud`en >`eqousin >ep`i
t`a <'etera.
\end{greektext}
}

Soluzione migliore sarebbe di usare LaTeX+Unicode, magari con font migliori (come psgreek), con un metodo di inserimento italiano+greco.