Both Special:LonelyPages and Special:DeadendPages look like they show ALL pages. (Though with bots currently out of action that's not a pressing matter… Might go away by itself anyway.) — 6×9 (Talk) 08:03, September 5, 2009 (UTC)
Running a maintenance script that will fix this. Might take some time. --TOR 16:28, September 10, 2009 (UTC)
Can you define "some time" please TOR, only this would still not appear to be resolved. ♫♫Яєdxx ♪♫♪♫♪ActionsWords 07:53, September 15, 2009 (UTC)
Again, can we have an update on this please TOR, only a month has almost passed and I would've thought the maintenance script had finished running by now... ♫♫Яєdxx ♪♫♪♫♪ActionsWords 12:33, October 2, 2009 (UTC) Can you also amend the "Lyric Wiki" on the Special pages to be LyricWikiplease. Thanks.
Checking on this again ... --KyleH (talk) 22:16, October 5, 2009 (UTC)
We're running a maintenance script now to fix this issue. The script has to iterate through all of LyricWiki's 1.2 million pages (almost 10x more than any other single wiki we host), so it's going to take a while. I don't really have a specific time-frame for a while though. I'll keep you updated on the status. --KyleH (talk) 18:08, October 9, 2009 (UTC)
This maintenance script has finished running; however it missed a few pages that were created or edited since we performed the import. TOR has a list of those pages and will be fixing them soon. --KyleH (talk) 18:09, October 13, 2009 (UTC)
No, it isn't showing all pages. And it can be fixed by making nulledits, I made some to some test-pages one hour or two hours before the cache was updated and the pages do not show up anymore now. I'll take care of it, because I do not touch pages as LWT does (which likely would work as well), but my AutoFix script actually does some work and that's even better than keeping the unwanted parts of a page ;) But maybe I'll ask Senv if he wants to write a bit for LWT that does the same work like my "AutoFix" script, including the opening and saving of pages. - Chris 11:17, December 31, 2009 (UTC)
Status update: I made it to write a bunch of scripts that all together take care of Special:DeadendPages. Will need a certain amount of time, though, because I can only edit 1000 pages per day at the top. Or is there any way to either make the list longer than 1000 pages (which would be preferred) or change the cache refreshing-rate to higher than 1 times per day? - Chris 00:34, February 17, 2010 (UTC)
Half a year later: Touchfix has done its work - S:DEP is empty except for the pages that really belong in there. - Chris 16:54, August 1, 2010 (UTC)
For all non-urgent code changes, we branch on Mondays and deploy on Wednesdays (meaning that any code that was done by Monday will go live on Wednesday... any code that misses the cutoff will go live the next Wednesday). This section is to track changes that've been made but aren't live yet.So don't expect these to work until the Wednesday after they were put here :)
It seems searching for songs with russian titles is case sensitive -- e.g. [1] works, but [2] doesn't. Can this be fixed?
Elenril 08:14, December 12, 2009 (UTC)
Another thing I noticed: [3] (the same like your working but not xml) doesn't set up the correct urls in the headline, it does for the link below the lyrics. — The preceding unsigned comment was added by LWChris (talk • contribs), 17:16, 12 December 2009.
In addition to the problems 6x9 and myself reported to you earlier about this, there is also another bug that has since become apparent to me whereby Wikify sometimes decides to capitalise letters. ♫♫Яєdxx ♪♫♪♫♪ActionsWords 15:23, 3 February 2009 (UTC)
Just to check, are there more bugs that I've lost track of? All that I'm currently aware of are the bugs in this thread and the ones that were recently fixed.
What I've noticed is that often (but not always) letters after apostrophes are capitalised. (Maybe that's what Red meant?) AFAIK it's only started doing so recently (probably introduced along with the capping-accented-letters fix). I haven't done any tests to see when they are capped and when not, though… — 6x9 (Talk) 15:49, 3 February 2009 (UTC)
Yes it's after an apostrophe. It is also inappropriately uncapping letters in the middle of words, e.g. McIver it makes Mciver in the link (but not in the second parameter). ♫♫Яєdxx ♪♫♪♫♪ActionsWords 12:05, 8 February 2009 (UTC)
@Sean: Yes seems it's after all apostrophes. Any way you can fix this?
@6: I like Humbug's thinking on album art.. ♫♫Яєdxx ♪♫♪♫♪ActionsWords 04:37, 22 March 2009 (UTC)
Inappropriate capping of letters (like your example: "McIver" vs "Mciver") may look ugly, but that's the capping policy.--KaputteMenschen 15:52, 11 April 2009 (UTC)
Filing a ticket :( Does this mean I have to fix this now? *shivver* This one is one of those things that's way trickier than it looks (because of unicode characters).
I think I followed the thread up about Orphaned pages or Lonely pages not including pages which are ONLY redirected to, but I can't remember where that thread is :( Until I run into it again, I'll re-start it here. The solution I was thinking of is two Special pages:
A list of all of the pages which are only linked to by redirects. That special page would have the weakness that some ACTUAL legitimate pages may link to those redirects which then link to these "hidden" lonely pages.
A second page (done now :D) which shows all real articles which link to redirects. The idea is that we can go through this page with a bot (which I haven't written yet, hopefully someone else will? ^^ please) and fix those original pages to link to the destination. This way, the previous SpecialPage which links to hidden lonely pages will actually only include pages which aren't linked to directly. This special page is done... it caches 1,000 results at a time for 2 hours: Special:Linkstoredirects.
