siunitx 1.1 (28sep08)

git-svn-id: svn://tug.org/texlive/trunk@10777 c570f23f-e606-0410-a88d-b1316a301751

27 files changed, 869 insertions, 169 deletions

diff --git a/Master/texmf-dist/doc/latex/siunitx/si-colour.pdf b/Master/texmf-dist/doc/latex/siunitx/si-colour.pdf
deleted file mode 100644
index 0a95a1935a1..00000000000
--- a/Master/texmf-dist/doc/latex/siunitx/si-colour.pdf
+++ /dev/null
diff --git a/Master/texmf-dist/doc/latex/siunitx/siunitx-manual.pdf b/Master/texmf-dist/doc/latex/siunitx/siunitx-manual.pdf
deleted file mode 100644
index 34bd6644f36..00000000000
--- a/Master/texmf-dist/doc/latex/siunitx/siunitx-manual.pdf
+++ /dev/null
diff --git a/Master/texmf-dist/doc/latex/siunitx/siunitx.pdf b/Master/texmf-dist/doc/latex/siunitx/siunitx.pdf
index f6c4808697a..e15849347ac 100644
--- a/Master/texmf-dist/doc/latex/siunitx/siunitx.pdf
+++ b/Master/texmf-dist/doc/latex/siunitx/siunitx.pdf
diff --git a/Master/texmf-dist/source/latex/siunitx/siunitx.dtx b/Master/texmf-dist/source/latex/siunitx/siunitx.dtx
index 0f82125a68e..ab6979af8ad 100644
--- a/Master/texmf-dist/source/latex/siunitx/siunitx.dtx
+++ b/Master/texmf-dist/source/latex/siunitx/siunitx.dtx
@@ -51,9 +51,7 @@
 %    documentation in PDF format.  Simply decompress siunitx.tds.dtx
 %    in your local TeX directory, run your hash program (texhash,
 %    initexmf --update-fndb, etc.) and everything will be ready to
-%    go.  The user documentation for the package is called
-%    siunitx-manual.pdf; the file siunitx.pdf includes the user
-%    manual and the fully-indexed source code.
+%    go.  
 %
 % Unpacking:
 %    (a) If siunitx.ins is present:
@@ -205,9 +203,10 @@ dcolumn, rccol and numprint packages.
 % Various adjustments are needed to the packages loaded.  xfrac is
 % only loaded if it is available.
 \usepackage[final]{graphicx}
-\IfFileExists{xfrac.sty}
-  {\usepackage{xfrac}}
-  {}
+% Currently, xfrac is not working with expl3!
+%\IfFileExists{xfrac.sty}
+%  {\usepackage{xfrac}}
+%  {}
 \usepackage{jawltxdoc}
 \usepackage{textcomp,tikz,pgfplots,tocloft,datatool}
 \usepackage[latin1]{inputenc}
@@ -223,7 +222,7 @@ dcolumn, rccol and numprint packages.
 %</driver>
 % \fi
 %
-%\CheckSum{9558}
+%\CheckSum{10072}
 %
 % \CharacterTable
 %  {Upper-case    \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z
@@ -305,7 +304,7 @@ dcolumn, rccol and numprint packages.
 %  kibi,byte,molar,Molar,mmHg,mebi,torr,parsec,lightyear,clight,
 %  eVperc,addplot,color,milli,cm,Hz,kg,DeclareRobustCommand,
 %  pascal,dalton,addtolocale,DTLnewdb,DTLnewrow,DTLnewdbentry,
-%  DTLforeach,DTLiffirstrow,DTLmul}}
+%  DTLforeach,DTLiffirstrow,DTLmul,newqualifier,hour}}
 %
 %\title{\currpkg\ --- A comprehensive (SI) units package^^A
 %  \thanks{This file describes version \fileversion, last revised
@@ -781,11 +780,14 @@ dcolumn, rccol and numprint packages.
 % later.
 %
 %\subsection{Literal units}
+%\changes{v1.1}{2008/07/23}{Unit processor extended to interpret
+%  \texttt{\textunderscore} correctly}
 % Units can be input in two ways, inspired by \SIstyle and \SIunits.
-% The \SIstyle-like method uses literal input.  Four characters have
+% The \SIstyle-like method uses literal input.  Five characters have
 % a special meaning:
 %\begin{itemize}
-%  \item ``|^|'' The superscript character is used without the usual
+%  \item ``|^|'' and ``|_|''  The superscript and subscript
+%    characters can be used without the usual
 %    need for surrounding maths characters (|$|);^^A$
 %  \item ``|.|'' and ``|,|'': the full stop (point) symbol and comma
 %    are made active, and produce the current contents of the
@@ -797,6 +799,7 @@ dcolumn, rccol and numprint packages.
 %\begin{LaTeXexample}
 %  \SI{10}{kg.m.s^{-2}}\\
 %  \SI{1.453}{g/cm^3}\\
+%  \SI{6.345}{kg_{pol}.mmol_{cat}^{-1}}\\
 %  \SI{33.562}{cd~s}\\
 %  \SI[unitsep=medium]{100}{m.s^{-2}}
 %\end{LaTeXexample}
@@ -1346,6 +1349,23 @@ dcolumn, rccol and numprint packages.
 %  \newprefix[binary]{\kibi}{Ki}{10}
 %\end{LaTeXexample}
 %
+%\changes{v1.1}{2008/09/22}{Qualifiers introduced}
+%\DescribeMacro{\newqualifier}
+%\DescribeMacro{\renewqualifier}
+%\DescribeMacro{\providequalifier}
+% It is possible to create unit ``qualifiers'', which add subscript
+% descriptions to units.  Although not encouraged, this is sometimes
+% necessary to make the meaning clear.  No qualifiers are pre-defined
+% by \currpkg, and so the user needs to declare them using
+% \cs{newqualifier}. Currently, only subscript qualifiers can be
+% created; a more extended set of options is planned for the next
+% release of \currpkg.
+%\begin{LaTeXexample}
+%  \newqualifier{\polymer}{pol}
+%  \newqualifier{\catalyst}{cat}
+%  \si{\kg\polymer\per\mole\catalyst\per\hour}
+%\end{LaTeXexample}
+%
 %\section{Specialist units}
 % In some subject area, there are units which are in common use even
 % though they are outside of the SI system.  Unlike the units
@@ -1674,6 +1694,7 @@ dcolumn, rccol and numprint packages.
 %\subsection{Number formatting}
 %\DescribeOption{numdigits}
 %\DescribeOption{numdecimal}
+%\DescribeOption{numdiv}
 %\DescribeOption{numexp}
 %\DescribeOption{numsign}
 %\DescribeOption{numaddn}
@@ -1686,13 +1707,14 @@ dcolumn, rccol and numprint packages.
 % \numprint package, \opt{numexp} (the list of exponent markers)
 % recognises \defaultopt{deDE} as valid by default.  \opt{numsign}
 % contains the sign markers for numbers
-% (\defaultopt{+-\cs{pm}\cs{mp}}).  \opt{numaddn} and
-% \opt{numgobble} both control which other characters do not give an
-% error when present in a number.  \opt{numaddn} contains valid
-% characters which should be included in the final output ``as is'',
-% whereas \opt{numgobble} lists the characters that are completely
-% ignored.  In all cases, the content of the options is a simple
-% string, for example \opt{numdigits=1234567890}.
+% (\defaultopt{+-\cs{pm}\cs{mp}}).  \opt{numdiv} holds the division
+% marker in numbers (\defaultopt{/}). \opt{numaddn} and \opt{numgobble}
+% both control which other characters do not give an error when present
+% in a number.  \opt{numaddn} contains valid characters which should be
+% included in the final output ``as is'', whereas \opt{numgobble} lists
+% the characters that are completely ignored.  In all cases, the
+% content of the options is a simple string, for example
+% \opt{numdigits=1234567890}.
 %
 %\DescribeOption{decimalsymbol}
 %\DescribeOption{digitsep}
@@ -1842,15 +1864,38 @@ dcolumn, rccol and numprint packages.
 % In contrast to the \opt{padnumber} option, the package can alter
 % the precision of the input number if the \opt{fixdp} option is set.
 % The will fix the decimal places of the output to the number stored
-% in the \opt{dp} option.  The later should be a positive integer.
+% in the \opt{dp} option.  The later should be a positive integer or
+% zero.
 %\begin{LaTeXexample}
 %  \sisetup{fixdp,dp=3}
 %  \num{1}\\
 %  \num{53.9}\\
 %  \num{4.56783}\\
 %  \num{-1.2942}\\
-%  \num{-1.2949}\\
-%  \num{9.9999}
+%  \num{-1.2959}\\
+%  \num{9.9999}\\
+%  \num{2.1264e90}\\
+%  \num[dp=0]{12.345}
+%\end{LaTeXexample}
+%
+%\DescribeOption{fixsf}
+%\changes{v1.1}{2008/08/22}{New \opt{fixsf} option}
+%\DescribeOption{sf}
+%\changes{v1.1}{2008/08/22}{New \opt{sf} option}
+% In contrast to the \opt{dp} option, the \opt{fixsf} and \opt{sf}
+% options control the number of significant figures in the output.
+% This never adds extra digits to the output, as significant figures
+% are concerned with accuracy.
+%\begin{LaTeXexample}
+%  \sisetup{fixsf,sf=3}
+%  \num{1}\\
+%  \num{53.9}\\
+%  \num{4.56783}\\
+%  \num{-1.2942}\\
+%  \num{-1.2959}\\
+%  \num{9.9999}\\
+%  \num{2.1264e90}\\
+%  \num{773322}
 %\end{LaTeXexample}
 %
 %\subsection{Angle formatting}
@@ -1941,7 +1986,7 @@ dcolumn, rccol and numprint packages.
 % whitespace after shorter mantissa components.  In the second
 % method, no additional space is added after the mantissa, and the
 % exponents do no line up (\ref{tab:alignexp}).  This is controlled
-% by the \opt{tabexpalign} option, which can be set to
+% by the \opt{tabalignexp} option, which can be set to
 % \defaultopt{true} or \opt{false}.
 %\begin{LaTeXexample}[float]
 %  \begin{table}
@@ -2004,6 +2049,33 @@ dcolumn, rccol and numprint packages.
 %  \end{table}
 %\end{LaTeXexample}
 %
+%\DescribeOption{tabparseonly}
+% It is possible to turn off the automatic alignment of numbers
+% while retaining the ability to parse the |S| column input.
+% Setting \opt{tabparseonly} to \opt{true} will still parse the
+% column contents, but will not align the decimal markers.  The
+% alignment of the cell is governed by \opt{tabnumalign}
+% (\ref{tab:parse}).
+%\begin{LaTeXexample}[float]
+%  \begin{table}
+%    \centering
+%    \caption{The \opt{tabparseonly} option}
+%    \label{tab:parse}
+%    \begin{tabular}{%
+%      S[tabparseonly]%
+%      S[tabparseonly,tabnumalign=left]%
+%      S[tabparseonly,tabnumalign=right]}
+%      \toprule
+%      {Header} & {Header} & {Header} \\
+%      \midrule
+%      14.2    & 14.2    & 14.2    \\
+%      1.23456 & 1.23456 & 1.23456 \\
+%      1.2e3   & 1.2e3   & 1.2e3   \\
+%      \bottomrule
+%    \end{tabular}
+%  \end{table}
+%\end{LaTeXexample}
+%
 %\subsection{Units}
 %\DescribeOption{per}
 % Most of the unit options are concerned with the processing of named
@@ -2330,6 +2402,8 @@ dcolumn, rccol and numprint packages.
