User talk:PuppyOnTheRadio

From Uncyclopedia, the content-free encyclopedia.
Jump to: navigation, search


How's the image and article importing going? Of more importance, certain images almost certainly should be prioritized to copy over here since Wikia will be deleting or Zombiebaron will be censoring them on the main site. I manually uploaded some images that will either be deleted or censored (all NSFW) that are used in articles; the ones not used in any articles here are of lower priority, of course. Simsilikesims (talk) 00:50, 11 October 2012 (UTC)

Article importing is up to date as of 2012-10-07, and a few slightly more up to date. As for image updating - something has gone screwy somewhere. I'm going through and doing things in alpha-numeric order, as Category:Pages with broken file links isn't updating the way it should, and Pages with prefix is coming up with false positives. For instance, File:!!!!!.png doesn't exist here or on main site, but it shows here. I have no idea why. So it's a slow process to work out what's needed. Bugging me up the wall! PuppyOnTheRadio sig.gif                                                 
Got it! Category:Pages with broken file links should be updating automagically, but it won't remove an article/category/file from it's list until the page is edited. Going to the pages in there you should find out what articles are missing images. Then editing the page (adding a line break in somewhere, or an extra space) will force it to drop off the category page. Just check that the page does not have the cat still showing prior to saving. If it does you'll probably need to upload a file that is still missing. As for why pages that don't exist are listing in the Pages with prefix - haven't experimented yet, but I have some ideas. PuppyOnTheRadio sig.gif                                                 
Derp! That file page had been imported here from main site, but the file no longer exists on main site. That means it shows up as a page, but not as a file. So I can delete a bunch of pages that have files missing and no alternative source to grab them from. PuppyOnTheRadio sig.gif                                                 
Got it! User:PuppyOnTheRadio/test4 gives me a list of files. Those that don't have an image uploaded come up as redlinks. Took a while but got there. PuppyOnTheRadio sig.gif                                         

Your signature should include a date, mon. ~BizzeeBeever (speak) Nov 14, '12 3:55pm

Also, how go ...things? ~BizzeeBeever (speak) Nov 14, '12 3:57pm
Lawl, how dare you ask for JS help on the main site!--Somebody20:29, 17 November 2012 (UTC)

Can You, please, do Pee Review for my article Awkward misdialing? Thanks.

Ok then, we're in bidness[edit]

Will now test ...stuff. Blottobot (talk) 23:07, 26 November 2012 (UTC)

  1. Copy this to here. (It has been updated since you last did that.)
  2. Go here
  3. Click the button what says "click me."
  4. ???
  5. Profit
  6. If it breaks, don't call me. :D Blottobot (talk)

More stuff imported and to import[edit]

I recently imported Green ink, along with several UnNews articles, and many of them are missing pictures. Any chance you could update the missing images list? Simsilikesims (talk) 05:44, 21 December 2012 (UTC)

Just started again from image #1 on the main site. Still resisting updating the articles at the moment. They were updated up to a few months ago, and I'll get back onto that track soon. Due to Blotto's script help image import will be a doddle. PuppyOnTheRadio sig.gif                                                 

The script[edit]

What happens when it gets to a page that doesn't exist? It's been a few weeks since I looked at the code, so I don't remember what it should do. It wasn't intended to import non-existent pages, anyway. Blottobot (talk) 02:00, 25 December 2012 (UTC)

