to the wiki

[ree]

The wiki says ljArchive doesn't work with Dreamwidth. That's news to me; within the last week, I imported an old LJ to Dreamwidth and promptly used ljArchive to back it up.

Checked my new backup. Yup, everything's there, and it's not the original LJ version - the comments all say ext_### instead of the LJ URL or username.

I logged into the wiki to change it, but then I realized that I don't know why it was marked "no" for not working. Did someone mark it "no" because ljArchive, known to be picky and stop working at the drop of a hatpin, failed on their first try and they never retried? Or is there some upcoming code push that will break all the API stuff; even though ljArchive works now, maybe it's preemptively labeled non-working because that will soon be true?

Anxious Ree is trepidatious. I decided not to edit the wiki.

For what it's worth: Windows XP, ljArchive 0.9.7, works fine for me. I've had ljArchive toss a wobbly before when it found an incomplete HTML tag in an entry; fixing my markup allowed ljArchive to work again. (I think if you don't have a > for every <, it breaks ljArchive's XML parser.)

If you know, off the top of your head, of an upcoming API breaker, would you reply here, please? I know there are down-the-line plans to retool all the API stuff, but I didn't think there was anything committed just yet.

Comments

[afuna]

This conclusion is with exor674's help --- it might have been because there was some old (now-fixed) issues surrounding xposting which broke the API.

Unfortunately, we haven't been able to test the hypotheses! But it seems to fit the timeline.

[ree]

Thanks! (I could have sworn I already replied to this. Bah, time.)

[afuna]

(Ahahah that happens to me a lot too. *sympathizes*)

[willow]

Surfing in via search engine; how do you know how to even FIND the possible incomplete HTML? And what if it's in a comment and an imported one at that, that you have no control over.

[ree]

I am the opposite of timely! Sorry about that - sometimes real life eats me.

When I wrote this, I was updating my LJ maybe 2-3 times a week and backing up twice a month, so if the previous backup worked and the new one wouldn't, there were only a few new entries to paw through. I am terrible about wanting to include esoteric code in my entries sometimes, so for me, it was usually as simple as remembering "right, I used crazy HTML on that one poem thing" and running that entry through an HTML validator until I fixed everything that I could.

I never had ljArchive choke on formatting in a comment. That might be down to me not getting a lot of comments (certainly almost zero comments with HTML in them).

Sorry I couldn't be more helpful. Best of luck in your journal keeping and backing-up!