%%%%% \iffalse meta-comment
%
%% File: unicodefonttable.dtx (C) Copyright 2019-2024 Frank Mittelbach
%
% It may be distributed and/or modified under the conditions of the
% LaTeX Project Public License (LPPL), either version 1.3c of this
% license or (at your option) any later version. The latest version
% of this license is in the file
%
% https://www.latex-project.org/lppl.txt
%
%
% The development version of the bundle can be found below
%
% https://github.com/FrankMittelbach/fmitex/
%
% for those people who are interested or want to report an issue.
%
\def\unicodefonttabledate {2024/06/06}
\def\unicodefonttableversion{1.0i}
%<*driver>
\documentclass{l3doc-TUB2024}
% some fixes local to this documentation
\ExplSyntaxOn
\cs_new:Npn\IfImplementationShownTF
{\bool_if:NTF \g__codedoc_typeset_implementation_bool}
\ExplSyntaxOff
\makeatletter
\def\meta@font@select{\normalfont\itshape}
\IndexPrologue{%
\IfImplementationShownTF{\section*{Index}%
Numbers written in italic refer to the page
where the corresponding entry is described
or mentioned. Numbers underlined refer to the
code line of the
definition; numbers in Roman refer to the
code lines where the entry is used.}%
{\section{Index}%
Numbers written in italic refer to the page
where the corresponding entry is described
or mentioned.}}
\makeatother
\usepackage{graphicx,unicodefonttable}
\EnableCrossrefs
\CodelineIndex
\begin{document}
\DocInput{unicodefonttable.dtx}
\advance\signaturewidth by 40pt
\makesignature
\end{document}
%</driver>
%
% \fi
%
% \tracinglostchars=3
%
% \newcommand\key [1]{\texttt{#1}}
% \newcommand\kval[1]{\texttt{#1}}
%
% \newcommand\keysetup[1]{\noindent\marginpar{\raggedleft Key setup: #1}\ignorespaces}
% \renewcommand\keysetup[1]{\paragraph*{\hspace*{-8pc}Key setup (#1)}}
%
%
% \title{The \texttt{unicodefonttable} package\thanks{This is
% version \unicodefonttableversion\ of the package, dated
% \unicodefonttabledate; the license is LPPL.}}
% \author{Frank Mittelbach}
% \address{Mainz, Germany}
% \netaddress{https://www.latex-project.org}
% \personalURL{https://ctan.org/pkg/unicodefonttable}
% \raggedbottom
%
% \maketitle
%
%
% \begin{abstract}
% A package for typesetting font tables for larger fonts, e.g.,
% TrueType or OpenType Unicode fonts.
% To produce a one-off table, a standalone version is available as well.
% \end{abstract}
%
% \vspace*{\medskipamount}
%
% \tableofcontents
%
% \vspace*{\bigskipamount}
%
% \section{Introduction}
%
% When I started to write a new chapter for the third edition of
% \emph{The \LaTeX{} Companion} on modern fonts available for
% different \LaTeX{} engines, I was a bit surprised that I couldn't
% find a way to easily typeset tables showing the glyphs available in
% TrueType or OpenType fonts. The \pkg{nfssfont} package available
% with \LaTeX{} only supports fonts from the 8-bit world, but
% modern fonts that can be used with \XeTeX{} or \LuaTeX{} can
% contain thousands of glyphs and having a method to display what
% is available in them was important for me.
%
% I therefore set out to write my own little package and what
% started as an afternoon exercise ended up being this package,
% offering plenty of bells and whistles for typesetting such
% font tables.
%
% As there can be many glyphs in such fonts a tabular
% representation of them might run for several pages, so the
% package internally uses the \pkg{longtable} package to handle
% that.
%
% In most cases the glyphs inside the fonts are indexed by their
% Unicode numbers so it is natural to display them sorted by their
% position in the Unicode character set.
%
% Unicode is organized in named blocks such as ``Basic Latin'',
% ``Latin-1~Supplement'', etc., typically consisting of 265
% characters each.\footnote{Some blocks are smaller, while those
% containing the Asian ideographs are much larger.} It is therefore
% helpful to use these block names as subtitles within the table,
% to more easily find the information one is looking for.
%
% A common way to represent the number of a single Unicode
% character is \texttt{U+} followed by four (or more)
% hexadecimal digits. For example, \texttt{U+0041} represents the letter
% ``A'' and \texttt{U+20AC} the Euro currency symbol ``\texteuro''.
% We use this convention by showing a Unicode range of sixteen
% characters at the left of each table row,
% e.g., \texttt{U+0040 -
% 004F}, followed by the sixteen glyphs in the range. Thus that
% \IfImplementationShownTF{}{\pagebreak}\ignorespaces
% particular table row from the ``Basic Latin'' block would show
% something like
%
% \displayfonttable*[noheader,range-start=0040, range-end=004F,
% color=black!50]{Latin Modern Sans}
%
% If a Unicode character has no glyph representation in a given
% font then this is indicated by a special symbol (by default a
% colored hyphen). By default some color is used, but we've grayscaled
% the output for \TUB.
%
% In order to easily locate any Unicode character the table shows by
% default sixteen hex digits as a column heading. For example, to find
% Euro currency symbol (\texttt{U+20AC}) one first finds the
% right row, which is the range \texttt{U+20A0 - 20AF}, and then
% the \texttt{C} column in that row, and the glyph is there (or an
% indication that the font is missing that glyph;
% the line shows that for some of the other slots).
%
% \displayfonttable*[noheader,range-start=20A0, range-end=20AF,
% color=black!50]{Latin Modern Sans}
%
% It can be useful to compare two fonts with each other by
% filling the table with glyphs from a secondary font if the
% primary font is missing them. For example, the next display shows
% two rows of Latin Modern Math (black glyphs) and instead of showing
% a missing glyph symbol in most slots, we use the glyphs from New
% Computer Modern Math, which has a much larger glyph set
% (normally red glyphs with gray background but again, grayscaled for
% \TUB).
%
%
% \displayfonttable[noheader,nostatistics,display-block=none,
% compare-with=NewCMMath-Regular.otf,range-start=2A00,range-end=2A1F,
% color=black!50,compare-color=none]
% {latinmodern-math.otf}
%
%
% \section{The user interface}
%
% The package offers one command to typeset a font table. The
% appearance of the table can be customized by specifying key/value
% pairs.
%
% \begin{function}{\displayfonttable}
% \begin{syntax}
% \cs{displayfonttable} \texttt{*} \oarg{key/value-list} \Arg{font-name} \oarg{font-features}
% \end{syntax}
% The \meta{font-name} is the font to be displayed. This and the
% \meta{font-features} argument are passed to \pkg{fontspec}, thus they
% should follow the conventions of that package for specifying a
% font. The \meta{key/value-list} offers customization
% possibilities discussed below.
%
% The \cs{displayfonttable*} is a variant of the command, intended
% for use with 8-bit legacy fonts. It presets some keys, but
% otherwise behaves identically. The preset values are:
%\begin{verbatim}
% nostatistics, display-block=none, hex-digits=head, range-end=FF
%\end{verbatim}
% For details see the next section.
% \end{function}
%
%
%
% \begin{function}{\fonttablesetup}
% \begin{syntax}
% \cs{fonttablesetup} \Arg{key/value-list}
% \end{syntax}
% Instead of or in addition to specifying key/values to
% \cs{displayfonttable} it is possible to set them up as
% defaults. Inside \cs{displayfonttable} the defaults are applied first,
% so one can still overwrite their values for an individual table.
% \end{function}
%
%
% \begin{function}{\fonttableglyphcount}
% \begin{syntax}
% \cs{fonttableglyphcount}
% \end{syntax}
% While typesetting a font table the package keeps track of the
% number of glyphs it finds in the font. After the table has finished,
% this value is available in \cs{fonttableglyphcount} and it is, for
% example, used when statistics are produced. At the start of the
% next table it is reset to zero.
% \end{function}
%
%
%
%
% \subsection{Keys and their values}
%
% Several of the available keys are booleans accepting \texttt{true}
% or \texttt{false}. They usually exist in pairs so that one can
% specify the desired behavior without needing to provide a value,
% e.g., specifying \key{header} is equivalent to specifying
% \key{header}\texttt{=true} or \key{noheader}\texttt{=false}, etc.
% In the lists below the default settings are indicated by an
% underline.
%
% \begin{variable}{header,noheader,title-format,title-format-cont}
% The first set of keys is concerned with the overall look and
% feel of the generated table.
% \begin{description}
% \item[\underline{\key{header}}, \key{noheader}]
% These keys determine whether a header to the table is
% produced.
%
% \item[\key{title-format}, \key{title-format-cont}]
% These keys define what is provided as a header title or
% continuation title if the table consists of several pages.
% They expect code as their value. This code can contain \verb=#1=
% and \verb=#2= to denote the \meta{font-name} and
% \meta{font-features} arguments, respectively.
%
% By default a title using the \cs{caption} command is produced; on
% continuation titles, the \meta{font-features} are not shown.
% This is typeset as a \env{longtable} header row, so you either need to use
% \cs{multicolumn} or a \cs{caption} command\Dash otherwise everything
% ends up in the first column.
%
% \end{description}
%
% \end{variable}
%
%
% \begin{variable}{display-block,hex-digits, hex-digits-font,hex-digits-row-format,color}
% These keys handle the inner parts of the table.
%
% \begin{description}
% \item[\key{display-block}]
% The Unicode dataset is organized in named blocks that are typically 128 or
% 256 characters, though some are noticeably larger and a few are
% smaller. With the \key{display-block} key it is
% possible to specify if and how such blocks should be made visible.
% The following values are supported:
% \begin{description}
% \item[\underline{\kval{titles}}]
% Above each display block that contains glyphs the Unicode title
% of the block is displayed.
%
% \item[\kval{rules}]
% Display blocks are indicated only by a \cs{midrule}.
%
% \item[\kval{none}]
% Display blocks are not indicated at all.
% \end{description}
%
% \item[\key{hex-digits}]
% To ease reading the table, rows of hex digits are added to
% it. Where or if this happens is controlled by this key. Allowed
% values for it are the following:
% \begin{description}
% \item[\kval{block}]
% A row of hex digits is placed at the beginning of each Unicode
% block containing glyphs in the displayed font.
%
% \item[\kval{foot}]
% A row is added to the foot of each table page.
%
% \item[\underline{\kval{head}}]
% A row is added to the top of each table page.
%
% \item[\kval{head+foot}]
% A row is added to the top and the foot of each table page.
%
% \item[\kval{none}]
% All hex digit rows are suppressed.
% \end{description}%
%
% \item[\key{hex-digits-font}]
% The font to use for the hex digits, by default
% \underline{\cs{ttfamily}\cs{scriptsize}}.{\hfuzz=2.7pt\par}
%
% \item[\key{hex-digits-row-format}]
% This key defines the format for the hex digits shown on the left
% of each row. It accepts one argument hold the hex values for the
% row except for the last digit, e.g, \texttt{0A3} for the values
% from \texttt{0A30} to \texttt{0A3F}. The default formatting is
% \underline{\ttfamily U+\#10\cs{,}-\cs{,}\#1F} and without further
% adjustments it is automaticaly set
% in \cs{ttfamily}\cs{footnotesize} and in the color
% specified by the \key{color} key.
% A suitable value that takes up less space would be \texttt{U+\#1x}.
%
%
%
% \item[\key{color}]
% This key determines the color for parts of the table (hex digits
% and Unicode ranges). It can be either \kval{none} or a color
% specification as understood by the \cs{color} command.
