RationalWiki talk:Content dumps

I don't think you can make a CC x SA copy without retaining an authorship trail? Or does simply linking to RW meet the requirement? ħuman 21:18, 20 May 2010 (UTC)

Linking to RW or any page with a full author list meets with attribution requirements, or including the author list on such the site where it is being reused. tmtoulouse 21:28, 20 May 2010 (UTC)

I hope you're still making full backups though, it would be a shame to lose all our history when I make a typo at 3 am. -- Nx / talk 21:34, 20 May 2010 (UTC)

Is there any way for the dumps to automatically include permalinks to where they came from? On every page, perhaps? ħuman 21:36, 20 May 2010 (UTC)

Yes, there is an on-site full backup of the file system space and all the databases, and we now have a new off site backup service that gets rsynced to every night. The content dump is merely what it says it is and isn't part of the disaster recovery plan. tmtoulouse 21:39, 20 May 2010 (UTC)

As per permalinks, the dumps are an XML format used by MW, and created by an MW utility. I am sure it is possible but would require learning the XML formatting, figuring how out to insert something without breaking it, and rewriting the dump utility to do it. The "cost/reward" function of this endeavor has me skeptical. tmtoulouse 21:41, 20 May 2010 (UTC)

The revision id is included in the xml file, since the "current version only" dump is actually structurally the same as the full dump, except all pages have only one revision. I don't know what happens when you import it though - does it set the revision id of the page on the import wiki to the revision id in the xml file? In any case, it would be relatively easy to write an app to parse the xml file and insert something into every page, e.g. {{rwrevision|XXXXX}}, which would then link to the permadiff on rw. -- Nx / talk 21:50, 20 May 2010 (UTC)

That would be really nice to have. ħuman 01:42, 21 May 2010 (UTC)