%% This is part of the OpTeX project, see http://petr.olsak.net/optex
\_codedecl \langlist {Languages declaration <2022-10-11>} % preloaded in format
\_doc -----------------------------
\`\_preplang` `<lang-id> <LongName> <lang-tag> <hyph-tag> <lr-hyph>`
declares a new language. The parameters (separated by space) are
\begitems
* <lang-id>: language identifier. It should be derived from ISO 639-1 code
but additional letters can be eventually added because <lang-id> must be used
uniquely in the whole declaration list. The \^`\_preplang` macro creates
the language switch `\_<lang-id>lang` and defines also `\<lang-id>lang`
as a macro which expands to `\_<lang-id>lang`. For example, `\_preplang cs Czech ...`
creates `\_cslang` as the language switch and defines
`\def\cslang{\_cslang}`.
* <LongName>: full name of the language.
* <lang-tag>: language tag, which is used for setting language-dependent phrases
and sorting data. If a language have two or more hyphenation
patterns but a single phrases set, then we declare this language
more than once with the same <lang-tag> but different <lang-hyph>.
* <hyph-tag>: a part of the file name where the hyphenation patterns
are prepared in Unicode. The full file name is `hyph-<hyph-tag>.tex`.
If <hyph-tag> is `{}` then no hyphenation patterns are loaded.
* <lr-hyph>: two digits, they denote `\lefthyphenmin` and
`\righthyphenmin` values.
\enditems
\^`\_preplang` allocates a new internal number by \^`\newlanguage\_<lang-id>Patt`
which will be bound to the hyphenation patterns. But the patterns nor other language
data are not read at this moment. The `\_<lang-id>lang` is defined as \^`\_langinit`.
When the `\_<lang-id>lang` switch is used firstly in a document then the language is
initialized, i.e.\ hyphenation patterns and language-dependent data are read.
The `\_<lang-id>lang` is re-defined itself after such initialization.
\^`\_preplang` does also `\def\_ulan:<longname> {<lang-id>}`, this is needed
for the \^`\uselanguage` macro.
\_cod -----------------------------
\_def\_preplang #1 #2 #3 #4 #5#6{% lang-id LongName lang-tag hyph-tag lr-hyph
\_ifcsname _#1lang\_endcsname \_else
\_ea\_newlanguage\_csname _#1Patt\_endcsname
\_xdef\_langlist{\_langlist\_space#1(#2)}%
\_fi
\_lowercase{\_sxdef{_ulan:#2}}{#1}%
\_slet{_#1lang}{_relax}%
\_sxdef {#1lang}{\_cs{_#1lang}}%
\_sxdef {_#1lang}{\_noexpand\_langinit \_cs{_#1lang}#1(#2)#3[#4]#5#6}%
}
\_doc -----------------------------
The \^`\_preplang` macro adds `<lang-id>(<LongName>)` to the `\_langlist` macro
which is accessible by \`\langlist`. It can be used for reporting declared
languages.
\_cod -----------------------------
\_def\langlist{\_langlist}
\_def\_langlist{en(USEnglish)}
\_doc -----------------------------
All languages with hyphenation patterns provided by \TeX/live are
declared here. The language switches \`\cslang`, \`\sklang`,
\`\delang`, \`\pllang` and many others are declared.
You can declare more languages by \^`\_preplang` in your
document, if you want.\nl
The usage of \^`\_preplang` with <lang-id> already declared is allowed.
The language is re-declared in this case. This can be used in your document
before first usage of the `\<lang-id>lang` switch.