% The default is \underline{\texttt{blue}}.
%
% \end{description}
%
% \end{variable}
%
%
% \begin{variable}{statistics, nostatistics,statistics-font,statistics-format}
% The next set of keys allows altering the statistics that are produced.
% \begin{description}
%
% \item[\underline{\key{statistics}}, \key{nostatistics}]
% These keys determine whether some statistics are listed at
% the end of the table.
%
% \item[\key{statistics-font}]
% The font used to typeset the statistics; the default is\\
% \underline{\cs{normalfont}\cs{small}}.
%
%
% \item[\key{statistics-format}]
% Code (text) to specify what should be typeset in the
% statistics. One can use \verb=#1= for the \meta{font-name} and
% \verb=#2= for the glyph count.
% The material is typeset on a single line at the end of the
% table. If several lines are needed you need to use \cs{parbox}
% or a similar construct.
% \end{description}
% \end{variable}
%
%
%
% \begin{variable}{glyph-width,missing-glyph, missing-glyph-font, missing-glyph-color}
% Another set of keys deals with customization on the glyph level.
% \begin{description}
% \item[\key{glyph-width}]
% All glyphs are typeset in a box with the same width, the default
% value is \underline{\texttt{6pt}} which is suitable for most 10pt
% fonts
% and make the table fit comfortably into the text width of
% a typical document.
%
% \item[\key{missing-glyph}]
% If a slot in a row doesn't have a glyph in the font you may still
% want display something to indicate this state. By giving the key
% a value any arbitrary glyph or material can be typeset. The
% default is to typeset a \kval{-} (hyphen) in a special color.
%
% Rows that contain no glyph whatsoever are not displayed at
% all. Instead a small vertical space is added to indicate the one
% or more rows are omitted.
%
% \item[\key{missing-glyph-font}]
%
% The font used for the missing glyphs (the default value is
% \underline{\cs{ttfamily}\cs{scriptsize}}).
%
% \item[\key{missing-glyph-color}]
% If not specified it uses the value specified with the \key{color}
% key. If you want a different color, e.g., \texttt{red}, you can
% use a color value or you can specify \kval{none} to use no coloring.
%
% \end{description}
%
% \end{variable}
%
%
% \begin{variable}{compare-with,compare-color, compare-bgcolor,statistics-compare-format}
%
% You can make comparisons between two fonts, which is useful, for
% example when dealing with incomplete math fonts and you need to
% see how well the symbols from one font blend with the supplementary
% symbols from another font.
%
% \begin{description}
% \item[\key{compare-with}]
%
% If given, the value is a \meta{comparison-font-name} that is used
% to supply missing glyphs. This means that if the \meta{font-name}
% to be displayed is missing a glyph in a slot, then the
% \meta{comparison-font-name} is checked, and if that font has the
% glyph in question, it will be displayed instead of showing a
% missing glyph indicator.
%
% \item[\key{compare-color}, \key{compare-bgcolor}]
%
% To distinguish real glyphs from missing but substituted glyphs,
% they can be colored specially (default \underline{\texttt{red}})
% and/or you can have their background colored (default is
% \underline{\texttt{black!10}}, i.e., a light gray).
%
% \item[\key{statistics-compare-format}]
%
% Code (text) to specify what should be typeset in the statistics
% when comparing two fonts. One can use \verb=#1= for the
% \meta{font-name} and \verb=#2= for its glyph count, \verb=#3= is
% the name of the comparison font, \verb=#4= its glyph count,
% \verb=#5= for the number of glyphs missing in this font and
% \verb=#6= the number of extra glyphs in it. This code is used
% instead of \key{statistics-format} when comparisons are made.
%
% The material is typeset on a single line at the end of the
% table. If several lines are needed you need to use \cs{parbox}
% or a similar construct.
%
% \end{description}
% \end{variable}
%
%
% \begin{variable}{range-start, range-end}
% Finally there are two keys for restricting the display range.
% \begin{description}
% \item[\key{range-start}, \key{range-end}]
% The full Unicode set of characters is huge and checking every
% slot to see if the current font contains a glyph in the slot
% takes a long time. If you know that font contains only a
% certain subset then you can speed up the table generation
% considerably by limiting the search (and consequently the output
% generation).
% The \key{range-start} specifies where to start with the search
% (default \underline{\texttt{0000}}) and \key{range-end} gives
% the last slot that is tested (default
% \underline{\texttt{FFFF}}).
%
% Thus, by default we restrict the display to slots below
% \texttt{10000}, because text fonts seldom contain glyphs in the
% higher planes. But if you want to see everything of the font (as
% far as supported by this package) and are prepared to wait for
% the higher
% planes to be scanned, you can go up to a value of \texttt{FFFFF}.
%
% However, please note that the \LuaTeX{} fontloader uses the
% \enquote{Supplementary Private Use Area-A}, which starts at
% \texttt{F0000}, as its own playground and places remapping into
% it, so by default you see random data instead of font data there. You
% either have to use the \XeTeX{} engine or load the font with
% \texttt{Renderer=HarfBuzz} in \LuaTeX{}.
%
% These keys are also quite useful in combination with the previous
% \key{compare-with} key, to display only, for example, the Greek
% letters and see how glyphs from two fonts blend with each other.
% \end{description}
% \end{variable}
%
%
%
% \subsection{A standalone interactive version}
%
% If you want to quickly display a single font, you can run
% \texttt{unicodefont.tex} through \LaTeX{} using \LuaTeX{} (or \XeTeX{})
% as the engine. Similar to
% \texttt{nfssfont.tex} (which is for 8-bit fonts with \pdfTeX) it
% asks you a few questions and then generates the font table for
% you. There are fewer configuration options available,
% but this workflow saves you writing a document to get a
% one-off table.
%
% Most font tables need several runs due to the use of
% \pkg{longtable}, which has to find the right width for the columns
% across several pages. The \texttt{unicodefont} file therefore
% remembers your selection from the previous run and asks you if
% you want to reapply it to speed up the process.
%
%
%
% \section{Notes on the table data}
%
% If you look at some parts of a Unicode font table you see a number of slots that
% do not show a \enquote{missing glyph} sign, but nonetheless appear
% to be empty. For example:
%
% \displayfonttable[noheader,nostatistics,display-block=none,
% range-start=0020,range-end=00BF,
% color=black!50]
% {TeX Gyre Pagella}
%
%
% The reason is that Unicode contains a lot of special
% spaces or otherwise invisible characters, e.g., \texttt{U+0020}
% is the normal space, \texttt{U+00A0} is a non-breaking space,
% \texttt{U+00AD} is a soft-hyphen (what \LaTeX{} users would
% indicate with \cs{-}), and so forth. Especially the row
% \texttt{U+2000-200F} in Table~6 looks strange as it appears to be totally
% empty, but in fact most of its slots contain spaces of different
% width.
%
%\displayfonttable[noheader,nostatistics,hex-digits=foot,
% range-start=2000,range-end=202F,
% color=black!50]
% {latinmodern-math.otf}
%
%
% Another somewhat surprising area is the \enquote{Mathematical
% Alphanumeric Symbols} block in math fonts, starting at \texttt{U+1D400}.
% There you see a number of missing characters, the first two being
% \texttt{U+1D455} (math italic small h) and \texttt{U+1D49D} (math
% script B).
%
% \displayfonttable[noheader,nostatistics,hex-digits=foot,
% range-start=1D400,range-end=1D4AF,
% color=black!50]
% {latinmodern-math.otf}
%
% In this case the reason is \emph{not} that the font
% fails to implement the characters, but that these characters have
% already been defined in earlier revisions of the Unicode standard in the
% lower Unicode plane. For example, the \enquote{h} is the
% Planck constant \texttt{U+210E} and \texttt{U+212C} is
% the script capital B, etc. The Unicode Consortium decided not
% to encode the \emph{same} character twice, hence the apparent
% holes.
%
%
% \IfImplementationShownTF{}
% {\appendix
% \def\theHsection{Appendix.\thesection}
% \addtocontents{toc}{\smallskip}
% \setlength\IndexMin{200pt}
% \PrintIndex
% }
%
%
% \section{Examples}
%
% In this section we show the results of a few calls to
% \cs{displayfonttable}.
% The tables are a bit easier to navigate if they use color in some
% places, but for \TUB{} this is not practical, so we
% use black and gray.
%
% Please note that this documentation was produced with
% \LuaTeX{}. If you reuse the examples with \XeTeX{}, you may have
% to specify the font names differently (i.e., following to the
% \pkg{fontspec} documentation for this engine).
%
%
%
% \subsection{Computer Modern Sans --- 7-bit font}
%
% Our first example is the original Computer Modern Sans, with
% character codes $\le 127$.
%
% Command used: \begin{verbatim}
% \displayfonttable*[color=none, range-end=7F]{cmss10}
% \end{verbatim}
%
% \displayfonttable*[color=none, range-end=7F]{cmss10}
%
%
% \iftrue
%
%\subsection{\TeX\ Gyre Heros --- 8-bit font}
%
% This example shows the \TeX\ Gyre Heros 8-bit font,
% in the T1 encoding, with character codes $\le 255$.
% We used \key{hex-digits-row-format} to shorten the row titles on
% the left:
%\begin{verbatim}
% \displayfonttable*[color=none,hex-digits-row-format=U+#1]{ec-qhvr}
%\end{verbatim}
%
%\displayfonttable*[color=none,hex-digits-row-format=U+#1]{ec-qhvr}
%
% \fi
%
%
% \iftrue
%\subsection{Latin Modern Math --- 8-bit fonts}
%
% The traditional Latin Modern Math Italic, Symbol and Extension fonts.
% The symbol font (\texttt{lmsy10}) has two characters added to the
% Computer Modern symbol repertoire, seen in the last row of the table.
% Commands used:
%
%\begin{verbatim}
% \displayfonttable*[color=none]{lmmi10}
% \displayfonttable*[color=none]{lmsy10}
% \displayfonttable*[color=none]{lmex10}
%\end{verbatim}
%
%\displayfonttable*[color=none]{lmmi10}
%\displayfonttable*[color=none]{lmsy10}
%\displayfonttable*[color=none]{lmex10}
%
% \fi
%
%
% \iffalse % 5pp
%
%\subsection{\TeX{} Gyre Pagella (Palatino) oldstyle figures --- OTF font}
%
% This example shows Pagella with oldstyle numerals.
% Command used:
%\begin{verbatim}
%\displayfonttable{TeX Gyre Pagella}[Numbers=OldStyle]
%\end{verbatim}
%
%\displayfonttable{TeX Gyre Pagella}[Numbers=OldStyle]
%
% \fi
%
%
% ^^A \IfImplementationShownTF{}{\newpage}
%
% \subsection{Latin Modern Math compared to New Computer Modern Math}
%
% This example shows the extra symbols available in New Computer
% Modern Math in comparison to Latin Modern Math as the base font.
% We use the following setup (including settings for the
% grayscaled \TUB\ output, as an example of color overrides):
%\begin{verbatim}
% \displayfonttable[hex-digits=head+foot, range-end=1FFFF,
% compare-with=New Computer Modern Math,
% title-format=\caption{Latin Modern Math compared to
% New Computer Modern Math},
% title-format-cont=\caption{LM Math vs.\ NewCM Math,
% \emph{cont.}},
% compare-color=black, compare-bgcolor=black!5,
% missing-glyph-color=black!50, color=black!75]
% {Latin Modern Math}
%\end{verbatim}
% That is, glyphs only in \texttt{NewCM} are shown with a light gray
% background.
