X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/3d8876fcebe385593e503c80fd82539f2662a7d4..2f8f985a27faf25c5a535cbe67d098690668c0f9:/installhtml diff --git a/installhtml b/installhtml old mode 100755 new mode 100644 index b5406c5..9f6ef7e --- a/installhtml +++ b/installhtml @@ -4,9 +4,11 @@ use strict; use Config; # for config options in the makefile +use File::Path qw(remove_tree); +use File::Spec::Functions qw(rel2abs no_upwards); use Getopt::Long; # for command-line parsing use Cwd; -use Pod::Html; +use Pod::Html 1.23 qw(anchorify relativize_url); =head1 NAME @@ -14,16 +16,17 @@ installhtml - converts a collection of POD pages to HTML format. =head1 SYNOPSIS - installhtml [--help] [--podpath=:...:] [--podroot=] - [--htmldir=] [--htmlroot=] [--norecurse] [--recurse] - [--splithead=,...,] [--splititem=,...,] - [--libpods=,...,] [--verbose] + installhtml [--help] [--podpath=:...:] [--podroot=] + [--htmldir=] [--htmlroot=] [--norecurse] [--recurse] + [--splithead=,...,] [--splititem=,...,] + [--ignore=,...,] [--verbose] =head1 DESCRIPTION I converts a collection of POD pages to a corresponding -collection of HTML pages. This is primarily used to convert the pod -pages found in the perl distribution. +collection of HTML pages. This is used to convert the pod pages found in the +perl distribution. (It is not intended as a general-purpose +converter/installer of POD pages in HTML format. See L.) =head1 OPTIONS @@ -41,7 +44,7 @@ Default is current directory. =item B<--podpath> POD search path The list of directories to search for .pod and .pm files to be converted. -Default is `podroot/.'. +Default is 'podroot/.'. =item B<--recurse> recurse on subdirectories @@ -56,7 +59,7 @@ be a path relative to the filesystem, not the resulting URL. =item B<--htmlroot> URL base directory The base directory which all resulting HTML files will be visible at in -a URL. The default is `/'. +a URL. The default is '/'. =item B<--splithead> POD files to split on =head directive @@ -66,19 +69,20 @@ relative to podroot. =item B<--splititem> POD files to split on =item directive -Comma-separated list of all pod files to split by the =item directive. -The .pod suffix is optional. I does not do the actual -split, rather it invokes I to do the dirty work. As with ---splithead, these files should have names specified relative to podroot. +Comma-separated list of all pod files to split by the =item directive. The +.pod suffix is optional. I does not do the actual split, rather +it invokes I, a separate program in the Perl 5 core distribution, +to do the dirty work. As with --splithead, these files should have names +specified relative to podroot. =item B<--splitpod> Directory containing the splitpod program -The directory containing the splitpod program. The default is `podroot/pod'. +The directory containing the splitpod program. The default is 'podroot/pod'. -=item B<--libpods> library PODs for LEE links +=item B<--ignore> files to be ignored -Comma-separated list of "library" pod files. This is the same list that -will be passed to pod2html when any pod is converted. +Comma-separated of files that shouldn't be installed, given relative +to podroot. =item B<--verbose> verbose output @@ -97,7 +101,6 @@ perl documentation: --htmlroot=/perl/nmanual \ --splithead=pod/perlipc \ --splititem=pod/perlfunc \ - --libpods=perlfunc,perlguts,perlvar,perlrun,perlop \ --recurse \ --verbose @@ -105,8 +108,6 @@ perl documentation: Chris Hall Ehallc@cs.colorado.eduE -=head1 TODO - =cut my $usage; @@ -115,7 +116,7 @@ $usage =<:...: --podroot= --htmldir= --htmlroot= --norecurse --recurse --splithead=,..., --splititem=,..., - --libpods=,..., --verbose + --ignore=,..., --verbose --help - this message --podpath - colon-separated list of directories