 %  expproduct & List or literal & Product sign for exponents \\
 %  fixdp & Boolean & Switch for fixing decimal \\
 %   & & places of numbers \\
+%  fixsf & Boolean & Switch for fixing significant \\
+%    & & figures of numbers \\
 %  fraction & List & Method used when setting \opt{per=frac} \\
 %  inlinebold & List & Select how inline bold is tested \\
 %  load & Modules & Modules to load \\
@@ -2379,8 +2453,9 @@ dcolumn, rccol and numprint packages.
 %  numaddn & Literal & Additional input allowed in numbers \\
 %  numcloseerr & Literal & Character indicating end of \\
 %   & & numerical error\\
-%  numdecimal & Literal & Decimal symbols in numbers\\
+%  numdecimal & Literal & Decimal symbols in numbers \\
 %  numdigits & Literal & Digit characters in numbers \\
+%  numdiv & Literal & Division characters in numbers \\
 %  numexp & Literal & Exponent characters in numbers \\
 %  numgobble & Literal & Characters to ignore in numbers \\
 %  numopenerr & Literal & Character indicating start of \\
@@ -2408,6 +2483,8 @@ dcolumn, rccol and numprint packages.
 %  retainplus & Boolean & Retain explicit plus sign\\
 %  seperr & Boolean & Separate number and error \\
 %  sepfour & Boolean & Separate four-digit numbers \\
+%  sf & Integer & Number of significant figures \\
+%   & & to output numbers to \\
 %  sign & List or literal & Sign to add to numbers \\
 %  slash & List or literal & Symbol used for ``/'' \\
 %  stickyper & Boolean & Require \cs{per} only once \\
@@ -2420,6 +2497,7 @@ dcolumn, rccol and numprint packages.
 %   & & length given by \opt{tabformat} \\
 %  tabformat & Number & Space reserved in table for numbers \\
 %  tabnumalign & List & Alignment of |S| column numbers \\
+%  tabparseonly & Boolean & Do not align |S| columns \\
 %  tabtextalign & List & Positioning of text in |S| columns \\
 %  tabunitalign & List & Positioning of units in |s| columns \\
 %  textcelsius & Literal & ``$^{\circ}$C'' symbol in text mode \\
@@ -3236,7 +3314,7 @@ author.
 %    \begin{macrocode}
 \NeedsTeXFormat{LaTeX2e}
 \ProvidesPackage{siunitx}
-  [2008/09/03 v1.0l A comprehensive (SI) units package]
+  [2008/09/28 v1.1 A comprehensive (SI) units package]
 %    \end{macrocode}
 % The package requires \eTeX, so the usual test is made.
 %    \begin{macrocode}
@@ -3261,6 +3339,7 @@ author.
   \catcode\string`\string ` \the\catcode\string`\`\relax
   \catcode\string`\string = \the\catcode\string`\=\relax
   \catcode\string`\string ^ \the\catcode\string`\^\relax
+  \catcode\string`\string _ \the\catcode\string`\_\relax
   \catcode\string`\string ~ \the\catcode\string`\~\relax
   \catcode\string`\string : \the\catcode\string`\:\relax
   \catcode\string`\string - \the\catcode\string`\-\relax
@@ -3271,6 +3350,7 @@ author.
 \catcode\string`\` 12\relax
 \catcode`\= 12\relax
 \catcode`\^ 7\relax
+\catcode`\_ 8\relax
 \catcode`\~ \active\relax
 \@makeother{\:}
 \@makeother{\-}
@@ -4051,7 +4131,9 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\changes{v1.1}{2008/09/22}{Added \opt{numdiv} option}
 %\begin{macro}{\si@numdigits}
+%\begin{macro}{\si@numdiv}
 %\begin{macro}{\si@numdecimal}
 %\begin{macro}{\si@numexp}
 %\begin{macro}{\si@numprod}
@@ -4065,7 +4147,7 @@ author.
 % and with characters ignored with no output renamed as \opt{gobble}.
 %    \begin{macrocode}
 \si@opt@cmdkeys{numdigits,numdecimal,numexp,numgobble,numsign,%
-  numcloseerr,numopenerr,numaddn,numprod}
+  numcloseerr,numopenerr,numaddn,numprod,numdiv}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
@@ -4076,6 +4158,7 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\si@numvalid}
 %\begin{macro}{\si@numextra}
 % The various valid characters are collected together in a single
@@ -4086,7 +4169,7 @@ author.
 \newcommand*{\si@numextra}{%
   \si@numopenerr\si@numcloseerr\si@numaddn}
 \newcommand*{\si@numvalid}{\si@numgobble\si@numexp\si@numsign
-  \si@numdecimal\si@numdigits\si@numextra\si@numprod}
+  \si@numdecimal\si@numdigits\si@numextra\si@numprod\si@numdiv}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
@@ -4354,7 +4437,9 @@ author.
 \si@opt@boolkey{strictarc}
 %    \end{macrocode}
 %\end{macro}
+%\changes{v1.1}{2008/09/03}{Added \opt{tabparseonly} option}
 %\begin{macro}{\ifsi@tab@fixed}
+%\begin{macro}{\ifsi@tabparseonly}
 %\begin{macro}{\si@tabnumalign}
 %\begin{macro}{\si@tab@rfill@S}
 %\begin{macro}{\si@tab@lfill@S}
@@ -4363,6 +4448,7 @@ author.
 % is used, and the fills used for achieving this.
 %    \begin{macrocode}
 \newif\ifsi@tab@fixed
+\si@opt@boolkey{tabparseonly}
 \si@opt@choicekey{tabnumalign}
   {centredecimal,centerdecimal,right,left,centre,center}
   {\si@tab@fixedtrue
@@ -4391,6 +4477,7 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\ifsi@tabalignexp}
 % A switch for aligning exponents.
 %    \begin{macrocode}
@@ -4548,20 +4635,49 @@ author.
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\ifsi@fixdp}
+%\changes{v1.1}{2008/08/22}{Modified to work with \opt{fixsf} option}
 %\begin{macro}{\si@num@dp}
+%\changes{v1.1}{2008/08/22}{Modified to work with \opt{fixsf} option}
 % To allow control of rounding, two options are needed.  One sets how
 % many fixed digits to use, the second turns this function on and
 % off.
 %    \begin{macrocode}
-\si@opt@boolkey{fixdp}
+\si@opt@boolkey[%
+  \ifsi@fixdp
+    \expandafter\si@fixsffalse
+  \fi]{fixdp}
 \newcount\si@num@dp
 \si@opt@key{dp}{%
   \si@str@ifonlychrs{#1}{0123456789}
-    {}
+    {\si@num@dp#1\relax
+     \si@fixdptrue
+     \si@fixsffalse}
     {\si@log@err{Invalid input for dp option}
-      {The dp option must be given a positive integer}}%
-  \si@num@dp#1\relax
-  \si@fixdptrue}
+      {The dp option must be given a positive integer}}}
+%    \end{macrocode}
+%\end{macro}
+%\end{macro}
+%\begin{macro}{\ifsi@fixsf}
+%\begin{macro}{\si@num@sf}
+% A similar approach for fixing significant figures.
+%    \begin{macrocode}
+\si@opt@boolkey[%
+  \ifsi@fixsf
+    \expandafter\si@fixdpfalse
+  \fi]{fixsf}
+\newcount\si@num@sf
+\si@opt@key{sf}{%
+  \si@str@ifonlychrs{#1}{0123456789}
+    {\si@num@sf#1\relax
+     \ifnum\si@num@sf=\z@\relax
+       \si@log@warn{Setting sf=0 disables function}%
+       \si@fixsffalse
+     \else
+       \si@fixsftrue
+       \si@fixdpfalse
+     \fi}
+    {\si@log@err{Invalid input for sf option}
+      {The sf option must be given a positive integer}}}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
@@ -5799,6 +5915,7 @@ author.
     \def\,{}%
     \catcode`\~=\active\relax
     \catcode`\^=\active\relax
+    \catcode`\_=\active\relax
     \everyeof{\noexpand}%
     \endlinechar\m@ne
     \protected@xdef\si@tempa{\scantokens{#1}}%
@@ -6648,6 +6765,9 @@ author.
      \expandafter\si@num@checkerr
    \fi}%
 %    \end{macrocode}
+%\changes{v1.1}{2008/08/22}{Fixed-length code generalised}
+%\changes{v1.1}{2008/08/26}{Length-fixing only applies to
+%  manitssa}
 % If both parts of the number contain only digits, then any rounding
 % can be attempted if there is no error part.  Otherwise, the input
 % must be left alone.
@@ -6656,8 +6776,10 @@ author.
   \expandafter\si@str@ifonlychrs\expandafter{\si@tempa}
     {0123456789}
     {\ifx\@empty\si@num@err\@empty
-       \ifsi@fixdp
-         \expandafter\expandafter\expandafter\si@num@fixdp
+       \renewcommand*{\si@tempb}{mant}%
+       \renewcommand*{\si@tempc}{#1}%
+       \ifx\si@tempb\si@tempc
+         \expandafter\expandafter\expandafter\si@num@fixlength
        \fi
      \fi}{}%
 %    \end{macrocode}
@@ -6894,6 +7016,8 @@ author.
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\si@num@largeerr}
+%\changes{v1.0m}{2008/09/16}{Fixed problem with error separation
+%  crossing decimal boundary}
 %\begin{macro}{\si@num@lerr}
 %\begin{macro}{\si@num@movedigit}
 % When the error has more digits than the decimal part, some
@@ -6904,7 +7028,7 @@ author.
   \si@tempcntb\si@tempcnta\relax
   \si@num@lerr
   \protected@edef\si@num@err{%
-    \si@tempa\ensuremath{\si@decimalsymbol}\si@tempb}}
+    \si@tempa\expandafter\@car\si@numdecimal\@nil\si@tempb}}
 \newcommand*{\si@num@lerr}{%
   \si@num@cntdigits{\si@tempb}%
   \ifnum\si@tempcnta=\si@tempcntb\relax\else
@@ -6918,6 +7042,19 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\begin{macro}{\si@num@fixlength}
+% The checks for fixing number length are spun out here.
+%    \begin{macrocode}
+\newcommand*{\si@num@fixlength}{%
+  \ifsi@fixdp
+    \expandafter\si@num@fixdp
+  \else
+    \ifsi@fixsf
+      \expandafter\expandafter\expandafter\si@num@fixsf
+    \fi
+  \fi}
+%    \end{macrocode}
+%\end{macro}
 %\begin{macro}{\si@num@fixdp}
 % The test for fixing decimal places starts by counting up the number
 % of decimal digits.  The number is then padded, rounded or left
@@ -6951,9 +7088,19 @@ author.
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
+%\begin{macro}{\si@num@rndtarget}
+%\changes{v1.1}{2008/08/25}{New counter}
+% A scratch counter is created to allow some code sharing between
+% fixing decimal places and significant figures.
+%    \begin{macrocode}
+\newcount\si@num@rndtarget
+%    \end{macrocode}
+%\end{macro}
 %\begin{macro}{\si@num@round}
 %\begin{macro}{\si@num@prernd}
 %\begin{macro}{\si@num@postrnd}
+%\begin{macro}{\si@num@rndstart}
+%\changes{v1.1}{2008/08/25}{Separated out from \cs{si@num@round}}
 % Rounding a number is more complicated.  The main macro here relies
 % on several others, and is simply a set-up and hand-off system.
 %    \begin{macrocode}
@@ -6961,6 +7108,9 @@ author.