%
% We also extended the range to cover \texttt{U+10000} to
% \texttt{U+1FFFF} in order to include the Unicode Math alphabets.
%
% \displayfonttable[hex-digits=head+foot, range-end=1FFFF,
% compare-with=New Computer Modern Math,
% title-format=\caption{Latin Modern Math compared to
% New Computer Modern Math},
% title-format-cont=\caption{LM Math vs.\ NewCM Math,
% \emph{cont.}},
% compare-color=black, compare-bgcolor=black!5,
% missing-glyph-color=black!50, color=black!75]
% {Latin Modern Math}
%
%
%\subsection{Garamond Libre's Byzantine Musical Symbols}
%
% As a final example we exhibit the Byzantine Musical Symbols as provided by
% Garamond Libre.
% Command used:
%\begin{verbatim}
% \displayfonttable[range-start=1D000, range-end=1D0FF,
% hex-digits=block,
% missing-glyph-color=black!50, color=black!75,
% statistics-format=Total number of glyphs in
% this block of #1 is #2]
% {Garamond Libre}
%\end{verbatim}
%
% Note that we have altered the text produced by the statistics,
% because the default is somewhat misleading if only a portion of
% the font is displayed. This produces the following table:
%
% \displayfonttable[range-start=1D000, range-end=1D0FF,
% hex-digits=foot,
% missing-glyph-color=black!50, color=black!75,
% statistics-format=Total number of glyphs in
% this block of #1 is #2]
% {Garamond Libre}
%
%
%
% \IfImplementationShownTF
% {\StopEventually{\setlength\IndexMin{200pt} \PrintIndex }}
% {\StopEventually{}}
%
%
%
% \section{The package implementation}
%
% \hfuzz=10pt ^^A lots of slightly overfull tt lines
%
% \begin{macrocode}
%<*package>
% \end{macrocode}
%
% By default the package uses coloring to improve the table
% appearance and therefore requires a color package.
% \begin{macrocode}
\RequirePackage{xcolor}
% \end{macrocode}
%
% \begin{macrocode}
%<@@=fmuft>
% \end{macrocode}
%
% We need the package \pkg{xparse} for specifying the document-level
% interface commands and \pkg{l3keys2e} to use the \pkg{expl3} key
% value methods within \LaTeXe{}. These packages automatically
% require \pkg{expl3} so there is no need to load that explicitly.
% Actually, \pkg{expl3}, \pkg{l3keys2e} and the \pkg{xparse}
% functionality is now all part of the \LaTeX{} kernel so the next
% line is actually not needed at all with a current \LaTeX{} kernel, but
% in order to support older installations we keep it for now.
%
% \begin{macrocode}
\RequirePackage{xparse,l3keys2e}
% \end{macrocode}
%
% Here we introduce the package and specify its version number:
% \begin{macrocode}
\ProvidesExplPackage{unicodefonttable}
{\unicodefonttabledate}
{\unicodefonttableversion}
{Producing font tables for Unicode and other fonts}
% \end{macrocode}
%
% This is a tmp fix for code in \texttt{lttabbing.dtx} until that
% is corrected there (probably 2024/11).
% \begin{macrocode}
\cs_set:Npn \tbl_crcr:n #1 {
\int_compare:nNnT \g__tbl_col_int > 0
{
\tbl_count_missing_cells:n {#1}
}
\crcr
}
% \end{macrocode}
%
% \newcommand\hex[1]{$\langle\textit{hex}_{#1}\rangle$}
%
%
%
%
%
% \subsection{User interface commands}
%
%
% Throughout the implementation we will define a number of keys (and
% their allowed values). We introduce them at the point where they are
% used, so they are sprinkled throughout the code.\footnote{This fits
% with the way this package was developed. I first implemented a
% single rigid table layout without configuration possibilities and
% then thought about which parts I wanted to have flexible. I then
% replaced the rigid code with code that is affected by setting
% key/value pairs.}
% \begin{macro}{\fonttablesetup}
% To set up user defaults for the keys we provide a standard
% interface. The command \cs{unicodefonttabletablesetup} expects a
% key/value list and can be called as often as necessary.
% \begin{macrocode}
\NewDocumentCommand \fonttablesetup { m }
{ \keys_set:nn {@@} {#1} \ignorespaces }
% \end{macrocode}
% \end{macro}
%
%
%
% \begin{macro}{\displayfonttable}
% The document-level command for generating a font table.
% \begin{macrocode}
\NewDocumentCommand\displayfonttable {s O{} m o}{%
\IfBooleanTF #1
{
% \end{macrocode}
% For the starred form we preset a number of keys with values
% suitable when displaying 8-bit legacy fonts.
% With such fonts Unicode block headers make little
% sense (as the fonts do not conform to the Unicode layout and
% since they have at most 265 glyphs). It is therefore also unnecessary to
% loop over the whole Unicode range of the first plane.
% If necessary all of them can still be overwritten in the optional argument.
% \begin{macrocode}
\@@_display_fonttable:nnn
{nostatistics,display-block=none,hex-digits=head,range-end=FF,#2}
{#3}{#4}
}
{
\@@_display_fonttable:nnn {#2}{#3}{#4}
}
}
% \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\@@_display_fonttable:nnn}
% This command is the main workhorse of the
% package. It produces a \texttt{longtable} containing all font
% glyphs with 16 glyphs per row. The first optional argument is
% used to configure the table through key/value pairs, the
% mandatory argument is the font name to display (in
% \texttt{fontspec} conventions) and the final optional argument is
% the font feature list if any. If the latter is not provided it
% will get a special value (\texttt{--NoValue--}) assigned by
% \texttt{xparse}, which is something that can be tested for.
% \begin{macrocode}
\cs_new:Npn \@@_display_fonttable:nnn #1#2#3 {
\group_begin:
% \end{macrocode}
% First initialize the font that should be displayed (perhaps with a
% feature list) and then update the key/value list using \verb=#1=.
% \begin{macrocode}
\fontspec{#2}[#3]
\keys_set:nn{@@}{#1}
% \end{macrocode}
% If the \LuaTeX{} engine is used without HarfBuzz and the display
% range includes code points above \texttt{U+EFFFF} the output shows
% remappings and not what is in the font, so we issue a warning.
% \changes{v1.0g}{2022/11/12}{Test for luatex without harfbuzz and
% private area A and warn (gh/8)}
% \begin{macrocode}
\bool_lazy_and:nnT
{ \sys_if_engine_luatex_p: }
{ \int_compare_p:nNn { "EFFFF } < { "\l_@@_range_end_tl } }
{
\directlua{token.put_next(token.create(font.getfont(font.current()).hb~
and~ 'use_none:n'~ or~ 'use:n'))}
{ \msg_warning:nn {unicodefonttable}{noharfbuzz} }
}
% \end{macrocode}
% If the user has asked for a comparsion to some other font we need to set this up:
% \begin{macrocode}
\tl_if_empty:NTF \l_@@_compare_with_tl
{ \tl_clear:N \l_@@_compare_font_tl }
{
\setfontface \l_@@_compare_font_tl {\l_@@_compare_with_tl}[]
\cs_set_eq:NN \@@_handle_missing_glyph:n
\@@_handle_missing_glyph_compare:n
}
% \end{macrocode}
% Typesetting the font tables in twocolumn mode makes little sense
% due to their width, and if \env{longtable} is used it will
% complain. However there is one case where it should work: in a
% page-wide float. To make this happen we claim that we are not in
% twocolumn mode if the display is inside a vertical box.
% \changes{v1.0g}{2022/11/12}{Support use in twocolumn mode if inside
% a table* float (gh/7)}
% \begin{macrocode}
\if_mode_vertical: \if_mode_inner: \@twocolumnfalse \fi: \fi:
% \end{macrocode}
% Then we start the table with 17 columns. We use \texttt{longtable}
% if we produce a caption and \texttt{longtable*} if not (so that
% the table number is not increased, which would look odd if you
% have other tables in your document).
% \begin{macrocode}
\begin{longtable\bool_if:NF\l_@@_display_header_bool{*}}
{@{}r@{\quad}*{16}{c}@{}}
% \end{macrocode}
% Special headers and footers are set up first:
% \begin{macrocode}
\@@_setup_header_footer:nn{#2}{#3}
% \end{macrocode}
% Then we produce all table rows with the glyphs.
% \begin{macrocode}
\@@_produce_table_rows:
% \end{macrocode}
% At the very end we may typeset some statistics. This can't be
% done in the table footer, because the data is dynamic (e.g.,
% number of glyphs processed) and the table footers are static and
% do not change based on the table content.
% \begin{macrocode}
\@@_handle_table_ending:n {#2}
\end{longtable\bool_if:NF\l_@@_display_header_bool{*}}
\group_end:
}
% \end{macrocode}
% \end{macro}
%
%
% \begin{macrocode}
\msg_new:nnn {unicodefonttable}{noharfbuzz}
{ You~ asked~ for~ displaying~ glyphs~ with~ code \iow_newline:
points~ above~ U+EFFFF~ \msg_line_context: ,~ i.e.,~ from~ the~
'Supplementary~ Private~ Use~ Area-A'\iow_newline:
without~ specifying~ '[Renderer=Harfbuzz]'~ when~
loading~ the~ font.
\iow_newline:\iow_newline:
With~ LuaLaTeX,~ this~ Unicode~ region~ is~ used~
for~ remappings~ (if~ the~ HarfBuzz~ engine~ is~ not~ used).~
Thus,~ the~ results~ shown~ do~ not~ reflect~ what~
is~ in~ the~ font!
}
% \end{macrocode}
%
%
%
% \begin{macro}{\fonttableglyphcount}
% \begin{macro}{\g_@@_glyph_int,\g_@@_glyph_only_B_int,\g_@@_glyph_also_B_int}
% While generating the font table we count the number of glyphs we
% see (and typeset). The total is available in the command
% \cs{fonttableglyphcount} after the table got finished and will be reset to
% zero when the next table starts.
% \begin{macrocode}
\DeclareDocumentCommand \fonttableglyphcount {}
{ \int_use:N \g_@@_glyph_int }
% \end{macrocode}
%
% \begin{macrocode}
\int_new:N \g_@@_glyph_int
% \end{macrocode}
% When comparing fonts we also record data for the second font: the
% number of glyphs in both and the number of glyphs only in the
% second one.
% \begin{macrocode}
\int_new:N \g_@@_glyph_only_B_int
\int_new:N \g_@@_glyph_also_B_int
% \end{macrocode}
% \end{macro}
% \end{macro}
% \subsection{The overall table layout}
%
%
% \begin{macro}{\@@_setup_header_footer:nn}
% Setting up header and footer lines of the table.
% This macro receives the \textit{font name} and the \textit{font
% features} specified by the user as its arguments.
% \begin{macrocode}
\cs_new:Npn \@@_setup_header_footer:nn #1#2{
% \end{macrocode}
% On the first page of the table the header may show a caption or
% some other sort of title based on the value of
% \cs{l_@@_display_header_bool}. The formatting is handled by
% \cs{@@_format_table_title:nn} which can be customized through the
% key \key{title-format}.
% \begin{macrocode}
\bool_if:NT \l_@@_display_header_bool
{ \@@_format_table_title:nn{#1}{#2} \@@_debug_nl:n{T}\\*[6pt] }
% \end{macrocode}
% We may also want to display a line of hex digits. This is
% controlled through the key \key{hex-digits} that accepts different
% values: \kval{head}, \kval{foot}, \kval{head+foo}, \kval{block}
% (after a block title) or \kval{none}.
