User talk:SQL/Archives/2008/January

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Thanks

Thanks for your support in my successful RFA. AliveFreeHappy (talk) 07:15, 29 December 2007 (UTC)

Now let the pestering begin. I was looking at AIV and came across this [1]. I'm curious how you would handle it. The reporter makes it sound like a clearcut vandalism case, but while some look like vandalism, others could be construed as editorial disagreement, etc. It is clearly a single-purpose editor at the moment. I'm not sure the warnings were actually appropriate, but with the stream of edits MAYBE a short block. In addition, this was listed later [2] which again looks like minor content dispute, not yet at a blockable level. Comments? AliveFreeHappy (talk) 10:01, 29 December 2007 (UTC)

Another pester. Take a look at this [3] where a particular user went from drive-by tagging (which caused a lot of commotion at the pump. [4])to massive merge tagging and prod tagging. At what point do we call such behavior disruptive? AliveFreeHappy (talk) 12:40, 30 December 2007 (UTC)

Oh, crap. I missed these messages, while I was on wikibreak. I hope everything worked out in the end :(. Very sorry! SQL^{Query me!} 05:46, 11 January 2008 (UTC)

A request for your consideration regarding CAT:AOTR

...My guinea pigs and the "A"s through "O"s having felt this message was OK to go forward with (or at least not complained bitterly to me about it :) ), today it's the turn of the "P"s through "S"s! I'm hoping that more of you chaps/chapettes will point to their own criteria instead of mine :)... it's flattering but a bit scary! :) Also, you may want to check back to the table periodically, someone later than you in the alphabet may have come up with a nifty new idea. ++Lar: t/c 04:17, 7 January 2008 (UTC)

Bot error

Added a template inside another template.[5] Gimmetrow 00:20, 8 January 2008 (UTC)

Thanks for the heads up, I think it thought it was inside the banner shell. SQL^{Query me!} 03:29, 8 January 2008 (UTC)

User:SQL/Reflist

What's left of ther NFS server's hard drive now SQL^{Query me!} 06:13, 11 January 2008 (UTC)

I've just looked in on this after several months away, and was wondering what's up with it? Has automation been implemented or is there still the need for us mere humans to intercede? Pairadox (talk) 06:04, 11 January 2008 (UTC)

I've just been running into problems with it :( My NFS server went belly-up recently (hdd failure), and, I didn't have a backup of the code :( I'm sure there's still a problem, and, I'll work on getting it recoded in the near future :) Thanks, for the reminder! SQL^{Query me!} 06:13, 11 January 2008 (UTC)

All part of the Barnstar quality[6] service! Pairadox (talk) 07:08, 11 January 2008 (UTC)

I just noticed you removed the RBR notice at the top on your page. Does that mean it's too late to ask? Pairadox (talk) 15:04, 11 January 2008 (UTC)

It's probably best right now, to wait for the community to decide (again) on it... At least, for the time being, I'm holding off on granting any. IIRC, WP:RFR is still open, however. SQL^{Query me!} 15:09, 11 January 2008 (UTC)

Thanks, but I'd rather not go that route, given the hue and cry. Keep a low profile and all that. Pairadox (talk) 15:17, 11 January 2008 (UTC)

Signpost updated for January 2nd and 7th, 2008.

Weekly Delivery

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list. Ralbot (talk) 09:20, 11 January 2008 (UTC)

Your list.

Are those exit nodes, or nodes in general. If exit nodes, have you vetted out the ones that don't allow editing the wiki? Regards M-ercury at 04:01, January 14, 2008

It's just a quick list (wow, you're quick! :) ). I haven't looked at how to remove those that dissalow wiki edits. They are (I believe) all exit nodes, however. I used the dnsbl to check that. I'll look at how to figure out the exit policies shortly. However, it already takes 20-25 mins to run that script... SQL^{Query me!} 04:07, 14 January 2008 (UTC)

Would you clarify to /Tor nodes so that admins with good intentions don't go down that list blocking :) ? Regards, M-ercury at 04:08, January 14, 2008

