done implementing ʾiʿrāb ġayr munṣarif, done documenting. getting close to v1.3

5 files changed, 254 insertions, 179 deletions

diff --git a/arabluatex.dtx b/arabluatex.dtx
index 613257b..52dc19f 100644
--- a/arabluatex.dtx
+++ b/arabluatex.dtx
@@ -448,9 +448,10 @@
 % corresponding \arb[trans]{tanwIn} (\arb{BuN}, \arb{B|aN}\,,
 % \arb{TaN}, \arb{BaN_A} or \arb{BiN}) is generated. Finally, \meta{u,
 % a, i} at the commencement of a word indicate a \enquote{connective
-% \arb[trans]{'alif}\,} (\arb[trans]{'alifu 'l-wa.sli}), but |voc| mode
-% does not show the \arb[trans]{wa.slaT} above the \arb[trans]{'alif};
-% instead, the accompanying vowel is expressed (\arb{u a i}).
+% \arb[trans]{'alif}\,} (\arb[trans]{'alifu 'l-wa.sli}), but |voc|
+% mode does not show the \arb[trans]{wa.slaT} above the
+% \arb[trans]{'alif}; instead, the accompanying vowel may be expressed
+% at the beginning of a sentence (\arb{u} \arb{a} \arb{i}).
 %
 % \DescribeOption{fullvoc}\\ \label{fullvoc-mode}In addition to what
 % the |voc| mode does, |fullvoc| expresses the \arb[trans]{sukUn} and
@@ -863,7 +864,7 @@ vowels (see §~3).
 %
 % \paragraph{\arb[trans]{\cap{`amruNU}}, and the silent
 % \arb[novoc]{U}} To that name a silent \arb[novoc]{U} is added to
-% distinguish it from \arb[trans]{\cap{`umaruN}}: see \textcite[i. 12
+% distinguish it from \arb[trans]{\cap{`umar-u}}: see \textcite[i. 12
 % C]{Wright}. In no way this affects the sound of the
 % \arb[trans]{tanwIn}, so it has to be discarded in |trans| mode:
 % \begin{quote}
@@ -996,7 +997,7 @@ vowels (see §~3).
 %   \arb[voc]{'a^syA'-a} \arb[trans]{'a^syA'-a}, |.zim'aN|
 %   \arb[voc]{.zim'aN} \arb[trans]{.zim'aN}.
 %
-%   \textbf{\arb[trans]{ta_hfIfu 'l-hamzaTi}}: if the
+%   \textbf{\arb[trans]{ta_hfIf-u 'l-hamzaT-i}}: if the
 %   \arb[trans]{hamzaT} has \arb[trans]{jazmaT} and is preceded by
 %   \emph{\arb[trans]{'alif} hamzatum}, it must be changed into the
 %   letter of prolongation that is homogeneous with the preceding
@@ -1029,8 +1030,8 @@ vowels (see §~3).
 %</example>
 % \fi
 %
-% \package{arabluatex} otherwise determines \arb[trans]{al-'alifu
-% 'l-mamdUdaTu} by context analysis.
+% \package{arabluatex} otherwise determines \arb[trans]{al-'alif-u
+% 'l-mamdUdaT-u} by context analysis.
 % 
 % \begin{quote}
 %   |'is'AduN| \arb[voc]{'is'AduN} \arb[trans]{'is'AduN}, |'AkilUna|
@@ -1158,7 +1159,6 @@ vowels (see §~3).
 %   |fa-lammA| |ra'aW| |'l-najm-a| \arb[fullvoc]{fa-lammA ra'aW
 %   'l-najm-a} \arb[trans]{fa-lammA ra'aW 'l-najm-a}.
 %
-%   \newpage
 %   \textbf{Particles}:---
 %   \begin{compactenum}[(a)]
 %   \item \arb[trans]{li-}: \arb[trans]{'alif-u 'l-wa.sl-i} is omitted
@@ -1198,11 +1198,17 @@ vowels (see §~3).