 \newcommand*{\si@num@postrnd}{}
 \newcommand*{\si@num@round}{%
   \si@log@debug{Rounding to \the\si@num@dp\space digits}%
+  \si@num@rndtarget\si@num@dp\relax
+  \si@num@rndstart}
+\newcommand*{\si@num@rndstart}{%
   \si@num@reverse{\si@num@postdec}%
   \si@num@reverse{\si@num@predec}%
   \let\si@num@prernd\si@num@predec
@@ -6973,6 +7123,7 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\si@num@reverse}
 %\begin{macro}{\si@num@rev}
 %\darg{storage-macro}
@@ -6993,6 +7144,8 @@ author.
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\si@num@rnd}
+%\changes{v1.1}{2008/08/25}{Altered check for ending to generalise
+%  code}
 %\begin{macro}{\si@num@rndpre}
 %\begin{macro}{\si@num@rndpost}
 %\changes{v1.0b}{2008/06/23}{Corrected bug in rounding code}
@@ -7005,10 +7158,10 @@ author.
 % to achieve the same thing.
 %    \begin{macrocode}
 \newcommand*{\si@num@rnd}{%
-  \ifnum\si@tempcnta>\z@\relax
-    \expandafter\si@num@rndpost
-  \else
+  \ifx\@empty\si@num@postrnd\@empty
     \expandafter\si@num@rndpre
+  \else
+    \expandafter\si@num@rndpost
   \fi}
 \newcommand*{\si@num@rndpre}{%
   \expandafter\edef\expandafter\si@tempa\expandafter{%
@@ -7042,12 +7195,12 @@ author.
     \advance\si@tempcntb\@ne\relax
   \fi
   \si@switchfalse
-  \advance\si@num@dp\@ne\relax
-  \ifnum\si@tempcnta>\si@num@dp\relax
-    \advance\si@num@dp\m@ne\relax
+  \advance\si@num@rndtarget\@ne\relax
+  \ifnum\si@tempcnta>\si@num@rndtarget\relax
+    \advance\si@num@rndtarget\m@ne\relax
   \else
-    \advance\si@num@dp\m@ne\relax
-    \ifnum\si@tempcnta>\si@num@dp\relax
+    \advance\si@num@rndtarget\m@ne\relax
+    \ifnum\si@tempcnta>\si@num@rndtarget\relax
       \ifnum\si@tempcntb>4\relax
         \expandafter\expandafter\expandafter\si@switchtrue
       \fi
@@ -7060,11 +7213,175 @@ author.
     \fi
   \fi
   \advance\si@tempcnta\m@ne\relax
+%    \end{macrocode}
+% If \cs{si@tempcnta} is negative, this must be the significant
+% figures routine and one digit needs to be thrown away.
+%    \begin{macrocode}
+  \ifnum\si@tempcnta=\m@ne\relax
+    \si@num@throw
+  \fi
   \si@num@rnd}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\begin{macro}{\si@num@throw}
+% Throw one digit away from the end of the post-decimal part.
+%    \begin{macrocode}
+\newcommand*{\si@num@throw}{%
+  \si@num@reverse{\si@num@postdec}%
+  \edef\si@num@postdec{\expandafter\@cdr\si@num@postdec\@nil}%
+  \si@num@reverse{\si@num@postdec}}
+%    \end{macrocode}
+%\end{macro}
+%\begin{macro}{\si@num@fixsf}
+% Setting up a fixed number of significant figures is rather more
+% complicated than fixing the decimal places.  There are a number of
+% tests needed to see what needs to be done.
+%    \begin{macrocode}
+\newcommand*{\si@num@fixsf}{%
+  \si@num@cntdigits{\si@num@predec}%
+  \ifx\@empty\si@num@predec\@empty
+    \si@tempcnta\z@\relax
+  \fi
+%    \end{macrocode}
+% If the pre-decimal component has only a single digit, it could be
+% ``0'': if so, only the significant figures in the decimal part are
+% counted.
+%    \begin{macrocode}
+  \ifnum\si@tempcnta=\@ne\relax
+    \renewcommand*{\si@tempa}{0}%
+    \ifx\si@num@predec\si@tempa
+%    \end{macrocode}
+% The number of significant digits after the decimal point have to be
+% counted up. If there are enough digits there, then some rounding
+% will again be needed.
+%    \begin{macrocode}
+      \ifx\@empty\si@num@postdec\@empty\else
+        \si@tempcnta\z@\relax
+        \si@switchfalse
+        \expandafter\si@num@cntpostsf\si@num@postdec
+          \@empty\@empty
+        \ifnum\si@tempcnta>\si@num@sf\relax
+          \si@log@debug{Rounding to \the\si@num@sf\space
+            figures}%
+          \si@num@rndtarget\si@num@sf\relax
+          \si@num@rndstart
+        \fi
+      \fi
+    \else
+      \si@num@cntprepost
+    \fi
+  \else
+%    \end{macrocode}
+% If there are too many digits in the pre-decimal part, a check is
+% first made to see if there may be a need to round from the first
+% post-decimal digit. If not, the post decimal part is thrown away.
+%    \begin{macrocode}
+    \ifnum\si@tempcnta<\si@num@sf\relax
+      \si@num@cntprepost
+    \else
+      \advance\si@tempcnta\m@ne\relax
+      \si@switchfalse
+      \ifnum\si@tempcnta>\si@num@sf\relax\else
+        \ifx\@empty\si@num@postdec\@empty\else
+          \edef\si@num@postdec{\expandafter\@car
+            \si@num@postdec\@nil}%
+          \ifnum\si@num@postdec>4\relax
+            \expandafter\si@switchtrue
+          \fi
+        \fi
+      \fi
+      \advance\si@tempcnta\@ne\relax
+      \renewcommand*{\si@num@postdec}{}%
+      \si@num@reverse{\si@num@predec}%
+      \let\si@num@prernd\si@num@predec
+      \renewcommand*{\si@num@predec}{}%
+      \si@num@sfpre
+      \ifsi@switch
+        \edef\si@num@predec{1\si@num@predec}%
+      \fi
+    \fi
+  \fi}
+%    \end{macrocode}
+%\end{macro}
+%\begin{macro}{\si@num@cntpostsf}
+% Counting up significant figures in the post-decimal part needs an
+% auxiliary macro.
+%    \begin{macrocode}
+\def\si@num@cntpostsf#1#2\@empty{%
+  \ifsi@switch
+    \advance\si@tempcnta\@ne\relax
+  \else
+    \if 0#1\relax\else
+      \si@tempcnta\@ne\relax
+      \expandafter\expandafter\expandafter\si@switchtrue
+    \fi
+  \fi
+  \ifx\@empty#2\@empty\else
+    \si@num@cntpostsf#2\@empty\@empty
+  \fi}
+%    \end{macrocode}
+%\end{macro}
+%\begin{macro}{\si@num@cntprepost}
+% There are two routes to find that the number of digits in both
+% parts of the number are important.  In either case, rounding is
+% only needed if the two parts have too many digits.
+%    \begin{macrocode}
+\newcommand*{\si@num@cntprepost}{%
+  \si@tempcntb\si@tempcnta\relax
+  \si@num@cntdigits{\si@num@postdec}%
+  \advance\si@tempcnta\si@tempcntb\relax
+  \ifnum\si@tempcnta>\si@num@sf
+    \si@log@debug{Rounding to \the\si@num@sf\space figures}%
+    \si@num@rndtarget\si@num@sf\relax
+    \si@num@rndstart
+    \ifsi@switch
+      \si@num@throw
+    \fi
+  \fi}
+%    \end{macrocode}
+%\end{macro}
+%\begin{macro}{\si@num@sfpre}
+% The code here is very similar to that used to round to a fixed
+% number of decimal places.
+%    \begin{macrocode}
+\newcommand*{\si@num@sfpre}{%
+  \expandafter\edef\expandafter\si@tempa\expandafter{%
+    \expandafter\@car\si@num@prernd\@nil}%
+  \expandafter\edef\expandafter\si@num@prernd\expandafter{%
+    \expandafter\@cdr\si@num@prernd\@nil}%
+  \si@tempcntb\si@tempa\relax
+  \ifsi@switch
+    \advance\si@tempcntb\@ne\relax
+  \fi
+  \si@switchfalse
+  \advance\si@num@sf\@ne\relax
+  \ifnum\si@tempcnta>\si@num@sf\relax
+    \advance\si@num@sf\m@ne\relax
+    \edef\si@num@predec{0\si@num@predec}%
+  \else
+    \advance\si@num@sf\m@ne\relax
+    \ifnum\si@tempcnta>\si@num@sf\relax
+      \ifnum\si@tempcntb>4\relax
+        \expandafter\expandafter\expandafter\si@switchtrue
+      \fi
+      \edef\si@num@predec{0\si@num@predec}%
+    \else
+      \ifnum\si@tempcntb=10\relax
+        \si@tempcntb\z@\relax
+        \expandafter\expandafter\expandafter\si@switchtrue
+      \fi
+      \edef\si@num@predec{\the\si@tempcntb\si@num@predec}%
+    \fi
+  \fi
+  \advance\si@tempcnta\m@ne\relax
+  \ifx\@empty\si@num@prernd\@empty\else
+    \expandafter\si@num@sfpre
+  \fi}
+%    \end{macrocode}
+%\end{macro}
+%
 %\begin{macro}{\si@num@int}
 %\darg{integer-part}
 % The formatting code for separating thousands is taken more-or-less
@@ -8052,15 +8369,19 @@ author.
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\si@tab@numout}
+%\changes{v1.1}{2008/09/03}{Extended for non-aligned output}
 % If a number is found, then some secondary processing is needed to
 % format it correctly.
 %    \begin{macrocode}
 \newcommand*{\si@tab@numout}{%
   \si@num@intabtrue
-  \ifsi@tab@fixed
-    \ifsi@tabautofit
-      \si@num@dp\si@tab@mantpostcnt\relax
-      \expandafter\expandafter\expandafter\si@fixdptrue
+  \ifsi@tabparseonly\else
+    \ifsi@tab@fixed
+      \ifsi@tabautofit
+        \si@num@dp\si@tab@mantpostcnt\relax
+        \expandafter\expandafter\expandafter\expandafter
+          \expandafter\expandafter\expandafter\si@fixdptrue
+      \fi
     \fi
   \fi
   \expandafter\si@num\expandafter{\the\si@tab@toks}%
@@ -8086,16 +8407,21 @@ author.
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\si@tab@format}
+%\changes{v1.1}{2008/09/03}{Extended for non-aligned output}
 % The formatting set up is taken from \pkg{dcolumn}, but with
 % control of the output form the stored information.  The choice
 % of a variable (decimal-centred) column or fixed width boxes is
 % made.
 %    \begin{macrocode}
 \newcommand*{\si@tab@format}{%
-  \ifsi@tab@fixed
-    \expandafter\si@tab@fixed
+  \ifsi@tabparseonly
+    \expandafter\si@tab@outonly
   \else
-    \expandafter\si@tab@unfixed
+    \ifsi@tab@fixed
+      \expandafter\expandafter\expandafter\si@tab@fixed
+    \else
+      \expandafter\expandafter\expandafter\si@tab@unfixed
+    \fi
   \fi
 %    \end{macrocode}
 % A hack to get the correct colour everywhere without too much work.
@@ -8103,10 +8429,30 @@ author.
   \ifsi@colourvalues
     \si@fam@colourcmd{\si@valuecolour}%
   \fi
-  \box\si@tab@prebox\box\si@tab@midbox\box\si@tab@postbox%
+  \box\si@tab@prebox\box\si@tab@midbox\box\si@tab@postbox
     \box\si@tab@expbox}
 %    \end{macrocode}
 %\end{macro}
+%\begin{macro}{\si@tab@outonly}
+% The non-alignment option means that no real processing is done: only font
+% control is required.