Y Done SQL^{Query me!} 04:13, 14 January 2008 (UTC)

SxWiki -- sxGetURL()

Hi,

I added a new function, sxGetURL() to SxWiki, please see here (doc) and here (src). --Gutza ^{T T+} 09:24, 15 January 2008 (UTC)

Looks great! Thank you! :) SQL^{Query me!} 03:03, 16 January 2008 (UTC)

Bots

Hi. I see you do a lot aound WP:Bot requests. Could you please see to EditorBot before it gets archived? It would be much appreciated. Thanks. Editorofthewiki (talk) 22:05, 15 January 2008 (UTC)

I'll take a look at it, shortly :) SQL^{Query me!} 03:03, 16 January 2008 (UTC)

I think, that's a little more than I can do right now, sorry :( I'm only really approved for wikiproject tagging, and, I don't know that much about pulling data to put together articles, automatedly.... SQL^{Query me!} 03:30, 16 January 2008 (UTC)

Bot edits

Hi. Adding a WP template to a talk page is not a minor edit. Presumably you have a bot bit, that's enough to hide the edits in Recent Changes - you don't need to mark it minor too.

Also, I think your regexes are potentially going to miss a few templates, because you're not searching for {{template:TemplateName, which is also a valid template call. My bot's left hand side of the regex is: "\{\{\s*(?<tl>template\s*:)?\s*(?<tlname>". It's minor - there's very very few of these instances in the wild - but you can be sure it'll cause a double tag somewhere down the line :) HTH, no reply needed. --kingboyk (talk) 13:23, 16 January 2008 (UTC)

PS Could you have your bot be a little more descriptive with it's edit summaries please? i.e. at least tell us which template you are adding. It wouldn't be a bad idea to name the category too, if it's category based tagging, so that folks can easily see why that particular article got a WP tag. Example --kingboyk (talk) 13:30, 16 January 2008 (UTC)

I think I left it set to minor on accident (throwback from a long-gone testing phase), I'll fix that after this run, if I make another run, same with (probably) the edit summary. Also, I'm using a case-insensitive regex, so, it'd pick up on {{(t|T)emplate :) Thanks for the tips, however :) SQL^{Query me!} 13:32, 16 January 2008 (UTC)

Aha! I figured out why it's marking them as minor. Long ago, SxWiki took "0" as the argument to mark minor, now, it's $null. tagem.php is just old enough, to still use "0". Fixed. SQL^{Query me!} 13:42, 16 January 2008 (UTC)

Great stuff, tnx. Nice work with the PHP library btw! --kingboyk (talk) 15:50, 16 January 2008 (UTC)

I'm not sure you grasped what I meant with regards to "template:"? I think your regex wouldn't find this: {{template:WPBiography}}. I might be mistaken as I didn't study it closely. --kingboyk (talk) 18:47, 16 January 2008 (UTC)

Regarding the edit summary, how's this: [7]? As far as the template goes, do you mean, I should be searching for {{Template:WPMEXICO}} as well as {{WPMEXICO}}? If so, I think you're right, I misunderstood :) I'll make that change post haste, if that's the case. SQL^{Query me!} 03:21, 17 January 2008 (UTC)

SxWiki error

Hello. I am toying around with your SxWiki script, and encountered the following: A simple script (taken from your example) like this ..

 <?php
 define('NN', "\n\n");
 require 'lib/SxWiki.php';
 $url = 'http://shaxxer.org/~tws/wiki/';
 $sxLgTA = sxLogin("Tws", 'xxx');
 $sxLgID = $sxLgTA[uid];
 $sxLgT = $sxLgTA[token];
 $maxlag = "5";
 $epm = setEPM(100);
 $foocat = sxGetCat('Tests', "0");
 foreach($foocat as $singlefoo) {
      $oldpage = sxGetPage($singlefoo);
      sxPutPage($singlefoo, 'LOLBOT BOUNCE!', $oldpage.NN.'ROFLCOPTER!', null);
 }
 ?>
 

