#!/usr/bin/perl -w #$example_indent = " "; $example_indent = ""; #$ex_color = "#179C4A"; # dark green -- too light #$ex_color = "#14843F"; # dark green -- still too light #$ex_color = "#056819"; # dark green -- " $ex_color = "#82140F"; # dark red $usage = " texinfo2HTML [-h] [filename] Convert texinfo doc to HTML format, through processing of texinfo comments as follows: 1. Pass single line to HTML but not to other forms of manual ------------------------------------------------------------ \@c HTML some-stuff 2. Commented-out portions ------------------------- \@c HTML 3. Break file here ------------------ \@c HTML If it occurs on a line all by itself, causes this perlscript to chop files here. The filename will be as specified. The other_words will be used as the title. If neither the filename nor the other_words are present, then this script makes up filenames using numbers, e.g., gri1.html, gri2.html. If the filename is \".\", then this same naming scheme is used, but the titles are used. 4. Insert latex command ----------------------- \@c HTML Plants a comment to be stripped out as a latex-only command. BUGS: Does not search for closing braces properly, so that lines like blah \@code{blah blah the end is here} and more blah will not get the \@code{} item converted correctly. "; #require "ctime.pl"; use POSIX qw(ctime); use Getopt::Std; &getopts('h'); if ($opt_h) { print "$usage"; exit 0; } $debug = 0; #$date = &ctime(time); $date = &ctime(time); chop($date); $CH = 0; # chapter $SEC1 = 1; # section $SEC2 = 1; # subsection $SEC3 = 1; # subsubsection $allow_P = 0; $in_example = 0; $last_was_node = 0; print "\n"; print "\n"; sub get_a_line() { if ($_=<>) { s/\@noindent//; if (/^\@c\s*HTML\s*\s*$/) { $_ = <>; return $_; } } } return $_; } else { return ""; } } while(get_a_line()) { $line++ if $debug; print STDERR "$0 at line: $line\n" if ($debug && ($line == 2000 * int($line / 2000))); next if /\@cindex/; # Prevent unclosed braces if (!$in_example) { while (/{[^}]*$/o) { $a = $_; $a =~ s/\n/ /o; $_ = get_a_line(); $_ = $a . $_; } } else { $_ = "$example_indent$_"; } # Handle HTML inserts if (/^\@c HTML (.*)/o) { $_ = "$1 \n"; &sub_refs; s,\@value\{(.*)\},$value{$1},g; # set/value pair print; next; } # Process image commands (NOT...it's there already) if (/\@image\{(.*)\}/) { #print "\n"; next; } next if /^\@top\s/; next if /^\@include\s*/; # Ignore all other comments next if /^\@c\s/; next if /^\@comment\s*/; # Ignore unnumbered, since special case (see Makefile) next if /\@unnumbered/; # Ignore special-cases text next if /^Copyright/; # Ignore texinfo command to input files next if /\\input/; # Ignore some tex-type commands next if /\@summarycontents/; next if /\@contents/; next if /\@bye/; next if /\@message/; next if /\@global/; next if /\@setfilename/; next if /\@setchapternewpage/; next if /\@settitle/; next if /\@page/; next if /\@space/; next if /\@sp/; next if /\@defindex/; next if /\@vskip/; s/\@noindent//; # Pass contents of ifhtml blocks next if /\@ifhtml/; next if /\@end ifhtml/; # Pass contents of ifinfo blocks next if /\@ifinfo/; next if /\@end ifinfo/; # Pass contents of ifnotinfo blocks next if /\@ifnotinfo/; next if /\@end ifnotinfo/; # Ignore contents of @dircategory and @direntry block next if /\@dircategory/; if (/^\@direntry/o) { while(get_a_line()) { if (/^\@end direntry/o) { $_ = ""; last; } } } # Ignore contents of tex and iftex blocks if (/^\@tex/o) { while(get_a_line()) { if (/^\@end tex/o) { $_ = ""; last; } } } if (/^\@iftex/o) { while(get_a_line()) { if (/^\@end iftex/o) { $_ = ""; last; } } } # Ignore indices (for now) next if /\@defindex/; if (/\@[cfv]index\s*(.*)/o) { print "\n"; next; } next if /\@EGindex/; next if /\@printindex/; # # Save set/value pair. if (/\@set\s*(\S*)\s*(\S*)/o) { $value{$1} = $2; next; } s,\@value\{([^}]*)\},$value{$1},g; # Substitute set/value pair # # Fix some texinfo escapes s,&,&,og; s,>,>,og; s,<,<,og; # Special tweak to make 'Gri' look cooler. # s, Gri , GRI ,og; s,\@code\{\@\@\},\@code{TEXINFO2HTML-AT-AT},og; s,\@},TEXINFO2HTML-CLOSE-BRACE,og; # retain inside e.g. @code{} s,\@\@,TEXINFO2HTML-AT-AT,og; s,\@\{,{,og; s,\@TeX\{\},TeX,og; s,\@dots\{\},...,og; # Put in place-holders for some accents. I should check for all # of them, but for now, I'm just kludging in a couple, # to solve an immediate problem and to serve as a place-holder # for better coding. A good reference on accents is # http://www.w3.org/MarkUp/html-spec/html-spec_13.html # which I found on 2002-feb-24. s,\@'a,TEXINFO2HTML-ACCENT-ACUTE-a,og; s,\@'e,TEXINFO2HTML-ACCENT-ACUTE-e,og; s,\@'o,TEXINFO2HTML-ACCENT-ACUTE-o,og; if (/^\@titlepage/o) { while(get_a_line()) { s,\@value\{(.*)\},$value{$1},g; # set/value pair if (/\@author\s*(.*)/o) { $author = $1; $author =~ s/\@\@/\@/; } else { if (/^\@end titlepage/o) { last; } } } next; } while (/\@url\{([^}]*)\}/) { $the_url = $1; s:\@url\{[^}]*\}:\@code{$the_url}:; } while (/\@uref\{([^}]*)\}/) { @items = split(/,/, $1); if ($#items == 0) { s:\@uref\{[^}]*\}:$items[0]:; } elsif ($#items == 1) { s:\@uref\{[^}]*\}:$items[1]:; } elsif ($#items == 2) { s:\@uref\{[^}]*\}:\@code{$items[2]}:; } else { die "Cannot have more than 3 items in a 'uref' at \"$_\""; } } # # Convert cross-references if (/^\@node\s*([^,]*), ([^,]*), ([^,]*), ([^\n]*)/o) { $next_node = $2; $previous_node = $3; $parent_node = $4; # Start anchor, and set flag so end put after next heading. # Need this flag because xmosaic 1.0 does not accept # ; it requires

