wget -nH --mirror -l1 http://www.livejournal.com/users/[username] (plus whatever switch you'd want to prevent it from grabbing your friends list, and calendar etc.)
of course, that'd be a bit harsh on the lj server, so you'd probably want to throw a --limit-rate 10 on it and let it run over night. (or the next week)

oh, and the force html switch.

Don't forget to save the cookie to grab your protected entries.

Might be better to use curl.

yeah, you should probably point to your browsers cookie file, instead of command line username/userpass, since i think there is some MD5 thingy.
anyway, i don't know curl. though it may be a better option, since a ? in a url seems to confound wget.

That's why you quote the URL, always quote the URL =P It only "confounds" the shell, in that it's intended behavior, in bash atleast(it's valid syntax).

does that work on a recursive wget though?
it seems to have troubles with anything that assumes you are using a dynamic web browser.

I've never had issues, all the dynamicness is server side, as far as a client is concerned, all pages are static (with the exception of best cacheing practices).

humm. i'll have to try that out.
php is always problematic though. worst case was i ended up with 5 gb of php permutations.

That way I'm still dependent on LiveJournal.

if you get the page requisites, it should save everything in HTML format, using your layout. the -l2 should grab all icons and other trifles...
but maybe i am not understanding your question.

Well, I need LiveJournal to add new entries... generating static html pages directly from text snippets shouldn't be too hard to do. I would write such a tool for myself if not the most problematic point would be to create great designs... For example, look at blogger.com. They do that on the server side: you can hit a button and they regenerate the HTML pages -- you can even tell them to upload them somewhere via FTP. I'd like to have the HTML generation on the client side.

Btw, just copying the LiveJournal pages is not a good idea because it will result in a lot of stale links (comments won't work, for example, but I don't need the comment feature -- reply by email is way good enough for my purposes).

http://wordpress.org/ ?
or are you looking for a WYSIWYG client to format your posts ala livejournals formatting, then allowing you to take that data and transfer it to another, more stable forum?

it seems that your pretty much want a stylesheet to lay over the xml?
that seems to be a whole other service, or a feature request you are looking for.
in wich case it seems like this is better suited for
http://www.livejournal.com/suggestions/

http://wordpress.org/ ?
or are you looking for a WYSIWYG client to format your posts ala livejournals formatting, then allowing you to take that data and transfer it to another, more stable forum?

Not forum, just a webserver serving static html pages. WordPress requires PHP and MySQL and is in no way static.

it seems that your pretty much want a stylesheet to lay over the xml?Almost. Except that archive pages and TOCs should be generated.

the things you want are why they created dynamic websites . . . . Is there any reason you need this to be static?

logjam's xml export is way more friendly to lj servers.

i'd suggest using that and some xslt to render the pages.

i don't think logjam will ever support creating static html pages; it's just too much feature creep. i might accept a patch that does the xslt+scp on post, but i dunno.

I have written a small tool in Ruby that takes the synchronized copy and turns it into HTML.

Unfortunately, there's no way based on that to have links to the comment page since the LJ protocol doesn't let you get the URL of a given post.

Cool. Could you please put it on sourceforge?

This is cool!

Thanks!

First, a patch to let LogJam sync URLs for posts, and then A new version of the lj2html script that uses the URL

FYI, the LJ protocol now does give you the anum in the getevents mode, so you can calculate the URL of a post in your synch copy.

Oh sweet! I will be making very good use of this soon.

LogJam will export your journal as monthly XML files. You could render those however you like then?

That's exactly what I think of. But, like I already pointed out, the main part of such a thing would be the render engine/the designs.

The rendering part is easy. Just change newlines to
. Livejournal doesn't do much.

Blosxom can generate static pages.

I'm using blosxom now. Thx for the hint!

I'm only missing one thing: the generation of an archive index.

http://omnibus.uni-freiburg.de/~stierm/weblog/index.html