+%    \begin{macrocode}
+\newcommand*{\si@tab@outonly}{%
+  \si@log@debug{Using non-aligned S column}%
+  \setbox\si@tab@prebox=\hbox{}%
+  \setbox\si@tab@midbox=\hbox{}%
+  \setbox\si@tab@postbox=\hbox{}%
+  \setbox\si@tab@expbox=\hbox{}%
+    \protected@edef\si@tempa{%
+      \si@tab@out
+    \ifx\@empty\si@num@out\@empty\else
+      \ensuremath{{\si@decimalsymbol}}%
+    \fi
+    \si@num@out
+    \si@tab@expout}%
+  \expandafter\si@out@num\expandafter{\si@tempa}}
+%    \end{macrocode}
+%\end{macro}
 %\begin{macro}{\si@tab@unfixed}
 %\changes{v1.0l}{2008/09/03}{Fixed issue with decimal sign and
 %  integers}
@@ -8116,7 +8462,7 @@ author.
 % in boxes.  The wider one is then used to set up the column width,
 % by resizing the other one.
 %    \begin{macrocode}
-\newcommand*\si@tab@unfixed{%
+\newcommand*{\si@tab@unfixed}{%
   \si@log@debug{Using variable width S column}%
   \ifx\@empty\si@num@out\@empty
     \setbox\si@tab@midbox=\hbox
@@ -8246,8 +8592,7 @@ author.
   \setbox\si@tab@prebox=\hbox to\si@tab@predim{\hss\hfill
     \expandafter\si@out@num\expandafter{\si@tab@out}}%
   \ifx\@empty\si@num@out\@empty
-    \setbox\si@tab@postbox=\hbox to\si@tab@postdim
-      {\expandafter\si@out@num\expandafter{\si@num@out}\hfil}%
+    \setbox\si@tab@postbox=\hbox to\si@tab@postdim{\hfil}%
   \else
     \ifsi@tabalignexp\else
       \protected@edef\si@num@out{\si@num@out\si@tab@expout}%
@@ -8419,6 +8764,33 @@ author.
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\begin{macro}{\newqualifier}
+%\changes{v1.1}{2008/09/04}{New user macro}
+%\begin{macro}{\renewqualifier}
+%\changes{v1.1}{2008/09/04}{New user macro}
+%\begin{macro}{\providequalifier}
+%\changes{v1.1}{2008/09/04}{New user macro}
+%\darg{qualifier}
+%\darg{symbol}
+% Creating unit qualifiers is much the same method.
+%    \begin{macrocode}
+\newcommand*{\newqualifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@unt@defqual{#1}{#2}}
+    {\si@log@err{Qualifier \string#1 already defined!}\@eha}}
+\newcommand*{\renewqualifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@log@err{Qualifier \string#1 undefined}\@ehc
+     \si@unt@defqual{#1}{#2}}
+    {\si@unt@defqual{#1}{#2}}}
+\newcommand*{\providequalifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@unt@defqual{#1}{#2}}
+    {}}
+%    \end{macrocode}
+%\end{macro}
+%\end{macro}
+%\end{macro}
 %\begin{macro}{\ifsi@unt@num}
 % A flag is needed to tell the processor whether there is a number,
 % to get the correct spacing.
@@ -8647,8 +9019,8 @@ author.
     \renewcommand*{\si@unt@spstack}{}%
     \renewcommand*{\si@unt@stacka}{}%
     \renewcommand*{\si@unt@stackb}{}%
-    \renewcommand*{\si@unt@holdstacka}{}%
-    \renewcommand*{\si@unt@holdstackb}{}%
+    \renewcommand*{\si@unt@spacestacka}{}%
+    \renewcommand*{\si@unt@spacestackb}{}%
     \renewcommand*{\si@unt@lastadda}{space}%
     \renewcommand*{\si@unt@lastaddb}{space}}
 %    \end{macrocode}
@@ -8821,6 +9193,28 @@ author.
     {#2}{#3}}}
 %    \end{macrocode}
 %\end{macro}
+%\begin{macro}{\si@unt@defqual}
+%\darg{qualifier}
+%\darg{symbol}
+% Qualifiers are defined in a similar manner.
+%    \begin{macrocode}
+\newcommand*{\si@unt@defqual}[2]{%
+  \si@log@debug{Declaring qualifier \string#1 with\MessageBreak
+    meaning \string#2}%
+  \protected\def#1{%
+    \ifsi@unt@littest
+      \expandafter\@gobble
+    \else
+      \ifsi@unt@litout
+        \expandafter\expandafter\expandafter\si@unt@litqual
+      \else
+        \expandafter\expandafter\expandafter\si@unt@qual
+      \fi
+    \fi
+    {#2}}}
+%    \end{macrocode}
+%\end{macro}
+%
 %\begin{macro}{\si@unt@unithook}
 %\begin{macro}{\si@unt@unit}
 %\darg{number}
@@ -9107,6 +9501,7 @@ author.
 %\begin{macro}{\si@unt@stackpower}
 %\begin{macro}{\si@unt@stkpower}
 %\begin{macro}{\si@unt@stkpwr}
+%\begin{macro}{\si@unt@skpwr}
 % Adding powers to the stack should also clear the power list.  If
 % the number is already zero, then  of course nothing happens.
 %    \begin{macrocode}
@@ -9133,6 +9528,7 @@ author.
     \expandafter\si@unt@stkpwr
   \fi}
 %    \end{macrocode}
+%\changes{v1.1}{2008/09/04}{Altered to account for qualifiers}
 % Finally, the actual adding (set up to avoid problems with the
 % \cs{if} above).
 %    \begin{macrocode}
@@ -9142,17 +9538,27 @@ author.
     \expandafter\ifx
       \csname si@unt@lastadd\si@unt@checkstack\endcsname
       \si@tempa
-      \si@log@debug{Adding power
-        \strip@pt\si@unt@powerdim\space to output stack}%
-      \si@unt@addtostack{power}{^{\num[fixdp=false]{%
-        \strip@pt\si@unt@powerdim}}}%
+      \si@unt@skpwr
+    \else
+      \renewcommand*{\si@tempa}{qual}%
+      \expandafter\ifx
+        \csname si@unt@lastadd\si@unt@checkstack\endcsname
+        \si@tempa
+        \si@unt@skpwrqual
+      \fi
     \fi
   \fi
   \si@unt@powerdim\z@\relax}
+\newcommand*{\si@unt@skpwr}{%
+  \si@log@debug{Adding power
+    \strip@pt\si@unt@powerdim\space to output stack}%
+  \si@unt@addtostack{power}{^{\num[fixdp=false]{%
+    \strip@pt\si@unt@powerdim}}}}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\si@unt@invpower}
 % A macro to change the sign of the current power.
 %    \begin{macrocode}
@@ -9168,6 +9574,45 @@ author.
   \fi}
 %    \end{macrocode}
 %\end{macro}
+%\begin{macro}{\si@unt@litqual}
+%    \begin{macrocode}
+\newcommand*{\si@unt@litqual}[1]{(#1)}
+%    \end{macrocode}
+%\end{macro}
+%\begin{macro}{\si@unt@qual}
+%    \begin{macrocode}
+\newcommand*{\si@unt@qual}[1]{%
+  \ifsi@unt@first
+    \si@log@err{Qualifier before units}
+      {A unit qualifier can only be given after\MessageBreak
+       one or more units.  You tried to open a unit
+       with\MessageBreak \string#1}%
+  \fi
+%    \end{macrocode}
+% The space after the current unit is shuffled so it will come
+% after the qualifier.
+%    \begin{macrocode}
+  \expandafter\let\expandafter\si@unt@spacestack
+    \csname si@unt@spacestack\si@unt@checkstack\endcsname
+  \expandafter\let\csname si@unt@spacestack\si@unt@checkstack
+    \endcsname\@empty
+%    \end{macrocode}
+%\begin{macro}{\si@unt@skpwrqual}
+% Some protection against trying to add two powers.
+%    \begin{macrocode}
+  \renewcommand*{\si@tempa}{power}%
+    \expandafter\ifx
+    \csname si@unt@lastadd\si@unt@checkstack\endcsname\si@tempa
+    \let\si@unt@skpwrqual\relax
+  \else
+    \let\si@unt@skpwrqual\si@unt@skpwr
+  \fi
+  \si@unt@addtostack{qual}{_{#1}}%
+  \expandafter\let\csname si@unt@spacestack\si@unt@checkstack
+    \endcsname\si@unt@spacestack}
+%    \end{macrocode}
+%\end{macro}
+%\end{macro}
 %\begin{macro}{\ifsi@unt@per}
 %\begin{macro}{\ifsi@unt@perseen}
 %\begin{macro}{\per}
@@ -9317,15 +9762,18 @@ author.
     \@ne\relax}
 %    \end{macrocode}
 %\end{macro}
-%\begin{macro}{\si@unt@holdstacka}
-%\begin{macro}{\si@unt@holdstackb}
+%\begin{macro}{\si@unt@spacestack}
+%\begin{macro}{\si@unt@spacestacka}
+%\begin{macro}{\si@unt@spacestackb}
 % The stacked material needs somewhere to live.
 %    \begin{macrocode}
-\newcommand*{\si@unt@holdstacka}{}
-\newcommand*{\si@unt@holdstackb}{}
+\newcommand*{\si@unt@spacestack}{}
+\newcommand*{\si@unt@spacestacka}{}
+\newcommand*{\si@unt@spacestackb}{}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\si@unt@holdspace}
 %\begin{macro}{\si@unt@addstack}
 %\darg{token}
@@ -9339,20 +9787,18 @@ author.
   \ifx\si@tempa\si@tempb
     \ifx\@empty\si@unt@stackb\@empty
     \else
-      \expandafter\protected@edef
-      \csname si@unt@holdstack\si@unt@checkstack\endcsname{#1}%
+      \protected@edef\si@unt@spacestackb{#1}%
     \fi
   \else
-    \expandafter\protected@edef
-    \csname si@unt@holdstack\si@unt@checkstack\endcsname{#1}%
+    \protected@edef\si@unt@spacestacka{#1}%
   \fi}
 \newcommand*{\si@unt@addstack}[1]{%
   \expandafter\protected@edef
     \csname si@unt@stack\si@unt@checkstack\endcsname
     {\csname si@unt@stack\si@unt@checkstack\endcsname
-     \csname si@unt@holdstack\si@unt@checkstack\endcsname#1}%
+     \csname si@unt@spacestack\si@unt@checkstack\endcsname#1}%
   \expandafter\renewcommand\expandafter*\expandafter{%
-    \csname si@unt@holdstack\si@unt@checkstack\endcsname}{}}
+    \csname si@unt@spacestack\si@unt@checkstack\endcsname}{}}
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
@@ -9769,6 +10215,7 @@ author.
 %
 %\subsection{Output routine}
 %\begin{macro}{\si@out}
+%\changes{v1.1}{2008/07/22}{Extended handling to subscripts}
 %\darg{text}
 % With all of the setup done, the text can finally be typeset.  This
 % is done inside a \cs{text} block, which takes care of
@@ -9780,13 +10227,15 @@ author.
 \begingroup
   \catcode`\^=\active
   \catcode`\-=\active
+  \catcode`\_=\active
   \gdef\si@out#1{%
     \begingroup
       \catcode`\^=\active
+      \catcode`\_=\active
       \makeatletter
       \endlinechar\m@ne
 %    \end{macrocode}
-%\changes{1.0f}{2008/07/17}{Fixed bug with colour and spacing}
+%\changes{v1.0f}{2008/07/17}{Fixed bug with colour and spacing}
 % The various font families can now be set up.  First a check is made
 % in case there are nested calls to \cs{si@out@text}.
 %    \begin{macrocode}
@@ -9817,28 +10266,36 @@ author.