% \begin{macrocode}
\bool_if:NT \l_@@_header_hex_digits_bool
{ \@@_display_row_of_hex_digits: \@@_debug_nl:n{H}\\* }
\endfirsthead
% \end{macrocode}
% Headers for later table pages have a continuation title and
% maybe a row of hex digits.
% \begin{macrocode}
\bool_if:NT \l_@@_display_header_bool
{ \@@_format_table_cont:nn{#1}{#2} \@@_debug_nl:n{T}\\*[6pt] }
\bool_if:NT \l_@@_header_hex_digits_bool
{ \@@_display_row_of_hex_digits: \@@_debug_nl:n{H}\\* }
\endhead
% \end{macrocode}
% Footers of the table are either empty or show a row of hex digits.
% \begin{macrocode}
\bool_if:NT \l_@@_footer_hex_digits_bool
{ \@@_display_row_of_hex_digits: \@@_debug_nl:n{H}\\* }
\endfoot
% \end{macrocode}
% The footer of the last page of the table will always be
% empty. Any special row, such as a row of hex digits, will be
% provided in the table body. The reason is that we may want to
% display statistics at the very end of the table and those can't be
% placed into a static footer.
% \begin{macrocode}
\endlastfoot
}
% \end{macrocode}
% \end{macro}
%
% \begin{macro}{\l_@@_header_hex_digits_bool}
% \begin{macro}{\l_@@_footer_hex_digits_bool}
% \begin{macro}{\l_@@_blockwise_hex_digits_bool}
% Here are the booleans we use in the code.
% \begin{macrocode}
\bool_new:N \l_@@_header_hex_digits_bool
\bool_new:N \l_@@_footer_hex_digits_bool
\bool_new:N \l_@@_blockwise_hex_digits_bool
% \end{macrocode}
% \end{macro}
% \end{macro}
% \end{macro}
%
%
% \begin{macro}{\@@_display_row_of_hex_digits:}
% \begin{macro}{\@@_format_hex_digit:n}
% Producing a row of hex digits is simple.
% \begin{macrocode}
\cs_new:Npn \@@_display_row_of_hex_digits: {
& \@@_format_hex_digit:n{0} & \@@_format_hex_digit:n{1}
& \@@_format_hex_digit:n{2} & \@@_format_hex_digit:n{3}
& \@@_format_hex_digit:n{4} & \@@_format_hex_digit:n{5}
& \@@_format_hex_digit:n{6} & \@@_format_hex_digit:n{7}
& \@@_format_hex_digit:n{8} & \@@_format_hex_digit:n{9}
& \@@_format_hex_digit:n{A} & \@@_format_hex_digit:n{B}
& \@@_format_hex_digit:n{C} & \@@_format_hex_digit:n{D}
& \@@_format_hex_digit:n{E} & \@@_format_hex_digit:n{F} }
% \end{macrocode}
% Each digit is typeset in typewriter and in script size. We offer
% font and color
% customizations. Note that it is important to set an explicit
% family. Otherwise the hex digits are formatted using the current
% table font (which may or may not work at all).
% \begin{macrocode}
\cs_new:Npn \@@_format_hex_digit:n #1 {
\l_@@_hex_digits_font_tl \l_@@_color_tl #1 }
% \end{macrocode}
% \end{macro}
% \end{macro}
%
% \begin{macro}{\l_@@_color_tl}
% The token list to hold definition if set up.
% \begin{macrocode}
\tl_new:N \l_@@_color_tl
% \end{macrocode}
% \end{macro}
%
%
% \keysetup{overall table}
% Here are the definitions for the keys used in the code above:
% \begin{macrocode}
\keys_define:nn {@@} {
% \end{macrocode}
% The \key{header} key is a boolean that determines if a header
% title should be produced (default)
% \begin{macrocode}
,header .bool_set:N = \l_@@_display_header_bool
,header .default:n = true
,header .initial:n = true
% \end{macrocode}
% To ease the setup we also support the key \key{noheader} which is
% a short form for \texttt{header=false}.
% \begin{macrocode}
,noheader .bool_set_inverse:N = \l_@@_display_header_bool
,noheader .default:n = true
% \end{macrocode}
% The default for the \key{title-format} key is to produce a
% \cs{caption} listing the font name and any features (if
% given). Note the \cs{IfValueTF} command (provided by
% \texttt{xparse}) that checks if the second argument got any value
% or has the special \texttt{--NoValue--} value.
% \begin{macrocode}
,title-format .cs_set:Np = \@@_format_table_title:nn #1#2
,title-format .initial:n =
\IfValueTF{#2} { \caption{ #1~ (features:~ \texttt{\small#2}) } }
{ \caption{ #1 } }
% \end{macrocode}
% The default continuation title ignores the given features, so the
% formatting is somewhat simpler. It uses \verb=\caption[]{...}= to
% make a caption that doesn't alter the table number.
% \changes{v1.0f}{2021/10/29}{Make documentation and code match:
% it should be \texttt{title-format-cont}}
% \begin{macrocode}
,title-format-cont .cs_set:Np = \@@_format_table_cont:nn #1#2
,title-format-cont .initial:n = \caption[]{#1~ \emph{cont.}}
% \end{macrocode}
% The key \key{hex-digits} is implemented as a choice, where each
% allowed value sets different booleans that are then used in the code.
% \begin{macrocode}
,hex-digits .choice:
,hex-digits / block .code:n =
\bool_set_true:N \l_@@_blockwise_hex_digits_bool
\bool_set_false:N \l_@@_header_hex_digits_bool
\bool_set_false:N \l_@@_footer_hex_digits_bool
,hex-digits / foot .code:n =
\bool_set_true:N \l_@@_footer_hex_digits_bool
\bool_set_false:N \l_@@_header_hex_digits_bool
\bool_set_false:N \l_@@_blockwise_hex_digits_bool
,hex-digits / head .code:n =
\bool_set_true:N \l_@@_header_hex_digits_bool
\bool_set_false:N \l_@@_footer_hex_digits_bool
\bool_set_false:N \l_@@_blockwise_hex_digits_bool
,hex-digits / head+foot .code:n =
\bool_set_true:N \l_@@_header_hex_digits_bool
\bool_set_true:N \l_@@_footer_hex_digits_bool
\bool_set_false:N \l_@@_blockwise_hex_digits_bool
,hex-digits / none .code:n =
\bool_set_false:N \l_@@_header_hex_digits_bool
\bool_set_false:N \l_@@_footer_hex_digits_bool
\bool_set_false:N \l_@@_blockwise_hex_digits_bool
,hex-digits .initial:n = head
% \end{macrocode}
% The font for hex digits are set with \key{hex-digits-font}.
% \begin{macrocode}
,hex-digits-font .tl_set:N = \l_@@_hex_digits_font_tl
,hex-digits-font .initial:n = \ttfamily \scriptsize
% \end{macrocode}
% Customizing the row header (on the left) can be done with this
% key. Defaults for font, fontsize, and color is set on the outside, but can, of
% course, be overwritten inside if that is desired.
% \changes{v1.0g}{2022/11/12}{Add key hex-digits-row-format to allow
% customizing the row title on the left (gh/3)}
% \begin{macrocode}
,hex-digits-row-format .cs_set:Np = \@@_format_row_hex_digits:n #1
,hex-digits-row-format .initial:n = U+#1 0 \, - \, #1 F
% \end{macrocode}
% The \key{color} key is used in most places that get colored; some
% have their own key but default to the main color.
% \begin{macrocode}
,color .choice:
,color / none .code:n = \tl_clear:N \l_@@_color_tl
,color / unknown .code:n = \tl_set:Nn \l_@@_color_tl { \color {#1} }
,color .initial:n = blue
}
% \end{macrocode}
%
%
%
%
% \begin{macro}{\@@_handle_table_ending:n}
% At the end of the table we may want to display a final row of
% hex digits and perhaps some statistics, i.e., the number of
% typeset glyphs.
% \begin{macrocode}
\cs_new:Npn \@@_handle_table_ending:n #1 {
% \end{macrocode}
%
% \begin{macrocode}
\@@_debug_nl:n{H} \\*
\bool_if:NT \l_@@_footer_hex_digits_bool
{ \@@_display_row_of_hex_digits: \@@_debug_nl:n{H} \\* }
\bool_if:NT \l_@@_display_statistics_bool
{ \\*[2pt]
\multicolumn{17}{l}{ \l_@@_stats_font_tl
% \end{macrocode}
% If we do font comparison, we use a different command for
% displaying statistics and pass more data to it.
% \begin{macrocode}
\tl_if_empty:NTF \l_@@_compare_with_tl
{
\@@_format_stats:nn{#1}{\fonttableglyphcount}
}
{
\@@_format_compare_stats:nnnnnn{#1}{\fonttableglyphcount}
{ \l_@@_compare_with_tl }
% \end{macrocode}
% The extra arguments are total glyph number in second font, glyphs
% missing in second font and glyphs only in second font.
% \begin{macrocode}
{ \int_eval:n { \int_use:N\g_@@_glyph_also_B_int +
\int_use:N\g_@@_glyph_only_B_int }
}
{ \int_eval:n { \fonttableglyphcount -
\int_use:N\g_@@_glyph_also_B_int }
}
{ \int_use:N\g_@@_glyph_only_B_int }
}
% \end{macrocode}
% We don't know exactly how wide the table is (and nor does the
% user) but one may need to use \cs{parbox} when formatting
% the statistic line(s). So we back up a bit (rather random) which
% allows us to use \verb=\parbox{\linewidth}{...= in the key
% without thinking too much about it.
% \begin{macrocode}
\hspace*{-3cm}
}
}
}
% \end{macrocode}
%
% \keysetup{for statistics}
% Here are the keys used above. By default we produce statistics.
% \begin{macrocode}
\keys_define:nn {@@} {
,statistics .bool_set:N = \l_@@_display_statistics_bool
,statistics .default:n = true
,statistics .initial:n = true
% \end{macrocode}
% the key \key{nostatistics} is just short for \texttt{statistics=false}:
% \begin{macrocode}
,nostatistics .bool_set_inverse:N = \l_@@_display_statistics_bool
,nostatistics .default:n = true
% \end{macrocode}
% The default font we use is \cs{normalfont}. Again we need to
% supply a family to avoid getting the font used in the table body.
% \begin{macrocode}
,statistics-font .tl_set:N = \l_@@_stats_font_tl
,statistics-font .initial:n = \normalfont\small
% \end{macrocode}
% And here we have the default text. There is only space for a
% single line. If more text is needed one needs to provide some
% explicit \cs{parbox}.
% \changes{v1.0g}{2022/11/12}{Change default text so that it makes
% more sense if only a portion of the font is displayed (gh/4)}
% \begin{macrocode}
,statistics-format .cs_set:Np = \@@_format_stats:nn #1#2
,statistics-format .initial:n = Total~ number~ of~ glyphs~ shown~ from~ #1:~#2
}
% \end{macrocode}
% \end{macro}
%
%
%
%
%
% \begin{macro}{\@@_debug_nl:n}
% While developing the code I had a bit of trouble getting the line
% endings correct, so I added a little macro that made them visible
% (displaying its argument in the table margin when the key
% \key{debug} is used. By default it does nothing.
% \begin{macrocode}
\cs_new:Npn \@@_debug_nl:n #1 {}
% \end{macrocode}
%
% \keysetup{debugging}
% This key is really internal and is therefore not documented above
% (and its behavior may changes over time).