.. will give me the error "Warning: preg_match() [function.preg-match]: Unknown modifier 'p' in /xxx/SxWiki.php" for lines 75-79. Since I copied the source from the wiki page and line numbers may vary, it's those 4 lines:

preg_match('/\<input type\=\\\'hidden\\\' value\=\"(.*)\" name\=\"wpStarttime\" \/\>/i', $response, $starttime);
preg_match('/\<input type\=\\\'hidden\\\' value\=\"(.*)\" name\=\"wpEdittime\" \/\>/i', $response, $edittime);
preg_match('/\<input name\=\"wpAutoSummary\" type\=\"hidden\" value\=\"(.*)\" \/\>/i', $response, $autosum);
preg_match('/\<input type\=\'hidden\' value\=\"(.*)\" name\=\"wpEditToken\" \/\>/i', $response, $token);

I first thought I mind find this out myself, but then i noticed i don't even see a p modifier at those lines, so i though i'd come here and ask you. Likely my mistake, as noone else seems to be having that problem. Thanks. ~ | twsx | ^talk_cont | ~ 15:34, 16 January 2008 (UTC)

PS: In case it matters, PHP version is 5.2.0. ~ | twsx | ^talk_cont | ~ 15:37, 16 January 2008 (UTC)

WMF servers use html_tidy, which, messes with the way code displays on-wiki. It looks like you copied it directly from here... The only good way to get it right now, is to click "edit this page", and, get the code from inside that form. That should be what's causing your problems. (I need to get off my duff and move it to sourceforge....) SQL^{Query me!} 15:41, 16 January 2008 (UTC)

Aye, could have thought of this myself. Thanks a lot, works now. And thanks for the whole thing anyway. :) ~ | twsx | ^talk_cont | ~ 15:43, 16 January 2008 (UTC)

Not a problem! Glad it works for ya :) SQL^{Query me!} 15:45, 16 January 2008 (UTC)

Since I'm already here, something else. It seems, the script is not logging in properly for me. It does say "Logged in as Tws" on the command line, yet the recent changes on the wiki show IP edits[8]. ~ | twsx | ^talk_cont | ~ 15:47, 16 January 2008 (UTC)

Does cookiejar.txt exist? If it does, you might try blanking it / deleting it. If not, you may need to create it. I've had that problem a couple times myself :) You may also want to make sure the process running the bot has sufficient permissions to read/write to it. SQL^{Query me!} 16:11, 16 January 2008 (UTC)

Yup, all good now. I think I should be good now, thanks again. :P ~ | twsx | ^talk_cont | ~ 21:58, 16 January 2008 (UTC)

Notta problem :) That error took me ages to track down, when it happened to me :) Glad I could be a help! SQL^{Query me!} 23:28, 16 January 2008 (UTC)

Help! I've been thwarted!

Hey, as instructed, I need to tell you that 128.61.115.234 has stopped me from running, with an edit summary of . SQLBot (talk) 01:05, 17 January 2008 (UTC)

*sigh* That switch isn't a toy. SQL^{Query me!} 01:55, 17 January 2008 (UTC)

Signpost updated for January 14th, 2008.

Weekly Delivery

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list. Ralbot (talk) 08:57, 18 January 2008 (UTC)

Place your bets?

I'm betting User:Lindsayw427 is a meatpuppet rather than a sock. I'd offer a friendly wager, but the smart money says we'll probably never know either way >.< --jonny-mt 09:33, 17 January 2008 (UTC)

Heh, fair enough... Given the timing between the user switchoff, sock seemed more likely to me :) (I was seriously expecting a nasty note regarding a deletion! It seems every time I get the shredder out, someone gets mad.... Thanks, for the humor! :P ) SQL^{Query me!} 09:35, 17 January 2008 (UTC)

But of course! And I know the pain...that's why I wrote User:jonny-mt/CSD and stuck a big banner at the top of my talk page. I haven't been yelled at in a while, so either it's working or I'm just going soft.... --jonny-mt 09:46, 17 January 2008 (UTC)

Ooh! Niice! I may just have to steal that! SQL^{Query me!} 09:49, 17 January 2008 (UTC)

