Issue: Flood Check Breaks Import

1- Please link to a page that wasn't imported.

2-This is a duplicate: http://www.vaultwiki.org/issues/1578/

3-Fixed for the next build.

Here is a small list of some of the source pages that never showed up:
http://en.runicgamesfansite.com/rgwi...itle=Alchemist
http://en.runicgamesfansite.com/rgwi...itle=Destroyer
http://en.runicgamesfansite.com/rgwi...tle=Vanquisher
http://en.runicgamesfansite.com/rgwi....php?title=Set
http://en.runicgamesfansite.com/rgwi...?title=Fishing

Did you have vBMediaWiki installed at the time of the attempted import?

After thinking about this for a day, the only reason I can think that these articles would be left out is: they are all created by the same user and triggered the flood-check.

In any case, the flood check is a bug and is disabled for the next build. But then while performing the flood-check fix, there was also another bug where the flood-check would not have been triggered unless the period you set is longer than ~5 seconds (likely), because only the first article created on each page load would be attributed to the correct author (the rest to the admin).

However, if you didn't have either vBWiki or vBMediaWiki installed, then these are all moot points.

If you believe this was your issue, you can remove the flood-check by editing vault/import/mediawiki.php, find (x2):
Add after:

Nope, Vaultwiki is the only wiki I've ever installed on VB. If I attempt to import the same wiki again would it make duplicate pages for the page already imported and edited?

My feature request brought to mind another possible reason for this: could the vbulletin character limit have been the source of the problem? It does appear to only be the largest articles that didn't make it in.

Yes, if the article is longer than the character limit, then it would not be imported. For reference, the feature request Webbstre refers to is here: http://www.vaultwiki.org/issues/1609/

After reviewing the DM code, the character limit is not checked if the is_automated flag is set. So if you make the aforementioned edit, you should be able to import the articles without worrying about the character limit (you just won't be able to edit them until a wiki-only setting is added).

You should be able to attempt the wiki import again because the wiki should not allow multiple articles with the same name (the duplicates would not be imported).

Since the automation check also fixes this case, I'm marking this issue as fixed again.

Ok, I will try another import later today. Thanks again.

As mentioned in the separate thread ( http://www.vaultwiki.org/issues/1613/#note7520 ) I am continuing this discussion, as successful further imports still resulted in the missing pages not being imported from the original wiki. Character limits have been changed and I don't use any other wiki software/add-ons, so I am at a loss as to what to try next. If you would like I can give you the information to access the new site via PM, as it is in a completely separate location and login information from the test domain that I sent you before.

Please send the new information. Is it live? To diagnose this issue without any more guessing, I'll have to perform a test import with stricter error reporting.

It's not exactly live yet, you'll have to do a minor hosts file edit to make it so your browser redirects to the correct location. I'll send you the necessary info.

Well, for a few of the un-imported pages at least it wasn't a VaultWiki error. Not sure if you were importing from more than 1 database, but those articles had the same ID number of some of your images that were already imported, but didn't exist in the MediaWiki database.

Anyway, there were some other errors which could have contributed, mainly that the output character set wasn't being set properly for non-default languages (like if you upgraded your board from vB 3 you probably still use ISO-8859-1 rather than the expected UTF-8).

I'm marking this as fixed, because once I told the importer to ignore duplicate IDs it imported the missing pages.

Thanks for looking into it for me! Perhaps the first import was interrupted, then the second import (which was the one that had all the images) created the ID numbers that were necessary to finish the first import?

I'm curious about the character output thing though. I've never used VB3, and I believe everything was set to UTF-8... where was that setting located?

Thanks again!

Sorry to keep bringing this up, but something else strange happened with that last import: all the newly imported pages had their discussion/comments page and wiki page switched. It's easy enough to copy and paste it to fix it, but I thought I would inform you in case this happens again to someone else in the future.