 %    \begin{macrocode}
   \gdef\si@out@text#1{%
     \let^\si@out@sp
+    \let_\si@out@sb
     \let\textsuperscript\si@out@sp
+    \let\textsubscript\si@out@sb
     \catcode`\-=\active\relax
     \let-\si@out@minus
     #1}
   \gdef\si@out@maths#1{%
     \let^\sp
+    \let_\sb
     \let\textsuperscript\sp
+    \let\textsubscript\sb
     $\si@fam@maths{#1}$}
 \endgroup
 %    \end{macrocode}
 %\end{macro}
 %\end{macro}
 %\begin{macro}{\si@out@sp}
+%\begin{macro}{\si@out@sb}
 %\darg{text}
 % \cs{textsuperscript} gives slightly different alignment of numbers
 % to using |^| in text mode.  To avoid this, a slightly different
-% definition is used.  Elsewhere \cs{textsuperscript} is used, as
-% the code above sorts out the text/maths mode issues.
+% definition is used.  Elsewhere \cs{textsuperscript} is used, as the
+% code above sorts out the text/maths mode issues. The same is done
+% for subscripts.
 %    \begin{macrocode}
 \newcommand*{\si@out@sp}[1]{\ensuremath{^{\text{#1}}}}
+\newcommand*{\si@out@sb}[1]{\ensuremath{_{\text{#1}}}}
 %    \end{macrocode}
 %\end{macro}
+%\end{macro}
 %\begin{macro}{\si@out@minus}
 % An active minus sign is needed.
 %    \begin{macrocode}
@@ -10042,9 +10499,10 @@ author.
   negcolour=red,
   noload={},
   numaddn={},%(
-  numcloseerr=),%
+  numcloseerr=),
   numdecimal={.,},
   numdigits=0123456789,
+  numdiv=/,
   numexp=eEdD,
   numgobble={},
   numopenerr=(,%)
@@ -10075,6 +10533,7 @@ author.
   tabautofit=false,
   tabformat=3.2,
   tabnumalign=centredecimal,
+  tabparseonly=false,
   tabtextalign=centre,
   tabunitalign=left,
   textrm=rmfamily,
@@ -10177,7 +10636,7 @@ author.
 % powers.
 %    \begin{macrocode}
 \ProvidesFile{si-prefix.cfg}
-  [2008/09/03 v1.0l siunitx: SI Multiple prefixes]
+  [2008/09/28 v1.1 siunitx: SI Multiple prefixes]
 \newprefix{\yocto}{-24}{y}
 \newprefix{\zepto}{-21}{z}
 \newprefix{\atto}{-18}{a}
@@ -10281,7 +10740,7 @@ author.
 % is an awkward one, but here the UK standard is used.
 %    \begin{macrocode}
 \ProvidesFile{si-named.cfg}
-  [2008/09/03 v1.0l siunitx: SI Named units]
+  [2008/09/28 v1.1 siunitx: SI Named units]
 \newunit{\becquerel}{Bq}
 \newunit{\coulomb}{C}
 \newunit{\farad}{F}
@@ -10377,7 +10836,7 @@ author.
 % This requires \file{si-prefix.cfg} and \file{si-named.cfg}.
 %    \begin{macrocode}
 \ProvidesFile{si-prefixed.cfg}
-  [2008/09/03 v1.0l siunitx: SI Prefixed units]
+  [2008/09/28 v1.1 siunitx: SI Prefixed units]
 \requiresiconfigs{prefix,named,accepted,physical}
 %    \end{macrocode}
 %\begin{macro}{\picometre}
@@ -10663,7 +11122,7 @@ author.
 % block for the appropriate option.  First currents.
 %    \begin{macrocode}
 \ProvidesFile{si-abbr.cfg}
-  [2008/09/03 v1.0l siunitx: Abbreviated units]
+  [2008/09/28 v1.1 siunitx: Abbreviated units]
 \requiresiconfigs{prefix,named,accepted,physical}
 \newunit{\pA}{\pico\ampere}
 \newunit{\nA}{\nano\ampere}
@@ -10875,7 +11334,7 @@ author.
 % These are defined here, although some are in very general use.
 %    \begin{macrocode}
 \ProvidesFile{si-addn.cfg}
-  [2008/09/03 v1.0l siunitx: SI Additional units]
+  [2008/09/28 v1.1 siunitx: SI Additional units]
 \newunit{\angstrom}{\si@sym@ringA}
 \newunit{\are}{a}
 \newunit{\hectare}{\hecto\are}
@@ -10924,7 +11383,7 @@ author.
 %\begin{macro}{\percent}
 %    \begin{macrocode}
 \ProvidesFile{si-accepted.cfg}
-  [2008/09/03 v1.0l siunitx: SI Accepted units]
+  [2008/09/28 v1.1 siunitx: SI Accepted units]
 \newunit{\minute}{min}
 \newunit{\hour}{h}
 \newunit{\Day}{d}
@@ -10973,7 +11432,7 @@ author.
 % for a negative kern does make things a bit complicated.
 %    \begin{macrocode}
 \ProvidesFile{si-physical.cfg}
-  [2008/09/03 v1.0l siunitx: SI Physically-measured units]
+  [2008/09/28 v1.1 siunitx: SI Physically-measured units]
 \newcommand*{\si@eVspacea}{\text{\kern-\si@eVcorra}}%
 \newcommand*{\si@eVspaceb}{\text{\kern-\si@eVcorrb}}%
 \newunit{\electronvolt}{e\protect\si@eVspacea V\protect%
@@ -11009,7 +11468,7 @@ author.
 % own definitions.
 %    \begin{macrocode}
 \ProvidesFile{si-synchem.cfg}
-  [2008/09/03 v1.0l siunitx: Units for synthetic chemists]
+  [2008/09/28 v1.1 siunitx: Units for synthetic chemists]
 \requiresiconfigs{prefix}
 \newunit{\mmHg}{mmHg}
 \newunit{\molar}{\mole\per\cubic\deci\metre}
@@ -11030,7 +11489,7 @@ author.
 % Some units inspired by \pkg{hepunits}.
 %    \begin{macrocode}
 \ProvidesFile{si-hep.cfg}
-  [2008/09/03 v1.0l siunitx: Units for high-energy physics]
+  [2008/09/28 v1.1 siunitx: Units for high-energy physics]
 \requiresiconfigs{prefix,named}
 %    \end{macrocode}
 %\begin{macro}{\micron}
@@ -11112,7 +11571,7 @@ author.
 % For astronomy, the \cs{parsec} unit is needed.
 %    \begin{macrocode}
 \ProvidesFile{si-astro.cfg}
-  [2008/09/03 v1.0l siunitx: Units for astronomy]
+  [2008/09/28 v1.1 siunitx: Units for astronomy]
 \newunit{\parsec}{pc}
 \newunit{\lightyear}{ly}
 %    \end{macrocode}
@@ -11135,7 +11594,7 @@ author.
 % \SIunits. First, the binary prefixes.
 %    \begin{macrocode}
 \ProvidesFile{si-binary.cfg}
-  [2008/09/03 v1.0l siunitx: Binary units]
+  [2008/09/28 v1.1 siunitx: Binary units]
 \newprefix[binary]{\kibi}{10}{Ki}
 \newprefix[binary]{\mebi}{20}{Mi}
 \newprefix[binary]{\gibi}{30}{Gi}
@@ -11178,7 +11637,7 @@ author.
 % default values.\footnote{The package author is in the UK.}
 %    \begin{macrocode}
 \ProvidesFile{si-UK.cfg}
-  [2008/09/03 v1.0l siunitx: UK locale]
+  [2008/09/28 v1.1 siunitx: UK locale]
 \sisetup{
   unitsep=thin,
   expproduct=times,
@@ -11195,7 +11654,7 @@ author.
 % The same as for the UK.
 %    \begin{macrocode}
 \ProvidesFile{si-USA.cfg}
-  [2008/09/03 v1.0l siunitx: USA locale]
+  [2008/09/28 v1.1 siunitx: USA locale]
 \sisetup{
   unitsep=thin,
   expproduct=times,
@@ -11212,7 +11671,7 @@ author.
 % Germany, hopefully.
 %    \begin{macrocode}
 \ProvidesFile{si-DE.cfg}
-  [2008/09/03 v1.0l siunitx: Germany locale]
+  [2008/09/28 v1.1 siunitx: Germany locale]
 \sisetup{
   unitsep=cdot,
   valuesep=thin,
@@ -11229,7 +11688,7 @@ author.
 % Design decisions taken from the same section of \SIstyle.
 %    \begin{macrocode}
 \ProvidesFile{si-ZA.cfg}
-  [2008/09/03 v1.0l siunitx: South Africa locale]
+  [2008/09/28 v1.1 siunitx: South Africa locale]
 \sisetup{
   unitsep=cdot,
   valuesep=thin,
@@ -11261,7 +11720,7 @@ author.
 % \units for conflicting options, so users probably expect none.
 %    \begin{macrocode}
 \ProvidesFile{si-units.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of units]
+  [2008/09/28 v1.1 siunitx: Emulation of units]
 \si@emulating{units}{1998/08/04 v0.9b}
 \si@ifloaded{SIunits}
   {\si@emclash{units}{SIunits}\endinput}{}
@@ -11331,7 +11790,8 @@ author.
 % clash.
 %    \begin{macrocode}
 \ProvidesFile{si-unitsdef.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of unitsdef]
+  [2008/09/28 v1.1 siunitx:
+   Emulation of unitsdef]
 \si@emulating{unitsdef}{2005/01/04 v0.2}
 \si@ifloaded{SIunits}
   {\si@emclash{unitsdef}{SIunits}\endinput}{}
@@ -11779,7 +12239,8 @@ author.
 % the macros in \SIstyle as given in the manual for that package.
 %    \begin{macrocode}
 \ProvidesFile{si-sistyle.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of SIstyle]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     SIstyle]
 \si@emulating{sistyle}{2006/12/20 v2.3}
 \sisetup{%
   sepfour=true,
@@ -11962,7 +12423,8 @@ author.
 % \SIunits emulation starts in much the same way.
 %    \begin{macrocode}
 \ProvidesFile{si-SIunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of SIunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     SIunits]
 \si@emulating{SIunits}{2007/12/02 v1.36}
 \sisetup{
   unitsep=thick,
@@ -12523,7 +12985,8 @@ author.
 % which are not really to be encouraged.
 %    \begin{macrocode}
 \ProvidesFile{si-hepunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of hepunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     hepunits]
 \si@emulating{hepunits}{2007/09/27}
 \requiresiconfigs{SIunits,accepted,prefix,hep}
 %    \end{macrocode}
@@ -12677,7 +13140,8 @@ author.
 % a little emulation and a few macros needed.
 %    \begin{macrocode}
 \ProvidesFile{si-fancynum.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of fancynum]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     fancynum]
 \si@emulating{fancynum}{2000/08/08 0.92}
 \sisetup{decimalsymbol=cdot,digitsep=comma}
 %    \end{macrocode}
@@ -12738,7 +13202,8 @@ author.
 % test is needed.
 %    \begin{macrocode}
 \ProvidesFile{si-fancyunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of fancyunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     fancyunits]
 \si@emulating{fancyunits}{2007/02/01 v1.0.1}
 \si@ifloaded{SIunits}
   {\si@log@err{SIunits emulation loaded\MessageBreak before
@@ -12950,14 +13415,6 @@ author.