% \begin{macrocode}
\keys_define:nn {@@} {
debug .code:n = \cs_set:Npn \@@_debug_nl:n ##1
{\rlap{\normalfont\scriptsize \qquad ##1}} }
% \end{macrocode}
% \end{macro}
%
%
%
%
%
%
%
% \subsection{The producing the table content}
%
%
% The body of the table consists of rows with sixteen glyphs each
% and to produce it we loop through all possible Unicode points
% starting at \texttt{U+0000} and ending with \texttt{U+FFFF}.
%
% This is implemented with a four-level nested loop that runs through the
% values \texttt{0}, \texttt{1}, \ldots, \texttt{F} with the
% current hex value in each of the four positions stored in some variable.
% \begin{macro}{\g_@@_hex_H_tl,\g_@@_hex_A_tl,
% \g_@@_hex_B_tl,\g_@@_hex_C_tl}
% \cs{g_@@_hex_H_tl} is a bit special because, it is initially not
% zero, but empty, so that slots in the lower plane are denoted by 4
% hex digits.
% We really only need three further variables, as the value in the
% innermost loop can used directly.
% \begin{macrocode}
\tl_new:N \g_@@_hex_H_tl % higher plane
\tl_new:N \g_@@_hex_A_tl
\tl_new:N \g_@@_hex_B_tl
\tl_new:N \g_@@_hex_C_tl
% \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\c_@@_hex_digits_clist}
% Here is the sequence we loop through on each level, except the
% one for the outer level.
% \begin{macrocode}
\clist_const:Nn\c_@@_hex_digits_clist{0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F}
% \end{macrocode}
% \end{macro}
%
%
%
% \begin{macro}{\@@_produce_table_rows:,\@@_handle_hex_H:n,
% \@@_handle_hex_A:n,\@@_handle_hex_B:n,
% \@@_handle_hex_C:n,\@@_handle_hex_D:n}
% The overall code layout is then fairly simply:
% \begin{macrocode}
\cs_new:Npn \@@_produce_table_rows: {
% \end{macrocode}
% First to some general initialization
% \begin{macrocode}
\@@_initialize_table_rows:
% \end{macrocode}
% and then loop we start the loop. The outer level is a bit special
% as currently Unicode has only slots allocated in plane 0, 1, 2
% and E (well, and F, but that is a private area) so we loop only
% over those and instead of \texttt{0} we use an empty value.
% Not covered is the whole of plane 16 which too is now a
% private area.
%
% \begin{macrocode}
\clist_map_function:nN { { } , 1, 2, E, F } \@@_handle_hex_H:n }
% \end{macrocode}
%
% Most fonts do not have glyphs in the higher planes, which is why
% by default we don't loop using a nonempty \cs{@@_handle_hex_H:n}.
% But if the user wants to scan and display the higher slots they
% can by setting \key{range-end} appropriatly.
%
% So after setting \cs{@@_handle_hex_H:n} we loop over
% \cs{c_@@_hex_digits_clist} for the next
% hex digit (which we call \enquote{A}).
% \begin{macrocode}
\cs_new:Npn \@@_handle_hex_H:n #1 { \tl_gset:Nn\g_@@_hex_H_tl{#1}
\clist_map_function:NN \c_@@_hex_digits_clist \@@_handle_hex_A:n }
% \end{macrocode}
%
% Handling \enquote{A} means storing its value for later use and
% then start a loop for setting the second (or third on higher planes) hex digits:
% \begin{macrocode}
\cs_new:Npn \@@_handle_hex_A:n #1 { \tl_gset:Nn\g_@@_hex_A_tl{#1}
\clist_map_function:NN \c_@@_hex_digits_clist \@@_handle_hex_B:n }
% \end{macrocode}
%
% Same game for \enquote{B} and \enquote{C}\footnote{Actually this
% is a white lie. In reality we do a lot of extra stuff when
% handling \enquote{C} so later one we give a second definition for
% \cs{@@_handle_hex_C:n} but for understanding the overall picture
% the simpler one shown here is better.}:
% \begin{macrocode}
\cs_new:Npn \@@_handle_hex_B:n #1 { \tl_gset:Nn\g_@@_hex_B_tl{#1}
\clist_map_function:NN \c_@@_hex_digits_clist \@@_handle_hex_C:n }
% \end{macrocode}
%
% \begin{macrocode}
\cs_new:Npn \@@_handle_hex_C:n #1 { \tl_gset:Nn\g_@@_hex_C_tl{#1}
\clist_map_function:NN \c_@@_hex_digits_clist \@@_handle_hex_D:n }
% \end{macrocode}
% In the innermost loop we now have the full Unicode number
% available, so there we have to decide what to do with it. This is
% done by \cs{@@_handle_hex_D:n} that receives the full number,
% e.g., \texttt{1A7C} or \texttt{1AD00}, as its argument.
% \begin{macrocode}
\cs_new:Npn \@@_handle_hex_D:n #1 {
\@@_handle_slot:x
{ " \g_@@_hex_H_tl \g_@@_hex_A_tl
\g_@@_hex_B_tl \g_@@_hex_C_tl #1 }
}
% \end{macrocode}
% \end{macro}
%
% \begin{macro}{\g_@@_row_tl}
% We first collect the glyphs for a whole row before deciding to
% typeset it, because if the row is entirely empty we want to omit
% it. The data for the row is collected slot by slot and the typesetting
% information (the glyph or the indication for a missing glyph is
% appended to \cs{g_@@_row_tl}.
% \begin{macrocode}
\tl_new:N \g_@@_row_tl
% \end{macrocode}
% \end{macro}
% \begin{macro}{\@@_handle_slot:n,\@@_handle_slot:x}
% If the current slot number under inspection contains a glyph in
% our font we want to typeset it. But we don't do this immediately,
% instead we build up the whole row and typeset it later. We
% therefore append a \verb=&= and the glyph (including the necessary
% formatting) to the token list \cs{g_@@_row_tl}.
% \begin{macrocode}
\cs_new:Npn \@@_handle_slot:n #1 {
\@@_if_uchar_exists:nTF { #1 }
{ \tl_gput_right:Nn \g_@@_row_tl
{ & \@@_format_glyph:n { \symbol{#1} } }
% \end{macrocode}
% We then increment the overall glyph count and record that we have
% seen at least one glyph in the current row. There is not much
% point in displaying rows that are completely empty; indeed,
% we'd end up with extremely large tables which are
% mostly empty.
% \begin{macrocode}
\int_gincr:N\g_@@_glyph_int
\bool_gset_true:N \g_@@_glyph_seen_bool
% \end{macrocode}
% If we do font comparison we also check if the glyph is in the
% second font and if so record that fact.
% \begin{macrocode}
\tl_if_empty:NF \l_@@_compare_font_tl
{
\group_begin:
\l_@@_compare_font_tl
\@@_if_uchar_exists:nT { #1 }
{ \int_gincr:N \g_@@_glyph_also_B_int }
\group_end:
}
}
% \end{macrocode}
% If the current slot has no glyph in the font we also add a
% \verb=&= followed by something that indicates the glyph is
% missing. If we do font comparison, it may show the glyph from
% the other font (if it exists there) in some special way to indicate which glyph
% should be in this slot.
% \begin{macrocode}
{ \@@_handle_missing_glyph:n {#1} }
}
% \end{macrocode}
%
% \begin{macrocode}
\cs_generate_variant:Nn \@@_handle_slot:n {x}
% \end{macrocode}
% \end{macro}
%
% \begin{macro}{\@@_handle_missing_glyph:n,
% \@@_handle_missing_glyph_std:n,
% \@@_handle_missing_glyph_compare:n}
%
% In the standard case we typeset a special symbol to indicate that the glyph is missing.
% For this case we provide some customization through keys:
% \cs{l_@@_missing_glyph_tl} holds the symbol for a missing glyph
% (default: a hyphen). It is typeset in a specific color and we allow for
% setting it in a special font. The actual symbol number in
% \verb=#1= is not needed in this scenario.
% \begin{macrocode}
\cs_new:Npn \@@_handle_missing_glyph_std:n #1 {
\tl_gput_right:Nn \g_@@_row_tl
{ &
\@@_format_glyph:n {
% \colorbox{black!30} % <--- povide interface
{\l_@@_missing_glyph_color_tl
\l_@@_missing_glyph_font_tl
\l_@@_missing_glyph_tl }
}
}
}
% \end{macrocode}
%
% \keysetup{missing glyphs}
% Here are the keys for customizing the missing glyph representation.
% \begin{macrocode}
\keys_define:nn {@@} {
missing-glyph-color .choice:
,missing-glyph-color / none .code:n =
\tl_clear:N \l_@@_missing_glyph_color_tl
,missing-glyph-color / unknown .code:n =
\tl_set:Nn \l_@@_missing_glyph_color_tl { \color {#1} }
%
,missing-glyph-font .tl_set:N = \l_@@_missing_glyph_font_tl
,missing-glyph-font .initial:n = \ttfamily \scriptsize
,missing-glyph .tl_set:N = \l_@@_missing_glyph_tl
,missing-glyph .initial:n = - }
% \end{macrocode}
%
%
% The default definition for the color is to use the same as the
% one specified by the \key{color} key. We therefore define the
% default outside of the \pkg{l3keys} method.
% \begin{macrocode}
\tl_new:N \l_@@_missing_glyph_color_tl
\tl_set:Nn \l_@@_missing_glyph_color_tl {\l_@@_color_tl}
% \end{macrocode}
%
% This is the version that handles a missing glyph by checking
% the \key{compare-with} font to see if that font contains the
% glyph.
% If yes, the substitute glyph will be typeset, otherwise the missing
% glyph symbol is shown by calling \cs{@@_handle_missing_glyph_std:n}.
% \begin{macrocode}
\cs_new:Npn \@@_handle_missing_glyph_compare:n #1 {
\group_begin:
% \end{macrocode}
% Locally switch to the other font, then check for the glyph:
% \begin{macrocode}
\l_@@_compare_font_tl
\@@_if_uchar_exists:nTF { #1 }
{
% \end{macrocode}
% If available, format it (together with the \texttt{\&}) but use a
% special color and perhaps a background color.
% \begin{macrocode}
\tl_gput_right:Nn \g_@@_row_tl
{ &
\@@_format_glyph:n
{ \l_@@_compare_bgcolor_tl { \l_@@_compare_color_tl
\l_@@_compare_font_tl
\symbol {#1} }
}
}
% \end{macrocode}
% Having seen a glyph only in the second font we record this fact.
% \begin{macrocode}
\int_gincr:N \g_@@_glyph_only_B_int
% \end{macrocode}
% Also tell the algorithm that we have seen a glyph to typeset. If
% we don't do this then a row consisting of only substitute glyphs is not
% typeset. However, we don't update the glyph count, because this
% is not a glyph from the main font we display.
% \begin{macrocode}
\bool_gset_true:N \g_@@_glyph_seen_bool
}
% \end{macrocode}
% If the alternate font doesn't have the glyph either we
% typeset the missing glyph symbol.
% \begin{macrocode}
{ \@@_handle_missing_glyph_std:n {} }
\group_end:
}
% \end{macrocode}
%
% \keysetup{comparison}
%
% In order to display glyphs from a secondary font we need a
% secondary color for the glyph itself and possibly some background color.