The (most recent) discussion was on the Community Portal… until I archived it yesterday. Sorry. The first half is here and the rest a bit lower. (Was that an earthquake, or Echo and Senv jumping up and down in joy?) — 6x9 (Talk) 00:45, 30 March 2009 (UTC)
Thanks, Sean; one note regarding Special:Linkstoredirects - don't you think it would be better to make items of this list unique?--Senvaikis (talk) 05:39, 30 March 2009 (UTC)
I think that all pairs are unique. Or did you mean make it so that everything in the left is unique or everything in the right is unique? If it's primarily going to be used by bots, as long as the pairs are unique it should be good. One interesting possible bug I noticed is that some of the links in the left column are red which makes me think that the pagelinks table has bad (old) data of links from pages that don't exist anymore.
Considering there's a "Backgammon" as well as several JPGs and PNGs in there… not to mention EchoSierra/Sandbox/1-3000 (while User:EchoSierra/Sandbox/1-3000 clearly exists)… I'd guess the left column strips the namespace off. — 6x9 (Talk) 21:55, 30 March 2009 (UTC)
I'd figured that was you :-) I GOT TO PATROL ONE OF SEAN'S EDITS! *hops up and down in glee* — 6x9 (Talk) 02:50, 31 March 2009 (UTC)
Finally got around to fixing those links. Still need to make that other special page. If anyone is feeling botty... this first page is a good candidate! Special:Linkstoredirects ^^ lol
(OUTDENT) Summary of where this stands: we still need a SpecialPage which is a list of all of the pages which are only linked to by redirects. That special page would have the weakness that some ACTUAL legitimate pages may link to those redirects which then link to these "hidden" lonely pages. The fact that we have Special:Linkstoredirects (which can hopefully continue to be cleaned out by the community as well as bots) should make that weakness not matter much though.
Hey there, I'm quite curious about the Special:CategoryIntersection. It seems not to be an extension enabled on request by wikia staff but a custom one by you. So could you tell me how I can use/enable this extension on another wiki? If it's a custom code, I'd love to get my hands on it and adapt it more for my use (of course with credits and such). Thanks in advantage! BFH-Wiki 11:08, March 24, 2012 (UTC)
Ah, should add that to the system if it's not in there. What wiki do you want to turn it on for?
On this wiki, eventually with a third category textfield. Altering the url with &category_3/4/5=... works fine, too, so it's good enough for now ;)
Thanks again, BFH-Wiki 13:07, March 25, 2012 (UTC)
Added it so that it can be changed by Community Support and then set it so it will go live automatically for battlefieldheroes. The change should go live at the next release in 2 days (Wed, March 28th, 2012).
Wow, that was faster and more expansive than expected. Wikia can count itself lucky to have staff members like you! Great work!BFH-Wiki 20:49, March 27, 2012 (UTC)
Looks like they don't like non-ascii characters – works as advertised with Prodigy, but not with Кино. See the lower half of this discussion. — 6×9 (Talk) 19:40, 29 April 2009 (UTC)
Wow, this one is rough. It seems PHP has a tough time decoding strings with %0 in them. None of the workarounds I found online seem to be working... but it can definitely be done because Кино (Kino):Дети Проходных Дворов works just fine. -Sean Colombo 06:40, December 23, 2009 (UTC)
Hi, could you provide an API in REST with JSON response that could list the albums per artist and the songs per album, for example:
pseudo javascript*
getAlbumsbyArtist(artist) - the response would include the artist URL here on lyricwiki, his/her name and then the list of albums with respective URL and name
getSongsbyAlbum(album,artist) - the response would include the both artist and album URL here on lyricwiki, his/her name and then the list of songs with respective URL, name.
Thanks.
Best regards.
— The preceding unsigned comment was added by 89.180.89.25 (talk), 11.50 29th May 2009 (UTC).
The only thing I miss is a relieable source for the editcount, the parser hook #ec: doesn't exist anymore and Special:EditCountisn't the best thing... You cannot sum 2 results from it up by using #expr::
{{#expr:{{Special:EditCount/Raygun Shaun}}+{{Special:EditCount/Dotmoo}}}} = Expression error: Unrecognised punctuation character "".
{{Special:EditCount/Raygun Shaun}} = 125
{{Special:EditCount/Dotmoo}} = 138
{{#expr:33+121}} = 154
...last works as it should. - Chris 21:21, October 29, 2009 (UTC)
I was about to point out that #ec works, because {{Userboxes/EditCount}} works… until I saw that Überfuzzy had changed it… — 6×9 (Talk) 06:16, October 30, 2009 (UTC)
Yeah, this is where I've got that "Special:EditCount" thingy from. - Chris 09:04, October 30, 2009 (UTC)
Bit of a bummer really. I also found that {{SUBST:Special:EditCount/LYRIC-Humbug}} doesn't work because it can't be SUBST'd. ♫LYRIC-Humbugwords♪deeds☘ 10:52, November 1, 2009 (UTC)
I think the keyowrds for the mainpage could be chosen a lot better than they are now. Current keywords:
lyrics,LyricWiki,lyricwiki,Main Page,Main Page/Header/en,Community Portal/Editor's Corner/en,Karhu,Karhu:Sinfonia For The Blunt Sword (2008),Main Page/de,Main Page/el,Main Page/es,Main Page/fi,Main Page/fr,Main Page/ja
Only the first two are sensible (afaik keywords are treated case-insensitive). Let me comment on the other ones:
Main Page
The first page displayed when visiting a general URL usually is the main page, why point that out?
Who would search for "Main Page"?
Main Page/Header/en
Why should the header be interesting?
Who would search for "Header" or "Header/en"?
Community Portal/Editor's Corner/en
The keywords are to represent your LyricWiki project, and the editor's corner is not what forms the page
Karhu
Why such an unknown artist as keyword?