containing .pod and @@ -123,12 +124,9 @@ Usage: $0 --help --podpath=:...: --podroot= --podroot - filesystem base directory from which all relative paths in podpath stem (default is .). --htmldir - directory to store resulting html files in relative - to the filesystem (\$podroot/html by default). + to the filesystem (\$podroot/html by default). --htmlroot - http-server base directory from which all relative paths in podpath stem (default is /). - --libpods - comma-separated list of files to search for =item pod - directives in as targets of C<> and implicit links (empty - by default). --norecurse - don't recurse on those subdirectories listed in podpath. (default behavior). --recurse - recurse on those subdirectories listed in podpath @@ -139,14 +137,14 @@ Usage: $0 --help --podpath=:...: --podroot= splitpod. --splitpod - directory where the program splitpod can be found (\$podroot/pod by default). + --ignore - comma-separated list of files that shouldn't be installed. --verbose - self-explanatory. END_OF_USAGE -my (@libpods, @podpath, $podroot, $htmldir, $htmlroot, $recurse, @splithead, - @splititem, $splitpod, $verbose, $pod2html); +my (@podpath, $podroot, $htmldir, $htmlroot, $recurse, @splithead, + @splititem, $splitpod, $verbose, $pod2html, @ignore); -@libpods = (); @podpath = ( "." ); # colon-separated list of directories containing .pod # and .pm files to be converted. $podroot = "."; # assume the pods we want are here @@ -157,28 +155,26 @@ $recurse = 0; # default behavior @splititem = (); # don't split any files by default $splitpod = ""; # nothing for now. -$verbose = 0; # whether or not to print debugging info +$verbose = 0; # whether or not to print debugging info $pod2html = "pod/pod2html"; usage("") unless @ARGV; # Overcome shell's p1,..,p8 limitation. -# See vms/descrip_mms.template -> descrip.mms for invokation. +# See vms/descrip_mms.template -> descrip.mms for invocation. if ( $^O eq 'VMS' ) { @ARGV = split(/\s+/,$ARGV[0]); } -use vars qw($opt_htmldir $opt_htmlroot $opt_podroot $opt_splitpod - $opt_verbose $opt_help $opt_podpath $opt_splithead $opt_splititem - $opt_libpods $opt_recurse); +our %Options; # parse the command-line -my $result = GetOptions( qw( +my $result = GetOptions( \%Options, qw( help podpath=s podroot=s htmldir=s htmlroot=s - libpods=s + ignore=s recurse! splithead=s splititem=s @@ -204,11 +200,10 @@ $splitpod = "$podroot/pod" unless $splitpod; # ignored in the conversion process. these are files that have been # process by splititem or splithead and should not be converted as a # result. -my @ignore = (); my @splitdirs; -# split pods. its important to do this before convert ANY pods because -# it may effect some of the links +# split pods. It's important to do this before convert ANY pods because +# it may affect some of the links @splitdirs = (); # files in these directories won't get an index split_on_head($podroot, $htmldir, \@splitdirs, \@ignore, @splithead); split_on_item($podroot, \@splitdirs, \@ignore, @splititem); @@ -229,6 +224,7 @@ foreach my $dir (@splititem) { } foreach my $dir (@splithead) { + (my $pod = $dir) =~ s,^.*/,,; $dir .= ".pod" unless $dir =~ /(\.pod|\.pm)$/; # let pod2html create the file runpod2html($dir, 1); @@ -240,29 +236,31 @@ foreach my $dir (@splithead) { # read in everything until what would have been the first =head # directive, patching the index as we go. - open(H, "<$file.html") || + open(H, '<', "$file.html") || die "$0: error opening $file.html for input: $!\n"; $/ = ""; my @data = (); while () { - last if /NAME=/; - $_ =~ s{HREF="#(.*)">}{ - my $url = "$file/$1.html" ; - $url = Pod::Html::relativize_url( $url, "$file.html" ) - if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) ; - "HREF=\"$url\">" ; - }eg; + last if m!

