Add script to find undocumented API

Also tweaks to find-doc-nits, including name/synopsis checking.
Ironically, it also reports on duplicated doc names :)

Reviewed-by: Richard Levitte <levitte@openssl.org>

2 files changed, 159 insertions, 28 deletions

diff --git a/util/find-doc-nits.pl b/util/find-doc-nits.pl
index 4c6f0dbdd8..b0fab95353 100644..100755
--- a/util/find-doc-nits.pl
+++ b/util/find-doc-nits.pl
@@ -13,6 +13,9 @@ use strict;
 use Pod::Checker;
 use File::Find;
 use File::Basename;
+use Getopt::Std;
+
+our($opt_s);
 
 my $temp = '/tmp/docnits.txt';
 my $OUT;
@@ -28,6 +31,46 @@ my %default_sections =
       crypto => 3,
       ssl    => 3 );
 
+# Cross-check functions in the NAME and SYNOPSIS section.
+sub name_synopsis()
+{
+    my $id = shift;
+    my $filename = shift;
+    my $contents = shift;
+
+    # If it's a generic page (all lowercase), or apps, skip it.
+    return if $filename =~ /[a-z]+\.pod/;
+    return if $filename =~ m@/apps/@;
+
+    # Get NAME section and all words in it.
+    return unless $contents =~ /=head1 NAME(.*)=head1 SYNOPSIS/ms;
+    my $tmp = $1;
+    $tmp =~ tr/\n/ /;
+    $tmp =~ s/-.*//g;
+    $tmp =~ s/,//g;
+    my %names;
+    foreach my $n ( split ' ', $tmp ) {
+        $names{$n} = 1;
+    }
+
+    # Find all functions in SYNOPSIS
+    return unless $contents =~ /=head1 SYNOPSIS(.*)=head1 DESCRIPTION/ms;
+    my $syn = $1;
+    foreach my $line ( split /\n+/, $syn ) {
+        next if $line =~ /typedef/;
+        next if $line =~ /STACK_OF/;
+        next unless $line =~ /([A-Za-z0-9_]+)\(/;
+        print "$id $1 missing from NAME section\n"
+            unless defined $names{$1};
+        $names{$1} = 2;
+    }
+
+    foreach my $n ( keys %names ) {
+        next if $names{$n} == 2;
+        print "$id $n missing from SYNOPSIS\n";
+    }
+}
+
 sub check()
 {
     my $filename = shift;
@@ -42,22 +85,27 @@ sub check()
     }
 
     my $id = "${filename}:1:";
-    print $OUT "$id doesn't start with =pod\n"
+
+    &name_synopsis($id, $filename, $contents);
+
+    print "$id doesn't start with =pod\n"
         if $contents !~ /^=pod/;
-    print $OUT "$id doesn't end with =cut\n"
+    print "$id doesn't end with =cut\n"
         if $contents !~ /=cut\n$/;
-    print $OUT "$id more than one cut line.\n"
+    print "$id more than one cut line.\n"
         if $contents =~ /=cut.*=cut/ms;
-    print $OUT "$id missing copyright\n"
+    print "$id missing copyright\n"
         if $contents !~ /Copyright .* The OpenSSL Project Authors/;
-    print $OUT "$id copyright not last\n"
+    print "$id copyright not last\n"
         if $contents =~ /head1 COPYRIGHT.*=head/ms;
-    print $OUT "$id head2 in All uppercase\n"
+    print "$id head2 in All uppercase\n"
         if $contents =~ /head2\s+[A-Z ]+\n/;
-    print $OUT "$id period in NAME section\n"
-        if $contents =~ /NAME.*\.\n.*SYNOPSIS/ms;
-    print $OUT "$id POD markup in NAME section\n"
-        if $contents =~ /NAME.*[<>].*SYNOPSIS/ms;
+    print "$id extra space after head\n"
+        if $contents =~ /=head\d\s\s+/;
+    print "$id period in NAME section\n"
+        if $contents =~ /=head1 NAME.*\.\n.*=head1 SYNOPSIS/ms;
+    print "$id POD markup in NAME section\n"
+        if $contents =~ /=head1 NAME.*[<>].*=head1 SYNOPSIS/ms;
 