\_cod -----------------------------
% lang-id LongName lang-tag hyph-tag lr-hyph
\_preplang enus USenglishmax en en-us 23
% Europe:
\_preplang engb UKenglish en en-gb 23
\_preplang be Belarusian be be 22
\_preplang bg Bulgarian bg bg 22
\_preplang ca Catalan ca ca 22
\_preplang hr Croatian hr hr 22
\_preplang cs Czech cs cs 23
\_preplang da Danish da da 22
\_preplang nl Dutch nl nl 22
\_preplang et Estonian et et 23
\_preplang fi Finnish fi fi 22
\_preplang fis schoolFinnish fi fi-x-school 11
\_preplang fr French fr fr 22
\_preplang de nGerman de de-1996 22
\_preplang deo oldGerman de de-1901 22
\_preplang gsw swissGerman de de-ch-1901 22
\_preplang elm monoGreek el el-monoton 11
\_preplang elp Greek el el-polyton 11
\_preplang grc ancientGreek grc grc 11
\_preplang hu Hungarian hu hu 22
\_preplang is Icelandic is is 22
\_preplang ga Irish ga ga 23
\_preplang it Italian it it 22
\_preplang la Latin la la 22
\_preplang lac classicLatin la la-x-classic 22
\_preplang lal liturgicalLatin la la-x-liturgic 22
\_preplang lv Latvian lv lv 22
\_preplang lt Lithuanian lt lt 22
\_preplang mk Macedonian mk mk 22
\_preplang pl Polish pl pl 22
\_preplang pt Portuguese pt pt 23
\_preplang ro Romanian ro ro 22
\_preplang rm Romansh rm rm 22
\_preplang ru Russian ru ru 22
\_preplang srl Serbian sr-latn sh-latn 22
\_preplang src SerbianCyrl sr-cyrl sh-cyrl 22
\_preplang sk Slovak sk sk 23
\_preplang sl Slovenian sl sl 22
\_preplang es Spanish es es 22
\_preplang sv Swedish sv sv 22
\_preplang uk Ukrainian uk uk 22
\_preplang cy Welsh cy cy 23
% Others:
\_preplang af Afrikaans af af 12
\_preplang hy Armenian hy hy 12
\_preplang as Assamese as as 11
\_preplang eu Basque eu eu 22
\_preplang bn Bengali bn bn 11
\_preplang nb Bokmal nb nb 22
\_preplang cop Coptic cop cop 11
\_preplang cu churchslavonic cu cu 12
\_preplang eo Esperanto eo eo 22
\_preplang ethi Ethiopic ethi mul-ethi 11
\_preplang fur Friulan fur fur 22
\_preplang gl Galician gl gl 22
\_preplang ka Georgian ka ka 12
\_preplang gu Gujarati gu gu 11
\_preplang hi Hindi hi hi 11
\_preplang id Indonesian id id 22
\_preplang ia Interlingua ia ia 22
\_preplang kn Kannada kn kn 11
\_preplang kmr Kurmanji kmr kmr 22
\_preplang ml Malayalam ml ml 11
\_preplang mr Marathi mr mr 11
\_preplang mn Mongolian mn mn-cyrl 22
\_preplang nn Nynorsk nn nn 22
\_preplang oc Occitan oc oc 22
\_preplang or Oriya or or 11
\_preplang pi Pali pi pi 12
\_preplang pa Panjabi pa pa 11
\_preplang pms Piedmontese pms pms 22
\_preplang zh Pinyin zh zh-latn-pinyin 11
\_preplang sa Sanskrit sa sa 13
\_preplang ta Tamil ta ta 11
\_preplang te Telugu te te 11
\_preplang th Thai th th 23
\_preplang tr Turkish tr tr 22
\_preplang tk Turkmen tk tk 22
\_preplang hsb Uppersorbian hsb hsb 22
\_preplang he Hebrew he {} 00
\_doc -----------------------------
\`\_preplangmore` `<lang-id><space>{<text>}` declares more activities
of the language switch. The <text> is processed whenever
`\_<lang-id>lang` is invoked. If \^`\_preplangmore` is not declared
for given language then \`\_langdefault` is processed.\nl
You can implement selecting a required script for given language, for
example:
\begtt
\_preplangmore ru {\_frenchspacing \_setff{script=cyrl}\selectcyrlfont}
\_addto\_langdefault {\_setff{}\selectlatnfont}
\endtt
The macros `\selectcyrlfont` and `\selectlatnfont` are not defined in
\OpTeX/. If you follow this example, you have to define them after your
decision what fonts will be used in your specific situation.
