[linux.git] / scripts / get_abi.pl

#!/usr/bin/env perl
# SPDX-License-Identifier: GPL-2.0

use strict;
use warnings;
use utf8;
use Pod::Usage;
use Getopt::Long;
use File::Find;
use Fcntl ':mode';

my $help = 0;
my $man = 0;
my $debug = 0;
my $enable_lineno = 0;
my $prefix="Documentation/ABI";

#
# If true, assumes that the description is formatted with ReST
#
my $description_is_rst = 1;

GetOptions(
	"debug|d+" => \$debug,
	"enable-lineno" => \$enable_lineno,
	"rst-source!" => \$description_is_rst,
	"dir=s" => \$prefix,
	'help|?' => \$help,
	man => \$man
) or pod2usage(2);

pod2usage(1) if $help;
pod2usage(-exitstatus => 0, -verbose => 2) if $man;

pod2usage(2) if (scalar @ARGV < 1 || @ARGV > 2);

my ($cmd, $arg) = @ARGV;

pod2usage(2) if ($cmd ne "search" && $cmd ne "rest" && $cmd ne "validate");
pod2usage(2) if ($cmd eq "search" && !$arg);

require Data::Dumper if ($debug);

my %data;
my %symbols;

#
# Displays an error message, printing file name and line
#
sub parse_error($$$$) {
	my ($file, $ln, $msg, $data) = @_;

	$data =~ s/\s+$/\n/;

	print STDERR "Warning: file $file#$ln:\n\t$msg";

	if ($data ne "") {
		print STDERR ". Line\n\t\t$data";
	} else {
	    print STDERR "\n";
	}
}

#
# Parse an ABI file, storing its contents at %data
#
sub parse_abi {
	my $file = $File::Find::name;

	my $mode = (stat($file))[2];
	return if ($mode & S_IFDIR);
	return if ($file =~ m,/README,);

	my $name = $file;
	$name =~ s,.*/,,;

	my $fn = $file;
	$fn =~ s,Documentation/ABI/,,;

	my $nametag = "File $fn";
	$data{$nametag}->{what} = "File $name";
	$data{$nametag}->{type} = "File";
	$data{$nametag}->{file} = $name;
	$data{$nametag}->{filepath} = $file;
	$data{$nametag}->{is_file} = 1;
	$data{$nametag}->{line_no} = 1;

	my $type = $file;
	$type =~ s,.*/(.*)/.*,$1,;

	my $what;
	my $new_what;
	my $tag = "";
	my $ln;
	my $xrefs;
	my $space;
	my @labels;
	my $label = "";

	print STDERR "Opening $file\n" if ($debug > 1);
	open IN, $file;
	while(<IN>) {
		$ln++;
		if (m/^(\S+)(:\s*)(.*)/i) {
			my $new_tag = lc($1);
			my $sep = $2;
			my $content = $3;

			if (!($new_tag =~ m/(what|where|date|kernelversion|contact|description|users)/)) {
				if ($tag eq "description") {
					# New "tag" is actually part of
					# description. Don't consider it a tag
					$new_tag = "";
				} elsif ($tag ne "") {
					parse_error($file, $ln, "tag '$tag' is invalid", $_);
				}
			}

			# Invalid, but it is a common mistake
			if ($new_tag eq "where") {
				parse_error($file, $ln, "tag 'Where' is invalid. Should be 'What:' instead", "");
				$new_tag = "what";
			}

			if ($new_tag =~ m/what/) {
				$space = "";
				$content =~ s/[,.;]$//;

				push @{$symbols{$content}->{file}}, " $file:" . ($ln - 1);

				if ($tag =~ m/what/) {
					$what .= ", " . $content;
				} else {
					if ($what) {
						parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});

						foreach my $w(split /, /, $what) {
							$symbols{$w}->{xref} = $what;
						};
					}

					$what = $content;
					$label = $content;
					$new_what = 1;
				}
				push @labels, [($content, $label)];
				$tag = $new_tag;

				push @{$data{$nametag}->{symbols}}, $content if ($data{$nametag}->{what});
				next;
			}

			if ($tag ne "" && $new_tag) {
				$tag = $new_tag;

				if ($new_what) {
					@{$data{$what}->{label_list}} = @labels if ($data{$nametag}->{what});
					@labels = ();
					$label = "";
					$new_what = 0;

					$data{$what}->{type} = $type;
					if (!defined($data{$what}->{file})) {
						$data{$what}->{file} = $name;
						$data{$what}->{filepath} = $file;
					} else {
						if ($name ne $data{$what}->{file}) {
							$data{$what}->{file} .= " " . $name;
							$data{$what}->{filepath} .= " " . $file;
						}
					}
					print STDERR "\twhat: $what\n" if ($debug > 1);
					$data{$what}->{line_no} = $ln;
				} else {
					$data{$what}->{line_no} = $ln if (!defined($data{$what}->{line_no}));
				}

				if (!$what) {
					parse_error($file, $ln, "'What:' should come first:", $_);
					next;
				}
				if ($new_tag eq "description") {
					$sep =~ s,:, ,;
					$content = ' ' x length($new_tag) . $sep . $content;
					while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
					if ($content =~ m/^(\s*)(\S.*)$/) {
						# Preserve initial spaces for the first line
						$space = $1;
						$content = "$2\n";
						$data{$what}->{$tag} .= $content;
					} else {
						undef($space);
					}

