improved quoting feature in all modes; still documenting

5 files changed, 184 insertions, 30 deletions

diff --git a/arabluatex.dtx b/arabluatex.dtx
index eff2045..ce814f9 100644
--- a/arabluatex.dtx
+++ b/arabluatex.dtx
@@ -36,7 +36,7 @@
 \usepackage{dox}
 \doxitem{Option}{option}{options}
 \usepackage{fontspec,luatextra,realscripts}
-\defaultfontfeatures{RawFeature={+liga}}
+\defaultfontfeatures{RawFeature={+liga;+onum;+hlig}}
 \setmainfont{Libertinus Serif}
 \setsansfont{Libertinus Sans}
 \setmonofont[Scale=.9]{Libertinus Mono}
@@ -604,21 +604,22 @@ consonants, though three of them are also used as vowels
 %   \arb[fullvoc]{ramaW} \arb[trans]{ramaW}, |banaWA|
 %   \arb[fullvoc]{banaWA}, \arb[trans]{banaWA}.
 % \end{quote}
-%
-% \paragraph{\arb[trans]{'alif ma.h_dUfaT} and defective
-% \arb[trans]{U, I}} they are written \meta{\_a, \_i \_u}:
+% 
+% \paragraph{\arb[trans]{'alif ma.h_dUfaT}and defective \arb[trans]{U,
+% I}} They are written \meta{\_a, \_i \_u}:
 % \begin{quote}
-%   |al-l_ahu| \arb[voc]{al-l_ahu} \arb[trans]{al-l_ahu}, |'il_ahuN|
-%   \arb[voc]{'il_ahuN} \arb[trans]{'il_ahuN}, |al-ra.hm_anu|
-%   \arb[voc]{al-ra.hm_anu} \arb[trans]{al-ra.hm_anu}, |l_akin|
-%   \arb[voc]{l_akin} \arb[trans]{l_akin}, |h_ahunA|
-%   \arb[voc]{h_ahunA} \arb[trans]{h_ahunA},
-%   |.hunaynu| |bnu| |'is.h_aqa|
-%   \arb[voc]{.hunaynu bnu 'is.h_aqa} \arb[trans]{\cap{.hunaynu} bnu
-%   \cap{'is.h_aqa}}, |rabb_i| \arb[voc]{rabb_i} \arb[trans]{rabb_i},
-%   |al-`A.s_i| \arb[voc]{al-`A.s_i} \arb[trans]{al-\cap{`A.s_i}}.
+% |al-l_ahu| \arb[voc]{al-l_ahu} \arb[trans]{al-l_ahu}, |'il_ahuN|
+% \arb[voc]{'il_ahuN} \arb[trans]{'il_ahuN}.
+%
+% |al-ra.hm_anu| \arb[voc]{al-ra.hm_anu} \arb[trans]{al-ra.hm_anu},
+% |l_akin| \arb[voc]{l_akin} \arb[trans]{l_akin}, |h_ahunA|
+% \arb[voc]{h_ahunA} \arb[trans]{h_ahunA}, |.hunaynu| |bnu|
+% |'is.h_aqa| \arb[voc]{.hunaynu bnu 'is.h_aqa}
+% \arb[trans]{\cap{.hunaynu} bnu \cap{'is.h_aqa}}, |rabb_i|
+% \arb[voc]{rabb_i} \arb[trans]{rabb_i}, |al-`A.s_i|
+% \arb[voc]{al-`A.s_i} \arb[trans]{al-\cap{`A.s_i}}.
 % \end{quote}
-%
+% 
 % \paragraph{Silent \arb[novoc]{U}/\arb[novoc]{I}}
 % Some words ending with \arb[voc]{BAT} are usually written
 % \arb[voc]{BawT} or \arb[voc]{B_aUT} instead of \arb[voc]{BAT}: see
@@ -1107,6 +1108,106 @@ consonants, though three of them are also used as vowels
 % speaking, it acts as a \emph{toggle switch}. If something, in a
 % given mode, is supposed to be visible, \enquote{quoting} hides it;
 % conversely, if it is supposed not to, it makes it visible.
+%
+% As shown above, \enquote{quoting} means inserting one straight
+% double quote (|"|) \emph{before} the letter that is to be acted
+% upon. Its effects depend on the mode which is currently selected,
+% either |novoc|, |voc| or |fullvoc|:---
+%
+% \paragraph{\texttt{novoc}} In this mode, \enquote{quoting}
+% essentially means make visible something that ought not to be so.