\_cod -----------------------------
\_def\_preplangmore #1 #2{\_ea \_gdef \_csname _langspecific:#1\_endcsname{#2}}
\_preplangmore en {\_nonfrenchspacing}
\_preplangmore enus {\_nonfrenchspacing}
\_def\_langdefault {\_frenchspacing}
\_doc -----------------------------
The \`\_langreset` is processed before macros declared by \^`\_preplangmore`
or before \^`\_langdefault`. If you set something for
your language by \^`\_preplangmore` then use `\def\_langreset{<settings>}`
in this code too in order to return default values for all other languages.
See `cs` part of `lang-data.opm` file for an example.
\_cod -----------------------------
\_def\_langreset {}
\_doc -----------------------------
The default `\language=0` is US-English with original hyphenation patterns
preloaded in the format (see the end of section~\ref[plain]).
We define `\_enlang` and \`\enlang` switches.
Note that if no language switch is used in the document then
`\language=0` and US-English patterns are used, but \^`\nonfrenchspacing`
isn't set.
\_cod -----------------------------
\_chardef\_enPatt=0
\_sdef{_lan:0}{en}
\_sdef{_ulan:usenglish}{en}
\_def\_enlang{\_uselang{en}\_enPatt23} % \lefthyph=2 \righthyph=3
\_def\enlang{\_enlang}
\_doc -----------------------------
The list of declared languages are reported during format generation.
\_cod -----------------------------
\_message{Declared languages: \_langlist.
Use \_string\<lang-id>lang to initialize language,
\_string\cslang\_space for example.}
\_doc -----------------------------
Each language switch `\_<lang-id>lang` defined by \^`\_preplang` has its initial state\nl
\`\_langinit` `\<switch> <lang-id>(<LongName>)<lang-tag>[<hyph-tag>]<lr-hyph>`.
The \^`\_langinit` macro does:
\begitems
* The internal language <number> is extracted from `\_the\_<lang-id>Patt`.
* `\def \_lan:<number> {<lang-tag>}` for mapping from `\language` number to the <lang-tag>.
* loads `hyph-<hyph-tag>.tex` file with hyphenation patterns when `\language=<number>`.
* loads the part of `lang-data.opm` file with language-dependent phrases
using \^`\_langinput`.
* `\def \_<lang-id>lang {\_uselang{<lang-id>}\_<lang-id>Patt <lr-hyph>}`,
i.e. the switch redefines itself for doing a \"normal job" when the
language switch is used repeatedly.
* Runs itself (i.e. `\_<lang-id>lang`) again for doing the \"normal job" firstly.
\enditems
\_cod -----------------------------
\_def\_langinit #1#2(#3)#4[#5]#6#7{% \_switch lang-id(LongName)lang-tag[hyph-file]lr-hyph
\_sxdef{_lan:\_ea\_the\_csname _#2Patt\_endcsname}{#4}%
\_begingroup \_setbox0=\_vbox{% we don't want spaces in horizontal mode
\_setctable\_optexcatcodes
% loading patterns:
\_language=\cs{_#2Patt}\_relax
\_ifx^#5^\_else
\_wlog{Loading hyphenation for #3: \_string\language=\_the\_language\_space(#5)}%
\_let\patterns=\_patterns \_let\hyphenation=\_hyphenation \_def\message##1{}%
\_isfile {hyph-#5}\_iftrue \_input{hyph-#5}%
\_else \_opwarning{No hyph. patterns #5 for #3, missing package?}\_fi
\_fi
% loading language data:
\_langinput{#4}%
}\_endgroup
\xdef#1{\noexpand\_uselang{#2}\_csname _#2Patt\_endcsname #6#7}%
#1% do language switch
}
\_doc -----------------------------
\`\_uselang``{<lang-id>}\_<lang-id>Patt <pre-hyph><post-hyph>`
is used as \"normal job" of the switch.