     # Look for multiple consecutive openssl #include lines.
     # Consecutive because of files like md5.pod. Sometimes it's okay
@@ -68,7 +116,7 @@ sub check()
             foreach my $line ( split /\n+/, $1 ) {
                 if ( $line =~ m@include <openssl/@ ) {
                     if ( ++$count == 2 ) {
-                        print $OUT "$id has multiple includes\n";
+                        print "$id has multiple includes\n";
                     }
                 } else {
                     $count = 0;
@@ -77,6 +125,8 @@ sub check()
         }
     }
 
+    return unless $opt_s;
+
     # Find what section this page is in.  If run from "." assume
     # section 3.
     my $section = $default_sections{$dirname} || 3;
@@ -85,29 +135,28 @@ sub check()
     }
 
     foreach ((@{$mandatory_sections{'*'}}, @{$mandatory_sections{$section}})) {
-        print $OUT "$id doesn't have a head1 section matching $_\n"
+        print "$id doesn't have a head1 section matching $_\n"
             if $contents !~ /^=head1\s+${_}\s*$/m;
     }
 
+    open my $OUT, '>', $temp
+        or die "Can't open $temp, $!";
     podchecker($filename, $OUT);
+    close $OUT;
+    open $OUT, '<', $temp
+        or die "Can't read $temp, $!";
+    while ( <$OUT> ) {
+        next if /\(section\) in.*deprecated/;
+        print;
+    }
+    close $OUT;
+    unlink $temp || warn "Can't remove $temp, $!";
 }
 
-open $OUT, '>', $temp
-    or die "Can't open $temp, $!";
-foreach (@ARGV ? @ARGV : glob('*/*.pod')) {
+getopts('s');
+
+foreach (@ARGV ? @ARGV : glob('doc/*/*.pod')) {
     &check($_);
 }
-close $OUT;
-
-my $count = 0;
-open $OUT, '<', $temp
-    or die "Can't read $temp, $!";
-while ( <$OUT> ) {
-    next if /\(section\) in.*deprecated/;
-    $count++;
-    print;
-}
-close $OUT;
-unlink $temp || warn "Can't remove $temp, $!";
 
-exit $count;
+exit;
diff --git a/util/find-undoc-api.pl b/util/find-undoc-api.pl
new file mode 100755
index 0000000000..7b2cb973b7
--- /dev/null
+++ b/util/find-undoc-api.pl
@@ -0,0 +1,82 @@
+#! /usr/bin/env perl
+# Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
+#
+# Licensed under the OpenSSL license (the "License").  You may not use
+# this file except in compliance with the License.  You can obtain a copy
+# in the file LICENSE in the source distribution or at
+# https://www.openssl.org/source/license.html
+
+use strict;
+use warnings;
+
+use File::Spec::Functions;
+use File::Basename;
+#use File::Copy;
+#use File::Path;
+use lib catdir(dirname($0), "perl");
+use OpenSSL::Util::Pod;
+
+my %dups;
+
+sub parsenum()
+{
+    my $file = shift;
+    my @apis;
+
+    open my $IN, '<', $file
+        or die "Can't open $file, $!, stopped";
+
+    while ( <$IN> ) {
+        next if /\sNOEXIST/;
+        next if /EXPORT_VAR_AS_FUNC/;
+        push @apis, $1 if /([^\s]+).\s/;
+    }
+
+    close $IN;
+
+    print "# Found ", scalar(@apis), " in $file\n";
+    return sort @apis;
+}
+
+sub getdocced()
+{
+    my $dir = shift;
+    my %return;
+
+    foreach my $pod ( glob("$dir/*.pod") ) {
+        next if $pod eq 'doc/crypto/crypto.pod';
+        next if $pod eq 'doc/ssl/ssl.pod';
+        my %podinfo = extract_pod_info($pod);
+        foreach my $n ( @{$podinfo{names}} ) {
+            $return{$n} = $pod;
+            print "# Duplicate $n in $pod and $dups{$n}\n"
+                if defined $dups{$n};
+            $dups{$n} = $pod;
+        }
+    }
+
+    return %return;
+}
+
+sub printem()
+{
+    my $docdir = shift;
+    my $numfile = shift;
+    my %docced = &getdocced($docdir);
+    my $count = 0;
+
+    foreach my $func ( &parsenum($numfile) ) {
+        next if $docced{$func};
+
+        # Skip ASN1 utilities
+        next if $func =~ /^ASN1_/;
+
+        print $func, "\n";
+        $count++;
+    }
+    print "# Found $count missing from $numfile\n\n";
+}
+
+
+&printem('doc/crypto', 'util/libcrypto.num');
+&printem('doc/ssl', 'util/libssl.num');