Karhu is the exact opposite to a gold-artist (just red-links, only one album)
Karhu certainly is not the mainstream-search-term (Michael Jackson or the artists from the iTunes- or common charts would be)
Karhu:Sinfonia For The Blunt Sword (2008)
This album only contains redlink-songs and should NOT be the figurehead of LyricWiki
This album is from last year, so not the one people are currently searching for
Main Page/de, Main Page/el, Main Page/es, Main Page/fi, Main Page/fr & Main Page/ja
Who would search for that?
As keywords are used to categorize a page and therefore are important whether the site is displayed as a result for certain searches or not, sensible keywords would be:
The last 7 are to be updated every few weeks. What do you think, as member of LW:1? - Chris 21:14, November 7, 2009 (UTC)
I assume Karhu and their album somehow ended up in there because they're the Free Album of the Week… It might make more sense to use the iTunes Top Songs titles & artists for tags instead. — 6×9 (Talk) 21:37, November 7, 2009 (UTC)
Oh of course, I see. However you're right, either the tags are for popular searches or for general searchterms.
N. B.: The current tags brought LyricWiki to the 22nd rank in the Google results for "karhu sinfonia for the blunt sword"; not bad, but who'd rather search for that than for "free music download" in general? - Chris 21:47, November 7, 2009 (UTC)
Fantastic points all around, guys :) Those are pretty horrible keywords. They're currently generated automatically by MediaWiki based on the links on the Main Page. There are really two issues here:
The Main Page could use some work in what it's linking to at all. I have some ideas and will be throwing up a proposed new design within (I hope) the next couple of days to get some feedback. I spent a lot of Friday talking to someone who has made successful Main Pages which brought other wikis a lot more traffic, a lot more user-friendliness and even some better search-engine-respect (don't worry, there' no shady link-exchange stuff required or anything like that).
If the meta-tags are still lame after the re-design, then I'll have to dig my hands into the code and fix that.
Minor problem for when you need a break with the big stuff :-]Edit
User-blog:LYRIC-Kiefer needs to be moved to User-blog:Kiefer. (Hrmmm. How does one link to their blog? I had to us the one-bracket method. Two-brackets were a red link.) I'd like to blog about a few music things, but I'd like to keep everything together. TOR said he was going to do it about a month ago, but I'm guessing it got lost in all the other projects. Kiefer talk contribs admin 21:40, November 12, 2009 (UTC)
Wow, my talk page is so ridiculous. Fortunately, LWChris is whipping me into shape ;) anyway... filed a FogBugz ticket for it
RT #28217
. Community or TOR will fix it if that's easy for them... if not, I asked them to bounce it back to me so I can do it manually in the database. Sorry for the 2.5 year delay :P (geeze)
You do really wonder why there's no pages in "Albums released in ALBUMYEAR"? Well, some categories are obsolete since the new parser functions take care of them. By the way, is there any possibility to get rid of the fLetter thingy? Is there anything like a substring-function, that, combined with some if/else queries for "0-9" or "Symbol" could do the work for us, or is it too complex? - Chris 21:45, November 22, 2009 (UTC)
(149,498 now… growing back, but veeeeerrrrrryyyyyy slooooowwwwlllly.) I meant the language subcategories that show up in Unused even though most of them (if not all) aren't.
The fLetter thing could be automated, and it wouldn't be too complex either – but it would take a huuuuuge sorry for all the multiple vowels, Sean #switch, with cases for each Chinese, each Japanese and each Korean character, each symbol and each number. Sean mentioned the possibility of a dedicated fLetter parser function once, though (hint hint). — 6×9 (Talk) 05:05, November 23, 2009 (UTC)
May I give the hint, that if there's the possibility to cut out a single character there likely is another function to check whether a character is present in a certain string (usually anything like if String.IndexOf(Character)>0) which would result in 1 string and 1 [el]if-query for every language... ;)
In JavaScript it'd look a little bit like this:
var fLetter=pagename.substr(0,1);
var fLettersChinese="...", fLettersKorean="...", ..., fLettersNumber="0123456789";
if (fLettersChinese.indexOf(fLetter)>0) then fLetter="Chinese" else
if (fLettersKorean.indexOf(fLetter)>0) then fLetter="Korean" else
...
if (fLettersNumber.indexOf(fLetter)>0) then fLetter="0-9";
That's possible, but still rather expensive (not something you'd want to use on x*10^5 pages), and you'll need a complete list of "symbols" as well. In regex this could be done easily with ranges (or so I've been told), but installing the regex extension would be rather dangerous (or so I've been told). Dunno how the FLETTER magic words work, but if they could be translated into a parser function (with added CJK/acc.char. detection) that might be the cleanest (and least expensive) solution. — 6×9 (Talk) 18:11, November 24, 2009 (UTC)
Related to the ticket that all of the categories are messed up:
Just wanted to ask whether it would be possible to reformat the {{Song Of The Day}} template back to the way I had it, by which I primarily mean remove the subject headings, i.e. "Reason:", "Video:", etc. from the main page. The reason I ask is because I think it looked much better being in same format as AOTW. ♫♫Яєdxx ♪♫♪♫♪ActionsWords 06:08, December 26, 2009 (UTC)
Yup, that's do-able. Right now it doesn't look as good.. it's just this way for now since it involves just basically pasting the nomination in from the queue rather than trying to cut it up into parts and reshape it (which will have to be done... I just have to poke at it for a bit to get that to work).
<LyricsResult>
<artist>Slipknot</artist>
<song>Before I Forget</song>
−
<lyrics>
Go!
Stapled shut - Inside an outside world and I'm
Sealed in tight - Bizarre but right at home
Claustrophobic - Closing in and I'm
Catastrophic - Not again
I'm smeared across the page - And doused in gasoline
I'll wear you like a stain - Yet I'm the one who[...]
</lyrics>
<url>http://lyrics.wikia.com/Slipknot:Before_I_Forget</url>
</LyricsResult>
Why is that [...]? Thank you in advance
Cheers,
Alexey.