 %   \arb[trans]{'awi ismu-hu}, |zayduN| |ibn-u| |`amriNU|
 %   \arb[fullvoc]{\cap{z}ayduN ibn-u \cap{`amriNU}}
 %   \arb[trans]{\cap{z}ayduN ibn-u
-%   \cap{`amriNU}},\footnote{\enquote{\arb[trans]{\cap{z}ayd} is the
-%   son of \arb[trans]{\cap{`a}mr}}: the second noun is not in
-%   apposition to the first, but forms part of the predicate. Hence
-%   \arb[voc]{zayduN ibn-u `amriNU} and not \arb[voc]{zayd-u bn-u
-%   `amriNU}, \enquote{Zayd, son of ʿAmr}.}  |imru'-u| |'l-qays-i|
+%   \cap{`amriNU}},\footnote{\label{fn:zayd-is-son}%
+%   \enquote{\arb[trans]{\cap{z}ayd} is the son of
+%   \arb[trans]{\cap{`a}mr}}: the second noun is not in apposition to
+%   the first, but forms part of the predicate. Hence \arb[voc]{zayduN
+%   ibn-u `amriNU} and not \arb[voc]{zayd-u bn-u `amriNU},
+%   \enquote{Zayd, son of ʿAmr}.}  |`umar-u| |ibn-u| |'l-_ha.t.tAb-i|
+%   \arb[fullvoc]{\cap{`umar}-u ibn-u \cap{'l-_ha.t.tAb-i}}
+%   \arb[trans]{\cap{`umar}-u ibn-u
+%   \cap{'l-_ha.t.tAb-i}},\footnote{\enquote{\arb[trans]{\cap{`umar}}
+%   is the son of \arb[trans]{\cap{al-_ha.t.tAb}}} (see
+%   \vref{fn:zayd-is-son}).}  |imru'-u| |'l-qays-i|
 %   \arb[fullvoc]{imru'-u 'l-qays-i} \arb[trans]{\cap{i}mru'-u
 %   \cap{'l-qays-i}}, |la-aymun-u| |'l-l_ah-i|
 %   \arb[fullvoc]{la-aymun-u 'l-l_ah-i} \arb[trans]{la-aymun-u
@@ -1219,7 +1225,9 @@ vowels (see §~3).
 %   'l-nAs-i}, |'abU| |'l-wazIr-i| \arb[fullvoc]{'abU 'l-wazIr-i}
 %   \arb[trans]{'abU 'l-wazIr-i}, |fI| |'l-ibtidA'-i| \arb[fullvoc]{fI
 %   'l-ibtidA'-i} \arb[trans]{fI 'l-ibtidA'-i}, |_dU 'l-i`lAl-i|
-%   \arb[fullvoc]{_dU 'l-i`lAl-i} \arb[trans]{_dU 'l-i`lAl-i}.
+%   \arb[fullvoc]{_dU 'l-i`lAl-i} \arb[trans]{_dU 'l-i`lAl-i},
+%   |maqh_A| |'l-'amIr-i| \arb[voc]{maqh_A 'l-'amIr-i}
+%   \arb[trans]{maqh_A 'l-'amIr-i}.
 % \end{quote}
 %
 % \subparagraph{\arb[trans]{'alif-u 'l-wa.sl-i} preceded by a diphthong}
@@ -1247,8 +1255,8 @@ vowels (see §~3).
 % cases \parencite[i. 22 A--C]{Wright}, it is encoded explicitly, like
 % so:---
 % \begin{quote}
-%   |'antumu| |'l-kA_dibUna| \arb[fullvoc]{'antumu 'l-kA_dibUna}
-%   \arb[trans]{'antumu 'l-kA_dibUna}, |ra'aytumu| |'l-rajul-a|
+%   |'antumu| |'l-kA_dib-Una| \arb[fullvoc]{'antumu 'l-kA_dib-Una}
+%   \arb[trans]{'antumu 'l-kA_dib-Una}, |ra'aytumu| |'l-rajul-a|
 %   \arb[fullvoc]{ra'aytumu 'l-rajul-a} \arb[trans]{ra'aytumu
 %   'l-rajul-a}, |mani| |'l-ka_d_dAb-u| \arb[fullvoc]{mani
 %   'l-ka_d_dAb-u} \arb[trans]{mani 'l-ka_d_dAb-u}, |qatalati|
@@ -1315,9 +1323,9 @@ vowels (see §~3).