% \begin{macrocode}
\tl_new:N \l_@@_compare_with_tl
\tl_new:N \l_@@_compare_color_tl
\tl_new:N \l_@@_compare_bgcolor_tl
% \end{macrocode}
%
% \begin{macrocode}
\keys_define:nn {@@}
{
,compare-with .tl_set:N = \l_@@_compare_with_tl
,compare-with .initial:n =
,compare-color .choice:
,compare-color / none .code:n
= \tl_clear:N \l_@@_compare_color_tl
,compare-color / unknown .code:n
= \tl_set:Nn \l_@@_compare_color_tl { \color {#1} }
,compare-color .initial:n = red
,compare-bgcolor .choice:
,compare-bgcolor / none .code:n
= \tl_clear:N \l_@@_compare_bgcolor_tl
,compare-bgcolor / unknown .code:n
= \tl_set:Nn \l_@@_compare_bgcolor_tl { \colorbox {#1} }
,compare-bgcolor .initial:n = black!10
% \end{macrocode}
% If we run a comparison we show different statistics that have
% their own key.
% \changes{v1.0g}{2022/11/12}{Change default text so that it makes
% more sense if only a portion of the font is displayed (gh/4)}
% \begin{macrocode}
,statistics-compare-format .cs_set:Np
= \@@_format_compare_stats:nnnnnn #1#2#3#4#5#6
,statistics-compare-format .initial:n
= \parbox{\linewidth}{
Total~ number~ of~ glyphs~ shown~ from~ \texttt{#1}:~#2\\
Comparison~ font~ \texttt{#3}~ has~ #5~ missing~ and~ #6~
extra~ glyphs}
}
% \end{macrocode}
%
%
% By default, i.e., if no font for comparison has been specified, we
% handle missing glyphs by displaying a missing glyph symbol.
% \begin{macrocode}
\cs_new_eq:NN \@@_handle_missing_glyph:n
\@@_handle_missing_glyph_std:n
% \end{macrocode}
% \end{macro}
%
% \begin{macro}{\@@_format_glyph:n}
% Every glyph is typeset in a box of equal width with the glyph
% centered and if necessary protruding on both sides.
% \begin{macrocode}
\cs_new:Npn \@@_format_glyph:n #1 {
\hbox_to_wd:nn {\l_@@_glyph_box_dim} { \hss #1 \hss } }
% \end{macrocode}
%
% \keysetup{glyph typesetting}
% The key to customize the width. The 6pt are fine for most cases.
% \begin{macrocode}
\dim_new:N\l_@@_glyph_box_dim
% \end{macrocode}
%
% \begin{macrocode}
\keys_define:nn {@@} {
glyph-width .dim_set:N = \l_@@_glyph_box_dim
,glyph-width .initial:n = 6pt
}
% \end{macrocode}
% \end{macro}
% \begin{macro}{\@@_if_uchar_exists:n}
% For testing whether or not a slot position contains a glyph we
% need to resort to low-level methods, because so far
% \texttt{expl3} doesn't offer an interface.
% \begin{macrocode}
\prg_set_conditional:Npnn \@@_if_uchar_exists:n #1 { TF , T }
{ \tex_iffontchar:D \tex_font:D #1 \scan_stop:
\prg_return_true:
\else:
\prg_return_false:
\fi:
}
% \end{macrocode}
% \end{macro}
%
%
%
%
% \subsection{Handling a single row}
%
%
% \begin{macro}{\@@_handle_hex_C:n}
% As promised here is the read definition for
% \cs{@@_handle_hex_C:n} in all its glory.
% \begin{macrocode}
\cs_set:Npn \@@_handle_hex_C:n #1 {
% \end{macrocode}
% We are now at the start of a new row (but with the last row not
% yet typeset) and this last row may need a Unicode block heading
% before it. This is the reason why we have to delay the
% typesetting, because in case the line doesn't contain any glyphs
% we want to typeset neither and that is only known after all
% slots in the row have been processed.
% \begin{macrocode}
\@@_maybe_typeset_a_row_and_display_a_block_title:
% \end{macrocode}
% We then store away the value for the third hex digit (denoted as
% C) in order to start with the next row.
% \begin{macrocode}
\tl_gset:Nn\g_@@_hex_C_tl{#1}
% \end{macrocode}
% Being at the start of a new row we might be at the start of a new
% Unicode block. If so we have to update the block title to add in
% front of the row when we typeset it (or in front of one of the
% next rows if the first rows in the is block have no glyphs). If
% we are still in the same block no update happens.
% \begin{macrocode}
\@@_update_block_title:n { \g_@@_hex_H_tl
\g_@@_hex_A_tl
\g_@@_hex_B_tl
\g_@@_hex_C_tl }
% \end{macrocode}
% We now check if this row is within the requested range, i.e.,
% greater than or equal to \cs{l_@@_range_start_tl} and not greater than
% \cs{l_@@_range_end_tl}.
% \begin{macrocode}
\int_compare:nNnF
{ " \g_@@_hex_H_tl \g_@@_hex_A_tl
\g_@@_hex_B_tl \g_@@_hex_C_tl 0 }
< { "\l_@@_range_start_tl }
{
\int_compare:nNnTF
{ " \g_@@_hex_H_tl \g_@@_hex_A_tl
\g_@@_hex_B_tl \g_@@_hex_C_tl 0 }
> { "\l_@@_range_end_tl }
% \end{macrocode}
% If we are past the \texttt{end-range} we break out the clist
% mapping, to avoind unnecessary repetition. This should be
% propagated back to the outer clists as well (not done).
% \begin{macrocode}
{ \clist_map_break: }
% \end{macrocode}
% If we are within range we process the slots in the row by first
% initializing \cs{g_@@_row_tl} with the row title (the info on the
% left) and then loop through all slots the row to append glyphs
% (or missing glyphs) to \cs{g_@@_row_tl} to build up everything we
% need to finally typeset it.
% \begin{macrocode}
{
\tl_gset:Nx \g_@@_row_tl
{
\exp_not:N \@@_format_row_title:n
{ \g_@@_hex_H_tl \g_@@_hex_A_tl
\g_@@_hex_B_tl \g_@@_hex_C_tl }
}
\clist_map_function:NN \c_@@_hex_digits_clist
\@@_handle_hex_D:n
}
}
}
% \end{macrocode}
% \end{macro}
%
%
% \begin{macro}{\@@_format_row_title:n}
% The function to format the row title on the left, as used above.
% \changes{v1.0g}{2022/11/12}{Add key hex-digits-row-format to allow
% customizing the row title on the left (gh/3)}
% \begin{macrocode}
\cs_new:Npn \@@_format_row_title:n #1 {
\texttt { \footnotesize \l_@@_color_tl \@@_format_row_hex_digits:n {#1} }
}
% \end{macrocode}
% \end{macro}
%
%
%
% \keysetup{ranges}
%
% For the range we have two keys, its start and the end. By default the
% whole range from 0 to FFFF is processed.
%
% \begin{macrocode}
\tl_new:N \l_@@_range_start_tl
\tl_new:N \l_@@_range_end_tl
% \end{macrocode}
%
% \begin{macrocode}
\keys_define:nn {@@}
{
,range-start .tl_set:N = \l_@@_range_start_tl
,range-start .initial:n = 0000
,range-end .tl_set:N = \l_@@_range_end_tl
,range-end .initial:n = FFFF
}
% \end{macrocode}
%
%
%
%
%
% \begin{macro}{\@@_maybe_typeset_a_row_and_display_a_block_title:}
% The function handles the just-finished row and, if the row does not
% consist only of missing glyphs, typesets it. If necessary it also typesets
% a Unicode block name first.
% \begin{macrocode}
\cs_new:Npn \@@_maybe_typeset_a_row_and_display_a_block_title: {
% \end{macrocode}
% We first check if the row had any real glyphs.
% \begin{macrocode}
\bool_if:NTF \g_@@_glyph_seen_bool
{
% \end{macrocode}
% If the row needs typesetting the fun part starts. We first look
% at the content of \cs{g_@@_block_title_tl}.
% \begin{macrocode}
\tl_if_empty:NTF \g_@@_block_title_tl
{
% \end{macrocode}
% It is empty we are in the middle of a block and we can ignore
% the Unicode title. However, we have to see if the previous row
% (or several) was missing (i.e., contained no glyphs). In that
% case we leave a little extra space, otherwise we just finish the
% previous row
% \begin{macrocode}
\bool_if:NTF \g_@@_row_missing_bool
{ \@@_debug_nl:n{A}\\[6pt] }
{ \@@_debug_nl:n{B}\\ }
}
{
% \end{macrocode}
% Otherwise we first have to typeset the Unicode block title (or
% whatever should happen instead).
% \begin{macrocode}
\typeout{ Processing~ \tl_use:N \g_@@_block_title_tl }
\bool_if:NTF \l_@@_display_block_bool
{
% \end{macrocode}
% If we are to typeset the title the action depends a bit on
% whether we are at the very first row or typesetting a later block.
% \begin{macrocode}
\bool_if:NTF \g_@@_first_row_bool
{
\bool_gset_false:N \g_@@_first_row_bool
\@@_debug_nl:n{C}\\[-4pt]
}
{
\@@_debug_nl:n{D}\\[8pt]
\noalign{\vskip 1pt plus 1pt} % space above block: customizable?
}
% \noalign{\smallskip} % space above block: customizable?
\multicolumn{17}{c}{\normalfont \bfseries
\tl_use:N \g_@@_block_title_tl}
% \end{macrocode}
% After the block title is typeset we may want to add a row of hex
% digits as well if that was requested, otherwise we only leave a
% bit of extra space.
% \begin{macrocode}
\bool_if:NTF \l_@@_blockwise_hex_digits_bool
{ \@@_debug_nl:n{E}\\*
\@@_display_row_of_hex_digits:
\@@_debug_nl:n{H}\\*[2pt]
}
{ \@@_debug_nl:n{F}\\*[2pt] }
}
{
% \end{macrocode}
% If the Unicode block title is not typeset we may still have to do
% someting special and again it differs if we at the very beginning
% of the table (because there we do nothing except changing the
% state of \cs{g_@@_first_row_bool}).
% \begin{macrocode}
\bool_if:NTF \g_@@_first_row_bool
{ \bool_gset_false:N \g_@@_first_row_bool }
{
\@@_debug_nl:n{G~ (new~ block)}
\l_@@_display_block_action_tl
}
}
% \end{macrocode}
% Once we are past the block title we clear it, so that it is not
% retypeset before the next row.
% \begin{macrocode}
\tl_gclear:N \g_@@_block_title_tl
}
% \end{macrocode}
% The final action is to typeset the row and reset the booleans (in
% case they were true; if they are false already then we do this
% unnecessarily, but that is probably faster than testing first).
% \begin{macrocode}
\bool_gset_false:N \g_@@_glyph_seen_bool
\bool_gset_false:N \g_@@_row_missing_bool
\tl_use:N \g_@@_row_tl
}
% \end{macrocode}
% Current row had no glyphs; remember that fact, and that is all we
% have to do in that case.
% \begin{macrocode}
{
\bool_gset_true:N \g_@@_row_missing_bool
}
}
% \end{macrocode}
% \end{macro}
%
%
% \subsection{Initialisation at the start of the table}
%
%
% \begin{macro}{\g_@@_first_row_bool,\g_@@_glyph_seen_bool,\g_@@_row_missing_bool}
% Declare the three booleans used in the code below. They will tell
% us answers to the following questions:
% \begin{itemize}
% \item Are we processing the first row?
% \item Have we seen any glyph so far (in the current row)?
% \item Did we have one or more missing rows recently?
% \end{itemize}
% \begin{macrocode}
\bool_new:N \g_@@_first_row_bool
\bool_new:N \g_@@_glyph_seen_bool
\bool_new:N \g_@@_row_missing_bool
% \end{macrocode}
% \end{macro}
% \begin{macro}{\@@_initialize_table_rows:}
% At the start of a table we are processing the first row
% and so we (obviously) haven't seen a glyph yet and there wasn't a
% missing row recently.