--Larkyphoto 14:40, January 8, 2010 (UTC)
Hey Alexey,
I hope you don't mind if it isn't Sean who answers your question; I guess you're still used to the old API or didn't expect the API to show different results than the pages?
Well, there's been a time when API-requests threw out the whole lyrics. But as the biggest music publishers recently increased their effords to tighten copyright-enforcement, their new contracts with LyricWiki contained an agreement to disable the programmatic access to the lyric-database of LW. I think it was Sean who at least saved the ability to get these first few words through API.
So these lines are the rest of what formerly was a source for many lyric programs, unfortunately... Can't fight it, though. So this is the reason for the cut lyrics. - Chris 18:24, January 8, 2010 (UTC)
@Sean: Would it be possible to have the API add a short info message after the truncated lyrics? I'm sure Alexey isn't the only user who wondered about this… — 6×9 (Talk) 18:54, January 8, 2010 (UTC)
Yeah, I think we'll have to do something like that, we're still getting frequent questions about this. We originally had a message right in the response, but some users complained. Maybe I'll just add an additional field to the response which explains briefly and has a link to more info.
Wonder if I had the honor of being the first downloader?
I noticed that in Search mode, the autocomplete includes pages that have been redirected on the wiki -- for example, searching for Feist, the first listing is her artist page, but then the second listing is Feist:(500) Days Of Summer... (and if you click on that, you get an unformatted version of the album page for (500) Days Of Summer (2009)) and the third is Feist:1,2,3,4 (which has been redirected to Feist:1234).
Also, I'm not sure about the the gray-on-black on the lyrics pages, or the title/artist being in a serif font (you don't see a lot of serifs on the iPhone!). Trainman 00:48, July 16, 2011 (UTC)
Sorry for the huge delay in responding. I did make app updates based on this post, but didn't get time to reply.
Anywho: redirects showing up in type-ahead is by-design... that helps if you don't know the right name of a song (one of the best features of redirects). The really bad side-effect you mentioned (unformatted album pages) was fixed in an update... now albums and "list" pages (like the Grammy Awards) are formatted lists just like artist discography pages.
I've been doing quite a lot of editing lately, and just finished up The xx. I decided to add all the singles, because I believe they should be there as well, and was wondering if there was a policy on this? Should singles be added, or not?
The short answer is, only if they contain songs not available on a regular album. Same goes for best-of compilations and live albums, unless they're notable (e.g. being multi-million sellers). There are sites dedicated to (and much better equipped for) providing complete discographies. We add discographies mainly so that song pages are easier to find, and that aim is best served if we keep them as short and to the point as possible. (That last bit's not policy though, just my opinion.)
This question has cropped up quite often, mostly on user talk pages. Maybe a new FAQ entry is due… — 6×9 (Talk) 14:25, October 3, 2011 (UTC)
When you say regular albums, do you mean including bonus discs released on deluxe edition in that definition?
Yes – if the songs in question are included on a bonus disc (or as bonus tracks), and that bonus disc is listed on the artist page, then it's preferable not to list the single as well. — 6×9 (Talk) 17:49, October 4, 2011 (UTC)
All right, I'll remember that for the future. I've already added the singles for The Xx, though, and the tracks on those singles can be found on the bonus disc of their only release, what should I do about them?
It's getting allmost impossible to use IE on LW, especially for anonymous users; here's one from bunch of errors:
User Agent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; GTB7.2; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; .NET4.0C; .NET4.0E; InfoPath.1; OfficeLiveConnector.1.3; OfficeLivePatch.0.0)
Timestamp: Fri, 13 Jan 2012 13:14:52 UTC
Message: Object doesn't support this action
Line: 227
Char: 71
Code: 0
URI: http://images1.wikia.nocookie.net/__am/46612/group/-/oasis_extensions_js
If warnings about errors are enabled in IE options, LW page scrolling becomes practically impossible...
That doesn't mean LW hasn't any problems on other browsers, but IE is a leader here certainly.
(Just for balance: this one - from FF 9.0.1:)
http://connect.facebook.net/en_US/all.js (f. stringify)
JSON.parse: unexpected character
Line 9
Senv, I reported this on (Thursday?) when you mentioned it but just wanted to follow up. Is it any better for you now? I checked it out in IE 9 on my computer, but IE 9 is often significantly different than IE 8.
looking at recent Janitors editions (1, 2), seems that "= problem", mentioned in Janitors talkpage, still remains unsolved. Or, what's more likely, - unnoticed; that's why I decided to 'bump' you right here ;).
I think it's finally fixed. Sorry for the delay. -Sean Colombo 13:34, April 26, 2012 (UTC)
Sorry for rearchiving this reminder, but the "ghost-artist" part of it still needs your attention; see latest additions (Lady Antebellum - or È®ç¾ä¹æ³). Tia, --Senvaikis (talk) 16:51, May 4, 2012 (UTC)
I'd like to talk to you about several things, so please join the chat when you've got time to. It'll take approx. 10-20 minutes. - Chris 19:18, April 18, 2012 (UTC)
Hi there,
I'm at the All Hands so we're booked all week except for a few seconds here and there... will pop on when I get a chance. Sorry for the delay! :)
I would like to improve the YouTube extension. In particular, I would like to add Dailymotion to the supported sites. It would also be interesting to support the parameter to start at the middle of the video ("t=" on YouTube, "start=" on Dailymotion).
Sarah Manley told me that you could help me to get started on modifying an extension, and then to have my changes integrated into Wikia. Right now I don't even know where I can find the code of the extension: I found this one and several others; I don't know which one is used on Wikia. And is it possible to test an extension somewhere on Wikia?