+% \begin{enumerate}[(a)]
+% \item Quoting a vowel, either short or long, makes the
+%   \arb[trans]{.dammaT}, \arb[trans]{fat.haT} or \arb[trans]{kasraT}
+%   appear above the appropriate consonant:---
+% \begin{quote}
+%   |`"ullima| |`ilma| |'l-hay'aTi| \arb[novoc]{`"ullima `ilma
+%   'l-hay'aTi} \arb[trans]{`"ullima `ilma 'l-hay'aTi}, |ya.gz"UA|
+%   \arb[novoc]{ya.gz"UA} \arb[trans]{ya.gz"UA}.
+% \end{quote}
+% \item The same applies when \enquote{quoting} the
+%   \arb[trans]{tanwIn}:---
+% \begin{quote}
+%   |wa-'innA| |sawfa| |tudriku-nA| |'l-manAyA| |muqadd"araT"aN|
+%   \arb[novoc]{wa-'innA sawfa tudriku-nA 'l-manAyA muqadd"araT"aN},
+%   \arb[trans]{wa-'innA sawfa tudriku-nA 'l-manAyA muqadd"araT"aN}.
+% \end{quote}
+% \item If no vowel follows the straight double quote, then a
+%   \arb[trans]{sukUn} is put above the preceding consonant:---
+% \begin{quote}
+%   |qAla isma`"| \arb[novoc]{qAla isma`"} \arb[trans]{qAla isma`"},
+%   |jA'at"| |hinduN| \arb[voc]{jA'at" hinduN} \arb[trans]{jA'at"
+%   hinduN}, |^sabIhuN| |bi-man| |q"u.ti`at"| |qadamA-hu|
+%   \arb[novoc]{^sabIhuN bi-man q"u.ti`at" qadamA-hu}
+%   \arb[trans]{^sabIhuN bi-man q"u.ti`at" qadamA-hu}.
+% \end{quote}
+% \item At the commencement of a word, the straight double quote is
+%   interpreted as \arb[trans]{'alifu 'l-wa.sli}:---
+% \begin{quote}
+%   |wa-"ust"u`mila| \arb[novoc]{wa-"ust"u`mila}
+%   \arb[trans]{wa-"ust"u`mila}, |huwa| |"inhazama| \arb[novoc]{huwa
+%   "inhazama} \arb[trans]{huwa "inhazama}, |al-"intiqA.du|
+%   \arb[novoc]{al-"intiqA.du} \arb[trans]{al-"intiqA.du}.
+% \end{quote}
+% \end{enumerate}
+%
+% \paragraph{\texttt{voc}}
+% In accordance with the general rule, in this mode, \enquote{quoting}
+% makes the vowels and the \arb[trans]{tanwIn} disappear, should this
+% feature be required for some reason:---
+% \begin{enumerate}[(a)]
+% \item Short and long vowels:---
+%   \begin{quote}
+%     |q"Ala q"A'iluN| \arb[voc]{q"Ala q"A'iluN} \arb[trans]{q"Ala
+%     q"A'iluN}, |ibnu 'abI 'u.saybi`aT"a| \arb[voc]{ibnu 'abI
+%     'u.saybi`aT"a} \arb[trans]{\cap{ibnu} \cap{'abI}
+%     \cap{'u.saybi`aT"a}}.
+%   \end{quote}
+% \item \arb[trans]{tanwIn}:---
+%   \begin{quote}
+%     |madInaT"aN| \arb[voc]{madInaT"aN} \arb[trans]{madInaT"aN},
+%     |bAb"aN| \arb[voc]{bAb"aN} \arb[trans]{bAb"aN}, |hud"aN_A|
+%     \arb[voc]{hud"aN_A} \arb[trans]{hud"aN_A}, |^say'"iN|
+%     \arb[voc]{^say'"iN} \arb[trans]{^say'"iN}.
+%   \end{quote}
+% \end{enumerate}
+% One may more usefully \enquote{quote} the initial vowels to write
+% the \arb[trans]{hamzaT} above the \arb[trans]{'alif} or insert a
+% straight double quote after a consonant not followed by a vowel to
+% make the \arb[trans]{sukUn} appear:---
+% \begin{enumerate}[(a)]
+% \item \arb[trans]{'alifu 'l-wa.sli}:---
+%   \begin{quote}
+%     |fI "istisqA'a| \arb[voc]{fI "istisqA'a} \arb[trans]{fI
+%     "istisqA'a}, |wa-"istisqA'u| \arb[voc]{wa-"istisqA'u}
+%     \arb[trans]{wa-"istisqA'u}, |qAla| |"uhrub| |fa-lan| |tuqtala|
+%     \arb[voc]{qAla "uhrub fa-lan tuqtala} \arb[trans]{qAla "uhrub
+%     fa-lan tuqtala}.