				} else {
					$data{$what}->{$tag} = $content;
				}
				next;
			}
		}

		# Store any contents before tags at the database
		if (!$tag && $data{$nametag}->{what}) {
			$data{$nametag}->{description} .= $_;
			next;
		}

		if ($tag eq "description") {
			my $content = $_;
			while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
			if (m/^\s*\n/) {
				$data{$what}->{$tag} .= "\n";
				next;
			}

			if (!defined($space)) {
				# Preserve initial spaces for the first line
				if ($content =~ m/^(\s*)(\S.*)$/) {
					$space = $1;
					$content = "$2\n";
				}
			} else {
				$space = "" if (!($content =~ s/^($space)//));
			}
			$data{$what}->{$tag} .= $content;

			next;
		}
		if (m/^\s*(.*)/) {
			$data{$what}->{$tag} .= "\n$1";
			$data{$what}->{$tag} =~ s/\n+$//;
			next;
		}

		# Everything else is error
		parse_error($file, $ln, "Unexpected content", $_);
	}
	$data{$nametag}->{description} =~ s/^\n+// if ($data{$nametag}->{description});
	if ($what) {
		parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});

		foreach my $w(split /, /,$what) {
			$symbols{$w}->{xref} = $what;
		};
	}
	close IN;
}

sub create_labels {
	my %labels;

	foreach my $what (keys %data) {
		next if ($data{$what}->{file} eq "File");

		foreach my $p (@{$data{$what}->{label_list}}) {
			my ($content, $label) = @{$p};
			$label = "abi_" . $label . " ";
			$label =~ tr/A-Z/a-z/;

			# Convert special chars to "_"
			$label =~s/([\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\xff])/_/g;
			$label =~ s,_+,_,g;
			$label =~ s,_$,,;

			# Avoid duplicated labels
			while (defined($labels{$label})) {
			    my @chars = ("A".."Z", "a".."z");
			    $label .= $chars[rand @chars];
			}
			$labels{$label} = 1;

			$data{$what}->{label} = $label;

			# only one label is enough
			last;
		}
	}
}

#
# Outputs the book on ReST format
#

# \b doesn't work well with paths. So, we need to define something else:
# Boundaries are punct characters, spaces and end-of-line
my $start = qr {(^|\s|\() }x;
my $bondary = qr { ([,.:;\)\s]|\z) }x;
my $xref_match = qr { $start(\/(sys|config|proc|dev|kvd)\/[^,.:;\)\s]+)$bondary }x;
my $symbols = qr { ([\x01-\x08\x0e-\x1f\x21-\x2f\x3a-\x40\x7b-\xff]) }x;

sub output_rest {
	create_labels();

	my $part = "";

	foreach my $what (sort {
				($data{$a}->{type} eq "File") cmp ($data{$b}->{type} eq "File") ||
				$a cmp $b
			       } keys %data) {
		my $type = $data{$what}->{type};

		my @file = split / /, $data{$what}->{file};
		my @filepath = split / /, $data{$what}->{filepath};

		if ($enable_lineno) {
			printf "#define LINENO %s%s#%s\n\n",
			       $prefix, $file[0],
			       $data{$what}->{line_no};
		}

		my $w = $what;

		if ($type ne "File") {
			my $cur_part = $what;
			if ($what =~ '/') {
				if ($what =~ m#^(\/?(?:[\w\-]+\/?){1,2})#) {
					$cur_part = "Symbols under $1";
					$cur_part =~ s,/$,,;
				}
			}

			if ($cur_part ne "" && $part ne $cur_part) {
			    $part = $cur_part;
			    my $bar = $part;
			    $bar =~ s/./-/g;
			    print "$part\n$bar\n\n";
			}

			printf ".. _%s:\n\n", $data{$what}->{label};

			my @names = split /, /,$w;
			my $len = 0;

			foreach my $name (@names) {
				$name =~ s/$symbols/\\$1/g;
				$name = "**$name**";
				$len = length($name) if (length($name) > $len);
			}

			print "+-" . "-" x $len . "-+\n";
			foreach my $name (@names) {
				printf "| %s", $name . " " x ($len - length($name)) . " |\n";
				print "+-" . "-" x $len . "-+\n";
			}

			print "\n";
		}

		for (my $i = 0; $i < scalar(@filepath); $i++) {
			my $path = $filepath[$i];
			my $f = $file[$i];

			$path =~ s,.*/(.*/.*),$1,;;
			$path =~ s,[/\-],_,g;;
			my $fileref = "abi_file_".$path;

			if ($type eq "File") {
				print ".. _$fileref:\n\n";
			} else {
				print "Defined on file :ref:`$f <$fileref>`\n\n";
			}
		}

		if ($type eq "File") {
			my $bar = $w;
			$bar =~ s/./-/g;
			print "$w\n$bar\n\n";
		}

		my $desc = "";
		$desc = $data{$what}->{description} if (defined($data{$what}->{description}));
		$desc =~ s/\s+$/\n/;

		if (!($desc =~ /^\s*$/)) {
			if ($description_is_rst) {
				# Remove title markups from the description
				# Having titles inside ABI files will only work if extra
				# care would be taken in order to strictly follow the same
				# level order for each markup.
				$desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;

				# Enrich text by creating cross-references

				my $new_desc = "";
				my $init_indent = -1;
				my $literal_indent = -1;

				open(my $fh, "+<", \$desc);
				while (my $d = <$fh>) {
					my $indent = $d =~ m/^(\s+)/;
					my $spaces = length($indent);
					$init_indent = $indent if ($init_indent < 0);
					if ($literal_indent >= 0) {
						if ($spaces > $literal_indent) {
							$new_desc .= $d;
							next;
						} else {
							$literal_indent = -1;
						}
					} else {
						if ($d =~ /()::$/ && !($d =~ /^\s*\.\./)) {
							$literal_indent = $spaces;
						}
					}