 %   |'a_hAfu| |mina| |'l-malik-i| |'lla_dI| |ya.zlimu| |'l-nAs-a|
 %   \arb[fullvoc]{'a_hAfu mina 'l-malik-i 'lla_dI ya.zlimu 'l-nAs-a}
 %   \arb[trans]{'a_hAfu mina 'l-malik-i 'lla_dI ya.zlimu 'l-nAs-a},
-%   |`udtu| |'l-^say_h-a| |'lladI| |huwa| |marI.duN|
-%   \arb[fullvoc]{`udtu 'l-^say_h-a 'lladI huwa marI.duN}
-%   \arb[trans]{`udtu 'l-^say_h-a 'lladI huwa marI.duN}, |mA| |'anA|
+%   |`udtu| |'l-^say_h-a| |'lla_dI| |huwa| |marI.duN|
+%   \arb[fullvoc]{`udtu 'l-^say_h-a 'lla_dI huwa marI.duN}
+%   \arb[trans]{`udtu 'l-^say_h-a 'lla_dI huwa marI.duN}, |mA| |'anA|
 %   |bi-'lla_dI| |qA'iluN| |la-ka| |^say'aN| \arb[fullvoc]{mA 'anA
 %   bi-'lla_dI qA'iluN la-ka ^say'aN} \arb[trans]{mA 'anA bi-'lla_dI
 %   qA'iluN la-ka ^say'aN}.
@@ -1329,11 +1337,11 @@ vowels (see §~3).
 % \end{quote}
 % The other forms are encoded regularly as |al-l| or |'l-l|:---
 % \begin{quote}
-%   |fa-'innA| |na_dkuru| |'l-.sawtayni| |'l-la_dayni| |rawaynA-humA|
-%   |`an| |ja.h.zaT-a| \arb[fullvoc]{fa-'innA na_dkuru 'l-.sawtayni
-%   'l-la_dayni rawaynA-humA `an \cap{ja.h.zaT-a}} \arb[trans]{fa-'innA
-%   na_dkuru 'l-.sawtayni 'l-la_dayni rawaynA-humA `an
-%   \cap{ja.h.zaT-a}}.
+%   |fa-'innA| |na_dkuru| |'l-.sawt-ayni| |'l-la_dayni| |rawaynA-humA|
+%   |`an| |ja.h.zaT-a| \arb[fullvoc]{fa-'innA na_dkuru 'l-.sawt-ayni
+%   'l-la_dayni rawaynA-humA `an \cap{ja.h.zaT-a}}
+%   \arb[trans]{fa-'innA na_dkuru 'l-.sawt-ayni 'l-la_dayni
+%   rawaynA-humA `an \cap{ja.h.zaT-a}}.
 %
 %   And also: |al-la_dAni| \arb[fullvoc]{al-la_dAni}
 %   \arb[trans]{al-la_dAni}, |al-la_dayni| \arb[fullvoc]{al-la_dayni}
@@ -1793,8 +1801,8 @@ vowels (see §~3).
 % \end{compactenum}
 % \DescribeMacro{\arbup} By default, \package{arabluatex} applies rule
 % \ref{ref:dmg-up-rend}. Once delimited by a set of Lua functions,
-% every \arb[trans]{tanwIn} is passed as an argument on to a
-% \cs{arbup} command which is set to \cs{textsuperscript}.
+% \arb[trans]{'i`rAb} is passed as an argument on to a \cs{arbup}
+% command which is set to \cs{textsuperscript}.
 %
 % \DescribeMacro{\NoArbUp} \DescribeMacro{\ArbUpDflt} \cs{NoArbUp} may
 % be used either in the preamble or at any point of the document in
@@ -1803,7 +1811,7 @@ vowels (see §~3).
 % any point of the document.
 %
 % \DescribeMacro{\SetArbUp} Finally, \cs{SetArbUp}\marg{formatting
-% directives} may be used to customize the way \arb[trans]{tanwIn} is
+% directives} may be used to customize the way \arb[trans]{'i`rAb} is
 % displayed. To take one example, here is how Arabic
 % \arb[trans]{'i`rAb} may be rendered as subscript text:---
 % \iffalse
@@ -1822,7 +1830,54 @@ muhaddamaTaN mi'_danatu-hu}: \arb[trans]{ra'aytu
 % As shown in the above example, |#1| is the token that is replaced
 % with the actual \arb[trans]{tanwIn} in the formatting directives of
 % the \cs{SetArbUp} command.