NAME

!; + $_ =~ s{href="#(.*)">}{ + my $url = "$file/@{[anchorify(qq($1))]}.html" ; + $url = relativize_url( $url, "$file.html" ) + if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' ); + "href=\"$url\">" ; + }egi; push @data, $_; - } + } close(H); - # now rewrite the file - open(H, ">$file.html") || + # now rewrite the file + open(H, '>', "$file.html") || die "$0: error opening $file.html for output: $!\n"; - print H "@data", "\n"; + print H @data, "\n\n\n\n\n"; close(H); } +remove_tree(@splitdirs, {safe=>1}); + ############################################################################## @@ -273,88 +271,66 @@ sub usage { sub parse_command_line { - usage() if defined $opt_help; - $opt_help = ""; # make -w shut up + usage() if defined $Options{help}; + $Options{help} = ""; # make -w shut up # list of directories - @podpath = split(":", $opt_podpath) if defined $opt_podpath; + @podpath = split(":", $Options{podpath}) if defined $Options{podpath}; # lists of files - @splithead = split(",", $opt_splithead) if defined $opt_splithead; - @splititem = split(",", $opt_splititem) if defined $opt_splititem; - @libpods = split(",", $opt_libpods) if defined $opt_libpods; + @splithead = split(",", $Options{splithead}) if defined $Options{splithead}; + @splititem = split(",", $Options{splititem}) if defined $Options{splititem}; - $htmldir = $opt_htmldir if defined $opt_htmldir; - $htmlroot = $opt_htmlroot if defined $opt_htmlroot; - $podroot = $opt_podroot if defined $opt_podroot; - $splitpod = $opt_splitpod if defined $opt_splitpod; - - $recurse = $opt_recurse if defined $opt_recurse; - $verbose = $opt_verbose if defined $opt_verbose; -} + $htmldir = $Options{htmldir} if defined $Options{htmldir}; + $htmlroot = $Options{htmlroot} if defined $Options{htmlroot}; + $podroot = $Options{podroot} if defined $Options{podroot}; + $splitpod = $Options{splitpod} if defined $Options{splitpod}; + $recurse = $Options{recurse} if defined $Options{recurse}; + $verbose = $Options{verbose} if defined $Options{verbose}; -sub absolute_path { - my($cwd, $path) = @_; - return "$cwd/$path" unless $path =~ m:/:; - # add cwd if path is not already an absolute path - $path = "$cwd/$path" if (substr($path,0,1) ne '/'); - return $path; + @ignore = map "$podroot/$_", split(",", $Options{ignore}) if defined $Options{ignore}; } sub create_index { my($html, $dir) = @_; - my(@files, @filedata, @index, $file); - my($lcp1,$lcp2); - + (my $pod = $dir) =~ s,^.*/,,; # get the list of .html files in this directory opendir(DIR, $dir) || die "$0: error opening directory $dir for reading: $!\n"; - @files = sort(grep(/\.html?$/, readdir(DIR))); + my @files = sort(grep(/\.html?$/, readdir(DIR))); closedir(DIR); - open(HTML, ">$html") || + open(HTML, '>', $html) || die "$0: error opening $html for output: $!\n"; # for each .html file in the directory, extract the index # embedded in the file and throw it into the big index. print HTML "
\n"; - foreach $file (@files) { - $/ = ""; + foreach my $file (@files) { - open(IN, "<$dir/$file") || - die "$0: error opening $dir/$file for input: $!\n"; - @filedata = ; - close(IN); + my $filedata = do { + open(my $in, '<', "$dir/$file") || + die "$0: error opening $dir/$file for input: $!\n"; + local $/ = undef; + <$in>; + }; # pull out the NAME section - my $name; - ($name) = grep(/NAME=/, @filedata); - ($lcp1,$lcp2) = ($name =~ m,/H1>\s(\S+)\s[\s-]*(.*?)\s*$,sm); - if (defined $lcp1 and $lcp1 eq '

') { # Uninteresting. Try again. - ($lcp1,$lcp2) = ($name =~ m,/H1>\s

\s(\S+)\s[\s-]*(.*?)\s*$,sm); - } + my($lcp1, $lcp2) = + ($filedata =~ + m#

NAME

\s*

\s*(\S+)\s+-\s+(\S.*?\S)

#); + defined $lcp1 or die "$0: can't find NAME section in $dir/$file\n"; + my $url= "$dir/$file" ; - if ( ! defined $opt_htmlroot || $opt_htmlroot eq '' ) { - $url = Pod::Html::relativize_url( "$dir/$file", $html ) ; + if ( ! defined $Options{htmlroot} || $Options{htmlroot} eq '' ) { + $url = relativize_url( $url, $html ) ; } - print HTML qq(); - print HTML "
$lcp1
$lcp2\n" if defined $lcp1; -# print HTML qq($lcp1
\n") if defined $lcp1; - - next; - - @index = grep(/.*/s, - @filedata); - for (@index) { - s/(\s*\s*)/$lcp2/s; - s,#,$dir/$file#,g; - # print HTML "$_\n"; - print HTML "$_\n

