ポイント

• WordPress用のxml-rpcのメソッド　wp.suggestCategories　を使用
• 記事の移行だけでメディアデータは持ってきていません

前回と同じく重複チェックには穴があるので要注意。

#!/usr/bin/perl

use strict;
use warnings;
use Encode;
use LWP::UserAgent;
use HTTP::Request;
use XML::LibXML;
use XMLRPC::Lite;
#use XMLRPC::Lite trace => 'debug' ;
use Data::Dumper;
use File::Basename;
use MIME::Base64;
use MIME::Parser;
use Getopt::Std;
use vars qw(%interface $username $passwd $target $postcnt);

BEGIN {
	$interface{'from'}
		= 'http://www.example.com/blog/xmlrpc.php';
	$interface{'to'}
		= 'http://www.example.net/blog/xmlrpc.php';
	$username = 'xxxxxx';
	$passwd  ='xxxxxx';
	$target  = '';
	$postcnt = 100;
}

# get original data
my $rpc = XMLRPC::Lite->new();
$rpc->proxy($interface{'from'});
my $res = $rpc->call("metaWeblog.getRecentPosts",
	1,	# blog ID
	$username,
	$passwd,
	$postcnt,
)->result;

$rpc->proxy($interface{'to'});
foreach my $content (@{$res}) {
	# get posts
	#print Dumper $content;
	my ($title, $description, $date, $excerpt, $wp_slug, $keyword);
	$title = $content->{'title'};
	$description = $content->{'description'};
	$date = $content->{'dateCreated'};
	$excerpt = $content->{'excerpt'};
	$wp_slug = $content->{'wp_slug'};
	$keyword = $content->{'mt_keywords'};

	# check data
	$target = "http://www.example.net/blog/uncategorized/"
		. $wp_slug
		. "/";
	my $ua = new LWP::UserAgent;
	my $req = new HTTP::Request(HEAD => $target);
	my $resp = $ua->request($req);
	if($resp->{'_rc'} eq '200'){
		print STDERR "Alredy convert URL: $target\n";
		next;
	}

	# edit posts

	# get category_id
	$res = $rpc->call("wp.suggestCategories",
		1,
		$username,
		$passwd,
		Encode::encode('utf8', $content->{'categories'}[0]),
	)->result;
	my $category_id = $res->[0]{'category_id'};

	$res = $rpc->call("metaWeblog.newPost",
		1,	# blog ID
		$username,
		$passwd,
		{
		'title'
			=> Encode::encode('utf8', "$title"),
		'description'
			=> Encode::encode('utf8', "$description"),
		'dateCreated'
			=> Encode::encode('utf8', "$date"),
		'mt_excerpt'
			=> Encode::encode('utf8', "$excerpt"),
		'mt_keywords'
			=> Encode::encode('utf8', "$keyword"),
		'wp_slug'
			=> Encode::encode('utf8', "$wp_slug"),
		#'custom_fields'=> [
		#	{
		#	'key'   => '',
		#	'value' => '',
		#	},
		#],
		},
		1
	)->result;
	my $postid = $res;

	$res = $rpc->call("mt.setPostCategories",
		$postid,
		$username,
		$passwd,
		[
			{
				"categoryId" => $category_id,
				"isPrimary"  => 1
			}
		]
	)-> result;
}

exit 1;

Programming Web Services With Xml-Rpc

著者/訳者：Simon St. Laurent

定価：￥ 2,802

Oreilly & Associates Inc

EAN：9780596001193

---

以下注意点

1. STDINがshiftjisなのはxmlの文字コードが実はshiftjisだったから
2. check wordpressの部分はサーバーの設定のよって違うと思うので適宜変更のこと
3. for文が降順なのはxmlのデータが生成時間の新しいものから並んでいるのでそれをwordpressでも引き継ぎたかったから

しかし、wordpressには罠が潜んでした・・・。
wordpressは存在しない投稿スラッグを指定した場合、指定された投稿スラッグに近い投稿スラッグがあればそちらにリダイレクトしてくれる。
たとえばこの記事のパーマリンクは
http://www.oganosin.net/blog/programming/xml-rpc-part2/
なんだけど
http://www.oganosin.net/blog/programming/xml-rpc-part/
でもアクセスできるはず。
なので下記のスクリプトの重複チェックには穴があります。