heading

# or it will not properly search the reference. I think this # is a bug, and reported it to the author in an email: # From kelley Wed Jun 30 16:31:13 1993 # To: marca\@ncsa.uiuc.edu chop; print "\n"; print "\n"; $last_was_node = 1; next; } &sub_refs; # # Convert headings; put the end anchor if needed if (/\@chapter/o || /\@section/o || /\@subsection/o || /\@subsubsection/o || /\@subsubsubsection/o || /\@appendixsubsec/o || /\@unnumbered/o || /\@appendixsubsec/o || /\@appendixsec/o || /\@appendix/o) { &sub_headings; &sub_emphasis; print; # # Last was a node, so put the end anchor in if ($last_was_node) { print "\nNavigation:\nnext,\nprevious,\nparent.\n"; $last_was_node = 0; } # # Done with this line next; } &sub_emphasis; # # Convert quotations s,\@quotation,
,; s,\@end quotation,
,; &process_examples(); # Convert Menus to UL (was DL before) if (/\@menu/o) { print "\n"; } if (/\@table\s*(.*)/o) { if ($1 eq "\@code") { $start_item = ""; $end_item = "" } if ($1 eq "\@emph") { $start_item = ""; $end_item = "" } print "
\n"; while(get_a_line()) { next if /\@sp/; next if /\@cindex/; s|\@anchor\{([^}]*)\}||g; &process_examples(); # Handle HTML inserts if (/^\@c HTML (.*)/o) { $_ = "$1 \n"; &sub_refs; s,\@value\{(.*)\},$value{$1},g; # set/value pair print; next; } if (/^\s*\@end\s*table/o) { $_ = ""; last; } if (/\s*\@item\s*(.*)/o) { $the_item = $1; $the_item =~ s:\@\{:{:og; $the_item =~ s:\@\}:}:og; $the_item =~ s:\@\@:\@:og; $the_item =~ s:\@code\{([^}]*)\}:`$1':og; print "
$start_item$the_item$end_item\n
"; } else { print "