The version of the extension we're using is /extensions/wikia/YouTube. There is no easy way for the public to test changes on Wikia yet (TOR is working on this, I think). So the best way to test the extension would be to install normal MediaWiki and then just download that code and include the /extensions/wikia/YouTube/YouTube.php file. I assume it doesn't have many other requirements.
Thanks for helping out & please let me know if you have other questions :)
I'm pretty sure this has happened for all wikias, but searching for something on the search bar doesn't take you directly to the page you search for anymore, instead, it takes you to a list of search options, which I don't like at all. However, it also seems that the search suggestions has been removed from this wiki specifically, though, although I may be mistaken. I just felt like telling you this because it really slows down my work here on the wiki, and was wondering if you could do anything about it.
Thanks in advance :) XxTimberlakexx 21:59, May 16, 2012 (UTC)
Hiya,
Thanks for the feedback :) it helps. The search change (to turn off automatically going to a page) was intentional, but there seems to be a lot of mixed feedback so they're pondering it more deeply. At the current moment, that is the intended behavior though.
As for the search-suggest... that's a temporary bug. It was re-written (to be better, faster, stronger ;)) but when it was released today, there was an issue after getting hammered by all the traffic of the site (and even more-so on wikis like LyricWiki which have eleventy-billion pages). Literally as I type this, some guys are working on fixing it 10 feet away. Should be fixed soon :)
Hope everything is going well w/you & that this clears things up a bit... thanks again for the feedback! :D
Things are going quite well, thanks for asking. Been doing a lot of Bronzing to artist pages, and I Bronzed my first artist page, The Fireman, last week! :D It irks me that some artist pages are ranked Bronze, though, when some of the songs and albums are still green. Luckily I consider Bronzing all their songs and albums before doing the artist page.
Anyways, glad to hear the search suggest is a temporary bug. I mean, taking away both the search suggest and going directly to a page you search for make it much harder for me to work on here, especially since the search-suggest helped me avoid getting sent to the list of search options. And what do you mean guys are fixing it 10 feet away? :O Great to hear it'll be fixed soon though, great news to me.
No problem for the feedback! :D
XxTimberlakexx 22:45, May 16, 2012 (UTC)
Just a quick note to say that you while we removed "go" search, we did add it as a personal preference so you can enable it for your own experience. Just go to Special:Preferences, click the "Under the Hood" tab, and choose "Enable Go-Search" in the advanced section. This will turn it back on for your personal experience. --Dopp (help forum | blog) 23:42, May 16, 2012 (UTC)
Hey Sean,
I opted in for Go Search like Dopp mentioned above, but starting this morning it takes me to the search options again, and the search-suggest disappeared again. And I checked, and Go Search is still on. I presume they're still working on it? XxTimberlakexx 21:23, May 21, 2012 (UTC)
Hi Sean, not sure if you're the best guy to come to, but anyhow...
I'm having a rather weird bug when viewing an album page on the Wikiamobile skin (the one used by smartphone browsers). Where the album art should appear in the infobox instead has a copy of the actual page. You can see an example of this by navigating your (computer) browser to this url. I'm not exactly sure what's causing it. Do you mind checking it out?
Actually, I just found the line of code causing it. In Template:AlbumHeader, there is a section of code <div style="display:none">{{:File:{{{Cover}}}}}</div> The mobile skin strips styling, so is causing the page to be outputted as well. What is the use of that piece of code, and can it be worked around? ǝsʞpɐןǝ (talk page) 14:05, 21/05/2012
I think you mean the template {{Album}} (the naming of the basic templates is not very coherent round here...). I quickfixed the issue by highjacking the printfooter class to that element, as this was the only class that had applied display: none to it in the WikiaMobile skin. Unfortunately our own style isn't used and I don't know if there is any page like MediaWiki:WikiaMobile.css that would be evaluated.
I hope that this won't break printouts, because as printers have a lot of time to render a page, I hope they will not strip the styles and let the display: none overrule any other setting saying else, for example with @media print. - Chris 22:53, May 21, 2012 (UTC)
As to its use, it's to display a notice on the album page if the image has no {{albumcover}} template, and add a category so we can find it. — 6×9 (Talk) 17:00, May 22, 2012 (UTC)
I wanted to say that but I forgot, it was late when I fixed it :D - Chris 17:52, May 23, 2012 (UTC)
A script on this page may be busy, or it may have stopped responding. You can stop the script now, or you can continue to see if the script will complete.
Script: http://slot1.images.wikia.nocookie.net/__am/53335/group/-/oasis_shared_js:3
Does this sound like the same issue you're having?
- Firefox
- Shrink the browser width just enough so that the horizontal scrollbar at the bottom appears with almost no sliver of horizontal movement.
- Be logged in
Afterwards, wiggle around the toolbar lock zones on random pages.
Current behavior:
After the precondition is met, crossing the toolbar lock zone will freeze the browser.
Expected behavior:
Nothing unusual should occur.
Steps to reproduce:
1. Meet the pre-condition as stated above.
2. Wiggle back and forth around the toolbar lock zone.
Or does that not describe your browser & what's happening?
Thanks for response, Sean, but now, on my work-box browser, this message doesn't appear. Should recheck that later in the evening, on my laptop (both W7+FF12). (Behaviour was very similar to described above, as I can recall). --Senvaikis (talk) 07:16, May 23, 2012 (UTC)
PS (evening): as it was expected, - can't reproduce the situation, e.g. - ticket closed.
Sad to say, but it was too optimistic closing this ticket - same script bug with the same behaviour makes pagescrolling impossible again, freezing FF completely... --Senvaikis (talk) 16:03, June 6, 2012 (UTC)
I'm not sure if I'm doing something wrong, but I can't reproduce this behaviour? [4] - Chris 19:48, June 6, 2012 (UTC)
Nothing strange, 'cause I can't reproduce it also (again). It appears irregularly and vanishes silently as Cheshire Cat, leaving only Sean's smile (...senv's mirages again...). Just one question: where then Sean's description of the bug was taken from? :) --Senvaikis (talk) 20:47, June 6, 2012 (UTC)
There is currently a case open (31232) in our bug-tracking system with the description above. One of our front-end engineers thought it sounded like the same thing. That bug still isn't fixed, so it's still possible they're releated.