 \restylefloat{table}
 \labelformat{figure}{\figurename~#1}
 \labelformat{table}{\tablename~#1}
-\ifpdf
-  \usepackage{embedfile}
-  \embedfile[%
-    stringmethod=escape,%
-    mimetype=plain/text,%
-    desc={LaTeX docstrip source archive for package `\jobname'}%
-    ]{\jobname.dtx}
-\fi
 \IfFileExists{\jobname.sty}
   {\usepackage{\jobname}}{}
 \usepackage[numbered]{hypdoc}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-DE.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-DE.cfg
index 46f0c886d01..61767cdce16 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-DE.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-DE.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-DE.cfg}
-  [2008/09/03 v1.0l siunitx: Germany locale]
+  [2008/09/28 v1.1 siunitx: Germany locale]
 \sisetup{
   unitsep=cdot,
   valuesep=thin,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-SIunits.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-SIunits.cfg
index 3e3f452f090..3103921552f 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-SIunits.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-SIunits.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-SIunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of SIunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     SIunits]
 \si@emulating{SIunits}{2007/12/02 v1.36}
 \sisetup{
   unitsep=thick,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-UK.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-UK.cfg
index 738b61d8787..99d93b7e6d9 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-UK.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-UK.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-UK.cfg}
-  [2008/09/03 v1.0l siunitx: UK locale]
+  [2008/09/28 v1.1 siunitx: UK locale]
 \sisetup{
   unitsep=thin,
   expproduct=times,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-USA.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-USA.cfg
index ec429c18443..622fb8ddccf 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-USA.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-USA.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-USA.cfg}
-  [2008/09/03 v1.0l siunitx: USA locale]
+  [2008/09/28 v1.1 siunitx: USA locale]
 \sisetup{
   unitsep=thin,
   expproduct=times,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-ZA.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-ZA.cfg
index 6dc518eb08a..e0dba73265c 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-ZA.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-ZA.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-ZA.cfg}
-  [2008/09/03 v1.0l siunitx: South Africa locale]
+  [2008/09/28 v1.1 siunitx: South Africa locale]
 \sisetup{
   unitsep=cdot,
   valuesep=thin,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-abbr.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-abbr.cfg
index 4e18e3fc370..c93c7a7b75b 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-abbr.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-abbr.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-abbr.cfg}
-  [2008/09/03 v1.0l siunitx: Abbreviated units]
+  [2008/09/28 v1.1 siunitx: Abbreviated units]
 \requiresiconfigs{prefix,named,accepted,physical}
 \newunit{\pA}{\pico\ampere}
 \newunit{\nA}{\nano\ampere}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-accepted.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-accepted.cfg
index 4f050c7ebd9..2236969c914 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-accepted.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-accepted.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-accepted.cfg}
-  [2008/09/03 v1.0l siunitx: SI Accepted units]
+  [2008/09/28 v1.1 siunitx: SI Accepted units]
 \newunit{\minute}{min}
 \newunit{\hour}{h}
 \newunit{\Day}{d}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-addn.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-addn.cfg
index 52def9ca5ab..f6ef0e272a2 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-addn.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-addn.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-addn.cfg}
-  [2008/09/03 v1.0l siunitx: SI Additional units]
+  [2008/09/28 v1.1 siunitx: SI Additional units]
 \newunit{\angstrom}{\si@sym@ringA}
 \newunit{\are}{a}
 \newunit{\hectare}{\hecto\are}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-astro.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-astro.cfg
index a39cce154c5..95a702c7e4a 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-astro.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-astro.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-astro.cfg}
-  [2008/09/03 v1.0l siunitx: Units for astronomy]
+  [2008/09/28 v1.1 siunitx: Units for astronomy]
 \newunit{\parsec}{pc}
 \newunit{\lightyear}{ly}
 \endinput
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-binary.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-binary.cfg
index f90bae2cd62..0481dbd628d 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-binary.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-binary.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-binary.cfg}
-  [2008/09/03 v1.0l siunitx: Binary units]
+  [2008/09/28 v1.1 siunitx: Binary units]
 \newprefix[binary]{\kibi}{10}{Ki}
 \newprefix[binary]{\mebi}{20}{Mi}
 \newprefix[binary]{\gibi}{30}{Gi}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-fancynum.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-fancynum.cfg
index 849a1c52043..9c34f5dc04c 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-fancynum.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-fancynum.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-fancynum.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of fancynum]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     fancynum]
 \si@emulating{fancynum}{2000/08/08 0.92}
 \sisetup{decimalsymbol=cdot,digitsep=comma}
 \let\fnum\num
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-fancyunits.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-fancyunits.cfg
index f73010149df..89212451f54 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-fancyunits.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-fancyunits.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-fancyunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of fancyunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     fancyunits]
 \si@emulating{fancyunits}{2007/02/01 v1.0.1}
 \si@ifloaded{SIunits}
   {\si@log@err{SIunits emulation loaded\MessageBreak before
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-hep.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-hep.cfg
index 8825525b522..4a704aee51a 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-hep.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-hep.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-hep.cfg}
-  [2008/09/03 v1.0l siunitx: Units for high-energy physics]
+  [2008/09/28 v1.1 siunitx: Units for high-energy physics]
 \requiresiconfigs{prefix,named}
 \provideunit{\micron}{\micro\metre}
 \newunit{\mrad}{\milli\rad}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-hepunits.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-hepunits.cfg
index 1e57ff7c5dd..e0bb32063f8 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-hepunits.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-hepunits.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-hepunits.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of hepunits]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     hepunits]
 \si@emulating{hepunits}{2007/09/27}
 \requiresiconfigs{SIunits,accepted,prefix,hep}
 \ifsi@old@noprefixcmds\else
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-named.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-named.cfg
index ad9af4617d5..748cf20d7db 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-named.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-named.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-named.cfg}
-  [2008/09/03 v1.0l siunitx: SI Named units]
+  [2008/09/28 v1.1 siunitx: SI Named units]
 \newunit{\becquerel}{Bq}
 \newunit{\coulomb}{C}
 \newunit{\farad}{F}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-physical.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-physical.cfg
index 3b6ab1f6018..4cb1f8cce0a 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-physical.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-physical.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-physical.cfg}
-  [2008/09/03 v1.0l siunitx: SI Physically-measured units]
+  [2008/09/28 v1.1 siunitx: SI Physically-measured units]
 \newcommand*{\si@eVspacea}{\text{\kern-\si@eVcorra}}%
 \newcommand*{\si@eVspaceb}{\text{\kern-\si@eVcorrb}}%
 \newunit{\electronvolt}{e\protect\si@eVspacea V\protect%
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-prefix.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-prefix.cfg
index 2969dd8c71e..0330987ccfc 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-prefix.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-prefix.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-prefix.cfg}
-  [2008/09/03 v1.0l siunitx: SI Multiple prefixes]
+  [2008/09/28 v1.1 siunitx: SI Multiple prefixes]
 \newprefix{\yocto}{-24}{y}
 \newprefix{\zepto}{-21}{z}
 \newprefix{\atto}{-18}{a}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-prefixed.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-prefixed.cfg
index 99593b2561e..32030fa1f2b 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-prefixed.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-prefixed.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-prefixed.cfg}
-  [2008/09/03 v1.0l siunitx: SI Prefixed units]
+  [2008/09/28 v1.1 siunitx: SI Prefixed units]
 \requiresiconfigs{prefix,named,accepted,physical}
 \newunit{\picometre}{\pico\metre}
 \newunit{\nanometre}{\nano\metre}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-sistyle.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-sistyle.cfg