" if (/^$/o); &sub_refs; &sub_headings; &sub_emphasis; print; } } print "

\n"; } # # Convert lists s,\@itemize.*,,; s,\@enumerate,
    ,; s,\@end\s*enumerate,
,; s,\@itemx,
  • ,; s,\@item,
  • ,; s,\@bullet,,; # HTML does not use # # Multiple blanks lines are paragraph separators in texinfo if (/^$/o) { if ($allow_P) { print "

    "; } $allow_P = 0; } elsif (/ $title / $author \n"; print "\n"; print "\n"; sub process_examples() { if (/\@example/o) { s,\s*\@example,\n\n\n\n
    \n
    \n,;
    	$in_example = 1;
        } elsif (/^\s*\@end example/o) {
    	s,\s*\@end example,
    \n
    ,; $in_example = 0; } else { s,\@value\{([^}]*)\},$value{$1},g; # Substitute set/value pair s,\@},},og; s,\@\{,{,og; } } sub sub_refs { die "line $. of file: cannot have multiple refs on one line" if (/\@[px]*ref\{(.*)\}(.*)\@[px]*ref\{(.*)\}/); # anchors s|\@anchor\{([^}]*)\}||g; # Change e.g. # @xref{Viewing} # into # see Viewing #if (/\@ref/){print "AAA[$_]AAA\n"; s|\@ref\{([^}]*)\}|see $1|g; #print "BBB[$_]BBB\n";} s|\@xref\{([^}]*)\}|see $1|g; s|\@pxref\{([^}]*)\}|see $1.|g; while (/\@url\{([^}]*)\}/) { $the_url = $1; s:\@url\{[^}]*\}:\@code{$the_url}:; } while (/\@uref\{([^}]*)\}/) { @items = split(/,/, $1); if ($#items == 0) { s:\@uref\{[^}]*\}:$items[0]:; } elsif ($#items == 1) { s:\@uref\{[^}]*\}:$items[1]:; } elsif ($#items == 2) { s:\@uref\{[^}]*\}:\@code{$items[2]}:; } else { die "Cannot have more than 3 items in a 'uref' at \"$_\""; } } } sub sub_headings { if (/\@chapter/) { $CH++; $SEC1 = 0; s,\@chapter\s*(.*),

    $CH: $1

    ,; } elsif (/\@unnumbered/) { $CH++; $SEC1 = 0; s,\@unnumbered\s*(.*),

    $CH: $1

    ,; } elsif (/\@section/) { $SEC1++; $SEC2 = 0; s,\@section\s*(.*),

    $CH.$SEC1: $1

    ,; } elsif (/\@subsection/) { $SEC2++; $SEC3 = 0; s,\@subsection\s*(.*),

    $CH.$SEC1.$SEC2: $1

    ,; } elsif (/\@subsubsection/) { $SEC3++; s,\@subsubsection\s*(.*),

    $CH.$SEC1.$SEC2.$SEC3: $1

    ,; } } sub sub_emphasis { s,<<,<<,g; s,>>,>>,g; s,\@emph\{([^}]*)\},$1,g; s,\@strong\{([^}]*)\},$1,g; s,\@footnote\{([^}]*)\}, [$1],g; s,\@b\{([^}]*)\},$1,g; s,\@code\{([^}]*)\},`$1',g; s,\@\@,\@,g; s,\@samp\{([^}]*)\},`$1',g; s,\@key\{([^}]*)\},`$1',g; s,\@kbd\{([^}]*)\},`$1',g; s,\@file\{([^}]*)\},`$1',g; s,TEXINFO2HTML-CLOSE-BRACE,},g; s,TEXINFO2HTML-AT-AT,\@,g; s,TEXINFO2HTML-ACCENT-ACUTE-a,á,g; s,TEXINFO2HTML-ACCENT-ACUTE-e,é,g; s,TEXINFO2HTML-ACCENT-ACUTE-o,ó,g; }