+%   \end{quote}
+% \item \arb[trans]{sukUn}:---
+%   \begin{quote}
+%     |qAla| |uqtul"| |fa-lan| |tuqtala| \arb[voc]{qAla uqtul" fa-lan
+%     tuqtala} \arb[trans]{qAla uqtul" fa-lan tuqtala}, |mA| |jA'at"|
+%     |mini| |imra'aTiN| \arb[voc]{mA jA'at" mini imra'aTiN}
+%     \arb[trans]{mA jA'at" mini imra'aTiN}, |kam"| |qad"| |ma.dat"|
+%     |min"| |laylaTiN| \arb[voc]{kam" qad" ma.dat" min" laylaTiN}
+%     \arb[trans]{kam" qad" ma.dat" min" laylaTiN}.
+%   \end{quote}
+% \end{enumerate}
+%
+% \paragraph{\texttt{fullvoc}}
+% In this mode, \enquote{quoting} may be used to take away any short
+% vowel (or \arb[trans]{tanwIn}, as seen above) or any
+% \arb[trans]{sukUn}:---
+% \begin{quote}
+%   |al-jamru| |'l-.sayfiyyu| |'lla_dI| |kAna| \verb+bi-q"rAn"|nUna+
+%   \arb[fullvoc]{al-jamru 'l-.sayfiyyu 'lla_dI kAna
+%   \cap{bi-q"rAn"|nUna}} \arb[trans]{al-jamru 'l-.sayfiyyu 'lla_dI
+%   kAna \cap{bi-q"rAn"|nUna}}.
+% \end{quote}
+%
+% \subsubsection{Quoting the \arb[trans]{hamzaT}}
+% \label{sec:quoting-hamza}
 % 
 %
 % \StopEventually{}
diff --git a/arabluatex_fullvoc.lua b/arabluatex_fullvoc.lua
index a849cd1..4d18a81 100644
--- a/arabluatex_fullvoc.lua
+++ b/arabluatex_fullvoc.lua
@@ -188,6 +188,16 @@ tanwinfv = {
    {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
    {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
    -- assimilations (end)
+   -- quoted tanwīn (begin)
+   {a="(\"uN)", b=""},
+   {a="(B)(\"aN)", b="%1"},
+   {a="(\"aN)(_A)", b="ى"},
+   {a="(\"aN)(Y)", b="ى"},
+   {a="(T)(\"aN)", b="%1"},
+   {a="(ء)(\"aN)", b="%1"},
+   {a="([^TA])(\"aN)", b="%1ا"},
+   {a="(\"iN)", b=""},
+   -- quoted tanwīn (end)
    {a="(uN)", b="ٌ"},
    {a="(B)(aN)", b="%1ً"},
    {a="(aN)(_A)", b="ًى"},
diff --git a/arabluatex_novoc.lua b/arabluatex_novoc.lua
index 9aa141e..8a5a7ed 100644
--- a/arabluatex_novoc.lua
+++ b/arabluatex_novoc.lua
@@ -38,9 +38,12 @@ trigraphsnv = { -- trigraphs or more
    {a="^(a)([%_%^%.]?[tdrzsn])%-", b="ا%1ل"},
    {a="([%s%-])(a)([%_%^%.]?[tdrzsn])%-", b="%1ا%2ل"},
    -- al- + initial unstable hamza
+   {a="^(a)l%-(\")([uai])", b="ا%1لٱ%3"},
+   {a="([%s%-])(a)l%-(\")([uai])", b="%1ا%2لٱ%4"},
    {a="^(a)l%-([uai])", b="ا%1لا%2"},
    {a="([%s%-])(a)l%-([uai])", b="%1ا%2لا%3"},
    -- li-/la- + art. + initial unstable hamza is a special orthography
+   {a="l([ai])%-l%-(\")([uai])", b="ل%1لٱ%3"},
    {a="l([ai])%-l%-([uai])", b="ل%1لا%2"},
    -- al- + lunar consonant (i.e. what remains)
    {a="^(a)l%-", b="ا%1ل"},
@@ -58,6 +61,7 @@ trigraphsnv = { -- trigraphs or more
    -- li-/la + assim. art. + solar consonant is a special orthography
    {a="l([ai])%-([%_%^%.]?[tdrzsn])%-([%_%^%.]?[tdrzsn])", b="ل%1ل%3"},
    -- art. with waṣla + initial unstable hamza
+   {a="'l%-(\")([uai])", b="الٱ%2"},
    {a="'l%-([uai])", b="الا%1"},   
    -- art. with waṣla + lunar consonant (i.e. what remains)