% \begin{macrocode}
\cs_new:Npn \@@_initialize_table_rows: {
\bool_gset_true:N \g_@@_first_row_bool
\bool_gset_false:N \g_@@_glyph_seen_bool
\bool_gset_false:N \g_@@_row_missing_bool
% \end{macrocode}
% And clearly the glyph count for the font(s) is zero.
% \begin{macrocode}
\int_gzero:N \g_@@_glyph_int
\int_gzero:N \g_@@_glyph_only_B_int
\int_gzero:N \g_@@_glyph_also_B_int
}
% \end{macrocode}
% \end{macro}
%
%
%
%
% \subsection{Handling block titles}
%
% \begin{macro}{g_@@_block_title_tl}
% We keep the current block title in this token list.
% \begin{macrocode}
\tl_new:N \g_@@_block_title_tl
% \end{macrocode}
% \end{macro}
% \begin{macro}{\@@_update_block_title:n}
% A block title is updated when the hex digits A,B,C have a certain
% value, so this is nothing more than a huge case switch.
% \begin{macrocode}
\cs_new:Npn \@@_update_block_title:n #1 {
\tl_gset:Nx \g_@@_block_title_tl {
\int_case:nnF{ "#1 }
{
{ "000 }{ Basic~ Latin }
{ "008 }{ Latin-1~ Supplement }
{ "010 }{ Latin~ Extended-A }
{ "018 }{ Latin~ Extended-B }
{ "025 }{ IPA~ Extensions }
{ "02B }{ Spacing~ Modifier~ Letters }
{ "030 }{ Combining~ Diacritical~ Marks }
{ "037 }{ Greek~ and~ Coptic }
{ "040 }{ Cyrillic }
{ "053 }{ Armenian }
{ "059 }{ Hebrew }
{ "060 }{ Arabic }
{ "070 }{ Syriac }
{ "075 }{ Arabic~ Supplement }
{ "078 }{ Thaana }
{ "07C }{ NKo }
{ "090 }{ Devanagari }
{ "098 }{ Bengali }
{ "0A0 }{ Gurmukhi }
{ "0A8 }{ Gujarati }
{ "0B0 }{ Oriya }
{ "0B8 }{ Tamil }
{ "0C0 }{ Telugu }
{ "0C8 }{ Kannada }
{ "0D0 }{ Malayalam }
{ "0D8 }{ Sinhala }
{ "0E0 }{ Thai }
{ "0E8 }{ Lao }
{ "0F0 }{ Tibetan }
{ "100 }{ Myanmar }
{ "10A }{ Georgian }
{ "110 }{ Hangul~ Jamo }
{ "120 }{ Ethiopic }
{ "138 }{ Ethiopic~ Supplement }
{ "13A }{ Cherokee }
{ "140 }{ Unified~ Canadian~ Aboriginal~ Syllabics }
{ "168 }{ Ogham }
{ "16A }{ Runic }
{ "170 }{ Tagalog }
{ "172 }{ Hanunoo }
{ "174 }{ Buhid }
{ "176 }{ Tagbanwa }
{ "178 }{ Khmer }
{ "180 }{ Mongolian }
{ "190 }{ Limbu }
{ "195 }{ Tai~ Le }
{ "198 }{ New~ Tai~ Le }
{ "19E }{ Khmer~ Symbols }
{ "1A0 }{ Buginese }
{ "1B0 }{ Balinese }
{ "1D0 }{ Phonetic~ Extensions }
{ "1D8 }{ Phonetic~ Extensions~ Supplement }
{ "1DC }{ Combining~ Diacritical~ Marks~ Supplement }
{ "1E0 }{ Latin~ Extended~ Additional }
{ "1F0 }{ Greek~ Extended }
{ "200 }{ General~ Punctuation }
{ "207 }{ Superscripts~ and~ Subscripts }
{ "20A }{ Currency~ Symbols }
{ "20D }{ Combining~ Diacritical~ Marks~ for~ Symbols }
{ "210 }{ Letterlike~ Symbols }
{ "215 }{ Number~ Forms }
{ "219 }{ Arrows }
{ "220 }{ Mathematical~ Operators }
{ "230 }{ Miscellaneous~ Technical }
{ "240 }{ Control~ Pictures }
{ "244 }{ Optical~ Character~ Recognition }
{ "246 }{ Enclosed~ Alphanumerics }
{ "250 }{ Box~ Drawing }
{ "258 }{ Block~ Elements }
{ "25A }{ Geometric~ Shapes }
{ "260 }{ Miscellaneous~ Shapes }
{ "270 }{ Dingbats }
{ "27C }{ Miscellaneous~ Mathematical~ Symbols-A }
{ "27F }{ Supplemental~ Arrows-A }
{ "280 }{ Braille~ Patterns }
{ "290 }{ Supplemental~ Arrows-B }
{ "298 }{ Miscellaneous~ Mathematical~ Symbols-B }
{ "2A0 }{ Supplemental~ Mathematical~ Operators }
{ "2B0 }{ Miscellaneous~ Symbols~ and~ Arrows }
{ "2C0 }{ Glagolitic }
{ "2C6 }{ Latin~ Extended-C }
{ "2C8 }{ Coptic }
{ "2D0 }{ Georgian~ Supplement }
{ "2D3 }{ Tifinagh }
{ "2D8 }{ Ethiopic~ Extended }
{ "2E0 }{ Supplemental~ Punctuation }
{ "2E8 }{ CJK~ Radicals~ Supplement }
{ "2F0 }{ Kangxi~ Radicals }
{ "2FF }{ Ideographic~ Description~ Characters }
{ "300 }{ CJK~ Symbols~ and~ Punctuation }
{ "304 }{ Hiragana }
{ "30A }{ Katakana }
{ "310 }{ Bopomofo }
{ "313 }{ Hangul~ Compatibility~ Jamo }
{ "319 }{ Kanbun }
{ "31A }{ Bopomofo~ Extended }
{ "31C }{ CJK~ Strokes }
{ "31F }{ Katakana~ Phonetic~ Extensions }
{ "320 }{ Enclosed~ CJK~ Letters~ and~ Months }
{ "330 }{ CJK~ Compatibility }
{ "4DC }{ Yijing~ Hexagram~ Symbols }
{ "A00 }{ Yi~ Syllables }
{ "A49 }{ Yi~ Radicals }
{ "A70 }{ Modifier~ Tone~ Letters }
{ "A72 }{ Latin~ Extended-D }
{ "A80 }{ Syloti~ Nagri }
{ "A84 }{ Phags-pa }
{ "A88 }{ Saurashtra }
{ "A8E }{ Devanagari Extended }
{ "A90 }{ Kayah Li }
{ "A93 }{ Rejang }
{ "A96 }{ Hangul Jamo Extended-A }
{ "A98 }{ Javanese }
{ "A9E }{ Myanmar Extended-B }
{ "AA0 }{ Cham }
{ "AA6 }{ Myanmar Extended-A }
{ "AA8 }{ Tai Viet }
{ "AAE }{ Meetei Mayek Extensions }
{ "AB0 }{ Ethiopic Extended-A }
{ "AB3 }{ Latin Extended-E }
{ "AB7 }{ Cherokee Supplement }
{ "ABC }{ Meetei Mayek }
{ "AC0 }{ Hangul Syllables }
{ "D7B }{ Hangul Jamo Extended-B }
{ "D80 }{ High Surrogates }
{ "DB8 }{ High Private Use Surrogates }
{ "DC0 }{ Low Surrogates }
{ "E00 }{ Private~ Use~ Area }
{ "F90 }{ CJK~ Compatibility~ Ideographs }
{ "FB0 }{ Alphabetic~ Presentation~ Forms }
{ "FB5 }{ Arabic~ Presentation~ Forms-A }
{ "FE0 }{ Variation~ Selectors }
{ "FE1 }{ Vertical~ Forms }
{ "FE2 }{ Combining~ Half~ Marks }
{ "FE3 }{ CJK~ Compatibility~ Forms }
{ "FE5 }{ Small~ Form~ Variants }
{ "FE7 }{ Arabic~ Presentation~ Forms-B }
{ "FF0 }{ Halfwidth~ and Fullwidth~ Forms }
{ "FFF }{ Specials~ ... }
%% ... Plane 1 ...
{ "1000 }{ Linear~ B~ Syllabary }
{ "1008 }{ Linear~ B~ Ideograms }
{ "1010 }{ Aegean~ Numbers }
{ "1014 }{ Ancient~ Greek~ Numbers }
{ "1019 }{ Ancient~ Symbols }
{ "101D }{ Phaistos~ Disc }
{ "1028 }{ Lycian }
{ "102A }{ Carian }
{ "102E }{ Coptic~ Epact~ Numbers }
{ "1030 }{ Old~ Italic }
{ "1033 }{ Gothic }
{ "1035 }{ Old~ Permic }
{ "1038 }{ Ugaritic }
{ "103A }{ Old~ Persian }
{ "1040 }{ Deseret }
{ "1045 }{ Shavian }
{ "1048 }{ Osmanya }
{ "104B }{ Osage }
{ "1050 }{ Elbasan }
{ "1053 }{ Caucasian~ Albanian }
{ "1060 }{ Linear~ A }
{ "1080 }{ Cypriot~ Syllabary }
{ "1084 }{ Imperial~ Aramaic }
{ "1086 }{ Palmyrene }
{ "1088 }{ Nabataean }
{ "108E }{ Hatran }
{ "1090 }{ Phoenician }
{ "1092 }{ Lydian }
{ "1098 }{ Meroitic~ Hieroglyphs }
{ "109A }{ Meroitic~ Cursive }
{ "10A0 }{ Kharoshthi }
{ "10A6 }{ Old~ South~ Arabian }
{ "10A8 }{ Old~ North~ Arabian }
{ "10AC }{ Manichaean }
{ "10B0 }{ Avestan }
{ "10B4 }{ Inscriptional~ Parthian }
{ "10B6 }{ Inscriptional~ Pahlavi }
{ "10B8 }{ Psalter~ Pahlavi }
{ "10C0 }{ Old~ Turkic }
{ "10C8 }{ Old~ Hungarian }
{ "10E6 }{ Rumi~ Numeral~ Symbols }
{ "1100 }{ Brahmi }
{ "1108 }{ Kaithi }
{ "110D }{ Sora~ Sompeng }
{ "1110 }{ Chakma }
{ "1115 }{ Mahajani }
{ "1118 }{ Sharada }
{ "111E }{ Sinhala~ Archaic~ Numbers }
{ "1120 }{ Khojki }
{ "1128 }{ Multani }
{ "112B }{ Khudawadi }
{ "1130 }{ Grantha }
{ "1140 }{ Newa }
{ "1148 }{ Tirhuta }
{ "1158 }{ Siddham }
{ "1160 }{ Modi }
{ "1166 }{ Mongolian~ Supplement }
{ "1168 }{ Takri }
{ "1170 }{ Ahom }
{ "118A }{ Warang~ Citi }
{ "11A0 }{ Zanabazar~ Square }
{ "11A5 }{ Soyombo }
{ "11AC }{ Pau~ Cin~ Hau }
{ "11C0 }{ Bhaiksuki }
{ "11C7 }{ Marchen }
{ "11D0 }{ Masaram~ Gondi }
{ "1200 }{ Cuneiform }
{ "1240 }{ Cuneiform~ Numbers~ and~ Punctuation }
{ "1248 }{ Early~ Dynastic~ Cuneiform }
{ "1300 }{ Egyptian~ Hieroglyphs }
{ "1440 }{ Anatolian~ Hieroglyphs }
{ "1680 }{ Bamum~ Supplement }
{ "16A4 }{ Mro }
{ "16AD }{ Bassa~ Vah }
{ "16B0 }{ Pahawh~ Hmong }
{ "16F0 }{ Miao }
{ "16FE }{ Ideographic~ Symbols~ and~ Punctuation }
{ "1700 }{ Tangut }
{ "1880 }{ Tangut~ Components }
{ "1B00 }{ Kana~ Supplement }
{ "1B10 }{ Kana~ Extended-A }
{ "1B17 }{ Nushu }
{ "1BC0 }{ Duployan }
{ "1BCA }{ Shorthand~ Format~ Controls }
{ "1D00 }{ Byzantine~ Musical~ Symbols }
{ "1D10 }{ Musical~ Symbols }
{ "1D20 }{ Ancient~ Greek~ Musical~ Notation }
{ "1D30 }{ Tai~ Xuan~ Jing~ Symbols }
{ "1D36 }{ Counting~ Rod~ Numerals }
{ "1D40 }{ Mathematical~ Alphanumeric~ Symbols }
{ "1D80 }{ Sutton~ SignWriting }
{ "1E00 }{ Glagolitic~ Supplement }
{ "1E80 }{ Mende~ Kikakui }
{ "1E90 }{ Adlam }
{ "1EE0 }{ Arabic~ Mathematical~ Alphabetic~ Symbols }
{ "1F00 }{ Mahjong~ Tiles }
{ "1F03 }{ Domino~ Tiles }
{ "1F0A }{ Playing~ Cards }
{ "1F10 }{ Enclosed~ Alphanumeric~ Supplement }
{ "1F20 }{ Enclosed~ Ideographic~ Supplement }
{ "1F30 }{ Miscellaneous~ Symbols~ and~ Pictographs }
{ "1F60 }{ Emoticons }
{ "1F65 }{ Ornamental~ Dingbats }
{ "1F68 }{ Transport~ and~ Map~ Symbols }
{ "1F70 }{ Alchemical~ Symbols }
{ "1F78 }{ Geometric~ Shapes~ Extended }
{ "1F80 }{ Supplemental~ Arrows-C }
{ "1F90 }{ Supplemental~ Symbols~ and~ Pictographs }
{ "2000 }{ CJK~ Unified~ Ideographs~ Extension~ B }
{ "2A70 }{ CJK~ Unified~ Ideographs~ Extension~ C }
{ "2B74 }{ CJK~ Unified~ Ideographs~ Extension~ D }
{ "2B82 }{ CJK~ Unified~ Ideographs~ Extension~ E }
{ "2CEB }{ CJK~ Unified~ Ideographs~ Extension~ F }
{ "2F80 }{ CJK~ Compatibility~ Ideographs~ Supplement }
{ "E010 }{ Tags }
{ "E000 }{ Variation~ Selectors~ Supplement }
{ "F000 }{ Supplementary~ Private~ Use~ Area-A }
% higher up not covered!