-% 
+%
+% \paragraph{\arb[trans]{'i`rAb} boundaries}
+% Every declinable noun (\arb[trans]{mu`rab}) may be declined either
+% with or without \arb[trans]{tanwIn}, viz. \arb[trans]{mun.sarifuN}
+% or \arb[trans]{.gayr-u mun.sarifiN}. The former is automatically
+% parsed by \package{arabluatex}, whereas the latter has to be
+% delimited with an hyphen, like so:---
+% \begin{quote}
+%   \arb[trans]{\textbf{mun.sarif}}: |mu`allimuN|
+%   \arb[voc]{mu`allimuN} \arb[trans]{mu`allimuN}, |kA'inuN|
+%   \arb[voc]{kA'inuN} \arb[trans]{kA'inuN}, |kA'inAtuN|
+%   \arb[voc]{kA'inAtuN} \arb[trans]{kA'inAtuN}, |\cap{`amraNU}|
+%   \arb[voc]{\cap{`amraNU}} \arb[trans]{\cap{`amraNU}}, |fataN_A|
+%   \arb[voc]{fataN_A} \arb[trans]{fataN_A}.
+%
+%   \arb[trans]{\textbf{.gayr mun.sarif}}: |al-mu`allim-u|
+%   \arb[voc]{al-mu`allim-u} \arb[trans]{al-mu`allim-u}, |kitAb-Ani|
+%   \arb[voc]{kitAb-Ani} \arb[trans]{kitAb-Ani}, |ra^sa'-Ani|
+%   \arb[voc]{ra^sa'-Ani} \arb[trans]{ra^sa'-Ani}, |sAriq-Una|
+%   \arb[voc]{sAriq-Una} \arb[trans]{sAriq-Una}, |qA.d-Una|
+%   \arb[voc]{qA.d-Una} \arb[trans]{qA.d-Una}, |al-.zulm-Atu|
+%   \arb[voc]{al-.zulm-Atu} \arb[trans]{al-.zulm-Atu}.
+% \end{quote}
+%
+% \begin{quoting}
+%   \textsc{Rem.}~\emph{a.} As the \arb[trans]{tanwIn} is passed over
+%   in pronunciation when it is followed by the letters
+%   \arb[novoc]{r}, \arb[novoc]{l}, \arb[novoc]{m}, \arb[novoc]{w},
+%   \arb[novoc]{y} (see \vref{ref:assimilation}), it may be desirable
+%   to further distinguish it by putting it above the line, but not to
+%   do the same for \arb[trans]{.gayr mun.sarif} terminations. This
+%   can be achieved by simply omitting the hyphen before any
+%   \arb[trans]{.gayr mun.sarif} termination:---\\
+%   |kAna| |.ganiyyaN| |l_akinna-hu| |labisa| |^gubbaTaN| |mumazzaqaN|
+%   |'aydu-hA| \arb[voc]{kAna .ganiyyaN l_akinna-hu labisa ^gubbaTaN
+%   mumazzaqaN 'aydu-hA} \arb[trans]{kAna .ganiyyaN l_akinna-hu labisa
+%   ^gubbaTaN mumazzaqaN 'aydu-hA}.
+%
+%   \textsc{Rem.}~\emph{b.} Although the hyphen before the
+%   \arb[trans]{tanwIn} is optional as \package{arabluatex} always
+%   parses nouns with such termination, it may also be used to mark
+%   better the inflectional endings:---\\
+%   |mana`a| |'l-nAs-a| |kAffaT-aN| |min| |mu_hA.tabati-hi|
+%   |'a.had-uN| |bi-sayyidi-nA| \arb[voc]{mana`a 'l-nAs-a kAffaT-aN
+%   min mu_hA.tabati-hi 'a.had-uN bi-sayyidi-nA} \arb[trans]{mana`a
+%   'l-nAs-a kAffaT-aN min mu_hA.tabati-hi 'a.had-uN bi-sayyidi-nA}.