Be my guest! Just promise you'll use the lolcat. That thing cracks me up every time. --jonny-mt 09:52, 17 January 2008 (UTC)

NUMBEROFROLLBACKERS, NUMBEROFADMINS

I like your bot updates of NUMBEROFROLLBACKERS. My interest is in the historical record of the number (ie how many were there on such and such a date) - if possible could you add the number to the edit summary, that would make the trend easier to track. Also, I maintain User:NoSeptember/admincount to keep track of {{NUMBEROFADMINS}} in a historical way. It would be nice if your bot could take a snapshot of the number every couple of days (on a subpage or yours perhaps) to improve upon the manual effort I do on my page. Maybe such a record already exists? Cheers, NoSeptember 14:52, 18 January 2008 (UTC)

I'd actually considered doing both, and thought to myself "Who'd possibly want that?" :P They'd be cake to implement, I'll get right on it :) (First part done before I finished posting this [9])SQL^{Query me!} 18:21, 18 January 2008 (UTC)

Second bit done, runs every 3 days (if my crontab syntax is right), outputs to templates transcluded to User:SQL/AORStats. How's that work for ya? SQL^{Query me!} 19:03, 18 January 2008 (UTC)

Thank you very much for your prompt resolution. It would be useful if one of the snapshots was for the end of each month (23:59 31 January for example) so it is easy to plug the numbers into some charts I do. Are there any other categories we want to track (NUMBEROFUSERS maybe)? Cheers, NoSeptember 21:22, 19 January 2008 (UTC)

Sure, I can make it run at the end of each month as well (thus gauranteeing at least one snapshot at the end of each month), also, # of users tracking would be trivial to implement, I'll get right on it :) SQL^{Query me!} 06:05, 20 January 2008 (UTC)

Ok, all is done, I added tracking of user data, forced a monthly run, and restarted the subpages, so that they'd be even :) HTH! SQL^{Query me!} 06:22, 20 January 2008 (UTC)

U.S. Roads Newsletter, Issue 1

The U.S. Roads WikiProject Newsletter
Volume 2, Issue 1 • January 19, 2007 • About the Newsletter
Departments Project news Deletion debates Featured member From the editors Coming and going	Features B-class audit shows stunning results	State and national updates Leaderboard California Illinois New York New Jersey Oklahoma Pennsylvania Wisconsin
Archives • Newsroom • Full Issue • Shortcut: WP:USRD/NEWS

Want to help on next month's newsletter? Don't want to receive these in future? Want to change your method of delivery? – It's all here. —^Mitch32^contribs 20:22, 19 January 2008 (UTC)

SxWiki -- Login via POST

I made a small change in SxWiki's login function, as to make it use POST instead of GET (seems like that's become mandatory in MediaWiki recently): [10]. (As you have probably noticed, it's a rather hackish solution because I left the GET variables in place -- didn't want to experiment extensively, just made it work with the new MediaWiki code.) --Gutza ^{T T+} 17:43, 20 January 2008 (UTC)

Thanks! Great work as always! :) I meant to do that in the beginning, but, I was in a hurry to get on to the more exciting parts of the coding, and just never got back to sxLogin() :P SQL^{Query me!} 14:24, 22 January 2008 (UTC)

BTW, since you keep adding code, I've listed you as a developer :P [11] SQL^{Query me!} 14:26, 22 January 2008 (UTC)

/me flattered. --Gutza ^{T T+} 22:21, 25 January 2008 (UTC)

Signpost updated for January 21st, 2008.

Weekly Delivery

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list. Ralbot (talk) 00:31, 24 January 2008 (UTC)

Hello!