					$d =~ s,Documentation/(?!devicetree)(\S+)\.rst,:doc:`/$1`,g;

					my @matches = $d =~ m,Documentation/ABI/([\w\/\-]+),g;
					foreach my $f (@matches) {
						my $xref = $f;
						my $path = $f;
						$path =~ s,.*/(.*/.*),$1,;;
						$path =~ s,[/\-],_,g;;
						$xref .= " <abi_file_" . $path . ">";
						$d =~ s,\bDocumentation/ABI/$f\b,:ref:`$xref`,g;
					}

					# Seek for cross reference symbols like /sys/...
					@matches = $d =~ m/$xref_match/g;

					foreach my $s (@matches) {
						next if (!($s =~ m,/,));
						if (defined($data{$s}) && defined($data{$s}->{label})) {
							my $xref = $s;

							$xref =~ s/$symbols/\\$1/g;
							$xref = ":ref:`$xref <" . $data{$s}->{label} . ">`";

							$d =~ s,$start$s$bondary,$1$xref$2,g;
						}
					}
					$new_desc .= $d;
				}
				close $fh;


				print "$new_desc\n\n";
			} else {
				$desc =~ s/^\s+//;

				# Remove title markups from the description, as they won't work
				$desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;

				if ($desc =~ m/\:\n/ || $desc =~ m/\n[\t ]+/  || $desc =~ m/[\x00-\x08\x0b-\x1f\x7b-\xff]/) {
					# put everything inside a code block
					$desc =~ s/\n/\n /g;

					print "::\n\n";
					print " $desc\n\n";
				} else {
					# Escape any special chars from description
					$desc =~s/([\x00-\x08\x0b-\x1f\x21-\x2a\x2d\x2f\x3c-\x40\x5c\x5e-\x60\x7b-\xff])/\\$1/g;
					print "$desc\n\n";
				}
			}
		} else {
			print "DESCRIPTION MISSING for $what\n\n" if (!$data{$what}->{is_file});
		}

		if ($data{$what}->{symbols}) {
			printf "Has the following ABI:\n\n";

			foreach my $content(@{$data{$what}->{symbols}}) {
				my $label = $data{$symbols{$content}->{xref}}->{label};

				# Escape special chars from content
				$content =~s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g;

				print "- :ref:`$content <$label>`\n\n";
			}
		}

		if (defined($data{$what}->{users})) {
			my $users = $data{$what}->{users};

			$users =~ s/\n/\n\t/g;
			printf "Users:\n\t%s\n\n", $users if ($users ne "");
		}

	}
}

#
# Searches for ABI symbols
#
sub search_symbols {
	foreach my $what (sort keys %data) {
		next if (!($what =~ m/($arg)/));

		my $type = $data{$what}->{type};
		next if ($type eq "File");

		my $file = $data{$what}->{filepath};

		my $bar = $what;
		$bar =~ s/./-/g;

		print "\n$what\n$bar\n\n";

		my $kernelversion = $data{$what}->{kernelversion} if (defined($data{$what}->{kernelversion}));
		my $contact = $data{$what}->{contact} if (defined($data{$what}->{contact}));
		my $users = $data{$what}->{users} if (defined($data{$what}->{users}));
		my $date = $data{$what}->{date} if (defined($data{$what}->{date}));
		my $desc = $data{$what}->{description} if (defined($data{$what}->{description}));

		$kernelversion =~ s/^\s+// if ($kernelversion);
		$contact =~ s/^\s+// if ($contact);
		if ($users) {
			$users =~ s/^\s+//;
			$users =~ s/\n//g;
		}
		$date =~ s/^\s+// if ($date);
		$desc =~ s/^\s+// if ($desc);

		printf "Kernel version:\t\t%s\n", $kernelversion if ($kernelversion);
		printf "Date:\t\t\t%s\n", $date if ($date);
		printf "Contact:\t\t%s\n", $contact if ($contact);
		printf "Users:\t\t\t%s\n", $users if ($users);
		print "Defined on file(s):\t$file\n\n";
		print "Description:\n\n$desc";
	}
}

# Ensure that the prefix will always end with a slash
# While this is not needed for find, it makes the patch nicer
# with --enable-lineno
$prefix =~ s,/?$,/,;

#
# Parses all ABI files located at $prefix dir
#
find({wanted =>\&parse_abi, no_chdir => 1}, $prefix);

print STDERR Data::Dumper->Dump([\%data], [qw(*data)]) if ($debug);

#
# Handles the command
#
if ($cmd eq "search") {
	search_symbols;
} else {
	if ($cmd eq "rest") {
		output_rest;
	}

	# Warn about duplicated ABI entries
	foreach my $what(sort keys %symbols) {
		my @files = @{$symbols{$what}->{file}};

		next if (scalar(@files) == 1);

		printf STDERR "Warning: $what is defined %d times: @files\n",
		    scalar(@files);
	}
}

__END__

=head1 NAME

abi_book.pl - parse the Linux ABI files and produce a ReST book.

=head1 SYNOPSIS

B<abi_book.pl> [--debug] [--enable-lineno] [--man] [--help]
	       [--(no-)rst-source] [--dir=<dir>] <COMAND> [<ARGUMENT>]

Where <COMMAND> can be:

=over 8

B<search> [SEARCH_REGEX] - search for [SEARCH_REGEX] inside ABI

B<rest>                  - output the ABI in ReST markup language

B<validate>              - validate the ABI contents

=back

=head1 OPTIONS

=over 8

=item B<--dir>

Changes the location of the ABI search. By default, it uses
the Documentation/ABI directory.

=item B<--rst-source> and B<--no-rst-source>

The input file may be using ReST syntax or not. Those two options allow
selecting between a rst-compliant source ABI (--rst-source), or a
plain text that may be violating ReST spec, so it requres some escaping
logic (--no-rst-source).

=item B<--enable-lineno>

Enable output of #define LINENO lines.