+% \end{quoting}
+%
 % \subsection{Examples}
 % \label{sec:examples-translit}
 % Here follows in transliteration the story of
@@ -2226,7 +2281,7 @@ wa-ya.sIru ta.hta 'l-jild-i
 %    \end{macrocode}
 % \end{macro}
 % \begin{macro}{\arbup}
-% \changes{v1.3}{2016/05/19}{\arb[trans]{tanwIn} is now written as
+% \changes{v1.3}{2016/05/19}{\arb[trans]{'i`rAb} is now written as
 % superscript text in \texttt{dmg} mode by default.}
 % \begin{macro}{\NoArbUp}
 % \begin{macro}{\ArbUpDflt}
diff --git a/arabluatex_fullvoc.lua b/arabluatex_fullvoc.lua
index f9d4e60..e4a9495 100644
--- a/arabluatex_fullvoc.lua
+++ b/arabluatex_fullvoc.lua
@@ -53,8 +53,9 @@ hamzafv = {
    {a="'i'([^uaiUAI])", b="إI%1"},   
    -- madda (historic writing below)
    {a="'a'([^uaiUAI])", b="آ%1"},
---tmp   {a="'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
-   {a="([^UIui])'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="([^uiUI])\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="^\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
+   {a="(%W)\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
    {a="(A)(')(uN?)$", b="aآء%3"},
    {a="(A)(')(uN?)(%W)", b="aآء%3%4"},
    {a="(A)(')(iN?)$", b="aآء%3"},
@@ -148,8 +149,9 @@ hamzafveasy = { -- differences marked below with 'easy'
    {a="'i'([^uaiUAI])", b="إI%1"},   
    -- madda (historic writing below)
    {a="'a'([^uaiUAI])", b="آ%1"},
---tmp   {a="'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
-   {a="([^UIui])'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="([^uiUI])\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="^\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
+   {a="(%W)\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
    --easy (begin)
    {a="(A)(')(uN?)$", b="aاء%3"},
    {a="(A)(')(uN?)(%W)", b="aاء%3%4"},
@@ -215,73 +217,73 @@ hamzafveasy = { -- differences marked below with 'easy'
 }
 
 tanwinfv = {
-   {a="uNU", b="ٌو"},
-   {a="aNU", b="ًوا"},
-   {a="iNU", b="ٍو"},
-   {a="([uai]N)(%s)([uai])", b="%1%2ٱ"},
-   {a="(aN[%_]?[AY])(%s)([uai])", b="%1%2ٱ"},
+   {a="%-?uNU", b="ٌو"},
+   {a="%-?aNU", b="ًوا"},
+   {a="%-?iNU", b="ٍو"},
+   {a="%-?([uai]N)(%s)([uai])", b="%1%2ٱ"},
+   {a="%-?(aN[%_]?[AY])(%s)([uai])", b="%1%2ٱ"},
    -- assimilations (begin)
-   {a="(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
-   {a="(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
-   {a="(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
-   {a="(T)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
-   {a="(ء)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
-   {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
-   {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
+   {a="%-?(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
+   {a="%-?(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
+   {a="%-?(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
+   {a="(T)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+   {a="(ء)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+   {a="([^TA])%-?(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
+   {a="%-?(iNI?)(%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=""},
+   {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="%-?(\"iNI?)", b=""},
    -- quoted tanwīn (end)
-   {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="(iNI?)", b="ٍ"}
+   {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="%-?(iNI?)", b="ٍ"}
 }
 
 tanwinfveasy = { -- no assimilations (see below)
-   {a="uNU", b="ٌو"},
-   {a="aNU", b="ًوا"},