It sets `\language`, `\lefthyphenmin`, `\righthyphenmin`.
Finally, it runs data from \^`\_preplangmore` or runs \^`\_langdefault`.
\_cod -----------------------------
\_def\_uselang#1#2#3#4{\_language=#2\_lefthyphenmin=#3\_righthyphenmin=#4\_relax
\_langreset \_def\_langreset{}\_trycs{_langspecific:#1}{\_langdefault}%
}
\_doc -----------------------------
The \`\uselanguage` `{<LongName>}` macro is defined here
(for compatibility with e-plain users). Its parameter is case insensitive.
\_cod -----------------------------
\_def\_uselanguage#1{\_def\_tmp{#1}\_lowercase{\_cs{_\_trycs{_ulan:#1}{0x}lang}}}
\_sdef{_0xlang}{\_opwarning{\_string\uselanguage{\_tmp}: Unknown language name, ignored}}
\_public \uselanguage ;
\_doc -----------------------------
\secc [langdata] Data for various languages
The \"language data" include declarations of
rules for sorting (see section~\ref[makeindex]),
language-dependent phrases and quotation marks (see section~\ref[langphrases]).
The language data are collected in the single `lang-data.opm` file.
Appropriate parts of this file is read by \^`\_langinput{<lang-tag>}`.
First few lines of the file looks like:
{\maxlines=47 \printdoc lang-data.opm }%
There are analogical declaration for more languages here. Unfortunately, this
file is far for completeness. I welcome you send me a part of declaration
for your language.
If your language is missing in this file then a warning is reported during
language initialization. You can create your private declaration in your
macros (analogical as in the `lang-data.opm` file
but without the \^`\_langdata` prefix).
Then you will want to remove the warning about missing data. This can be
done by \`\nolanginput``{<lang-tag>}` given before initialization of your
language.
The whole file `lang-data.opm` is not preloaded in the format because
I suppose a plenty languages here and I don't want to waste the \TeX/ memory
by these declarations. Each part of this file prefixed by
\`\_langdata` `<lang-tag> {<LongName>}` is read separately
when \`\_langinput``{<lang-tag>}` is used. And it is used
in the \^`\_langinit` macro (i.e.\ when the language is initialized),
so the appropriate part of this file is read automatically on demand.
If the part of the `lang-data.opm` concerned by <lang-tag> is read already
then `\_li:<lang-tag>` is set to `R` and we don't read this part of the file again.
\_cod -----------------------------
\_def\_langinput #1{%
\_unless \_ifcsname _li:#1\_endcsname
\_bgroup
\_edef\_tmp{\_noexpand\_langdata #1 }\_everyeof\_ea{\_tmp{}}%
\_long \_ea\_def \_ea\_tmp \_ea##\_ea1\_tmp{\_readlangdata{#1}}%
\_globaldefs=1
\_ea\_tmp \_input{lang-data.opm}%
\_ea\_glet \_csname _li:#1\_endcsname R%
\_egroup
\_fi
}
\_def\_readlangdata #1#2{%
\_ifx^#2^\_opwarning{Missing data for language "#1" in lang-data.opm}%
\_else \_wlog{Reading data for the language #2 (#1)}%
\_fi
}
\_def\_langdata #1 #2{\_endinput}
\_def\_nolanginput #1{\_ea\_glet \_csname _li:#1\_endcsname N}
\_public \nolanginput ;
\_doc -----------------------------
Data of two preferred languages are preloaded in the format:
\_cod -----------------------------
\_langinput{en} \_langinput{cs}
\_endcode
2022-10-11 \_langreset introduced
2022-02-19 \_langinput moved here
2022-02-17 released, original file was hyphen-lan.opm