index 3f7977ed19e..18a516494a4 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-sistyle.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-sistyle.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-sistyle.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of SIstyle]
+  [2008/09/28 v1.1 siunitx: Emulation of
+     SIstyle]
 \si@emulating{sistyle}{2006/12/20 v2.3}
 \sisetup{%
   sepfour=true,
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-synchem.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-synchem.cfg
index 97048463852..d1dff21c562 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-synchem.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-synchem.cfg
@@ -14,7 +14,7 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-synchem.cfg}
-  [2008/09/03 v1.0l siunitx: Units for synthetic chemists]
+  [2008/09/28 v1.1 siunitx: Units for synthetic chemists]
 \requiresiconfigs{prefix}
 \newunit{\mmHg}{mmHg}
 \newunit{\molar}{\mole\per\cubic\deci\metre}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-units.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-units.cfg
index 799b0f99142..ce3c24034aa 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-units.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-units.cfg
@@ -15,7 +15,7 @@
 %% 
 \si@ifloaded{units}{\endinput}{}
 \ProvidesFile{si-units.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of units]
+  [2008/09/28 v1.1 siunitx: Emulation of units]
 \si@emulating{units}{1998/08/04 v0.9b}
 \si@ifloaded{SIunits}
   {\si@emclash{units}{SIunits}\endinput}{}
diff --git a/Master/texmf-dist/tex/latex/siunitx/config/si-unitsdef.cfg b/Master/texmf-dist/tex/latex/siunitx/config/si-unitsdef.cfg
index c0a7bebdc86..322f7b5691e 100644
--- a/Master/texmf-dist/tex/latex/siunitx/config/si-unitsdef.cfg
+++ b/Master/texmf-dist/tex/latex/siunitx/config/si-unitsdef.cfg
@@ -14,7 +14,8 @@
 %%  ----------------------------------------------------------------
 %% 
 \ProvidesFile{si-unitsdef.cfg}
-  [2008/09/03 v1.0l siunitx: Emulation of unitsdef]
+  [2008/09/28 v1.1 siunitx:
+   Emulation of unitsdef]
 \si@emulating{unitsdef}{2005/01/04 v0.2}
 \si@ifloaded{SIunits}
   {\si@emclash{unitsdef}{SIunits}\endinput}{}
diff --git a/Master/texmf-dist/tex/latex/siunitx/siunitx.sty b/Master/texmf-dist/tex/latex/siunitx/siunitx.sty
index 3d5eb1260d7..b0d0bf5285b 100644
--- a/Master/texmf-dist/tex/latex/siunitx/siunitx.sty
+++ b/Master/texmf-dist/tex/latex/siunitx/siunitx.sty
@@ -15,7 +15,7 @@
 %% 
 \NeedsTeXFormat{LaTeX2e}
 \ProvidesPackage{siunitx}
-  [2008/09/03 v1.0l A comprehensive (SI) units package]
+  [2008/09/28 v1.1 A comprehensive (SI) units package]
 \begingroup
   \@ifundefined{eTeXversion}
     {\PackageError{siunitx}
@@ -30,6 +30,7 @@
   \catcode\string`\string ` \the\catcode\string`\`\relax
   \catcode\string`\string = \the\catcode\string`\=\relax
   \catcode\string`\string ^ \the\catcode\string`\^\relax
+  \catcode\string`\string _ \the\catcode\string`\_\relax
   \catcode\string`\string ~ \the\catcode\string`\~\relax
   \catcode\string`\string : \the\catcode\string`\:\relax
   \catcode\string`\string - \the\catcode\string`\-\relax
@@ -40,6 +41,7 @@
 \catcode\string`\` 12\relax
 \catcode`\= 12\relax
 \catcode`\^ 7\relax
+\catcode`\_ 8\relax
 \catcode`\~ \active\relax
 \@makeother{\:}
 \@makeother{\-}
@@ -377,11 +379,11 @@
 \si@opt@key{textsf}{\sisetup{unittextsf=#1,valuetextsf=#1}}
 \si@opt@key{texttt}{\sisetup{unittexttt=#1,valuetexttt=#1}}
 \si@opt@cmdkeys{numdigits,numdecimal,numexp,numgobble,numsign,%
-  numcloseerr,numopenerr,numaddn,numprod}
+  numcloseerr,numopenerr,numaddn,numprod,numdiv}
 \newcommand*{\si@numextra}{%
   \si@numopenerr\si@numcloseerr\si@numaddn}
 \newcommand*{\si@numvalid}{\si@numgobble\si@numexp\si@numsign
-  \si@numdecimal\si@numdigits\si@numextra\si@numprod}
+  \si@numdecimal\si@numdigits\si@numextra\si@numprod\si@numdiv}
 \si@opt@boolkey{seperr}
 \si@opt@boolkey{trapambigerr}
 \si@opt@cmdkeys{openerr,closeerr}
@@ -531,6 +533,7 @@
 \si@opt@boolkey{astroang}
 \si@opt@boolkey{strictarc}
 \newif\ifsi@tab@fixed
+\si@opt@boolkey{tabparseonly}
 \si@opt@choicekey{tabnumalign}
   {centredecimal,centerdecimal,right,left,centre,center}
   {\si@tab@fixedtrue
@@ -639,15 +642,35 @@
    \ifx\si@tempa\si@tempb
      \let\si@tab@lfill@s\relax
    \fi}
-\si@opt@boolkey{fixdp}
+\si@opt@boolkey[%
+  \ifsi@fixdp
+    \expandafter\si@fixsffalse
+  \fi]{fixdp}
 \newcount\si@num@dp
 \si@opt@key{dp}{%
   \si@str@ifonlychrs{#1}{0123456789}
-    {}
+    {\si@num@dp#1\relax
+     \si@fixdptrue
+     \si@fixsffalse}
     {\si@log@err{Invalid input for dp option}
-      {The dp option must be given a positive integer}}%
-  \si@num@dp#1\relax
-  \si@fixdptrue}
+      {The dp option must be given a positive integer}}}
+\si@opt@boolkey[%
+  \ifsi@fixsf
+    \expandafter\si@fixdpfalse
+  \fi]{fixsf}
+\newcount\si@num@sf
+\si@opt@key{sf}{%
+  \si@str@ifonlychrs{#1}{0123456789}
+    {\si@num@sf#1\relax
+     \ifnum\si@num@sf=\z@\relax
+       \si@log@warn{Setting sf=0 disables function}%
+       \si@fixsffalse
+     \else
+       \si@fixsftrue
+       \si@fixdpfalse
+     \fi}
+    {\si@log@err{Invalid input for sf option}
+      {The sf option must be given a positive integer}}}
 \si@opt@boolkey{tabautofit}
 \si@opt@boolkey{xspace}
 \si@opt@boolkey
@@ -1240,6 +1263,7 @@
     \def\,{}%
     \catcode`\~=\active\relax
     \catcode`\^=\active\relax
+    \catcode`\_=\active\relax
     \everyeof{\noexpand}%
     \endlinechar\m@ne
     \protected@xdef\si@tempa{\scantokens{#1}}%
@@ -1700,8 +1724,10 @@
   \expandafter\si@str@ifonlychrs\expandafter{\si@tempa}
     {0123456789}
     {\ifx\@empty\si@num@err\@empty
-       \ifsi@fixdp
-         \expandafter\expandafter\expandafter\si@num@fixdp
+       \renewcommand*{\si@tempb}{mant}%
+       \renewcommand*{\si@tempc}{#1}%
+       \ifx\si@tempb\si@tempc
+         \expandafter\expandafter\expandafter\si@num@fixlength
        \fi
      \fi}{}%
   \si@num@ifextra{\si@num@predec}{}
@@ -1837,7 +1863,7 @@
   \si@tempcntb\si@tempcnta\relax
   \si@num@lerr
   \protected@edef\si@num@err{%
-    \si@tempa\ensuremath{\si@decimalsymbol}\si@tempb}}
+    \si@tempa\expandafter\@car\si@numdecimal\@nil\si@tempb}}
 \newcommand*{\si@num@lerr}{%
   \si@num@cntdigits{\si@tempb}%
   \ifnum\si@tempcnta=\si@tempcntb\relax\else
@@ -1847,6 +1873,14 @@
 \def\si@num@movedigit#1#2\@empty{%
   \protected@edef\si@tempa{\si@tempa#1}%
   \protected@edef\si@tempb{#2}}
+\newcommand*{\si@num@fixlength}{%
+  \ifsi@fixdp
+    \expandafter\si@num@fixdp
+  \else
+    \ifsi@fixsf
+      \expandafter\expandafter\expandafter\si@num@fixsf
+    \fi
+  \fi}
 \newcommand*{\si@num@fixdp}{%
   \si@num@cntdigits{\si@num@postdec}%
   \ifx\@empty\si@num@postdec\@empty
@@ -1865,10 +1899,14 @@
 \newcommand*{\si@num@pd}{%
   \advance\si@tempcnta\@ne\relax
   \protected@edef\si@num@postdec{\si@num@postdec0}}
+\newcount\si@num@rndtarget
 \newcommand*{\si@num@prernd}{}
 \newcommand*{\si@num@postrnd}{}
 \newcommand*{\si@num@round}{%
   \si@log@debug{Rounding to \the\si@num@dp\space digits}%
+  \si@num@rndtarget\si@num@dp\relax
+  \si@num@rndstart}
+\newcommand*{\si@num@rndstart}{%
   \si@num@reverse{\si@num@postdec}%
   \si@num@reverse{\si@num@predec}%
   \let\si@num@prernd\si@num@predec
@@ -1887,10 +1925,10 @@
     \si@num@rev#2\@empty\@empty
   \fi}
 \newcommand*{\si@num@rnd}{%
-  \ifnum\si@tempcnta>\z@\relax
-    \expandafter\si@num@rndpost
-  \else
+  \ifx\@empty\si@num@postrnd\@empty
     \expandafter\si@num@rndpre
+  \else
+    \expandafter\si@num@rndpost
   \fi}
 \newcommand*{\si@num@rndpre}{%
   \expandafter\edef\expandafter\si@tempa\expandafter{%
@@ -1924,12 +1962,12 @@
     \advance\si@tempcntb\@ne\relax
   \fi
   \si@switchfalse
-  \advance\si@num@dp\@ne\relax
-  \ifnum\si@tempcnta>\si@num@dp\relax
-    \advance\si@num@dp\m@ne\relax
+  \advance\si@num@rndtarget\@ne\relax
+  \ifnum\si@tempcnta>\si@num@rndtarget\relax
+    \advance\si@num@rndtarget\m@ne\relax
   \else
-    \advance\si@num@dp\m@ne\relax
-    \ifnum\si@tempcnta>\si@num@dp\relax
+    \advance\si@num@rndtarget\m@ne\relax
+    \ifnum\si@tempcnta>\si@num@rndtarget\relax
       \ifnum\si@tempcntb>4\relax
         \expandafter\expandafter\expandafter\si@switchtrue
       \fi
@@ -1942,7 +1980,120 @@
     \fi
   \fi
   \advance\si@tempcnta\m@ne\relax
+  \ifnum\si@tempcnta=\m@ne\relax
+    \si@num@throw
+  \fi
   \si@num@rnd}
+\newcommand*{\si@num@throw}{%
+  \si@num@reverse{\si@num@postdec}%
+  \edef\si@num@postdec{\expandafter\@cdr\si@num@postdec\@nil}%
+  \si@num@reverse{\si@num@postdec}}
+\newcommand*{\si@num@fixsf}{%
+  \si@num@cntdigits{\si@num@predec}%
+  \ifx\@empty\si@num@predec\@empty
+    \si@tempcnta\z@\relax
+  \fi
+  \ifnum\si@tempcnta=\@ne\relax
+    \renewcommand*{\si@tempa}{0}%
+    \ifx\si@num@predec\si@tempa
+      \ifx\@empty\si@num@postdec\@empty\else
+        \si@tempcnta\z@\relax
+        \si@switchfalse
+        \expandafter\si@num@cntpostsf\si@num@postdec
+          \@empty\@empty
+        \ifnum\si@tempcnta>\si@num@sf\relax
+          \si@log@debug{Rounding to \the\si@num@sf\space
+            figures}%
+          \si@num@rndtarget\si@num@sf\relax
+          \si@num@rndstart
+        \fi
+      \fi
+    \else
+      \si@num@cntprepost
+    \fi
+  \else
+    \ifnum\si@tempcnta<\si@num@sf\relax
+      \si@num@cntprepost
+    \else
+      \advance\si@tempcnta\m@ne\relax
+      \si@switchfalse
+      \ifnum\si@tempcnta>\si@num@sf\relax\else
+        \ifx\@empty\si@num@postdec\@empty\else
+          \edef\si@num@postdec{\expandafter\@car
+            \si@num@postdec\@nil}%
+          \ifnum\si@num@postdec>4\relax
+            \expandafter\si@switchtrue
+          \fi
+        \fi
+      \fi
+      \advance\si@tempcnta\@ne\relax
+      \renewcommand*{\si@num@postdec}{}%
+      \si@num@reverse{\si@num@predec}%
+      \let\si@num@prernd\si@num@predec
+      \renewcommand*{\si@num@predec}{}%
+      \si@num@sfpre
+      \ifsi@switch
+        \edef\si@num@predec{1\si@num@predec}%
+      \fi
+    \fi
+  \fi}
+\def\si@num@cntpostsf#1#2\@empty{%
+  \ifsi@switch
+    \advance\si@tempcnta\@ne\relax