-   {a="iNU", b="ٍو"},
-   {a="([uai]N)(%s)([uai])", b="%1%2ٱ"},
-   {a="(aN[%_]?[AY])(%s)([uai])", b="%1%2ٱ"},
+   {a="%-?uNU", b="ٌو"},
+   {a="%-?aNU", b="ًوا"},
+   {a="%-?iNU", b="ٍو"},
+   {a="%-?([uai]N)(%s)([uai])", b="%1%2ٱ"},
+   {a="%-?(aN[%_]?[AY])(%s)([uai])", b="%1%2ٱ"},
    -- assimilations (begin)
---easy   {a="(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
---easy   {a="(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
---easy   {a="(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
---easy   {a="(T)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
---easy   {a="(ء)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
---easy   {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
---easy   {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
+--easy   {a="%-?(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
+--easy   {a="%-?(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
+--easy   {a="%-?(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
+--easy   {a="(T)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+--easy   {a="(ء)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+--easy   {a="([^TA])%-?(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
+--easy   {a="%-?(iNI?)(%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=""},
+   {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="%-?(\"iNI?)", b=""},
    -- quoted tanwīn (end)
-   {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="(iNI?)", b="ٍ"}
+   {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="%-?(iNI?)", b="ٍ"}
 }
 
 trigraphsfv = { -- trigraphs or more
@@ -457,10 +459,10 @@ digraphsfv = {
    {a="yy", b="يّ"},
    -- sukūn begin
    --tmp: next line is added
-   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])%-(%_?[uaiUAIY])", b="%1%2"},
+   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])%-([uaiUAI])", b="%1%2"},
    {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])$", b="%1ْ"},
    {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])([%s])", b="%1ْ%2"},
-   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])%-(%_?[^uaiUAIًٌٍ])", b="%1ْ%2"},
+   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])([%_]?[^%_uaiUAIYًٌٍ])", b="%1ْ%2"},
    -- take out sukūn in cases of assimilation
    {a="(n)(ْ)(%s)([روي])", b="%1%3%4"},
    {a="(n)(ْ)(%s)([ل])", b="%1%3%4"},
@@ -559,7 +561,7 @@ digraphsfveasy = { -- see the diffenrences under 'easy' marker below
    -- so there be no need to edit them:
    {a="ْ", b=""}, 
    --tmp: next line is added
---   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])%-([uai])", b="%1%2"},
+--   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])%-([uaiUAI])", b="%1%2"},
 --   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])$", b="%1ْ"},
 --   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])([%s])", b="%1ْ%2"},
 --   {a="([%_%^%.]?[Bbtjghxdrzs%`fqklmnwy])([%_]?[^%_uaiUAIًٌٍ])", b="%1ْ%2"},
@@ -654,4 +656,3 @@ singlefveasy = { -- see the differences under 'easy' tag below
    {a="([^0-9])%-([^0-9])", b="%1%2"},
    {a="B", b="ـ"},
 }
-
diff --git a/arabluatex_novoc.lua b/arabluatex_novoc.lua
index 16f3b51..f069f5a 100644
--- a/arabluatex_novoc.lua
+++ b/arabluatex_novoc.lua
@@ -23,36 +23,36 @@ along with this program.  If not, see
 --]]
 
 tanwinnv = {
-   {a="uNU", b="و"},
-   {a="aNU", b="وا"},
-   {a="iNU", b="و"},
+   {a="%-?uNU", b="و"},
+   {a="%-?aNU", b="وا"},
+   {a="%-?iNU", b="و"},
    -- assimilations (begin). These are good but may not apply here.
