recode manual pages to UTF-8

[debhelper.git] / dh_installman

diff --git a/dh_installman b/dh_installman
index 6ee000919ec9509952a51c3082372286a080949c..55c573701d5f33af3b441f51584812ce892cbb17 100755 (executable)
--- a/dh_installman
+++ b/dh_installman
@@ -27,17 +27,21 @@ extension of ".3perl"). If your .TH line is incorrect or missing, the program
 may guess wrong based on the file extension.
 
 It also supports translated man pages, by looking for extensions
-like .ll.8 and .ll_LL.8
+like .ll.8 and .ll_LL.8, or by use of the --language switch.
 
 If dh_installman seems to install a man page into the wrong section or with
 the wrong extension, this is because the man page has the wrong section
 listed in its .TH line. Edit the man page and correct the section, and
-dh_installman will follow suit.  See to L<man(7)> for details about the .TH
+dh_installman will follow suit.  See L<man(7)> for details about the .TH
 section. If dh_installman seems to install a man page into a directory
 like /usr/share/man/pl/man1/, that is because your program has a
 name like "foo.pl", and dh_installman assumes that means it is translated
-into Polish. There is currently no support for resolving this ambiguity;
-programs in debian should proably not have extensions like that anyway.
+into Polish. Use --language=C to avoid this.
+
+dh_installman will use man to guess the character encoding of each manual
+page and convert it to UTF-8. If the guesswork fails for some reason, you
+can override it using an encoding declaration. See L<manconv(1)> for
+details.
 
 Any man page filenames specified as parameters will be installed into the
 first package dh_installman is told to act on. By default, this is the
@@ -60,6 +64,11 @@ is acting on contain ".so" links. If so, it changes them to symlinks.
 Install all files specified by command line parameters in ALL packages
 acted on.
 
+=item B<--language>=ll
+
+Use this to specify that the man pages being acted on are written in the
+specified language.
+
 =item I<manpage ...>
 
 Install these man pages into the first package acted on. (Or in all
@@ -82,6 +91,8 @@ my @sofiles;
 my @sodests;
 
 foreach my $package (@{$dh{DOPACKAGES}}) {
+       next if is_udeb($package);
+
        my $tmp=tmpdir($package);
        my $file=pkgfile($package,"manpages");
        my @manpages;
@@ -114,7 +125,7 @@ foreach my $package (@{$dh{DOPACKAGES}}) {
                        open (IN, $page) or die "$page: $!";
                }
                while (<IN>) {
-                       if (/^\.TH\s+\S+\s+(\d+\S*)/) {
+                       if (/^\.TH\s+\S+\s+"?(\d+[^"\s]*)"?/) {
                                $section=$1;
                                last;
                        }
@@ -136,26 +147,43 @@ foreach my $package (@{$dh{DOPACKAGES}}) {
                my ($instname)=$basename=~m/^(.*)\./;
        
                my $destdir="$tmp/usr/share/man/man$realsection/";
-               # Translated man pages are typically specified by adding the
-               # language code to the filename, so detect that and
-               # redirect to appropriate directory, stripping the code.
-               my ($langcode)=$basename=~m/.*\.([a-z][a-z](?:_[A-Z][A-Z])?)\.(?:[1-9]|man)/;
+               my $langcode;
+               if (! defined $dh{LANGUAGE} || ! exists $dh{LANGUAGE}) {
+                       # Translated man pages are typically specified by adding the
+                       # language code to the filename, so detect that and
+                       # redirect to appropriate directory, stripping the code.
+                       ($langcode)=$basename=~m/.*\.([a-z][a-z](?:_[A-Z][A-Z])?)\.(?:[1-9]|man)/;
+               }
+               elsif ($dh{LANGUAGE} ne 'C') {
+                       $langcode=$dh{LANGUAGE};
+               }
+               
                if (defined $langcode && $langcode ne '') {
-                       $destdir="$tmp/usr/share/man/$langcode/man$realsection/";
                        # Strip the language code from the instname.
                        $instname=~s/\.$langcode$//;
                }
+               
+               if (defined $langcode && $langcode ne '') {
+                       $destdir="$tmp/usr/share/man/$langcode/man$realsection/";
+               }
                $destdir=~tr:/:/:s; # just for looks
+               my $instpage="$destdir/$instname.$section";
 
-               if (! -e "$destdir/$instname.$section" && 
-                   ! -l "$destdir/$instname.$section") {
-                       if (! -d $destdir) {
-                               doit "install","-d",$destdir;
-                       }
-                       doit "install","-p","-m644",$page,
-                               "$destdir$instname.$section$gz";
+               next if -l $instpage;
+               next if compat(5) && -e $instpage;
+
+               if (! -d $destdir) {
+                       doit "install","-d",$destdir;
                }
-               
+               if ($gz) {
+                       complex_doit "zcat \Q$page\E > \Q$instpage\E";
+               }
+               else {
+                       doit "install","-p","-m644",$page,$instpage;
+               }
+               complex_doit "man --recode UTF-8 \Q$instpage\E > \Q$instpage.new\E";
+               doit "chmod",644,"$instpage.new";
+               doit "mv","$instpage.new",$instpage;
        }
 
        # Now the .so conversion.
@@ -181,10 +209,20 @@ sub find_so_man {
        }
 
        # Test first line of file for the .so thing.
-       open (SOTEST,$_) || die "$_: $!";
+       if (/\.gz$/) {
+               open (SOTEST, "zcat $_|") or die "$_: $!";
+       }
+       else {
+               open (SOTEST,$_) || die "$_: $!";
+       }
        my $l=<SOTEST>;
        close SOTEST;
-       if ($l=~m/\.so\s+(.*)/) {
+
+       if (! defined $l) {
+               error("failed to read $_");
+       }
+       
+       if ($l=~m/\.so\s+(.*)\s*/) {
                my $solink=$1;
                # This test is here to prevent links like ... man8/../man8/foo.8
                if (basename($File::Find::dir) eq