+  \else
+    \if 0#1\relax\else
+      \si@tempcnta\@ne\relax
+      \expandafter\expandafter\expandafter\si@switchtrue
+    \fi
+  \fi
+  \ifx\@empty#2\@empty\else
+    \si@num@cntpostsf#2\@empty\@empty
+  \fi}
+\newcommand*{\si@num@cntprepost}{%
+  \si@tempcntb\si@tempcnta\relax
+  \si@num@cntdigits{\si@num@postdec}%
+  \advance\si@tempcnta\si@tempcntb\relax
+  \ifnum\si@tempcnta>\si@num@sf
+    \si@log@debug{Rounding to \the\si@num@sf\space figures}%
+    \si@num@rndtarget\si@num@sf\relax
+    \si@num@rndstart
+    \ifsi@switch
+      \si@num@throw
+    \fi
+  \fi}
+\newcommand*{\si@num@sfpre}{%
+  \expandafter\edef\expandafter\si@tempa\expandafter{%
+    \expandafter\@car\si@num@prernd\@nil}%
+  \expandafter\edef\expandafter\si@num@prernd\expandafter{%
+    \expandafter\@cdr\si@num@prernd\@nil}%
+  \si@tempcntb\si@tempa\relax
+  \ifsi@switch
+    \advance\si@tempcntb\@ne\relax
+  \fi
+  \si@switchfalse
+  \advance\si@num@sf\@ne\relax
+  \ifnum\si@tempcnta>\si@num@sf\relax
+    \advance\si@num@sf\m@ne\relax
+    \edef\si@num@predec{0\si@num@predec}%
+  \else
+    \advance\si@num@sf\m@ne\relax
+    \ifnum\si@tempcnta>\si@num@sf\relax
+      \ifnum\si@tempcntb>4\relax
+        \expandafter\expandafter\expandafter\si@switchtrue
+      \fi
+      \edef\si@num@predec{0\si@num@predec}%
+    \else
+      \ifnum\si@tempcntb=10\relax
+        \si@tempcntb\z@\relax
+        \expandafter\expandafter\expandafter\si@switchtrue
+      \fi
+      \edef\si@num@predec{\the\si@tempcntb\si@num@predec}%
+    \fi
+  \fi
+  \advance\si@tempcnta\m@ne\relax
+  \ifx\@empty\si@num@prernd\@empty\else
+    \expandafter\si@num@sfpre
+  \fi}
 \newcommand*{\si@num@int}[1]{%
   \renewcommand*{\si@num@predec}{}%
   \ifsi@sepfour
@@ -2488,10 +2639,13 @@
 \newcount\si@tempcntb
 \newcommand*{\si@tab@numout}{%
   \si@num@intabtrue
-  \ifsi@tab@fixed
-    \ifsi@tabautofit
-      \si@num@dp\si@tab@mantpostcnt\relax
-      \expandafter\expandafter\expandafter\si@fixdptrue
+  \ifsi@tabparseonly\else
+    \ifsi@tab@fixed
+      \ifsi@tabautofit
+        \si@num@dp\si@tab@mantpostcnt\relax
+        \expandafter\expandafter\expandafter\expandafter
+          \expandafter\expandafter\expandafter\si@fixdptrue
+      \fi
     \fi
   \fi
   \expandafter\si@num\expandafter{\the\si@tab@toks}%
@@ -2501,17 +2655,35 @@
 \newbox\si@tab@postbox
 \newbox\si@tab@expbox
 \newcommand*{\si@tab@format}{%
-  \ifsi@tab@fixed
-    \expandafter\si@tab@fixed
+  \ifsi@tabparseonly
+    \expandafter\si@tab@outonly
   \else
-    \expandafter\si@tab@unfixed
+    \ifsi@tab@fixed
+      \expandafter\expandafter\expandafter\si@tab@fixed
+    \else
+      \expandafter\expandafter\expandafter\si@tab@unfixed
+    \fi
   \fi
   \ifsi@colourvalues
     \si@fam@colourcmd{\si@valuecolour}%
   \fi
-  \box\si@tab@prebox\box\si@tab@midbox\box\si@tab@postbox%
+  \box\si@tab@prebox\box\si@tab@midbox\box\si@tab@postbox
     \box\si@tab@expbox}
-\newcommand*\si@tab@unfixed{%
+\newcommand*{\si@tab@outonly}{%
+  \si@log@debug{Using non-aligned S column}%
+  \setbox\si@tab@prebox=\hbox{}%
+  \setbox\si@tab@midbox=\hbox{}%
+  \setbox\si@tab@postbox=\hbox{}%
+  \setbox\si@tab@expbox=\hbox{}%
+    \protected@edef\si@tempa{%
+      \si@tab@out
+    \ifx\@empty\si@num@out\@empty\else
+      \ensuremath{{\si@decimalsymbol}}%
+    \fi
+    \si@num@out
+    \si@tab@expout}%
+  \expandafter\si@out@num\expandafter{\si@tempa}}
+\newcommand*{\si@tab@unfixed}{%
   \si@log@debug{Using variable width S column}%
   \ifx\@empty\si@num@out\@empty
     \setbox\si@tab@midbox=\hbox
@@ -2589,8 +2761,7 @@
   \setbox\si@tab@prebox=\hbox to\si@tab@predim{\hss\hfill
     \expandafter\si@out@num\expandafter{\si@tab@out}}%
   \ifx\@empty\si@num@out\@empty
-    \setbox\si@tab@postbox=\hbox to\si@tab@postdim
-      {\expandafter\si@out@num\expandafter{\si@num@out}\hfil}%
+    \setbox\si@tab@postbox=\hbox to\si@tab@postdim{\hfil}%
   \else
     \ifsi@tabalignexp\else
       \protected@edef\si@num@out{\si@num@out\si@tab@expout}%
@@ -2672,6 +2843,19 @@
   \si@ifdefinable{#2}
     {\si@unt@defpower[#1]{#2}{#3}}
     {}}
+\newcommand*{\newqualifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@unt@defqual{#1}{#2}}
+    {\si@log@err{Qualifier \string#1 already defined!}\@eha}}
+\newcommand*{\renewqualifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@log@err{Qualifier \string#1 undefined}\@ehc
+     \si@unt@defqual{#1}{#2}}
+    {\si@unt@defqual{#1}{#2}}}
+\newcommand*{\providequalifier}[2]{%
+  \si@ifdefinable{#1}
+    {\si@unt@defqual{#1}{#2}}
+    {}}
 \newif\ifsi@unt@num
 \newcommand*{\si@unt@unitarg}{}
 \newcommand*{\si@unt@SIopts}{}
@@ -2792,8 +2976,8 @@
     \renewcommand*{\si@unt@spstack}{}%
     \renewcommand*{\si@unt@stacka}{}%
     \renewcommand*{\si@unt@stackb}{}%
-    \renewcommand*{\si@unt@holdstacka}{}%
-    \renewcommand*{\si@unt@holdstackb}{}%
+    \renewcommand*{\si@unt@spacestacka}{}%
+    \renewcommand*{\si@unt@spacestackb}{}%
     \renewcommand*{\si@unt@lastadda}{space}%
     \renewcommand*{\si@unt@lastaddb}{space}}
 \newcommand*{\si@unt@final}{%
@@ -2873,6 +3057,20 @@
       \fi
     \fi
     {#2}{#3}}}
+\newcommand*{\si@unt@defqual}[2]{%
+  \si@log@debug{Declaring qualifier \string#1 with\MessageBreak
+    meaning \string#2}%
+  \protected\def#1{%
+    \ifsi@unt@littest
+      \expandafter\@gobble
+    \else
+      \ifsi@unt@litout
+        \expandafter\expandafter\expandafter\si@unt@litqual
+      \else
+        \expandafter\expandafter\expandafter\si@unt@qual
+      \fi
+    \fi
+    {#2}}}
 \newcommand*{\si@unt@unithook}{}
 \newcommand*{\si@unt@unit}[3]{%
   \ifnum\si@unt@depthcnt=\m@ne\relax
@@ -3034,19 +3232,50 @@
     \expandafter\ifx
       \csname si@unt@lastadd\si@unt@checkstack\endcsname
       \si@tempa
-      \si@log@debug{Adding power
-        \strip@pt\si@unt@powerdim\space to output stack}%
-      \si@unt@addtostack{power}{^{\num[fixdp=false]{%
-        \strip@pt\si@unt@powerdim}}}%
+      \si@unt@skpwr
+    \else
+      \renewcommand*{\si@tempa}{qual}%
+      \expandafter\ifx
+        \csname si@unt@lastadd\si@unt@checkstack\endcsname
+        \si@tempa
+        \si@unt@skpwrqual
+      \fi
     \fi
   \fi
   \si@unt@powerdim\z@\relax}
+\newcommand*{\si@unt@skpwr}{%
+  \si@log@debug{Adding power
+    \strip@pt\si@unt@powerdim\space to output stack}%
+  \si@unt@addtostack{power}{^{\num[fixdp=false]{%
+    \strip@pt\si@unt@powerdim}}}}
 \newcommand*{\si@unt@invpower}{%
   \si@tempdima\si@unt@powerdim\relax
   \si@unt@powerdim -\si@tempdima\relax
   \ifdim\si@unt@powerdim=\p@\relax
     \si@unt@powerdim\z@\relax
   \fi}
+\newcommand*{\si@unt@litqual}[1]{(#1)}
+\newcommand*{\si@unt@qual}[1]{%
+  \ifsi@unt@first
+    \si@log@err{Qualifier before units}
+      {A unit qualifier can only be given after\MessageBreak
+       one or more units.  You tried to open a unit
+       with\MessageBreak \string#1}%
+  \fi
+  \expandafter\let\expandafter\si@unt@spacestack
+    \csname si@unt@spacestack\si@unt@checkstack\endcsname
+  \expandafter\let\csname si@unt@spacestack\si@unt@checkstack
+    \endcsname\@empty
+  \renewcommand*{\si@tempa}{power}%
+    \expandafter\ifx
+    \csname si@unt@lastadd\si@unt@checkstack\endcsname\si@tempa
+    \let\si@unt@skpwrqual\relax
+  \else
+    \let\si@unt@skpwrqual\si@unt@skpwr
+  \fi
+  \si@unt@addtostack{qual}{_{#1}}%
+  \expandafter\let\csname si@unt@spacestack\si@unt@checkstack
+    \endcsname\si@unt@spacestack}
 \newif\ifsi@unt@per
 \newif\ifsi@unt@perseen
 \si@newrobustcmd*{\si@per}{%
@@ -3122,28 +3351,27 @@
   \expandafter\advance
     \csname si@unt@unitcnt\si@unt@checkstack\endcsname
     \@ne\relax}
-\newcommand*{\si@unt@holdstacka}{}
-\newcommand*{\si@unt@holdstackb}{}
+\newcommand*{\si@unt@spacestack}{}
+\newcommand*{\si@unt@spacestacka}{}
+\newcommand*{\si@unt@spacestackb}{}
 \newcommand*{\si@unt@holdspace}[1]{%
   \renewcommand*{\si@tempa}{b}%
   \edef\si@tempb{\si@unt@checkstack}%
   \ifx\si@tempa\si@tempb
     \ifx\@empty\si@unt@stackb\@empty
     \else
-      \expandafter\protected@edef
-      \csname si@unt@holdstack\si@unt@checkstack\endcsname{#1}%
+      \protected@edef\si@unt@spacestackb{#1}%
     \fi
   \else
-    \expandafter\protected@edef
-    \csname si@unt@holdstack\si@unt@checkstack\endcsname{#1}%
+    \protected@edef\si@unt@spacestacka{#1}%
   \fi}
 \newcommand*{\si@unt@addstack}[1]{%
   \expandafter\protected@edef
     \csname si@unt@stack\si@unt@checkstack\endcsname
     {\csname si@unt@stack\si@unt@checkstack\endcsname
-     \csname si@unt@holdstack\si@unt@checkstack\endcsname#1}%
+     \csname si@unt@spacestack\si@unt@checkstack\endcsname#1}%
   \expandafter\renewcommand\expandafter*\expandafter{%
-    \csname si@unt@holdstack\si@unt@checkstack\endcsname}{}}
+    \csname si@unt@spacestack\si@unt@checkstack\endcsname}{}}
 \newcommand*{\si@unt@stackout}{%
   \si@unt@litouttrue
   \ifsi@frac
@@ -3375,9 +3603,11 @@
 \begingroup
   \catcode`\^=\active
   \catcode`\-=\active
+  \catcode`\_=\active
   \gdef\si@out#1{%
     \begingroup
       \catcode`\^=\active
+      \catcode`\_=\active
       \makeatletter
       \endlinechar\m@ne
       \ifsi@fam@set\else
@@ -3396,16 +3626,21 @@
     \check@mathfonts}
   \gdef\si@out@text#1{%
     \let^\si@out@sp
+    \let_\si@out@sb
     \let\textsuperscript\si@out@sp
+    \let\textsubscript\si@out@sb
     \catcode`\-=\active\relax
     \let-\si@out@minus
     #1}
   \gdef\si@out@maths#1{%
     \let^\sp
+    \let_\sb
     \let\textsuperscript\sp
+    \let\textsubscript\sb
     $\si@fam@maths{#1}$}
 \endgroup
 \newcommand*{\si@out@sp}[1]{\ensuremath{^{\text{#1}}}}
+\newcommand*{\si@out@sb}[1]{\ensuremath{_{\text{#1}}}}
 \newcommand*{\si@out@minus}{\ensuremath{-}}
 \newif\ifsi@out@num
 \newcommand*{\si@out@num}[1]{%
@@ -3510,9 +3745,10 @@
   negcolour=red,
   noload={},
   numaddn={},%(
-  numcloseerr=),%
+  numcloseerr=),
   numdecimal={.,},
   numdigits=0123456789,
+  numdiv=/,
   numexp=eEdD,
   numgobble={},
   numopenerr=(,%)
@@ -3543,6 +3779,7 @@
   tabautofit=false,
   tabformat=3.2,
   tabnumalign=centredecimal,
+  tabparseonly=false,
   tabtextalign=centre,
   tabunitalign=left,
   textrm=rmfamily,