=item B<--debug>

Put the script in verbose mode, useful for debugging. Can be called multiple
times, to increase verbosity.

=item B<--help>

Prints a brief help message and exits.

=item B<--man>

Prints the manual page and exits.

=back

=head1 DESCRIPTION

Parse the Linux ABI files from ABI DIR (usually located at Documentation/ABI),
allowing to search for ABI symbols or to produce a ReST book containing
the Linux ABI documentation.

=head1 EXAMPLES

Search for all stable symbols with the word "usb":

=over 8

$ scripts/get_abi.pl search usb --dir Documentation/ABI/stable

=back

Search for all symbols that match the regex expression "usb.*cap":

=over 8

$ scripts/get_abi.pl search usb.*cap

=back

Output all obsoleted symbols in ReST format

=over 8

$ scripts/get_abi.pl rest --dir Documentation/ABI/obsolete

=back

=head1 BUGS

Report bugs to Mauro Carvalho Chehab <[email protected]>

=head1 COPYRIGHT

Copyright (c) 2016-2019 by Mauro Carvalho Chehab <[email protected]>.

License GPLv2: GNU GPL version 2 <http://gnu.org/licenses/gpl.html>.

This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

=cut
Commit	Line	Data
c25ce589	1	#!/usr/bin/env perl
ecb351f1	2	# SPDX-License-Identifier: GPL-2.0
bbc249f2 MCC	3
bbc249f2 MCC	4	use strict;
234948bf	5	use warnings;
55e5414f	6	use utf8;
bbc249f2 MCC	7	use Pod::Usage;
	8	use Getopt::Long;
	9	use File::Find;
	10	use Fcntl ':mode';
	11
234948bf MCC	12	my $help = 0;
	13	my $man = 0;
	14	my $debug = 0;
	15	my $enable_lineno = 0;
33e3e991	16	my $prefix="Documentation/ABI";
bbc249f2	17
11ce90a4 MCC	18	#
	19	# If true, assumes that the description is formatted with ReST
	20	#
2fcce37a	21	my $description_is_rst = 1;
11ce90a4	22
bbc249f2 MCC	23	GetOptions(
bbc249f2 MCC	24	"debug\|d+" => \$debug,
61439c4a	25	"enable-lineno" => \$enable_lineno,
11ce90a4	26	"rst-source!" => \$description_is_rst,
33e3e991	27	"dir=s" => \$prefix,
bbc249f2 MCC	28	'help\|?' => \$help,
	29	man => \$man
	30	) or pod2usage(2);
	31
	32	pod2usage(1) if $help;
	33	pod2usage(-exitstatus => 0, -verbose => 2) if $man;
	34
33e3e991	35	pod2usage(2) if (scalar @ARGV < 1 \|\| @ARGV > 2);
bbc249f2	36
33e3e991 MCC	37	my ($cmd, $arg) = @ARGV;
33e3e991 MCC	38
7ce7b89b	39	pod2usage(2) if ($cmd ne "search" && $cmd ne "rest" && $cmd ne "validate");
33e3e991	40	pod2usage(2) if ($cmd eq "search" && !$arg);
bbc249f2 MCC	41
	42	require Data::Dumper if ($debug);
	43
	44	my %data;
234948bf	45	my %symbols;
bbc249f2 MCC	46
	47	#
	48	# Displays an error message, printing file name and line
	49	#
	50	sub parse_error($$$$) {
	51	my ($file, $ln, $msg, $data) = @_;
	52
75442fb0 MCC	53	$data =~ s/\s+$/\n/;
	54
	55	print STDERR "Warning: file $file#$ln:\n\t$msg";
	56
	57	if ($data ne "") {
	58	print STDERR ". Line\n\t\t$data";
	59	} else {
	60	print STDERR "\n";
	61	}
bbc249f2 MCC	62	}
	63
	64	#
	65	# Parse an ABI file, storing its contents at %data
	66	#
	67	sub parse_abi {
	68	my $file = $File::Find::name;
	69
	70	my $mode = (stat($file))[2];
	71	return if ($mode & S_IFDIR);
	72	return if ($file =~ m,/README,);
	73
	74	my $name = $file;
	75	$name =~ s,.*/,,;
	76
a4ea67bc MCC	77	my $fn = $file;
	78	$fn =~ s,Documentation/ABI/,,;
	79
	80	my $nametag = "File $fn";
d0ebaf51 MCC	81	$data{$nametag}->{what} = "File $name";
	82	$data{$nametag}->{type} = "File";
	83	$data{$nametag}->{file} = $name;
33e3e991	84	$data{$nametag}->{filepath} = $file;
d0ebaf51	85	$data{$nametag}->{is_file} = 1;
61439c4a	86	$data{$nametag}->{line_no} = 1;
d0ebaf51	87
bbc249f2 MCC	88	my $type = $file;
	89	$type =~ s,./(.)/.*,$1,;
	90
	91	my $what;
	92	my $new_what;
234948bf	93	my $tag = "";
bbc249f2	94	my $ln;
6619c661	95	my $xrefs;
4e6a6234	96	my $space;
d0ebaf51	97	my @labels;
234948bf	98	my $label = "";
bbc249f2 MCC	99
	100	print STDERR "Opening $file\n" if ($debug > 1);
	101	open IN, $file;
	102	while(<IN>) {
	103	$ln++;
4e6a6234	104	if (m/^(\S+)(:\s)(.)/i) {
bbc249f2	105	my $new_tag = lc($1);
4e6a6234 MCC	106	my $sep = $2;
4e6a6234 MCC	107	my $content = $3;
bbc249f2	108
7ce7b89b	109	if (!($new_tag =~ m/(what\|where\|date\|kernelversion\|contact\|description\|users)/)) {
bbc249f2	110	if ($tag eq "description") {
4e6a6234 MCC	111	# New "tag" is actually part of