#!/usr/bin/perl

use strict;
use warnings;
use Encode;
use encoding "utf-8", STDIN => "shiftjis", STDOUT => "utf-8";
use LWP::UserAgent;
use HTTP::Request;
use XML::LibXML;
use XMLRPC::Lite;
#use XMLRPC::Lite trace => 'debug' ;
use Data::Dumper;
use File::Basename;
use MIME::Base64;
use MIME::Parser;
use Getopt::Std;
use vars qw($interfaceurl $blogid $username $passwd $target);

BEGIN {
	$interfaceurl = "http://www.example.com/wp/xmlrpc.php";
	$blogid = 1;
	$username = 'xxxxx';
	$passwd   = 'xxxxx';
	$target = '';
}

sub read_bin_file(){
	my ($filename) = @_;
	#--------------------
	# read in the picture
	my $fh = IO::Handle->new();
	open($fh, $filename) or die "$! $filename";
	local($/) = undef;  # slurp
	binmode($fh);
	my $bits  = <$fh>;
	close($fh);
	return $bits;
}

# get XML data
my $ua = new LWP::UserAgent;
my $req = new HTTP::Request(GET =>
	"http://www.example.com/data.xml");
my $resp = $ua->request($req);

my $parser = XML::LibXML->new();
my $doc = $parser->parse_string($resp->content);

my $rpc = XMLRPC::Lite->new();
my @items = $doc->getElementsByTagName('item');
for(my $i=$#items; $i>=0; $i--){
	my ($wp_slug,$description,$excerpt,$img);
	# edit data

	# check wordpress
	my $posturl = "http://www.example.com/wp/uncategorize/"
		. $wp_slug
		. "/";
	my $req = new HTTP::Request(HEAD => $posturl);
	my $resp = $ua->request($req);
	if($resp->{'_rc'} eq '200'){
		print STDERR "Alredy convert URL: $posturl\n";
		next;
	}

	my $rpc = XMLRPC::Lite->new();
	$rpc->proxy($interfaceurl);
	my $res = $rpc->call("metaWeblog.newMediaObject",
		$blogid,
		$username,
		$passwd,
		{
			'bits' => XMLRPC::Data->type('base64',
				&read_bin_file($img)),
			'type' => "image/jpeg",
			'name' => basename($img),
		}
	)->result;
	$img = $res->{'url'};

	$res = $rpc->call("metaWeblog.newPost",
		$blogid,
		$username,
		$passwd,
		{
		'title' =>
			Encode::encode('utf8', "$title"),
		'description'  =>
			Encode::encode('utf8', "$description"),
		'dateCreated'  =>
			Encode::encode('utf8', "$date"),
		'mt_excerpt'  =>
			Encode::encode('utf8', "$excerpt"),
		'wp_slug'  => $wp_slug,
		'custom_fields'=> [
			{
			'key'   => '',
			'value' => '',
			},
		],
		},
		1
	)->result;

	my $postid = $res;
	$res = $rpc->call("mt.setPostCategories",
		$postid,
		$username,
		$passwd,
		[
			{
				"categoryId" => 1,
				"isPrimary"  => 1
			}
		]
	)-> result;
}
exit 1;

#!/usr/bin/perl

use strict;
use warnings;
use Encode;
use XMLRPC::Lite;
#use XMLRPC::Lite trace => 'debug' ;
use Data::Dumper;
use vars qw($interface $username $passwd $target $postscnt);

BEGIN {
	$interface = 'http://www.example.com/blog/xmlrpc.php';
	$username  = 'xxxxx';
	$passwd    = 'xxxxx';
	$postscnt  = 100;
}

my $rpc = XMLRPC::Lite->new();
$rpc->proxy($interface);
#$rpc->authorization_basic('id', 'pwd');
my $res = $rpc->call("metaWeblog.getRecentPosts",
	1,      # blog ID
	$username,
	$passwd,
	$postscnt,
)->result;
print Dumper $res;
exit 1;