#!/usr/bin/env perl
#***************************************************************************
# _ _ ____ _
# Project ___| | | | _ \| |
# / __| | | | |_) | |
# | (__| |_| | _ <| |___
# \___|\___/|_| \_\_____|
#
# Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
#
# This software is licensed as described in the file COPYING, which
# you should have received as part of this distribution. The terms
# are also available at https://curl.se/docs/copyright.html.
#
# You may opt to use, copy, modify, merge, publish, distribute and/or sell
# copies of the Software, and permit persons to whom the Software is
# furnished to do so, under the terms of the COPYING file.
#
# This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
# KIND, either express or implied.
#
# SPDX-License-Identifier: curl
#
###########################################################################
=begin comment
This script converts an nroff file to curldown
Example: cd2nroff [options] <file.md> > <file.3>
Note: when converting .nf sections, this tool does not know if the
section is code or just regular quotes. It then assumes and uses ~~~c
for code.
=end comment
=cut
my $nroff2cd = "0.1"; # to keep check
sub single {
my ($f)=@_;
open(F, "<:crlf", "$f") ||
return 1;
my $line;
my $title;
my $section;
my $source;
my @seealso;
my @desc;
my $header; # non-zero when TH is passed
my $quote = 0; # quote state
while(<F>) {
$line++;
my $d = $_;
if($_ =~ /^.\\\"/) {
# a comment we can ignore
next;
}
if(!$header) {
if($d =~ /.so (.*)/) {
# this is basically an include, so do that
my $f = $1;
# remove leading directory
$f =~ s/(.*?\/)//;
close(F);
open(F, "<:crlf", "$f") || return 1;
}
if($d =~ /^\.TH ([^ ]*) (\d) \"(.*?)\" ([^ \n]*)/) {
# header, this needs to be the first thing after leading comments
$title = $1;
$section = $2;
# date is $3
$source = $4;
# if there are enclosing quotes around source, remove them
$source =~ s/[\"\'](.*)[\"\']\z/$1/;
$header = 1;
print <<HEAD
---
c: Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
SPDX-License-Identifier: curl
Title: $title
Section: $section
Source: $source
HEAD
;
}
next;
}
if($quote) {
if($d =~ /^\.SH/) {
#end of quote without an .fi
$quote = 0;
push @desc, "~~~\n";
}
elsif($d =~ /^\.fi/) {
#end of quote
$quote = 0;
push @desc, "~~~\n";
next;
}
else {
# double-backslashes converted to single ones
$d =~ s/\\\\/\\/g;
push @desc, $d;
next;
}
}
if($d =~ /^\.SH (.*)/) {
my $word = $1;
# if there are enclosing quotes, remove them first
$word =~ s/[\"\'](.*)[\"\']\z/$1/;
if($word eq "SEE ALSO") {
# we just slurp up this section
next;
}
push @desc, "\n# $word\n\n";
}
elsif($d =~ /^\.(RS|RE)/) {
# ignore these
}
elsif($d =~ /^\.IP (.*)/) {
my $word = $1;
# if there are enclosing quotes, remove them first
$word =~ s/[\"\'](.*)[\"\']\z/$1/;
push @desc, "\n## $word\n\n";
}
elsif($d =~ /^\.IP/) {
# .IP with no text we just skip
}
elsif($d =~ /^\.BR (.*)/) {
# only used for SEE ALSO
my $word = $1;
# remove trailing comma
$word =~ s/,\z//;
for my $s (split(/,/, $word)) {
# remove all double quotes
$s =~ s/\"//g;
# tream leading whitespace
$s =~ s/^ +//g;
push @seealso, $s;
}
}
elsif($d =~ /^\.I (.*)/) {
push @desc, "*$1*\n";
}
elsif($d =~ /^\.B (.*)/) {
push @desc, "**$1**\n";
}
elsif($d =~ /^\.nf/) {
push @desc, "~~~c\n";
$quote = 1;
}
else {
# embolden
$d =~ s/\\fB(.*?)\\fP/**$1**/g;
# links to "curl.*()" are left bare since cd2nroff handles them
# specially
$d =~ s/\\fI(curl.*?\(3\))\\fP/$1/ig;
# emphasize
$d =~ s/\\fI(.*?)\\fP/*$1*/g;
# emphasize on a split line
$d =~ s/\\fI/*/g;
# bold on a split line
$d =~ s/\\fB/**/g;
# remove backslash amp
$d =~ s/\\&//g;
# remove backslashes
$d =~ s/\\//g;
# fix single quotes
$d =~ s/\(aq/'/g;
# fix double quotes
$d =~ s/\(dq/\"/g;
push @desc, $d;
}
}
close(F);
print "See-also:\n";
for my $s (sort @seealso) {
print " - $s\n" if($s);
}
print "---\n";
print @desc;
return !$header;
}
exit single($ARGV[0]);