	112	# description. Don't consider it a tag
	113	$new_tag = "";
7d7ea8d2	114	} elsif ($tag ne "") {
bbc249f2 MCC	115	parse_error($file, $ln, "tag '$tag' is invalid", $_);
	116	}
	117	}
	118
2c0700e7 MCC	119	# Invalid, but it is a common mistake
2c0700e7 MCC	120	if ($new_tag eq "where") {
75442fb0	121	parse_error($file, $ln, "tag 'Where' is invalid. Should be 'What:' instead", "");
2c0700e7 MCC	122	$new_tag = "what";
	123	}
	124
bbc249f2	125	if ($new_tag =~ m/what/) {
4e6a6234	126	$space = "";
234948bf MCC	127	$content =~ s/[,.;]$//;
234948bf MCC	128
c7ba3334 MCC	129	push @{$symbols{$content}->{file}}, " $file:" . ($ln - 1);
c7ba3334 MCC	130
bbc249f2 MCC	131	if ($tag =~ m/what/) {
	132	$what .= ", " . $content;
	133	} else {
234948bf MCC	134	if ($what) {
	135	parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});
	136
	137	foreach my $w(split /, /, $what) {
c7ba3334	138	$symbols{$w}->{xref} = $what;
234948bf MCC	139	};
234948bf MCC	140	}
4e6a6234	141
bbc249f2	142	$what = $content;
d0ebaf51	143	$label = $content;
bbc249f2 MCC	144	$new_what = 1;
bbc249f2 MCC	145	}
d0ebaf51	146	push @labels, [($content, $label)];
bbc249f2	147	$tag = $new_tag;
6619c661	148
234948bf	149	push @{$data{$nametag}->{symbols}}, $content if ($data{$nametag}->{what});
bbc249f2 MCC	150	next;
	151	}
	152
7d7ea8d2	153	if ($tag ne "" && $new_tag) {
4e6a6234	154	$tag = $new_tag;
bbc249f2	155
4e6a6234	156	if ($new_what) {
234948bf	157	@{$data{$what}->{label_list}} = @labels if ($data{$nametag}->{what});
d0ebaf51 MCC	158	@labels = ();
d0ebaf51 MCC	159	$label = "";
4e6a6234	160	$new_what = 0;
bbc249f2	161
4e6a6234	162	$data{$what}->{type} = $type;
c7ba3334 MCC	163	if (!defined($data{$what}->{file})) {
	164	$data{$what}->{file} = $name;
	165	$data{$what}->{filepath} = $file;
	166	} else {
	167	if ($name ne $data{$what}->{file}) {
	168	$data{$what}->{file} .= " " . $name;
	169	$data{$what}->{filepath} .= " " . $file;
	170	}
	171	}
4e6a6234	172	print STDERR "\twhat: $what\n" if ($debug > 1);
c7ba3334 MCC	173	$data{$what}->{line_no} = $ln;
	174	} else {
	175	$data{$what}->{line_no} = $ln if (!defined($data{$what}->{line_no}));
4e6a6234	176	}
bbc249f2	177
4e6a6234 MCC	178	if (!$what) {
	179	parse_error($file, $ln, "'What:' should come first:", $_);
	180	next;
	181	}
f82a8a74 MCC	182	if ($new_tag eq "description") {
f82a8a74 MCC	183	$sep =~ s,:, ,;
11ce90a4	184	$content = ' ' x length($new_tag) . $sep . $content;
f82a8a74 MCC	185	while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
	186	if ($content =~ m/^(\s)(\S.)$/) {
	187	# Preserve initial spaces for the first line
11ce90a4	188	$space = $1;
f82a8a74 MCC	189	$content = "$2\n";
	190	$data{$what}->{$tag} .= $content;
	191	} else {
	192	undef($space);
4e6a6234	193	}
e9bca891	194
4e6a6234 MCC	195	} else {
	196	$data{$what}->{$tag} = $content;
	197	}
bbc249f2 MCC	198	next;
bbc249f2 MCC	199	}
bbc249f2 MCC	200	}
bbc249f2 MCC	201
4e6a6234	202	# Store any contents before tags at the database
d0ebaf51 MCC	203	if (!$tag && $data{$nametag}->{what}) {
d0ebaf51 MCC	204	$data{$nametag}->{description} .= $_;
6619c661 MCC	205	next;
6619c661 MCC	206	}
bbc249f2	207
4e6a6234	208	if ($tag eq "description") {
e9bca891 MCC	209	my $content = $_;
e9bca891 MCC	210	while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
f82a8a74 MCC	211	if (m/^\s*\n/) {
	212	$data{$what}->{$tag} .= "\n";
	213	next;
	214	}
	215
	216	if (!defined($space)) {
e9bca891	217	# Preserve initial spaces for the first line
f82a8a74	218	if ($content =~ m/^(\s)(\S.)$/) {
e9bca891	219	$space = $1;
f82a8a74	220	$content = "$2\n";
4e6a6234 MCC	221	}
4e6a6234 MCC	222	} else {
4e6a6234	223	$space = "" if (!($content =~ s/^($space)//));
4e6a6234	224	}
f82a8a74 MCC	225	$data{$what}->{$tag} .= $content;
f82a8a74 MCC	226
4e6a6234 MCC	227	next;
4e6a6234 MCC	228	}
bbc249f2 MCC	229	if (m/^\s(.)/) {
	230	$data{$what}->{$tag} .= "\n$1";
	231	$data{$what}->{$tag} =~ s/\n+$//;
	232	next;
	233	}
	234
	235	# Everything else is error
75442fb0	236	parse_error($file, $ln, "Unexpected content", $_);
bbc249f2	237	}
234948bf MCC	238	$data{$nametag}->{description} =~ s/^\n+// if ($data{$nametag}->{description});
	239	if ($what) {