\n"; - } + print HTML qq(

); + print HTML "$lcp1
$lcp2
\n"; } print HTML "
\n"; @@ -371,7 +347,7 @@ sub split_on_head { foreach $pod (@splithead) { # figure out the directory name and filename $pod =~ s,^([^/]*)$,/$1,; - $pod =~ m,(.*?)/(.*?)(\.pod)?$,; + $pod =~ m,(.*)/(.*?)(\.pod)?$,; $dirname = $1; $filename = "$2.pod"; @@ -391,11 +367,12 @@ sub split_on_item { print "splitting files by item.\n" if $verbose && $#splititem >= 0; $pwd = getcwd(); - my $splitter = absolute_path($pwd, "$splitpod/splitpod"); + my $splitter = rel2abs("$splitpod/splitpod", $pwd); + my $perl = rel2abs($^X, $pwd); foreach my $pod (@splititem) { # figure out the directory to split into $pod =~ s,^([^/]*)$,/$1,; - $pod =~ m,(.*?)/(.*?)(\.pod)?$,; + $pod =~ m,(.*)/(.*?)(\.pod)?$,; $dirname = "$1/$2"; $filename = "$2.pod"; @@ -404,15 +381,14 @@ sub split_on_item { # split the pod push(@$splitdirs, "$podroot/$dirname"); - if (! -d "$podroot/$dirname") { - mkdir("$podroot/$dirname", 0755) || + -d "$podroot/$dirname" and remove_tree("$podroot/$dirname", {safe=>1}); + mkdir("$podroot/$dirname", 0755) || die "$0: error creating directory $podroot/$dirname: $!\n"; - } chdir("$podroot/$dirname") || die "$0: error changing to directory $podroot/$dirname: $!\n"; die "$splitter not found. Use '-splitpod dir' option.\n" unless -f $splitter; - system("perl", $splitter, "../$filename") && + system($perl, $splitter, "../$filename") && warn "$0: error running '$splitter ../$filename'" ." from $podroot/$dirname"; } @@ -434,7 +410,7 @@ sub splitpod { # read the file in paragraphs $/ = ""; - open(SPLITIN, "<$pod") || + open(SPLITIN, '<', $pod) || die "$0: error opening $pod for input: $!\n"; @filedata = ; close(SPLITIN) || @@ -454,7 +430,7 @@ sub splitpod { # L<> links as necessary. my %heads = (); foreach $i (0..$#poddata) { - $heads{htmlize($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/; + $heads{anchorify($1)} = 1 if $poddata[$i] =~ /=head[1-6]\s+(.*)/; } # create a directory of a similar name and store all the @@ -463,9 +439,9 @@ sub splitpod { my $dir = $pod; $dir =~ s/\.pod//g; push(@$splitdirs, "$poddir/$dir"); + -d "$poddir/$dir" and remove_tree("$poddir/$dir", {safe=>1}); mkdir("$poddir/$dir", 0755) || - die "$0: could not create directory $poddir/$dir: $!\n" - unless -d "$poddir/$dir"; + die "$0: could not create directory $poddir/$dir: $!\n"; $poddata[0] =~ /^\s*=head[1-6]\s+(.*)/; $section = ""; @@ -486,15 +462,15 @@ sub splitpod { # determine an appropriate filename (this must correspond with # what pod2html will try and guess) # $poddata[$i] =~ /^\s*=head[1-6]\s+(.*)/; - $file = "$dir/" . htmlize($section) . ".pod"; + $file = "$dir/" . anchorify($section) . ".pod"; # create the new .pod file print "\tcreating $poddir/$file\n" if $verbose; - open(SPLITOUT, ">$poddir/$file") || + open(SPLITOUT, '>', "$poddir/$file") || die "$0: error opening $poddir/$file for output: $!\n"; $poddata[$i] =~ s,L<([^<>]*)>, - defined $heads{htmlize($1)} ? "L<$dir/$1>" : "L<$1>" - ,ge; + defined $heads{anchorify($1)} ? "L<$dir/$1>" : "L<$1>" + ,ge; print SPLITOUT $poddata[$i]."