gobozgz / wxr2blogger Goto Github PK

OVERVIEW
--------

This program converts a WordPress WXR file into a series of Atom
files suitable for importing into Blogger.  As of 8/26/08 you need
to be using Blogger in Draft to access the import functionality.

Post contents and dates are preserved.  Comment contents, author
details and dates are preserved.  See the KNOWN ISSUES section below
for caveats.


USAGE
-----

Command line:

  % wxr2blogger wxrfile author id

  wxrfile is the WXR file from the WordPress export feature.
  author is the name of the author of the posts: e.g. "Joe Soap"
  id is the Blogger ID of the author of the posts.

The output is left in files named "output-XXX.xml", where XXX is a
number.

You can get your Blogger ID by going into "View Profile" from your
Blogger dashboard and looking at the URL in your browser's location
bar.  It's the big number at the end of the URL.  Copy it exactly
as you see it, leading zeros and all.


GENERAL COMMENTS
----------------

  * This program is licensed under version 2 of the GNU General
    Public License.

  * This code was written and tested on Mac OS X 10.5 only.  I don't
    believe I'm using any modules that you won't find on every
    modern Python distribution, but let me know if you run into
    problems running this on Linux or some other OS.

  * Send any patches or bug reports to [email protected]


KNOWN ISSUES
------------

  * If you see an error message like this:

        xml.parsers.expat.ExpatError: unbound prefix: line 1421, column 0

    then you've run into a problem with malformed WXR output from
    WordPress.  Unfortunately, until the WordPress developers address
    this, you will need to use a workaround.  Look at the top of
    your WXR file - you should see something that looks like this:

        <rss version="2.0"
               xmlns:content="http://purl.org/rss/1.0/modules/content/"
               xmlns:wfw="http://wellformedweb.org/CommentAPI/"
               xmlns:dc="http://purl.org/dc/elements/1.1/"
               xmlns:wp="http://wordpress.org/export/1.0/"
        >

    Add the following in before the last ">":

               xmlns:excerpt="http://www.example.com/the/excerpt/namespace"

    so that the it looks like this:

        <rss version="2.0"
               xmlns:content="http://purl.org/rss/1.0/modules/content/"
               xmlns:wfw="http://wellformedweb.org/CommentAPI/"
               xmlns:dc="http://purl.org/dc/elements/1.1/"
               xmlns:wp="http://wordpress.org/export/1.0/"
               xmlns:excerpt="http://www.example.com/the/excerpt/namespace"
        >

    Then try again.  See the WordPress ticket here:

        http://trac.wordpress.org/ticket/7624

    and the wxr2blogger ticket here:

        http://code.google.com/p/wxr2blogger/issues/detail?id=1

  * Blogger seems to choke with large numbers of post in one Atom
    file, which is why this program generates a series of Atom files
    instead.  Feel free to experiment with the number of posts.
    Look at top of the wxr2blogger script for the line that reads:

        POSTCOUNT = 4

    Change that to a larger number and see if Blogger accepts that.

  * Blogger stops importing new entries after a while: looks like
    there's a daily limit.  Unfortunately, the actual import appears
    to succeed, and you only notice something has gone wrong when
    trying to view a post fails.  Make sure you view all posts after
    importing to ensure you haven't hit this limit.  If you have,
    wait a day (although I've noticed a few hours seems to work,
    too) and continue importing then.

  * Before publishing your blog entries, you might want to go through
    them and make sure the HTML came across as you expected.  WXR
    files don't have any images or attachments in them, so you need
    to manually add them back in.  For images, that's fairly
    straight-forward.  For attachments, it's a bit harder as Blogger
    doesn't support attachments.

  * I haven't tested every scenario, so if you hit a failure, let
    me know.  One area that might cause problems is with categories
    that have special characters (like ampersands) in them.  Let
    me know if you hit this and what you did to fix it.

  * All posts are assumed to have been created by the author specified
    on the command line.  This would probably be relatively
    straight-forward to fix with some form of
    WordPress-author-to-Blogger-ID mapping.  If you're interested
    in doing this, let me know.

  * There are no comments in the code.  Some day I might get to
    fixing that.