	240	parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});
	241
	242	foreach my $w(split /, /,$what) {
c7ba3334	243	$symbols{$w}->{xref} = $what;
234948bf MCC	244	};
234948bf MCC	245	}
bbc249f2 MCC	246	close IN;
	247	}
	248
234948bf MCC	249	sub create_labels {
234948bf MCC	250	my %labels;
bbc249f2	251
234948bf MCC	252	foreach my $what (keys %data) {
234948bf MCC	253	next if ($data{$what}->{file} eq "File");
4e6a6234	254
234948bf	255	foreach my $p (@{$data{$what}->{label_list}}) {
d0ebaf51 MCC	256	my ($content, $label) = @{$p};
	257	$label = "abi_" . $label . " ";
	258	$label =~ tr/A-Z/a-z/;
	259
	260	# Convert special chars to "_"
	261	$label =~s/([\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\xff])/_/g;
	262	$label =~ s,_+,_,g;
	263	$label =~ s,_$,,;
	264
2e7ce055 MCC	265	# Avoid duplicated labels
	266	while (defined($labels{$label})) {
	267	my @chars = ("A".."Z", "a".."z");
	268	$label .= $chars[rand @chars];
	269	}
	270	$labels{$label} = 1;
	271
234948bf	272	$data{$what}->{label} = $label;
d0ebaf51 MCC	273
	274	# only one label is enough
	275	last;
6619c661	276	}
234948bf MCC	277	}
	278	}
	279
	280	#
	281	# Outputs the book on ReST format
	282	#
	283
50ebf8f4 MCC	284	# \b doesn't work well with paths. So, we need to define something else:
	285	# Boundaries are punct characters, spaces and end-of-line
	286	my $start = qr {(^\|\s\|\() }x;
	287	my $bondary = qr { ([,.:;\)\s]\|\z) }x;
87ec9ea1	288	my $xref_match = qr { $start(\/(sys\|config\|proc\|dev\|kvd)\/[^,.:;\)\s]+)$bondary }x;
b0f9580a	289	my $symbols = qr { ([\x01-\x08\x0e-\x1f\x21-\x2f\x3a-\x40\x7b-\xff]) }x;
55e5414f	290
234948bf MCC	291	sub output_rest {
	292	create_labels();
	293
9d4fdda3 MCC	294	my $part = "";
9d4fdda3 MCC	295
234948bf MCC	296	foreach my $what (sort {
	297	($data{$a}->{type} eq "File") cmp ($data{$b}->{type} eq "File") \|\|
	298	$a cmp $b
	299	} keys %data) {
	300	my $type = $data{$what}->{type};
c7ba3334 MCC	301
	302	my @file = split / /, $data{$what}->{file};
	303	my @filepath = split / /, $data{$what}->{filepath};
234948bf MCC	304
	305	if ($enable_lineno) {
	306	printf "#define LINENO %s%s#%s\n\n",
c7ba3334	307	$prefix, $file[0],
234948bf MCC	308	$data{$what}->{line_no};
234948bf MCC	309	}
6619c661	310
234948bf	311	my $w = $what;
6619c661	312
c7ba3334	313	if ($type ne "File") {
9d4fdda3 MCC	314	my $cur_part = $what;
	315	if ($what =~ '/') {
	316	if ($what =~ m#^(\/?(?:[\w\-]+\/?){1,2})#) {
	317	$cur_part = "Symbols under $1";
	318	$cur_part =~ s,/$,,;
	319	}
	320	}
	321
	322	if ($cur_part ne "" && $part ne $cur_part) {
	323	$part = $cur_part;
	324	my $bar = $part;
	325	$bar =~ s/./-/g;
	326	print "$part\n$bar\n\n";
	327	}
	328
234948bf	329	printf ".. _%s:\n\n", $data{$what}->{label};
45f96517	330
234948bf	331	my @names = split /, /,$w;
45f96517 MCC	332	my $len = 0;
	333
	334	foreach my $name (@names) {
b0f9580a	335	$name =~ s/$symbols/\\$1/g;
c01d62d3	336	$name = "$name";
45f96517 MCC	337	$len = length($name) if (length($name) > $len);
	338	}
	339
45f96517 MCC	340	print "+-" . "-" x $len . "-+\n";
	341	foreach my $name (@names) {
	342	printf "\| %s", $name . " " x ($len - length($name)) . " \|\n";
	343	print "+-" . "-" x $len . "-+\n";
	344	}
45f96517	345
c7ba3334 MCC	346	print "\n";
	347	}
	348
	349	for (my $i = 0; $i < scalar(@filepath); $i++) {
	350	my $path = $filepath[$i];
	351	my $f = $file[$i];
	352
	353	$path =~ s,./(./.*),$1,;;
	354	$path =~ s,[/\-],_,g;;
	355	my $fileref = "abi_file_".$path;
	356
	357	if ($type eq "File") {
c7ba3334	358	print ".. _$fileref:\n\n";
c7ba3334 MCC	359	} else {
	360	print "Defined on file :ref:`$f <$fileref>`\n\n";
	361	}
234948bf	362	}
bbc249f2	363
a4ea67bc MCC	364	if ($type eq "File") {
	365	my $bar = $w;
	366	$bar =~ s/./-/g;
	367	print "$w\n$bar\n\n";
	368	}
	369
234948bf MCC	370	my $desc = "";
	371	$desc = $data{$what}->{description} if (defined($data{$what}->{description}));
	372	$desc =~ s/\s+$/\n/;
bbc249f2	373
4e6a6234	374	if (!($desc =~ /^\s*$/)) {
11ce90a4	375	if ($description_is_rst) {
daaaf58a MCC	376	# Remove title markups from the description
	377	# Having titles inside ABI files will only work if extra
	378	# care would be taken in order to strictly follow the same
	379	# level order for each markup.