Hi, This is a test!!! SQLBot (talk) 04:35, 24 January 2008 (UTC)

Question

I've been tagging and assessing articles on the "WikiProject Texas" and I have run across dozens of articles that your bot has tagged with "WikiProject Mexico". Generally they are Mexican Americans (but not always) in many cases several generations removed from Mexico. There is a "WikiProject Mexican-Americans" so I am wondering if this is a bot glitch, or Santa Ana's revenge via Wikipedia, or just painting with a broad brush? It doesn't bother me one way or another, but it does seem Mexican American would be a better fit. Jacksinterweb (talk) 23:16, 24 January 2008 (UTC)

I was given a set of categories, to tag, by a WP:MEXICO member [12], they all seemed relevant enough. That same person, promised to clean up the irrelevant ones a week or so ago. It seems, out of the thousands tagged, there've either been a couple articles that didn't belong in the cats (nothing I can do about that), or, a couple cats that looked OK to me (While I am actually somewhat familiar with Mexico, I'm by no means an expert.). I've actually got another 300+ cats to go through, but, I've stopped for the time being, pending a cat cleanup (my list is here: User:SQL/Sandbox), from a person more experienced than I. I probably stopped a week or two ago. Anyhow, in a nutshell, the bot's not bugged, it's doing what it was supposed to, the list, however, was a little buggy, it seems. (Anytime, however, one of us gets a list of this size, there are invariably a few members that shouldn't have been in the cat, etc, as explained above).

Anyhow, if there's anything I can do for you, don't hesitate to ask! (Or, if I didn't explain it very well) SQL^{Query me!} 02:14, 25 January 2008 (UTC)

I keep adding the template because...

If I don't keep adding the template, people will assume that everyone is happy with the state of affairs that lead to WP:RFR being enacted against consensus - making a similar situation likely to happen again. Until we reach a point where it is passed with consensus, I'm going to keep adding it. -Halo (talk) 21:11, 26 January 2008 (UTC)

Unblocked.

per your note on my talk page. Matthew Brown (Morven) (T:C) 20:55, 27 January 2008 (UTC)

TOR block of 84.19.182.23

I've unblocked it based on your information. Thanks for the update. Jayjg ^(talk) 21:40, 27 January 2008 (UTC)

SxWiki -- SxGetPrefix()

I originally thought this function was supposed to get the pages which start with a given prefix -- instead, it returns all pages, starting the list from the given parameter. Is that what you intended? If so, shouldn't the function+method be renamed? If not, shouldn't it be rewritten? --Gutza ^{T T+} 20:55, 29 January 2008 (UTC)

Yeah, it's *supposed* to only match articles starting with a certain prefix, Sorta like Special:Prefixindex does. I only really ever used it once, over at wikisource, and, it seemed to do the job ok... I'll take a look at it in a bit :) SQL^{Query me!} 05:03, 30 January 2008 (UTC)

It doesn't look like you're accomplishing that -- you're using action=query&list=allpages, which is documented here: mw:API:Query - Lists#allpages / ap (done) ("Returns a list of pages in a given namespace starting at from, ordered by page title."). --Gutza ^{T T+} 09:41, 30 January 2008 (UTC)

Indeed, but, it only returns pages, with the prefix specified, due to this segment:

189                   foreach($pref['query']['allpages'] as $pref_l) {
  190                          if(!preg_match($searchpf, $pref_l['title'])) {
  191                                  $done = 1;
  192                          } else {
  193                                  array_push($result, $pref_l['title']);
  194                          }
  195                  }

...$searchpf is basically "/^$prefixname/", and, $done throws it out of the while loop. However, I was just playing with it a bit, and, I've noticed, that it only really works on NS0. I'll have to get around to fixing that :( SQL^{Query me!} 12:41, 30 January 2008 (UTC)

Oh, then that's probably the problem -- the one and only test I made was for NS 2 (User), and the results were not what I had expected. In a different train of thoughts, you should consider replacing the preg_match with something along the lines of "substr($pref_l['title'],0,strlen($searchpf))==$searchpf" in order to avoid messing up the regexp with unintentional special characters in the pattern. --Gutza ^{T T+} 12:57, 30 January 2008 (UTC)

SxWiki 2.0?

