words.pl revision 34bff097c14521b5e57ce37279a34256e1f78aa5
3141ac1ae0a0b9f86af05f439bc79316451b94f3Carsten Fischer# ##### old intention below
3f69b6948966979163bdfe8331c38833d5d90ecdChristian Maeder# read file "words.input" from current directory and produce
acc7fe823b23e829cfbde7464f92557822a2a7d9Christian Maeder# "\wordline{<word>}\hline" lines. plus patterns written in lines
acc7fe823b23e829cfbde7464f92557822a2a7d9Christian Maeder# starting with "%" . THe pattern should contain a variable calles
3141ac1ae0a0b9f86af05f439bc79316451b94f3Carsten Fischer# "$word. It functions as input.
1caad5414a81bd3593e10ca6d089ebabac2d9ad8Christian Maeder# words.input contains words seperated
1caad5414a81bd3593e10ca6d089ebabac2d9ad8Christian Maeder# by space, newline or tabular creates or overwrites a file called
3141ac1ae0a0b9f86af05f439bc79316451b94f3Carsten Fischer# "generated-words.tex"
f323228f78695312bd7e881484e18df94c1e8f0eChristian Maeder# the fonts file can be created by this bash line:
f323228f78695312bd7e881484e18df94c1e8f0eChristian Maeder# for f in `ls /usr/share/texmf/tex/latex/psnfss/*.sty` ; do f=`basename $f| sed 's/\.sty//'`;echo '\usepackage{'$f'} ::: '$f ; done > fonts.input
bccea164bdfc2ddc3d1e20749bb5477a46eab3a6Christian Maeder#$ENV{'TEXINPUTS'} =$ENV{'TEXINPUTS'}."::".dirname($0);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian MaederCopyright : (c) Klaus L�ttich, Uni Bremen 2002-2004
bccea164bdfc2ddc3d1e20749bb5477a46eab3a6Christian MaederLicense : similar to LGPL, see HetCATS/LICENSE.txt or LIZENZ.txt
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder Created by a Perl-script (HetCATS/utils/words.pl)!
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maedermy %used_words = &process_the_words; # the words, LaTeX makros,
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # syllables (ligatures), letters widely
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # used in (Het)CASL
ea39a854498febb718cbdd6035fb935fd145daacEugen Kuksa# Split into sections that give the names of the Haskell maps.
acc7fe823b23e829cfbde7464f92557822a2a7d9Christian Maedermy @fonts = &read_fonts; # adds an "empty font" for LaTeX-default
acc7fe823b23e829cfbde7464f92557822a2a7d9Christian Maedermy %widths = (); # a table of width from various fonts
bccea164bdfc2ddc3d1e20749bb5477a46eab3a6Christian Maeder $widths{$font->[1]} = &process_one_font($font);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder foreach my $sec (keys %{$widths{$font->[1]}}) {
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder print "$sec :", join(",", @{${$widths{$font->[1]}}{$sec}}),"\n";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maedermy %word_widths = &calc_max_width(\%widths,\%used_words);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # a table of sections to table of
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # words per section to max width over all fonts
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder print "$sec :\n",
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder join(", ", (map {"$_: ".$word_widths{$sec}{$_}; }
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder&generate_haskell_FM(\%word_widths); # uses %word_widths
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder die "cannot create Haskell module \"LaTeX_maps.hs\"";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # generate a list of pairs for each section named after the section
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder my @words = sort (keys %{$word_widths->{$sec}});
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder push @two_letter_words, (grep {length($_) == 2;} @words);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder join(",", (map {"(\"".&escape_String($_)."\",".
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder int($word_widths->{$sec}{$_} * 0.351 * 1000).
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder my @long_words = sort (grep {length($_) > 2;} @words);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder grep {my $ret = $_ ne $last_word;$last_word = $_;$ret; }
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder (grep {m/^$c/}
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder print HASKELL "\nligatures :: Map String Bool\n",
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder "ligatures = fromList [",
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder join(",", map { "(\"".&escape_String($_)."\",True)";}
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder grep {my $ret = $_ ne $last_word;$last_word = $_;$ret; }
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder return join("", map {
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder if (m/^\\$/o) {
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder } elsif(m/^"$/o) { # "
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # substitute ������� with \196\214\220\223\228\246\252
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder $_ =~ s/�/\\196/o; $_ =~ s/�/\\214/o; $_ =~ s/�/\\220/o;
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder $_ =~ s/�/\\223/o;
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder $_ =~ s/�/\\228/o; $_ =~ s/�/\\246/o; $_ =~ s/�/\\252/o;
3141ac1ae0a0b9f86af05f439bc79316451b94f3Carsten Fischer } split(//o,$_[0]));
ea39a854498febb718cbdd6035fb935fd145daacEugen Kuksa return "\n$map_name :: Map String Int\n$map_name = fromList";
f323228f78695312bd7e881484e18df94c1e8f0eChristian Maeder return "\n$map_name :: Map Char [String] \n$map_name = fromList";
d5a644d593fc653a5476f9b3efffe1d34693f1e4Christian Maeder print "$sec: ";
3141ac1ae0a0b9f86af05f439bc79316451b94f3Carsten Fischer #print "$word: ";
f323228f78695312bd7e881484e18df94c1e8f0eChristian Maeder my $cur = $font_widths->{$font}->{$sec}->[$width_word_index];
01ad2c5a32e6cf168ac76d1b4b737ec4372db9f9Christian Maeder print "$font: ".int($cur * 0.351 * 1000)." " if $word eq "~";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder "\nWarning: max length of $word is undefined or zero\n"
f323228f78695312bd7e881484e18df94c1e8f0eChristian Maeder #exit if $width_word_index >= 5;
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # debugging: print "font: ".join(", ", @{$_[0]})."\n";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # generate two documents one human readable and one for the machine
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder # &gen_tex('width-table.tex.svmono.templ',
d5a644d593fc653a5476f9b3efffe1d34693f1e4Christian Maeder # &gen_tex('width-table.tex.svmono.templ',
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder basename($computer_tex_filename,'.tex').'.pdf';
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder return &get_widths($computer_pdf_filename); # seperated in sections
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder my $txt_filename = basename($pdf_filename,'.pdf').'.txt';
acc7fe823b23e829cfbde7464f92557822a2a7d9Christian Maeder system($PDFTOTEXT_BIN,"-raw",$pdf_filename);
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder #open WIDTH, "pdftotext $pdf_filename | egrep 'section: |wl: ' |"
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder# or die "cannot call pdftotext or egrep or cannot fork";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder or die "cannot read file \"$txt_filename\"";
6cc0d8b77759c557e7d9459cd2734625a4db78b9Christian Maeder m/section: (\w+)\+\+\+/o && do {
@widths = ();
close WIDTH;
return \%widths;
sub gen_tex {
my $output_filename =
while (<TEMPL>) {
print OUT $_;
close TEMPL;
close OUT;
return $output_filename;
sub read_fonts {
my @read_fonts = ();
while (<FONTS>) {
my @fnt_descrp = split /\s+:::\s+/;
close FONTS;
return @read_fonts;
sub process_the_words {
my @sec_words = ();
my %all_words = ();
if ($line =~ m/^%/o) {
chomp $line;
$line =~ s///o;
@sec_words = ();
$line =~ s///o;
} elsif ($line =~ m/^&/o) {
$line =~ s///o;
} elsif ($line =~ m/^~(.*)$/o) {
close WORDS;
close GENWORDS;
return %all_words;
sub sep_tabular {