Thanks for the additional info, hopefully we can figure out what's going on.
The problem is that it doesn't remember what it's processed (keeps happening). I think I fixed the file permissions (again) and re-ran. If it works, then it shouldn't do any updates tomorrow.
Bump rhymes with Jump. — 6×9 (Talk) 16:01, July 8, 2012 (UTC)
Yikes. Will try to fix it again tomorrow. -Sean Colombo 22:29, July 8, 2012 (UTC)
Looks fine, ran it by hand and it seems to have updated the file (which records progress) just fine. Must be some weird permissions thing. For the record, the log file was sean:501, then I chowned to sean:www-data and ran the script it as me, and it worked. The xml files are root:root, so I chowned the log file to be root:root now. If it's not still root:root, then maybe Chef is messing with the permissions.
Been keeping an eye on this and it looks like it's finally behaving. It's done a couple more updates, then remembered that they actually updated. -Sean Colombo (talk) 02:29, July 23, 2012 (UTC)
I want paste Chinese music lyric here, but it isn't possible:
When I hear your voice over the line.
All I wanna do is to keep you on my mind.
I only had one die. How long?
Don't wanna play the fool. How did we end up like this baby.
Just wanna play a cool. I knew always promise.
RAP:Won't you tell me baby boo what you want what you need may be,
I'll be the one that you need till I die baby.
Let me!Tell you about the things that you don't realize,
so let me televize my love,so you can fantasize.
Conjuction-Function But look at the situation.
Don't waste a time let's have a conversation.
It's kind of sad but that's our only option.
Let's keep it down low till I get there uh
连耳朵都肿啦 话说太多声音都哑啦 抱着电话有点爱睏 还捨不得挂
RAP:Yeah, that's how it is ,as I keep on reminess,
you know I keep it real on yes we've got to deal with this, but I'm sick of this,
pocker face acting stupid for the Show-biz Gee-biz but wait I'll be there,
When I'm done with this
RAP:My heart is always with you even though we're a part baby.
I know you're sad and blue but that can't take as a part.
Get away from the dark baby come and see the light.
Yuki,in my dream I could with in my Mitsubishi.
But times going are going and gone it's gone like the wind
that can't stop, but stop not at the bus stop, you I my drop top.
Yeah put time in lock. That's right.
You and me baby messing around in parking lot, don't wake me up.
想着我 Take the time to explain my mind
跟着我 You so fine I wanna make you mine
只想永远在一起 Can you be with me?
Live in luxury,and we be making precious history.
I keep your picture by my bed so I can know you're in my head.
Take my hand by your hand,and I will always be your man.
You gotta trust in me like I would trust in you,
baby close your eyes and my love will see you through.
RAP:If you keep this in your mind, we will make it through this time.
You will find, it's a sign, it's written in my rhyme.
Even though we are apart, you are always in my heart.
Honey stay with me forever and throuugh any thing endeavor.
Can we make it through this phase, I will see you through this maze.
If you put your faith in me, happy ending it will be.That's a promise.
Yuki, have a little patience,and in time, you'll see.
— The preceding unsigned comment was added by 123.122.92.28 (talk • contribs), 07:06, June 23, 2012 (UTC).
Hi, what is the name of that song, and why was it impossible to create the page? - Chris 22:29, June 23, 2012 (UTC)
— The preceding unsigned comment was added by Iamvincentliu (talk • contribs), 07:13, June 24, 2012 (UTC).
Right now it works for me... Please try if this problem was of temporary nature, or if it still persists for you and confirm it. - Chris 09:40, June 24, 2012 (UTC)
It works now, thank you. Vincent Liu 14:19, June 24, 2012 (UTC)
Sean! Sean! Sean! *poke poke* I've got an idea for a great Wikia extension you can surely create within a day or two! I need to talk to you! - Chris 19:19, July 11, 2012 (UTC)
Go on... haha. I'm in chat -Sean Colombo 21:01, July 11, 2012 (UTC)
I get the following message while trying to create page Walk Off The Earth:Man Down (2011) PHP fatal error in /usr/wikia/slot2/code/extensions/AbuseFilter/AbuseFilter.hooks.php line 29: Call to a member function getRawText() on a non-object
Any ideas? --Vincent Liu (talk) 06:04, July 20, 2012 (UTC)
What's been changed in file upload system? (Yesterday I've lost ~4 hours to understand why Lwts file uploading suddenly stopped working, while discovered that now parameter "wpEditToken" must be prefilled by HttpRequest...) Was this the only change? And would it be possible to have some kind of bulletin-board for informing (at least admins) about all such kind system changes?