---   {a="(uN)(%s)([rlmnwy])", b="|%2%3%3"},
---   {a="(aN)(_A)(%s)([rlmnwy])", b="ى%3%4%4"},
---   {a="(aN)(Y)(%s)([rlmnwy])", b="ى%3%4%4"},
---   {a="(T)(aN)(%s)([rlmnwy])", b="%1%3%4%4"},
---   {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"},
+--   {a="%-?(uN)(%s)([rlmnwy])", b="|%2%3%3"},
+--   {a="%-?(aN)(_A)(%s)([rlmnwy])", b="ى%3%4%4"},
+--   {a="%-?(aN)(Y)(%s)([rlmnwy])", b="ى%3%4%4"},
+--   {a="(T)%-?(aN)(%s)([rlmnwy])", b="%1%3%4%4"},
+--   {a="(ء)%-?(aN)(%s)([rlmnwy])", b="%1%3%4%4"},
+--   {a="([^TA])%-?(aN)(%s)([rlmnwy])", b="%1ا%3%4%4"},
+--   {a="%-?(iNI?)(%s)([rlmnwy])", b="|%2%3%3"},
    -- 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="ٍ"},
+   {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="%-?(\"iNI?)", b="ٍ"},
    -- "quoted" tanwīn (end)
-   {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=""},
+   {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="%-?(iNI?)", 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"},
diff --git a/arabluatex_trans.lua b/arabluatex_trans.lua
index 45fd3d5..76e297e 100644
--- a/arabluatex_trans.lua
+++ b/arabluatex_trans.lua
@@ -138,8 +138,15 @@ tanwintrdmg = {
    {a="(T)%-?(aN)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="t\\arbup{ani}%3%4"},
    {a="([^TA])%-?(aN)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="%1\\arbup{ani}%3%4"},
    {a="%-?(iN)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="\\arbup{ini}%2%3"},
+-- tmp (+ ibn)
+   {a="%-?(uN)(%s)(i)", b="\\arbup{uni}%2'"},
+   {a="%-?(aN)(_A)(%s)(i)", b="ạ\\arbup{ni}%3%'"},
+   {a="%-?(aN)(Y)(%s)(i)", b="ạ\\arbup{ni}%3%'"},
+   {a="(T)%-?(aN)(%s)(i)", b="t\\arbup{ani}%3%'"},
+   {a="([^TA])%-?(aN)(%s)(i)", b="%1\\arbup{ani}%3%'"},
+   {a="%-?(iN)(%s)(i)", b="\\arbup{ini}%2%'"},
    --
---   {a="uN", b="\\arbup{un}"}, (now included in last line in this table)
+--   {a="uN", b="\\arbup{un}"}, (now included in the last line of this table)
    {a="%-?(aN)(_A)", b="ạ\\arbup{n}"},
    {a="%-?(aN)(Y)", b="ạ\\arbup{n}"},
    {a="(T)%-?(\"?aN)", b="t\\arbup{an}"},
@@ -260,10 +267,16 @@ trigraphstrdmgeasy = { -- see the differences below under 'easy' tag
 digraphstrdmg = {
    {a="([uai]%-)(\"?[uai])", b="%1'"}, -- hyphen + initial alif without hamza
    {a="^(\"?[uai])", b="%1"},      -- initial alif without hamza
+   -- this is not necessary, take out for now:
 -- {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'"},
+--tmp (more complex, see next 4 l.)
+-- {a="([%_]?[AYa])(%s)(\"?[uai])", b="a%2'"},
+   {a="(%_A)(%s)(\"?[uai])", b="ạ%2'"},
+   {a="(Y)(%s)(\"?[uai])", b="ạ%2'"},
+   {a="(%_a)(%s)(\"?[uai])", b="a%2'"},
+   {a="(A)(%s)(\"?[uai])", b="a%2'"},
    {a="([%_]?[Uu])(%s)(\"?[uai])", b="u%2'"},
    {a="([%_]?[Ii])(%s)(\"?[uai])", b="i%2'"},
    -- ʾiʿrāb hyphen (begin)
@@ -281,7 +294,11 @@ digraphstrdmg = {
    -- shorten long vowels preceding ʾalif conjunctionis
    {a="(U)(A)", b="U"},
    {a="(aW)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="awu%2%3"},
-   {a="([%_]?[AYa])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="a%2%3"},
+--tmp {a="([%_]?[AYa])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="a%2%3"},
+   {a="(%_a)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="a%2%3"},
+   {a="(%_A)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="ạ%2%3"},
+   {a="(A)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="a%2%3"},
+   {a="(Y)(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="ạ%2%3"},
    {a="([%_]?[Uu])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="u%2%3"},
    {a="([%_]?[Ii])(%s)(['][%_%^%.]?[l'btjghxdrzs`fqkmnwy]%-)", b="i%2%3"},
    {a="%-%-", b=""},
diff --git a/arabluatex_voc.lua b/arabluatex_voc.lua
index 0097450..0331794 100644
--- a/arabluatex_voc.lua
+++ b/arabluatex_voc.lua
@@ -93,8 +93,9 @@ hamza = {
    {a="'i'([^uaiUAI])", b="إI%1"},   
    -- madda (historic writing below)
    {a="'a'([^uaiUAI])", b="آ%1"},
---tmp   {a="'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
-   {a="([^UIui])'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="([^uiUI])\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="^\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