    {a="'l%-", b="ال"},
@@ -86,7 +90,17 @@ tanwinnv = {
 --   {a="(ء)(aN)(%s)([rlmnwy])", b="%1%3%4%4"},
 --   {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ا%3%4%4"},
 --   {a="(iN)(%s)([rlmnwy])", b="|%2%3%3"},
-   --assimilations (end)
+   -- assimilations (end)
+   -- "quoted" tanwīn (begin)
+   {a="(\"uN)", b="ٌ"},
+   {a="(B)(\"aN)", b="%1ً"},
+   {a="(\"aN)(_A)", b="ًى"},
+   {a="(\"aN)(Y)", b="ًى"},
+   {a="(T)(\"aN)", b="%1ً"},
+   {a="(ء)(\"aN)", b="%1ً"},
+   {a="([^TA])(\"aN)", b="%1ًا"},
+   {a="(\"iN)", b="ٍ"},
+   -- "quoted" tanwīn (end)
    {a="(uN)", b=""},
    {a="(B)(aN)", b="%1"},
    {a="(aN)(_A)", b="ى"},
@@ -94,10 +108,18 @@ tanwinnv = {
    {a="(T)(aN)", b="%1"},
    {a="(ء)(aN)", b="%1"},
    {a="([^TA])(aN)", b="%1ا"},
-   {a="(iN)", b=""}
+   {a="(iN)", b=""},
+   -- initial straight double quote gives a connective ʾalif. This has
+   -- nothing to do with the tanwīn, but I put it here for time being.
+   {a="^\"", b="ٱ"},
+   {a="([%s%-])\"", b="%1ٱ"}
 }
 
 longvnv = {
+   {a="\"A", b="َا"},
+   {a="\"U", b="ُو"},
+   {a="\"I", b="ِي"},
+   {a="\"Y", b="aى"},
    {a="A", b="ا"},
    {a="U", b="و"},
    {a="I", b="ي"},
diff --git a/arabluatex_trans.lua b/arabluatex_trans.lua
index 684977a..46dbc33 100644
--- a/arabluatex_trans.lua
+++ b/arabluatex_trans.lua
@@ -158,20 +158,20 @@ tanwintrdmg = {
    {a="uN", b="un"},
    {a="(aN)(_A)", b="an"},
    {a="(aN)(Y)", b="an"},
-   {a="(T)(aN)", b="tan"},
-   {a="([^TA])(aN)", b="%1an"},
+   {a="(T)(\"?aN)", b="tan"},
+   {a="([^TA])(\"?aN)", b="%1an"},
    {a="iN", b="in"}
 }
 
 digraphstrdmg = {
-   {a="([uai]%-)([uai])", b="%1'"}, -- hyphen + initial alif without hamza
-   {a="^([uai])", b="%1"},      -- initial alif without hamza
+   {a="([uai]%-)(\"?[uai])", b="%1'"}, -- hyphen + initial alif without hamza
+   {a="^(\"?[uai])", b="%1"},      -- initial alif without hamza
 -- {a="([%_]?[uaiUAIY])(%s)([uai])", b="%1%2'"}, -- initial alif without hamza
-   {a="(aW)(%s)([uai])", b="awu%2%3"},
-   {a="([^%_][uai])(%s)([uai])", b="%1%2'"},
-   {a="([%_]?[AYa])(%s)([uai])", b="a%2'"},
-   {a="([%_]?[Uu])(%s)([uai])", b="u%2'"},
-   {a="([%_]?[Ii])(%s)([uai])", b="i%2'"},
+   {a="(aW)(%s)(\"?[uai])", b="awu%2%3"},
+   {a="([^%_][uai])(%s)(\"?[uai])", b="%1%2'"},
+   {a="([%_]?[AYa])(%s)(\"?[uai])", b="a%2'"},
+   {a="([%_]?[Uu])(%s)(\"?[uai])", b="u%2'"},
+   {a="([%_]?[Ii])(%s)(\"?[uai])", b="i%2'"},
    -- shorten long vowels preceding ʾalif conjunctionis
    {a="(U)(A)", b="U"},
    {a="(aW)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="awu%2%3"},
@@ -179,7 +179,7 @@ digraphstrdmg = {
    {a="([%_]?[Uu])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="u%2%3"},
    {a="([%_]?[Ii])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="i%2%3"},
    {a="%-%-", b=""},
-   {a="T([^uai])", b="%1"},
+--   {a="T([^uai])", b="%1"},
    {a="T(%s)", b="%1"},