P.S. One more curiosity - right now two other albs of the same artist (also used to be uncreatable) suddenly turned into normal state and were created without any cheating ;) --Senvaikis (talk) 12:53, July 23, 2012 (UTC)
Last news: the bug have spread on songs also... --Senvaikis (talk) 06:25, July 26, 2012 (UTC)
Not sure if this has already been noted (or if this is any help at all), but I've found that the error seems to happen when creating an album that has the same name as a song page that already exists. For example, these previously existing song pages prevented albums from being created:
Nice catch! Reviewed over 30 albums, created with fake-stage - all of them contained self-titled songs. To make it 100% clear I've made a little experiment with A Wilhelm Scream:Diver (2006). It was uncreatable until I temporary renamed song A Wilhelm Scream:Diver. Just after that album was created without any problems. So, your genious assumption was 100% right! I always knew women are more observant :) Thanks, --Senvaikis (talk) 16:35, July 27, 2012 (UTC)
... and I never knew she was female. Whoops... So what's the problem? Creating two similar named pages within a short period of time, creating a page that contains the title (which would be absolutely stupid of course), or is it just some SQL statement that goes postal because of a false positive match? - Chris (talk) 18:40, July 27, 2012 (UTC)
Can't be the time difference, The Turtles:You Baby was created 3 years ago. Maybe the implied-redirect-rule, which stealthily redirects folks from "Page Name (With Parenthesis)" to "Page Name" if the latter exists but the former doesn't, doesn't play nicely with the new MW version? Doubt it's some SQL error; abuse filter should jump into action *before* DB gets involved. — 6×9 (Talk) 19:05, July 27, 2012 (UTC)
I have already had implied redirect problems when I created different versions of songs, like with Deadmau5:Ghosts N Stuff. I'm not sure how I fixed it back then, but I'm sure it was either by appending action=edit or redirect=no to the url.
I've never seen the error mentioned above, when exactly do we get it? If you visit the page name, if you click create, if you save? - Chris 11:38, July 28, 2012 (UTC)
Obviously, it's trying to get the latest revision. So my assumption is: thanks to implied redirects, AbuseFilter thinks, we are submitting an edit for an existing page. Therefore, the extensions kicks in to inspect our edit. But as we are in fact creating a new article, $article->getRevision() is likely NULL (because there is no latest revision), and NULL->getRawText() results in the error described above.
I'm not sure what exactly makes AbuseFilter think the page already exists now (i. e. why that behaviour didn't occur earlier). Maybe with V 1.19 they redesigned the function to determine whether a page exists, now using code which falls for the implied redirects, opposed to a database query that asks for that particular title and returns the correct result (page does not exist). But that is just a guess. - Chris 16:47, July 29, 2012 (UTC)
But please, don't treat this note as instigation for a new "investigations". As it stands, now we are more like those blind men from one wise story, not real admins. I'd better like to hear some comment from any real admin, making all these changes. So, please, any "sighted" person, - tell us, blind men - what are we touching? I'm afraid only that "sighted" persons are all deaf-mute... ;) --Senvaikis (talk) 11:51, July 30, 2012 (UTC)
What a hell? One more tangle, this time in mwapi. Don't know if that was done under design, or this new behaviour is just one more new bug, but now mwapi fails to return correct result, if some query parameters are lowercased.
Two examples:
Let's try to get a list of artists in interval "Al" - "Ali". There are over 480 artists in this interval, but api returns... only "Alaska & Me"...
Seems that now api understands only uppercased query parameters (this query returns 482 items, as it was expected).
But why it stopped to understand lowercase, and is this the only change, made in api? And should developers start changing their tools, or it's better to wait for some fixes or at least comments?
Or am I missing something here?
Anyway - life is never boring in LW... :) --Senvaikis (talk) 19:44, July 24, 2012 (UTC)
It appears that LyricWiki was recently upgraded to MediaWiki 1.19 (it's being rolled out across all of wikia over a course of weeks/months because it requires a database change which takes a while on each wiki). My guess is that this and a bunch of other weirdness lately are related.
the SOTD page is kind of down, I'm afraid the DB is clogged. We need to run a maintenance script on it. Is there any way to do that, or can you export a dump of the sotdnoms table? - Chris 19:00, July 31, 2012 (UTC)
The SOTD page is still down… --Vincent Liu (talk) 07:14, August 9, 2012 (UTC)
Looks like we need to update it to work with MediaWiki 1.19.
PHP Fatal error: Call to protected method DatabaseMysql::doQuery() from context 'SOTD' (URL: lyrics.wikia.com/Special:SOTD/Admin) in /usr/wikia/slot2/code/extensions/3rdparty/LyricWiki/SongOfTheDay/Special_SOTD.body.php on line 265
Thank you, it's working again! --Vincent Liu (talk) 23:30, August 17, 2012 (UTC)
Hi Sean, could you please grantCubs Fan the right "sotdhelper"? I don't know why but it's apparently switchable by anything above bureacrats. - Chris 18:03, August 19, 2012 (UTC)
Done. Weird that the right needs more than Bureaucrat. I'm on a phone but will peek at that when I get too a computer. - Sean Colombo (talk) 04:26, August 20, 2012 (UTC)
is it possible to add a new customized page namespace for LyricWiki? I don't need it yet but it's crucial for my planning for T-Line to know whether we have the possibility to set up something like this. - Chris 10:35, August 30, 2012 (UTC)
[5] should answer the question about possibility. Or did you actually mean to ask Sean whether he'd do it? ;-) — 6×9 (Talk) 11:27, August 30, 2012 (UTC)
yep it is doable. Would need to be done in the code though. -Sean Colombo (talk) 13:57, August 30, 2012 (UTC)
Cool. You all feel free to comment on this. - Chris 16:16, August 30, 2012 (UTC)
Sean Colombo, I hereby present you with this Diamond Record because you were the one to get the ball rolling. Look at everything here, look at what grew from the littlest seed you planted not even 6.5 years ago: some site in the world wide web. By now, LyricWiki has become one amongst the top five(!) MediaWikis in the world. Look at what you've done, Sean, it is your digital legacy. You will be missed. Chris 01:24, September 11, 2012 (UTC)
:'( that's very sweet. I won't be far and hope I can stay in touch... but I'll still miss building this thing with you guys! -Sean Colombo (talk) 20:18, September 11, 2012 (UTC)
When I visit Goear links, Google gives a malicious website error. Anything wrong? --Vincent Liu (something to say?) 05:15, February 4, 2013 (UTC)
Google error message: [6] Maybe someone managed to hack their site? Should we hide the SF.goear links for now? — 6×9 (Talk) 12:43, February 4, 2013 (UTC)
Probably a good idea to hide the links for now. Usually that type of problem doesn't last very long once they realize it (because it's typically an advertiser that's injecting something bad). I'll contact GoEar to make sure that they know.