\n\n"; print SPLITOUT "=over 4\n\n"; print SPLITOUT "=item *\n\nBack to L<$dir/\"$prevsec\">\n\n" if $prevsec; @@ -513,34 +489,34 @@ sub splitpod { # sub installdir { my($dir, $recurse, $podroot, $splitdirs, $ignore) = @_; - my(@dirlist, @podlist, @pmlist, $doindex); - @dirlist = (); # directories to recurse on - @podlist = (); # .pod files to install - @pmlist = (); # .pm files to install + my @dirlist; # directories to recurse on + my @podlist; # .pod files to install + my @pmlist; # .pm files to install # should files in this directory get an index? - $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1); + my $doindex = (grep($_ eq "$podroot/$dir", @$splitdirs) ? 0 : 1); opendir(DIR, "$podroot/$dir") || die "$0: error opening directory $podroot/$dir: $!\n"; - # find the directories to recurse on - @dirlist = map { if ($^O eq 'VMS') {/^(.*)\.dir$/i; "$dir/$1";} else {"$dir/$_";}} - grep(-d "$podroot/$dir/$_" && !/^\.{1,2}/, readdir(DIR)) if $recurse; - rewinddir(DIR); - - # find all the .pod files within the directory - @podlist = map { /^(.*)\.pod$/; "$dir/$1" } - grep(! -d "$podroot/$dir/$_" && /\.pod$/, readdir(DIR)); - rewinddir(DIR); - - # find all the .pm files within the directory - @pmlist = map { /^(.*)\.pm$/; "$dir/$1" } - grep(! -d "$podroot/$dir/$_" && /\.pm$/, readdir(DIR)); + while(readdir DIR) { + no_upwards($_) or next; + my $is_dir = -d "$podroot/$dir/$_"; + next if $is_dir and not $recurse; + my $target + = $is_dir ? \@dirlist + : s/\.pod$// ? \@podlist + : s/\.pm$// ? \@pmlist + : undef + ; + push @$target, "$dir/$_" if $target; + } closedir(DIR); + if ($^O eq 'VMS') { s/\.dir$//i for @dirlist } + # recurse on all subdirectories we kept track of foreach $dir (@dirlist) { installdir($dir, $recurse, $podroot, $splitdirs, $ignore); @@ -549,12 +525,13 @@ sub installdir { # install all the pods we found foreach my $pod (@podlist) { # check if we should ignore it. - next if grep($_ eq "$podroot/$pod.pod", @$ignore); + next if $pod =~ m(/t/); # comes from a test file + next if grep($_ eq "$pod.pod", @$ignore); # check if a .pm files exists too - if (grep($_ eq "$pod.pm", @pmlist)) { - print "$0: Warning both `$podroot/$pod.pod' and " - . "`$podroot/$pod.pm' exist, using pod\n"; + if (grep($_ eq $pod, @pmlist)) { + print "$0: Warning both '$podroot/$pod.pod' and " + . "'$podroot/$pod.pm' exist, using pod\n"; push(@ignore, "$pod.pm"); } runpod2html("$pod.pod", $doindex); @@ -563,6 +540,7 @@ sub installdir { # install all the .pm files we found foreach my $pm (@pmlist) { # check if we should ignore it. + next if $pm =~ m(/t/); # comes from a test file next if grep($_ eq "$pm.pm", @ignore); runpod2html("$pm.pm", $doindex); @@ -594,18 +572,14 @@ sub runpod2html { # invoke pod2html print "$podroot/$pod => $htmldir/$html\n" if $verbose; -#system("./pod2html", - Pod::Html'pod2html( - #Pod::Html'pod2html($pod2html, + Pod::Html::pod2html( "--htmldir=$htmldir", "--htmlroot=$htmlroot", "--podpath=".join(":", @podpath), - "--podroot=$podroot", "--netscape", + "--podroot=$podroot", + "--header", ($doindex ? "--index" : "--noindex"), "--" . ($recurse ? "" : "no") . "recurse", - ($#libpods >= 0) ? "--libpods=" . join(":", @libpods) : "", "--infile=$podroot/$pod", "--outfile=$htmldir/$html"); die "$0: error running $pod2html: $!\n" if $?; } - -sub htmlize { htmlify(0, @_) }