SxWiki is a cool, useful little tool in its current incarnation. But we both know it's not what it's supposed to be, and it needs a rewrite in a more professional form (i.e. as a class). Are you still interested in maintaining it and collaborating towards a rewrite, or would you recommend I branch? --Gutza ^{T T+} 22:39, 25 January 2008 (UTC)

Actually, I intended it not to be classed, so that it would be easier for more intermediate users such as myself (Yep, I'd call myself beginner to intermediate when it comes to PHP :P ) :) When I first started out here, I tried and tried, to find a reasonably working PHP Wikipedia framework, that I could understand... I couldn't find one that suited me, was well-documented, and, worked, so, I set out to make one :) However, a 2.0 (or, maybe it could be called "Pro" or etc, rename the original to "Lite") branch would be awesome, especially, as a more efficient, well-documented, more powerful, more flexible PHP Wikipedia Bot framework is sorely needed (about as badly as a SF.Net project, and a decent CVS / SVN over there). BTW, thanks again, for all the help, so far! SQL^{Query me!} 07:11, 26 January 2008 (UTC)

Great, I'll come up with a mockup some time today. But I'm strongly for classes, so that's what I'll implement (not sure if that's what you were thinking of when you talked about a Pro version). We'll decide later what we call it. :-) --Gutza ^{T T+} 14:01, 26 January 2008 (UTC)

Yep! That's what I meant! Awesome! :) SQL^{Query me!} 19:27, 26 January 2008 (UTC)

Please consider adding me as a developer on sf (b_stancescu). --Gutza ^{T T+} 19:38, 26 January 2008 (UTC)

Y Done :) SQL^{Query me!} 20:26, 26 January 2008 (UTC)

Feel free to set the project up, btw, I registered it, but, I never really got around to figuring it out... I've gotta go to work now :( (I added you as a full project admin...) SQL^{Query me!} 20:29, 26 January 2008 (UTC)

Excellent, I'll do just that. I'll concentrate on the code though, so don't expect miracles on the SF front. :-) --Gutza ^{T T+} 20:34, 26 January 2008 (UTC)

Me again. The SF project was already properly set up, I didn't have to do anything. I just added some brief info in the Wiki section, describing what we're planning to do. Of course we'll end up tweaking a lot of stuff as the SF project matures, but it's enough for now. --Gutza ^{T T+} 22:05, 26 January 2008 (UTC)

You can follow my progress on CVS here: [13]. --Gutza ^{T T+} 23:12, 26 January 2008 (UTC)

Over and out for now. I've set up the CVS repository (see above), the API documentation site (see the Wiki on SF for the Pro version) and started working on the actual API documentation (see the link to the Programmer's Documentation on that Wiki). There still is a lot of work to be done, and I don't know when I'll be willing and able to continue (maybe tomorrow night, maybe in the next weekend). But I think we're on the right track, independent of that. In a (somewhat) different train of thoughts, please consider keeping the CVS repository on SF up to date with the changes on SxWiki Lite, we should ideally have to keep a single version up to date. --Gutza ^{T T+} 02:35, 27 January 2008 (UTC)

Excellent! Thank you so much! :) I'll have to figure out CVS, but, I've worked with it a little before :) I've been meaning for a long time, to simply move Lite over there anyhow, I put an ages old version up there a long time ago, I think, I'm going to get to it, package it with docs and whatnot, and just leave a "moved" link here. No rush on Pro, I developed Lite over many months, just whenever I felt like tinkering with it :)

Again, thank you for all your help! :) SQL^{Query me!} 05:54, 27 January 2008 (UTC)

Good news, I've almost reached alpha version with the Pro branch! I've tested the very basic functionality (login, getPage, putPage and getCat), and they work -- never tested anything else though. It would be great if you could review the code/docu, make a couple of tests of your own and give me some feedback!

You can get the code from CVS via web interface here: http://sxwiki.cvs.sourceforge.net/*checkout*/sxwiki/pro/src/SxWiki.php

The documentation is available here: http://bogdan.stancescu.ro/personal_projects/SxWiki/pro/doc/

Here's a piece of code that should work (after you fill in your credentials, of course):

<?
  require_once('SxWiki.php');
  $sx=new SxWiki();
  $sx->login('username','password');
  $pagename='User talk:SQL';
  $data=$sx->getPage($pagename);
  $sx->putPage($pagename,'testing',$data."\n\nYes!");
?>

I've only included the code sample above to give you an example, I sure hope you'll get creative! :-)

Oh, and we had the first bug submitted on SF yesterday (I fixed it).