-Sean Colombo (talk) 16:18, February 4, 2013 (UTC)
PS: Actually, it seems that the warning is gone & they're not listed as suspicious anymore
Errors from SOTD update at 08:00, February 22, 2013 (UTC) (only 0 songs left)Edit
in this particular case it was good the spam filter blocked your edits: the correct name for this song isn't "In The Blood /My Forgotten Friend/", but "In The Blood (My Forgotten Friend)", which we already have here: Sweet Noise:In The Blood (My Forgotten Friend). Please contribute your edits, corrections or additions to this page. Thank you, Chris 20:55, March 13, 2013 (UTC)
Well, the official track name, as it is written on the album cover, is "in the blood/my forgotten friend/". What policy do we use here? We use track names from last.fm, or from the album?
Do you have any visible reference for the album cover? I couldn't find it in the Internet.
As slashes do have a special meaning in wiki page titles (sub pages) and URLs in common (sub folders), I think we should keep the current page name. - Chris 19:27, March 15, 2013 (UTC)
Since I own a copy of this album I do. Pics of track listing to view here: http://bit.ly/149g5Y5
Indeed, these are slashes. We can now treat it two ways:
a) we can see the slashes as stylistic element, like using the † character instead of "t" for printed tracklists, whereas the track names on Amazon or iTunes use the "t". We then might "translate" these characters in the semantically correct characters without further things to do. In this specific case it's hard to find out if the slashes are stylistic, because there is no official digital plaintext tracklisting available (at least not on their home page, as far as I can tell).
b) we can treat it as the correct name, and have to decide whether we want to stick to that or the way people would usually search for the song.
Anyway, the fact that these are slashes does not help with my technical concerns (page name ending on "/"). We could maybe create a redirect here, so that people who search for the one thing will be redirected to the other. I'd like to consult at least one other admin first, though, in which direction this redirect should point. - Chris 21:40, March 19, 2013 (UTC)
Now that you mentioned getting plaintext tracklisting, I've put the Album in my PC and CDDA tags included on the disc list this track with parenthesis, even if I've seen the singer use slashes instead of quotes and parenthesis a lot. I don't know if it's his personal choice to write like this, or if it comes from his education (I believe he studied in Libya and/or Sudan, where he lived as child, if I remember correctly what he was saying in interviews and wrote on his facebook profile).
Besides now that I look at the last.fm page for this album, it's listed with parenthesis even there. So let's assume we have the correct name now.
Okay, I can try to set up a redirect just in case, but I'll do this tomorrow, because I do not have much time now and I want to be able to fix any weird side effects rightaway if such a redirect has any. - Chris 00:29, March 21, 2013 (UTC)
Ok, that would be perfect. Thanks. - L0ner (talk) 17:49, March 21, 2013 (UTC)
Okay, sorry, I tried it but even for me the spamfilter prohibits the creation of a page ending on /. I think it's because usually that'd cause the technical issues I've been talking of. The spamfilter can hardly know we have turned off the subpages for the content namespace. Sorry, I thought I could create such a redirect, but it's not possible. Maybe a system admin could, but I guess it's not worth the enormous efforts, is it? - Chris 00:04, March 22, 2013 (UTC)
I doubt it's got anything to do with subpages; slashes haven't caused problems in a loooooong while (and even then only in conjunction with ampersands). It's probably that the additional ending slash makes it look like a URL and therefore suspicious.
I already tried moving the page to "…/", but that wasn't allowed either. Probably a good thing, since we might not have been able to edit it afterwards… — 6×9 (Talk) 12:36, March 23, 2013 (UTC)
So just leave it just like it is now. - L0ner (talk) 16:11, March 23, 2013 (UTC)
Errors from SOTD update at 07:00, March 24, 2013 (UTC) (only 0 songs left)Edit
I saw your "ad" on the Wikia Developers Wiki. What kind of jQuery compression do we need?
Hi there :)
It's actually not compression that's needed... we need to port a bunch of old YUI code to jQuery. More info here: w:c:dev:Removing_YUI. I'm not sure if that's up to date anymore, but I assume it is.
Hey Sean! Long time no talk. I just wanted to quote to you something I asked EchoSierra about - she told me to come to you though since UberBot is your bot.
Quote by XxTimberlakexx: "Do you think there's a way to program UberBot to insert the SOTD template in the correct spot when it puts it on a song page? It always puts it below the Song template no matter what, so it always ends up above the AddAlb, YouTube embed, WP-Song, GameFeature, and B-Hits templates. I mean, it's the correct position for songs that don't have these templates, but there should be a way for UberBot to recognize these templates and put the SOTD badge under all of them once it detects them."
It's had this error for as long as I can remember. It's pretty hindering, as a human will always need to move it to the correct spot every day. Looking forward to your response ^^ XxTimberlakexx (talk) 20:33, May 10, 2013 (UTC)
Think I fixed it. It'll appear below youtube where that is, and if there isn't one... then it'll appear below the first one of these it finds: wp-song, gamefeature, b-hits, or addalb. If those aren't found, then it'll appear below Song. Tested it a little and it seems to work.
Neat! I'm surprised this wasn't reported to you sooner! I assumed it was something that couldn't be fixed because people would have noticed by now. Pretty surprising no one's said anything about it before.
Anyways, tomorrow's SOTD has YouTube and WP-Song, so we'll see if UberBot correctly positions it. XxTimberlakexx (talk) 02:20, May 12, 2013 (UTC)