	380	$desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;
	381
55e5414f MCC	382	# Enrich text by creating cross-references
55e5414f MCC	383
c27c2e34	384	my $new_desc = "";
2ae7bb57 MCC	385	my $init_indent = -1;
	386	my $literal_indent = -1;
	387
c27c2e34 MCC	388	open(my $fh, "+<", \$desc);
c27c2e34 MCC	389	while (my $d = <$fh>) {
2ae7bb57 MCC	390	my $indent = $d =~ m/^(\s+)/;
	391	my $spaces = length($indent);
	392	$init_indent = $indent if ($init_indent < 0);
	393	if ($literal_indent >= 0) {
	394	if ($spaces > $literal_indent) {
	395	$new_desc .= $d;
	396	next;
	397	} else {
	398	$literal_indent = -1;
	399	}
	400	} else {
	401	if ($d =~ /()::$/ && !($d =~ /^\s*\.\./)) {
	402	$literal_indent = $spaces;
	403	}
	404	}
	405
c27c2e34 MCC	406	$d =~ s,Documentation/(?!devicetree)(\S+)\.rst,:doc:`/$1`,g;
	407
	408	my @matches = $d =~ m,Documentation/ABI/([\w\/\-]+),g;
	409	foreach my $f (@matches) {
	410	my $xref = $f;
	411	my $path = $f;
	412	$path =~ s,./(./.*),$1,;;
	413	$path =~ s,[/\-],_,g;;
	414	$xref .= " <abi_file_" . $path . ">";
	415	$d =~ s,\bDocumentation/ABI/$f\b,:ref:`$xref`,g;
	416	}
55e5414f	417
c27c2e34 MCC	418	# Seek for cross reference symbols like /sys/...
c27c2e34 MCC	419	@matches = $d =~ m/$xref_match/g;
55e5414f	420
c27c2e34 MCC	421	foreach my $s (@matches) {
	422	next if (!($s =~ m,/,));
	423	if (defined($data{$s}) && defined($data{$s}->{label})) {
	424	my $xref = $s;
55e5414f	425
c27c2e34 MCC	426	$xref =~ s/$symbols/\\$1/g;
c27c2e34 MCC	427	$xref = ":ref:`$xref <" . $data{$s}->{label} . ">`";
55e5414f	428
c27c2e34 MCC	429	$d =~ s,$start$s$bondary,$1$xref$2,g;
c27c2e34 MCC	430	}
55e5414f	431	}
c27c2e34	432	$new_desc .= $d;
55e5414f	433	}
c27c2e34 MCC	434	close $fh;
c27c2e34 MCC	435
55e5414f	436
c27c2e34	437	print "$new_desc\n\n";
4e6a6234	438	} else {
11ce90a4	439	$desc =~ s/^\s+//;
bbc249f2	440
11ce90a4 MCC	441	# Remove title markups from the description, as they won't work
	442	$desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;
	443
	444	if ($desc =~ m/\:\n/ \|\| $desc =~ m/\n[\t ]+/ \|\| $desc =~ m/[\x00-\x08\x0b-\x1f\x7b-\xff]/) {
	445	# put everything inside a code block
	446	$desc =~ s/\n/\n /g;
	447
	448	print "::\n\n";
	449	print " $desc\n\n";
	450	} else {
	451	# Escape any special chars from description
	452	$desc =~s/([\x00-\x08\x0b-\x1f\x21-\x2a\x2d\x2f\x3c-\x40\x5c\x5e-\x60\x7b-\xff])/\\$1/g;
	453	print "$desc\n\n";
	454	}
4e6a6234	455	}
bbc249f2	456	} else {
d0ebaf51	457	print "DESCRIPTION MISSING for $what\n\n" if (!$data{$what}->{is_file});
bbc249f2	458	}
6619c661	459
234948bf	460	if ($data{$what}->{symbols}) {
d0ebaf51 MCC	461	printf "Has the following ABI:\n\n";
d0ebaf51 MCC	462
234948bf	463	foreach my $content(@{$data{$what}->{symbols}}) {
c7ba3334	464	my $label = $data{$symbols{$content}->{xref}}->{label};
d0ebaf51 MCC	465
	466	# Escape special chars from content
	467	$content =~s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g;
	468
	469	print "- :ref:`$content <$label>`\n\n";
	470	}
	471	}
a16ab14e MCC	472
	473	if (defined($data{$what}->{users})) {
	474	my $users = $data{$what}->{users};
	475
	476	$users =~ s/\n/\n\t/g;
	477	printf "Users:\n\t%s\n\n", $users if ($users ne "");
	478	}
	479
bbc249f2 MCC	480	}
	481	}
	482
33e3e991 MCC	483	#
	484	# Searches for ABI symbols
	485	#
	486	sub search_symbols {
	487	foreach my $what (sort keys %data) {
	488	next if (!($what =~ m/($arg)/));
	489
	490	my $type = $data{$what}->{type};
	491	next if ($type eq "File");
	492
	493	my $file = $data{$what}->{filepath};
	494
	495	my $bar = $what;
	496	$bar =~ s/./-/g;
	497
	498	print "\n$what\n$bar\n\n";
	499
234948bf MCC	500	my $kernelversion = $data{$what}->{kernelversion} if (defined($data{$what}->{kernelversion}));
	501	my $contact = $data{$what}->{contact} if (defined($data{$what}->{contact}));
	502	my $users = $data{$what}->{users} if (defined($data{$what}->{users}));
	503	my $date = $data{$what}->{date} if (defined($data{$what}->{date}));
	504	my $desc = $data{$what}->{description} if (defined($data{$what}->{description}));
	505
	506	$kernelversion =~ s/^\s+// if ($kernelversion);
	507	$contact =~ s/^\s+// if ($contact);
	508	if ($users) {
	509	$users =~ s/^\s+//;