Cheers,
Gutza ^{T T+} 14:30, 29 January 2008 (UTC)

Ooh, cool! I'll sync it with my codebase here :) (Not sure I've figured out how to commit yet) That is a lot shorter with classes :) Great work! SQL^{Query me!} 18:30, 29 January 2008 (UTC)

Me again. I tried blocking using my version of the code -- it doesn't work. Does it work with yours, or was it just a prototype? --Gutza ^{T T+} 16:05, 29 January 2008 (UTC)

Just tested it out (I'm bad about breaking things sometimes :P ) [14] Yep, still works for me. Not sure what might break it for you.... can you check to make sure it's pulling a block token? Also, I think there's been some funnyness with the API recently, could be related. SQL^{Query me!} 18:30, 29 January 2008 (UTC)

I'm testing on the Romanian Wikipedia (not an admin on en.wiki), and it doesn't work over there, although everything looks right, including the token... :-( I'll try with your code over there -- can you please try with my code over here? --Gutza ^{T T+} 19:23, 29 January 2008 (UTC)

Ahh, romanian, is it still unable to edit over there too? I'll try your blocking code here, in a sec SQL^{Query me!} 19:32, 29 January 2008 (UTC)

$target='SQLTestTarget';
  $data=$sx->blockUser($target, "1 hour", "Testing Sxwiki Pro block interface", NULL, NULL, NULL, NULL);

I tried that, but, no errors, no blocks. SQL^{Query me!} 19:38, 29 January 2008 (UTC)

Ok, so it's my code, somehow, thank you for confirming that. --Gutza ^{T T+} 19:48, 29 January 2008 (UTC)

Now here's a surprise: your code doesn't work on the Romanian Wikipedia either! Here's what I'm trying:

<?
  $username="Gutza";
  $password=(XXX);
 
  $wikipedia="http://ro.wikipedia.org";
 
  include("SxWiki.php"); //Include framework
  $url = $wikipedia.'/w/';
  $sxLgTA = sxLogin($username,$password); //Log in; dies on failure
 
  $sxLgID = $sxLgTA[uid];
  $sxLgT = $sxLgTA[token];
  $maxlag = "5"; //Set maxlag to 5
  $epm = setEPM(10); //set 10 edits per min.
  SxBlockUser('WelcomeBot',"1 hour", "Testing Sxwiki Lite block interface", NULL, NULL, NULL, NULL);
?>

Am I doing something wrong? (I log in fine, and the token's all right; I'm using this version of the code: [15].) --Gutza ^{T T+} 20:17, 29 January 2008 (UTC)

I wonder, if this isn't the same issue we ran into with editing at the rowiki? Can you dump $response, and render it, and see if it's presenting a captcha again? SQL^{Query me!} 05:07, 30 January 2008 (UTC)

I think the issue we ran into previously was related to double calls to urlencode() (we retrieved the name of the article via URL, which was already urlescaped, and passed it to SxPutPage which urlescaped it again). The first response is all right -- as I said, I log in fine and the token is ok. The second response however (after sending the POST request) seems to be the previous page all over again. I believe I'm passing the wrong parameters, or the MediaWiki installations are a bit different, for whatever reason.

I applied for admin status here in order to be able to compare the behavior of both installations -- it will take a week, but at least we won't have to exchange notes on such silly details. :-) --Gutza ^{T T+} 09:38, 30 January 2008 (UTC)

One thing I ran into testing it last time (I forget if I mentioned it or not), was, some POST fields like "Save", and whatnot, may need to be localized for the Romanian language. I tried to do so, but, may have failed... I'm not sure how well I've got cryllic support working here... Might be something to test :) SQL^{Query me!} 02:47, 31 January 2008 (UTC)

Signpost updated for January 28th, 2008.

Weekly Delivery

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list. Ralbot (talk) 04:16, 31 January 2008 (UTC)