It hangs. And I think there was an image page where half the url was chopped due to a bullet point in it (in your list) which it got to and just sat there for a while. But you're right - it's my perogative to ensure that I'm just giving it proper URLs. I extract them via C&P into excel and then a handy extract hyperlink function, which gives me the URL encoded version anyway. Much neater and resolves a few issues. PuppyOnTheRadio sig.gif                                                 
Ok. I think if an image does not exist on "the real Uncyclopedia", the script gets an error-xml response from the server, and then it just stops executing altogether because I didn't set a .fail() callback (because I am stupid). I'll see if I can fix that, presently. Or not.
Also, there appears to be a bug, not in the SCRIPT, but in MediaWiki. Certain files, when checked by the script, still show up as missing, so it goes and re-imports them. You can find these by looking for "uploaded a new version of" in the upload log. Far as I can see, there is no way to get around it; it's a bug in the MW API.
Oh, and Merry Christmas down there. If you're a Jew and are offended, I apologize. Not for offending you, but because you are a Jew. Blottobot (talk) 02:26, 25 December 2012 (UTC)
To help me figure out what went wrong, I'll need the addresses of the images that hung the script.Blottobot (talk)
No stress. It works enough that the odd bug just means I have to keep an eye on what I'm doing. As for the "uploaded new version"... I wonder if that's because the original file has been updated? Again, it's not an issue. It handled 5000 odd images over the course of 30 minutes. I'm going to start doing it 500 odd at time - easier for my DPL anyway. PuppyOnTheRadio sig.gif                                                 
Nope. Re-uploading is just a random uploading. Stupid API. PuppyOnTheRadio sig.gif                                                 
Uploading a new version of a file always seems to happen with the same ones, in my experience. I have ID'd a common denominator: the images have characters which need certain characters in their names to be URI-encoded (using the "%" escape). Examples: 1, 2, 3, 4, 5, 6, 7. Now I'm thinking there is a problem with my script.Blottobot (talk) 04:15, 25 December 2012 (UTC)
The special characters in those aren't URI encoded (& or +). If they are, would we have the same issue? Experiment for later. But minimal stress. PuppyOnTheRadio sig.gif                                                 
Had a thought. MW places special emphasis on particular combinations of characters. These two in a URI could screw up the way MW API would see a page. Had a situation where a noob uploaded a file once as File:''Norman' which messed up it's inclusion in an article, as it translated it as File:<i>Norman' in the article. Or something similar. Wonder if we're sitting against a similar constraint here. The only way I could “fix” that issue was to rename the file. PuppyOnTheRadio sig.gif                                                 
Whoops, missed your replies. The "+" and "&" in a filename generally have significance in query strings, not just in the MediaWiki software; how this impacts the software, I have no idea, but like you said, I'm not gonna stress over it too much. If I have a moment or two free this week, I'll take a swing at it. If I don't get back to you in the next week, hit my talk page.Blottobot (talk) 08:42, 29 December 2012 (UTC)
I ran a VBA/Excel script to drag out all of the file names (excluding the first 20,000) from main site into a single list. I then started to run a script to work out which files where "missing" from the mirror but hit a snag, so I have to rework the script to counter it. It then worked out the url for the image itself. (Oddly the image comes up from the file page differently fro the way it shows on [[Media:xxx]].) Once I have that script sorted it should automagically generate a list of missing images. Once I've done that I'm going to try and determine which images have been "censored" by Wikia, and try and grab the uncensored versions. Personally I think the censored images should have been categorised, but not my choice. I created a tool to help locate unused images to speed this process up. As far as I can see that has been all but completely ignored.
As for the "new" Uncyclopedia... I'm not going to become involved. It has been created with the core of the current admins on Uncyc. Which means the creation has been "politically" motivated. This suggests tht it's going to crumble fairly quickly. Especially as it's going to hit an issue with using content from the original site - covered by NC-BY-CC-SA - and also looking at using advertising to sustain it. While I'm annoyed at the way the wikia site has been slowly becoming less liberal, I think just packing up the bat and ball and going elsewhere is going to be counter-productive. There was a diplomatic way to counter what has happened.
And - unfortunately - it has also prompted me to not want to return to the current Uncyc. I'm happy to work within the fairly simplistic censorship we have on Wikia. We don't need to show explicit images to be funny. Unbooks:Where do babies come from? is a prime example. I was actually about to start to come back when the "new" site became announced. This has left me feeling... well, I can't say disillusioned, because I already was. I'll simply say my cynicism quotient has broken the meter. PuppyOnTheRadio sig.gif                                                 
Well, sorry to hear that. But the move wasn't "political" at all. Wikia's censorship may be "fairly simplistic", but we have no proof that it won't progress to systematic wiki-wide deletion of items on an Arbitrary List of Offensive Things. And diplomacy doesn't work with them. When the content warning went up (with NO advance notice), we begged, pleaded and cajoled (well, the others did; I just sent rambling rants because I'm a Real Funny Guy), and tried for a compromise, but got nothing other than silence or a firm "NO". Wikia is a business; is their product; if they want to turn it into a tepid Oh Internet! clone, that's fine, but we're not sticking around to find out. The current users figured that they could be technically inept clowns themselves, without having to worry about a corporate overlord disallowing use of the word "fuck", or shouting "tone it down!" because somebody complained about Cancer porn. (I am not even joking; the last time I spoke to her, Sannse claimed that article was among the most complained-about.)
On NC-BY-CC-SA front...we have the right to "fork" the wiki any time we want, and as all past contributors will have the right to use the new wiki with their old usernames and passwords, and all attributions will be linked with their accounts, we think we're in the clear. Advertising revenue is going to be used to pay for hosting and maintenance, not to build offices and PX systems or line wallets. Will it "crumble quickly"? ...Ok, maybe. But the move has been in the pipe for almost a year, and nobody has quit because "you all suck, fuck you die". (Excepting Skullthumper, and for that, we're blaming Minecraft and Reddit.) Blottobot (talk) 15:42, 30 December 2012 (UTC)