	510	$users =~ s/\n//g;
	511	}
	512	$date =~ s/^\s+// if ($date);
	513	$desc =~ s/^\s+// if ($desc);
33e3e991 MCC	514
	515	printf "Kernel version:\t\t%s\n", $kernelversion if ($kernelversion);
	516	printf "Date:\t\t\t%s\n", $date if ($date);
	517	printf "Contact:\t\t%s\n", $contact if ($contact);
	518	printf "Users:\t\t\t%s\n", $users if ($users);
c7ba3334	519	print "Defined on file(s):\t$file\n\n";
33e3e991 MCC	520	print "Description:\n\n$desc";
	521	}
	522	}
	523
61439c4a MCC	524	# Ensure that the prefix will always end with a slash
	525	# While this is not needed for find, it makes the patch nicer
	526	# with --enable-lineno
	527	$prefix =~ s,/?$,/,;
33e3e991	528
bbc249f2 MCC	529	#
	530	# Parses all ABI files located at $prefix dir
	531	#
	532	find({wanted =>\&parse_abi, no_chdir => 1}, $prefix);
	533
	534	print STDERR Data::Dumper->Dump([\%data], [qw(*data)]) if ($debug);
	535
	536	#
33e3e991	537	# Handles the command
bbc249f2	538	#
c7ba3334	539	if ($cmd eq "search") {
33e3e991	540	search_symbols;
c7ba3334 MCC	541	} else {
	542	if ($cmd eq "rest") {
	543	output_rest;
	544	}
	545
	546	# Warn about duplicated ABI entries
	547	foreach my $what(sort keys %symbols) {
	548	my @files = @{$symbols{$what}->{file}};
	549
	550	next if (scalar(@files) == 1);
bbc249f2	551
c7ba3334 MCC	552	printf STDERR "Warning: $what is defined %d times: @files\n",
	553	scalar(@files);
	554	}
	555	}
bbc249f2 MCC	556
	557	__END__
	558
	559	=head1 NAME
	560
	561	abi_book.pl - parse the Linux ABI files and produce a ReST book.
	562
	563	=head1 SYNOPSIS
	564
61439c4a MCC	565	B<abi_book.pl> [--debug] [--enable-lineno] [--man] [--help]
61439c4a MCC	566	[--(no-)rst-source] [--dir=<dir>] <COMAND> [<ARGUMENT>]
33e3e991 MCC	567
	568	Where <COMMAND> can be:
	569
	570	=over 8
	571
	572	B<search> [SEARCH_REGEX] - search for [SEARCH_REGEX] inside ABI
	573
7ce7b89b MCC	574	B<rest> - output the ABI in ReST markup language
	575
	576	B<validate> - validate the ABI contents
33e3e991 MCC	577
33e3e991 MCC	578	=back
bbc249f2 MCC	579
	580	=head1 OPTIONS
	581
	582	=over 8
	583
33e3e991 MCC	584	=item B<--dir>
	585
	586	Changes the location of the ABI search. By default, it uses
	587	the Documentation/ABI directory.
	588
11ce90a4 MCC	589	=item B<--rst-source> and B<--no-rst-source>
	590
	591	The input file may be using ReST syntax or not. Those two options allow
	592	selecting between a rst-compliant source ABI (--rst-source), or a
	593	plain text that may be violating ReST spec, so it requres some escaping
	594	logic (--no-rst-source).
	595
61439c4a MCC	596	=item B<--enable-lineno>
	597
	598	Enable output of #define LINENO lines.
	599
bbc249f2 MCC	600	=item B<--debug>
	601
	602	Put the script in verbose mode, useful for debugging. Can be called multiple
	603	times, to increase verbosity.
	604
	605	=item B<--help>
	606
	607	Prints a brief help message and exits.
	608
	609	=item B<--man>
	610
	611	Prints the manual page and exits.
	612
	613	=back
	614
	615	=head1 DESCRIPTION
	616
33e3e991 MCC	617	Parse the Linux ABI files from ABI DIR (usually located at Documentation/ABI),
	618	allowing to search for ABI symbols or to produce a ReST book containing
	619	the Linux ABI documentation.
	620
	621	=head1 EXAMPLES
	622
	623	Search for all stable symbols with the word "usb":
	624
	625	=over 8
	626
	627	$ scripts/get_abi.pl search usb --dir Documentation/ABI/stable
	628
	629	=back
	630
	631	Search for all symbols that match the regex expression "usb.*cap":
	632
	633	=over 8
	634
	635	$ scripts/get_abi.pl search usb.*cap
	636
	637	=back
	638
	639	Output all obsoleted symbols in ReST format
	640
	641	=over 8
	642
	643	$ scripts/get_abi.pl rest --dir Documentation/ABI/obsolete
	644
	645	=back
bbc249f2 MCC	646
	647	=head1 BUGS
	648
7ce7b89b	649	Report bugs to Mauro Carvalho Chehab <[email protected]>
bbc249f2 MCC	650
	651	=head1 COPYRIGHT
	652
7ce7b89b	653	Copyright (c) 2016-2019 by Mauro Carvalho Chehab <[email protected]>.
bbc249f2 MCC	654
	655	License GPLv2: GNU GPL version 2 <http://gnu.org/licenses/gpl.html>.
	656
	657	This is free software: you are free to change and redistribute it.
	658	There is NO WARRANTY, to the extent permitted by law.
	659
	660	=cut