+   {a="(%W)\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
    {a="(A)(')(uN?)$", b="aآء%3"},
    {a="(A)(')(uN?)(%W)", b="aآء%3%4"},
    {a="(A)(')(iN?)$", b="aآء%3"},
@@ -179,8 +180,9 @@ hamzaeasy = { -- differences marked below with 'easy'
    {a="'i'([^uaiUAI])", b="إI%1"},   
    -- madda (historic writing below)
    {a="'a'([^uaiUAI])", b="آ%1"},
---tmp   {a="'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
-   {a="([^UIui])'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="([^uiUI])\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
+   {a="^\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="آ%1"},
+   {a="(%W)\'a?A([%_%^%.]?[%`%'btjghdrzsfqklmnywAY])", b="%1آ%2"},
    --easy (begin)
    {a="(A)(')(uN?)$", b="aاء%3"},
    {a="(A)(')(uN?)(%W)", b="aاء%3%4"},
@@ -246,69 +248,69 @@ hamzaeasy = { -- differences marked below with 'easy'
 }
 
 tanwin = {
-   {a="uNU", b="ٌو"},
-   {a="aNU", b="ًوا"},
-   {a="iNU", b="ٍو"},
+   {a="%-?uNU", b="ٌو"},
+   {a="%-?aNU", b="ًوا"},
+   {a="%-?iNU", b="ٍو"},
    -- assimilations (begin)
-   {a="(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
-   {a="(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
-   {a="(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
-   {a="(T)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
-   {a="(ء)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
-   {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
-   {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
+   {a="%-?(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
+   {a="%-?(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
+   {a="%-?(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
+   {a="(T)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+   {a="(ء)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+   {a="([^TA])%-?(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
+   {a="%-?(iNI?)(%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=""},
+   {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="%-?(\"iNI?)", b=""},
    -- quoted tanwīn (end)
-   {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="(iNI?)", b="ٍ"}
+   {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="%-?(iNI?)", b="ٍ"}
 }
 
 tanwineasy = { -- 'easy' requires some lines to be taken out:
-   {a="uNU", b="ٌو"},
-   {a="aNU", b="ًوا"},
-   {a="iNU", b="ٍو"},
+   {a="%-?uNU", b="ٌو"},
+   {a="%-?aNU", b="ًوا"},
+   {a="%-?iNU", b="ٍو"},
    -- assimilations (begin)
---   {a="(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
---   {a="(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
---   {a="(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
---   {a="(T)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
---   {a="(ء)(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
---   {a="([^TA])(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
---   {a="(iN)(%s)([rlmnwy])", b="ٍ%2%3%3"},
+--   {a="%-?(uN)(%s)([rlmnwy])", b="ٌ%2%3%3"},
+--   {a="%-?(aN)(_A)(%s)([rlmnwy])", b="ًى%3%4%4"},
+--   {a="%-?(aN)(Y)(%s)([rlmnwy])", b="ًى%3%4%4"},
+--   {a="(T)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+--   {a="(ء)%-?(aN)(%s)([rlmnwy])", b="%1ً%3%4%4"},
+--   {a="([^TA])%-?(aN)(%s)([rlmnwy])", b="%1ًا%3%4%4"},
+--   {a="%-?(iNI?)(%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=""},
+   {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="%-?(\"iNI?)", b=""},
    -- quoted tanwīn (end)
-   {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="(iNI?)", b="ٍ"}
+   {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="%-?(iNI?)", b="ٍ"}
 }
 
 trigraphs = { -- trigraphs or more
@@ -452,7 +454,7 @@ digraphs = {
    --tmp-added hyphen + initial alif without hamza:
    {a="(%-)([uai])([%^%_%.%`]?)(%a)", b="%1ا%3%4"},
    {a="^([uai])", b="ا%1"},      -- initial alif without hamza
-   {a="(%s)([uai])", b="%1ا%2"}, -- initial alif without hamza
+   {a="(%s)([uai])", b="%1ا"}, -- initial alif without hamza
    {a="%-%-", b="ـ"},
    {a="ؤؤ", b="ؤّ"},
    {a="أأ", b="أّ"},