    {a="T$", b=""},
    {a="_t", b="ṯ"},
@@ -359,8 +359,8 @@ tanwintrloc = {
    {a="uN", b="un"},
    {a="(aN)(_A)", b="an"},
    {a="(aN)(Y)", b="an"},
-   {a="(T)(aN)", b="tan"},
-   {a="([^TA])(aN)", b="%1an"},
+   {a="(T)(\"?aN)", b="tan"},
+   {a="([^TA])(\"?aN)", b="%1an"},
    {a="iN", b="in"}
 }
 
@@ -372,7 +372,7 @@ digraphstrloc = {
    {a="uww", b="ūw"},
    {a="iyy", b="īy"},
    {a="([tkdsg])(h)", b="%1'%2"},
-   {a="T([^uai])", b="h%1"},
+--   {a="T([^uai])", b="h%1"},
    {a="T$", b="h"},
    {a="T(%W)", b="h%1"},
    {a="_t", b="th"},
diff --git a/arabluatex_voc.lua b/arabluatex_voc.lua
index ecdb067..f40c405 100644
--- a/arabluatex_voc.lua
+++ b/arabluatex_voc.lua
@@ -162,9 +162,12 @@ trigraphs = { -- trigraphs or more
    {a="^(a)([%_%^%.]?[tdrzsn])%-", b="ا%1ل%2"},
    {a="([%s%-])(a)([%_%^%.]?[tdrzsn])%-", b="%1ا%2ل%3"},
    -- al- + initial unstable hamza
+   {a="^(a)l%-(\")([uai])", b="ا%1لٱ%3"},
+   {a="([%s%-])(a)l%-(\")([uai])", b="%1ا%2لٱ%4"},
    {a="^(a)l%-([uai])", b="ا%1لا%2"},
    {a="([%s%-])(a)l%-([uai])", b="%1ا%2لا%3"},
    -- li-/la- + art. + initial unstable hamza is a special orthography
+   {a="l([ai])%-l%-(\")([uai])", b="ل%1لٱ%3"},
    {a="l([ai])%-l%-([uai])", b="ل%1لا%2"},
    -- al- + lunar consonant (i.e. what remains)
    {a="^(a)l%-", b="ا%1ل"},
@@ -186,6 +189,7 @@ trigraphs = { -- trigraphs or more
    -- li-/la + assim. art. + solar consonant is a special orthography
    {a="l([ai])%-([%_%^%.]?[tdrzsn])%-([%_%^%.]?[tdrzsn])", b="ل%1ل%3%3"},
    -- art. with waṣla + initial unstable hamza
+   {a="'l%-(\")([uai])", b="الٱ%2"},   
    {a="'l%-([uai])", b="الا%1"},   
    -- art. with waṣla + lunar consonant (i.e. what remains)
    {a="'l%-", b="ال"},
@@ -217,6 +221,16 @@ tanwin = {
    {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
    {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
    -- assimilations (end)
+   -- quoted tanwīn (begin)
+   {a="(\"uN)", b=""},
+   {a="(B)(\"aN)", b="%1"},
+   {a="(\"aN)(_A)", b="ى"},
+   {a="(\"aN)(Y)", b="ى"},
+   {a="(T)(\"aN)", b="%1"},
+   {a="(ء)(\"aN)", b="%1"},
+   {a="([^TA])(\"aN)", b="%1ا"},
+   {a="(\"iN)", b=""},
+   -- quoted tanwīn (end)
    {a="(uN)", b="ٌ"},
    {a="(B)(aN)", b="%1ً"},
    {a="(aN)(_A)", b="ًى"},
@@ -228,6 +242,9 @@ tanwin = {
 }
 
 digraphs = {
+   -- initial straight double quote gives a connective ʾalif
+   {a="^\"[uai]", b="ٱ"},
+   {a="([%s%-])\"[uai]", b="%1ٱ"},
    {a="(aW)(%s)([uai])", b="awuا%2%3"},
    {a="(%-)([uai])", b="%1ا%2"}, -- hyphen + initial alif without hamza
    {a="^([uai])", b="ا%1"},      -- initial alif without hamza
@@ -320,10 +337,14 @@ single = {
 }
 
 longv = {
+   {a="\"A", b="ا"},
+   {a="\"U", b="و"},
+   {a="\"I", b="ي"},
+   {a="\"Y", b="ى"},
    {a="A", b="َا"},
    {a="U", b="ُو"},
    {a="I", b="ِي"},
-   {a="Y", b="aى"},
+   {a="Y", b="aى"}
 }
 
 shortv = {