}
% \end{macrocode}
% If none of the above has matched we are somewhere within a block
% so we want keep the current name. However, since the case
% statement was executed within a |\tl_gset:Nx| we have to do this
% by passing the current block name back.
% \begin{macrocode}
{ \tl_use:N \g_@@_block_title_tl }
}
}
% \end{macrocode}
% \end{macro}
%
%
% \keysetup{display blocks}
%
% The Unicode blocks may get indicated in different ways: with titles,
% only through rules, or not at all. Here is the necessary setup.
%
% \begin{macrocode}
\bool_new:N \l_@@_display_block_bool
\tl_new:N \l_@@_display_block_action_tl
% \end{macrocode}
%
% \begin{macrocode}
\keys_define:nn {@@}
{
,display-block .choice:
,display-block / titles .code:n =
\bool_set_true:N \l_@@_display_block_bool
\tl_set:Nn \l_@@_display_block_action_tl {\\}
,display-block / rules .code:n =
\bool_set_false:N \l_@@_display_block_bool
\tl_set:Nn \l_@@_display_block_action_tl {\\ \midrule}
,display-block / none .code:n =
\bool_set_false:N \l_@@_display_block_bool
\tl_set:Nn \l_@@_display_block_action_tl {\\}
,display-block .initial:n = titles
}
% \end{macrocode}
%
% That's all of the programming using the L3 layer.
% \begin{macrocode}
\ExplSyntaxOff
% \end{macrocode}
% What remains is to require all packages needed \ldots
% \begin{macrocode}
\RequirePackage{longtable,booktabs,caption,fontspec}
% \end{macrocode}
% \ldots and executing all options passed to the
% package via \cs{usepackage}.
% \begin{macrocode}
\ProcessKeysPackageOptions{@@}
%</package>
% \end{macrocode}
%
%
%
%
%
% \section{The standalone \texttt{unicodefont.tex} file}
%
% \begin{macrocode}
%<*standalone>
\documentclass{article}
% \end{macrocode}
%
% \begin{macrocode}
\setlength\textwidth{470pt}
\setlength\oddsidemargin{0pt}
\addtolength\textheight{7\baselineskip}
\addtolength\topmargin{-3\baselineskip}
% \end{macrocode}
%
% \begin{macrocode}
\usepackage{unicodefonttable}
% \end{macrocode}
%
% \begin{macrocode}
\def\DEFAULTfontname{Latin Modern Roman}
\def\DEFAULTfontfeatures{}
\def\DEFAULTtableconfig{}
\def\DEFAULTunicodefont{}
\begin{document}
% \end{macrocode}
%
% \begin{macrocode}
\typeout{^^J}
% \end{macrocode}
%
% \begin{macrocode}
\ifx\generatetable\undefined
\else
\typein[\answer]{^^JReuse settings from last time (default yes)?^^J^^J%
[ font name = \DEFAULTfontname^^J
\space unicode? = \ifx\DEFAULTunicodefont\empty yes^^J
\space font features = \DEFAULTfontfeatures
\else no\fi^^J
\space table config = \DEFAULTtableconfig \space]}
\fi
% \end{macrocode}
%
% \begin{macrocode}
\ifx\answer\empty
\let\FontNameToTable\DEFAULTfontname
\let\IsUnicodeFont\DEFAULTunicodefont
\let\FontFeaturesToApply\DEFAULTfontfeatures
\let\TableConfigurationToApply\DEFAULTtableconfig
\else
% \end{macrocode}
%
% \begin{macrocode}
\typein[\FontNameToTable]%
{^^JInput external font name as understood by fontspec, e.g.,^^J%
'TeX Gyre Pagella' or 'lmroman10-regular.otf'%
\ifx\DEFAULTfontname\empty\else
^^J^^J[default \DEFAULTfontname]\fi:}
\ifx\FontNameToTable\empty \let\FontNameToTable\DEFAULTfontname \fi
% \end{macrocode}
%
% \begin{macrocode}
\typein[\IsUnicodeFont]%
{^^JIs this a Unicode font?^^J^^J%
\ifx\DEFAULTunicodefont\empty [default yes]\else [default no]\fi:}
% \end{macrocode}
%
% \begin{macrocode}
\ifx\IsUnicodeFont\empty
% \ifx\DEFAULTunicodefont\empty
% \else
\let\IsUnicodeFont\DEFAULTunicodefont
% \fi
\else
\ifx\DEFAULTunicodefont\empty
\else
\let\IsUnicodeFont\empty
\fi
\fi
% \end{macrocode}
%
% \begin{macrocode}
\ifx\IsUnicodeFont\empty
\typein[\FontFeaturesToApply]%
{^^JInput font feature key/value list to apply%
\ifx\DEFAULTfontfeatures\empty\else
^^J^^J[default \DEFAULTfontfeatures]\fi:}
\ifx\FontFeaturesToApply\empty \let\FontFeaturesToApply\DEFAULTfontfeatures \fi
\else
\let\FontFeaturesToApply\DEFAULTfontfeatures
\fi
% \end{macrocode}
%
% \begin{macrocode}
\typein[\TableConfigurationToApply]%
{^^JInput table configuration key/value list to apply%
\ifx\DEFAULTtableconfig\empty\else
^^J^^J[default
\expandafter\detokenize\expandafter{\DEFAULTtableconfig}]\fi:}
\ifx\TableConfigurationToApply\empty
\let\TableConfigurationToApply\DEFAULTtableconfig
\fi
% \end{macrocode}
%
% \begin{macrocode}
\edef\generatetable{\noexpand\displayfonttable
\ifx\IsUnicodeFont\empty\else *\fi
\ifx\TableConfigurationToApply\empty\else
[\expandafter\unexpanded\expandafter{\TableConfigurationToApply}]\fi
{\FontNameToTable}%
\ifx\FontFeaturesToApply\empty\else[\FontFeaturesToApply]\fi
}
% \end{macrocode}
%
% \begin{macrocode}
\fi
% \end{macrocode}
%
% \begin{macrocode}
\makeatletter
\protected@write\@auxout{}{\gdef\string\generatetable
{\expandafter\detokenize\expandafter{\generatetable}}}
\protected@write\@auxout{}{\gdef\string\DEFAULTfontname{\FontNameToTable}}
\protected@write\@auxout{}{\gdef\string\DEFAULTunicodefont{\IsUnicodeFont}}
\protected@write\@auxout{}{\gdef\string\DEFAULTfontfeatures{\FontFeaturesToApply}}
\protected@write\@auxout{}{\gdef\string\DEFAULTtableconfig
{\expandafter\detokenize\expandafter{\TableConfigurationToApply}}}
\makeatother
% \end{macrocode}
%
% \begin{macrocode}
\generatetable
\end{document}
%</standalone>
% \end{macrocode}
%
%
%
% \section{A samples file}
%
% \begin{macrocode}
%<*samples>
% \end{macrocode}
%
% \begin{macrocode}
%<<VERBATIMLINE
%!TEX program = lualatex
%VERBATIMLINE
% \end{macrocode}
%
% \begin{macrocode}
\documentclass{article}
\usepackage{xparse,color}
\usepackage{fontspec}
\setmainfont{Linux Biolinum O}
\setmonofont{SourceCodePro}
\usepackage{unicodefonttable}
\addtolength\textwidth{30pt}
\begin{document}
\listoftables
\section{Computer Modern --- 8bit font}
\displayfonttable*[color=none,
range-end = 7F,
]{cmr10}
%\section{Computer Modern Sans --- 8bit font} \displayfonttable*[]{cmss10}
\newpage
\section{TeX Gyre Heros (Helvetica) --- 8bit font}
\displayfonttable*[color=red,nostatistics=false,
hex-digits = head+foot,
range-end = FF,
]{ec-qhvr}
\newpage
\section{Latin Modern Sans --- OTF font}
\displayfonttable[
% display-block = rules,
% missing-glyph = \tiny\setlength\fboxsep{0pt}\fbox{$\times$},
hex-digits = block,
title-format-cont = \caption[]{\emph{continued}},
]{Latin Modern Sans}
\newpage
\section{\TeX{} Gyre Pagella (Palatino) oldstyle figures --- OTF font}
\displayfonttable{TeX Gyre Pagella}[Numbers=OldStyle]
\newpage
\section{Comparing Latin Modern Math with New Computer Modern Math}
\displayfonttable[compare-with=NewCMMath-Regular.otf, range-end=1FFFF]
{latinmodern-math.otf}
\end{document}
% \end{macrocode}
%
% \begin{macrocode}
%</samples>
% \end{macrocode}
%
% \Finale
%
\endinput
%%%%%%%%%%%%%%%%%%%