[00:09:24] how comes the api returnes full urls like this in json: "fullurl":"http:\/\/en.wikipedia.org\/wiki\/Help!_(album)" ? http://en.wikipedia.org/w/api.php?action=query&prop=info&format=json&inprop=url&redirects=&titles=Help!_%28album%29 [00:09:44] (i don't see why the / needs to be escaped) [00:11:16] jorn: cause JSON says it needs to? maybe? [00:11:32] print json.dumps({'foo':'http://dbpedia.org'}) [00:11:32] {"foo": "http://dbpedia.org"} [00:12:25] it's not escaped for me when viewed with JSONView [00:12:46] (but is when i view source) [00:13:24] yes, i got it with curl and it's escaped.... [00:13:47] but weird thing is when i use a json parser it seems to be unescaped correctly [00:13:58] I don't know why that is [00:14:06] But we're just using PHP's json_encode() so... [00:14:43] none of the samples at http://json.org/example.html are escaped like that [00:15:13] well i'm pretty sure it's not invalid. maybe just redundant [00:15:25] so jorn's parser needs a ficx? [00:16:44] it seems \/ is defined as escape in strings for / in json [00:16:44] but doesn't need to be used... [00:16:49] ok, sorry for my confusion [00:17:32] http://www.php.net/manual/en/json.constants.php#constant.json-unescaped-slashes [00:20:14] jeremyb would you know who would handle language codes? [00:28:04] jorn: RoanKattouw jeremyb: yeah, the / is escaped by default in PHP's json_encode implementation because of possible security issues otherwise. [00:28:11] e.g. a string can contain [00:28:21] {"foo": ""} [00:28:25] thats perfectly fine in JSON [00:29:04] but when echo'ed in it can be an injection vector [00:29:11] of course that should be html escaped additionally [00:29:28] for non-html outputs that option was introduced in PHP 5.4 [00:30:05] the json spec allows this additional escaping I believe, not required but allowed. [00:30:27] Krinkle: thanks for clearing that up [00:31:03] Krinkle: but does \/ actually prevent that? i guess not because \/ is special but because <\/script> means something different than [00:31:28] jeremyb: it prevents "it" yes [00:32:04] "it" being the fact that {"foo":"

Raw html here

"} would actually show an H1 [00:32:08] instead of being a javascript string [00:32:17] yeah, it's valid JS. so i can't see it being disallowed [00:32:24] Hm.. [00:32:36] anyway, the scape guards against that [00:32:50] yeah [00:32:51] that doesn't mean that browsers don't have their own guard against it as well, but that's something else [00:33:04] jeremyb: same thing [00:33:09] jeremyb: it is valid javascript [00:33:13] jeremyb: and valid json [00:33:20] am i lagging that much? [00:33:34] jeremyb: but when embedded in HTML, different rules apply [00:33:43] jeremyb: 1+1 [00:33:57] jeremyb: ping [00:34:41] lol [00:34:49] I guess you are lagging much [00:35:27] haha [00:36:32] Krinkle: http://dpaste.com/756191/plain/ [00:37:33] jeremyb: ok [00:38:02] are the msgs in a different order for you? [00:38:10] actaully, no [00:38:22] huh [00:41:59] 07 00:41:32 -!- Irssi: Join to #wikimedia-tech was synced in 920 secs [01:01:03] wtf, what is wrong with the internets [01:01:06] ? [01:02:22] hello jeremyb [01:02:31] like pinging out? [01:02:45] my quits were pings? [01:02:58] [2012-06-07, 02:59:32] * jeremyb (~jeremyb@wikimedia/jeremyb) Quit (Ping timeout: 245 seconds) [01:03:13] bad wireless maybe [01:03:15] and the other was a ping too? [01:03:20] nah [01:03:33] this is a VM in a colo [01:03:41] yes both were [01:04:03] k, thx [01:49:23] [42bdac0d] 2012-06-07 01:48:09: Fatal exception of type MWException [01:49:39] Just got that when trying to view my enwiki watchlist [01:49:45] Reload fixes [02:11:49] !log LocalisationUpdate completed (1.20wmf4) at Thu Jun 7 02:11:49 UTC 2012 [02:11:56] Logged the message, Master [02:19:52] !log LocalisationUpdate completed (1.20wmf3) at Thu Jun 7 02:19:52 UTC 2012 [02:19:57] Logged the message, Master [15:00:14] is there any way to get Pagecount data from a single project? [15:10:59] Alchimista, the files are per day, you need to filter [15:11:23] Platonides: but to get the data of a single project, i must grab the entire file, right? [15:12:41] yes, I don't think anyone is serving already-filtered ones [15:12:45] dumps halted or delayed? [15:14:01] Danny_B|backup: i didn't understood the diference [15:14:58] halted = not running at all, delayed = eg. one instance is down, so everything takes longer now... [15:15:02] Alchimista: ^^ [15:15:10] mark: around? [15:15:22] quick ipv6 question [15:15:30] how do you figure? [15:15:40] i see 3 running. all logged something recently [15:16:23] Akoopal: he's not around much the next couple days [15:16:48] Akoopal: maybe we can find someone else to answer? (or remote chance i could answer :P) [15:17:12] I was wondering if ipv6 is only via hurican electric atm [15:17:24] well it's bgp [15:17:47] my ipv4 route goes over telia, but ipv6 goes to the US, then over hurican electric and then back to emea [15:17:56] which is a bit of a detour :-) [15:18:23] are you native? [15:18:33] Danny_B|backup: any kind.. i need to trace the evolution on a small-micro-nano wikipedia, so any kind of information related to page-views can help [15:18:38] * jeremyb should go soon [15:18:44] jeremyb: yes [15:18:47] AS702 :-) [15:18:57] no [15:19:06] we have many ipv6 transit and peering [15:19:06] it's a mark ;) [15:19:31] hi mark :-) [15:19:45] Alchimista: seems like misunderstanding to me, i was asking about the current status of xml dumps [15:19:53] ugh. how do you know you're not awake? when you do `man as702` instead of `whois as702` [15:21:08] Danny_B|backup: ah, seems normal [19:54:53] !log kaldari synchronized php-1.20wmf4/extensions/LastModified/E3Experiments/js/ext.E3Experiments.Timestamp.js 'syncing js file for E3Experiments' [19:54:57] Logged the message, Master [20:32:20] !log kaldari synchronized wmf-config/InitialiseSettings.php 'turning on LastModified and E3Experiemnts for en.wiki' [20:32:25] Logged the message, Master [20:44:51] what is E3Experiemnts ?? [20:45:17] http://www.mediawiki.org/wiki/Extension:E3_Experiments - idk [20:46:34] !log kaldari synchronized php-1.20wmf4/extensions/LastModified 'syncing LastModified extension' [20:46:38] Logged the message, Master [21:18:38] eurgh [21:18:56] stupid...chatzilla. [21:33:25] apergos: ping [21:37:54] Vito: do i smell a mirror? ;) [21:38:34] jeremyb: you're right ^^ [21:38:50] <3 [21:39:06] and they have a good infrastructure [21:39:17] unfortunately they dismissed most of their international activities [21:39:34] international? [21:39:35] otherwise it would have been the best possible mirror [21:39:53] jeremyb: tiscali owned a carrier [21:40:25] you think it would be worth asking them about peering? [21:41:02] is no longer worth I think [21:41:19] https://en.wikipedia.org/wiki/Tiscali#Sales_and_acquisitions [21:41:23] but they're still a mirror? [21:41:58] that article seems to be stuck in 2009 [21:42:34] tell me they don't still sell dialup [21:42:39] RoanKattouw: can i bug you to look at the patchsets to the demux script? [21:42:51] ori-livneh: Link? [21:43:03] RoanKattouw: https://gerrit.wikimedia.org/r/#/c/10539/ [21:43:19] jeremyb: actually it is the second best cost-effective dial-up in Italy [21:43:22] *more [21:50:47] RoanKattouw: thanks! :) [22:13:40] there seems to be a weird bug on commons [22:13:44] multichill: ping [22:14:48] pong? [22:15:24] multichill: https://commons.wikimedia.org/wiki/File:Brenda_Asnicar_2012.jpg [22:15:29] are you able to delete this file? [22:15:38] (it's a blatant copyvio) [22:15:50] (even if is quite nice to see) [22:16:19] I tried to edit the page to put a {{speedy}} [22:16:25] but I got an error message [22:16:56] "Brenda Asnicar 2012.jpg" has been deleted (undelete). See the deletion log for a record of recent deletions. Depending on the reason for deletion, you may want to remove any links to this page. [22:17:37] took a while [22:19:19] so the description page had been already deleted [22:19:27] while it took a while to remove the image too? [22:19:56] https://commons.wikimedia.org/wiki/Special:Undelete/File:Brenda_Asnicar_2012.jpg [22:20:06] err I don't see any deleted description page [22:20:08] * Vito confused [22:21:14] https://commons.wikimedia.org/w/index.php?title=Special:Undelete&target=File%3ABrenda-Asnicar-2012-2.jpg×tamp=20120530195928 [22:21:18] @ Vito [22:21:40] Brooke: i'm confused. why can mediawikiwiki have rss use unrestricted but uawikimedia needs a whitelist? [22:21:42] idk [22:21:57] https://commons.wikimedia.org/wiki/Special:Undelete/File:Brenda_Asnicar_2012.jpg [22:22:09] Bsadowski1: ^ [22:22:47] I'm confused [22:23:07] Reedy, can you check to see if it failed to log the page deletion (not file)? [22:24:01] or maybe Reedy can answer that? (question above to Brooke re bug 27706) [22:36:52] Bsadowski1, Vito: I can't see what you're looking at sine I'm not an admin on Commons, but I've had issues with Commons not refreshing pages even after giving a purge command. Could commons be having the same problem with what you're looking at? [22:37:08] *since [22:37:12] We're talking about the Undelete page. [22:37:46] but both problems seem to be related to consistency [22:44:25] jeremyb: You just need to find and kill Reedy. [22:44:29] It's not too difficult. [22:46:21] Brooke: README? [22:53:23] Brooke: i guess wikimania's not too far away... [22:53:41] Brooke: anyway, can you get an answer for me? ;) [23:17:15] Brooke: you don't have a labs / gerrit account? [23:38:32] jeremyb: I'm not sure Reedy is going to Wikimania. [23:38:41] And no, I don't have an account. [23:38:43] It seemed very troublesome. [23:38:47] I wanted to be able to make my own. [23:38:57] Brooke: i'm pretty sure he is [23:39:18] Brooke: you should get one. i wanted you to review my change ;) [23:39:58] !g 10673 [23:39:58] https://gerrit.wikimedia.org/r/#q,10673,n,z [23:40:01] Brooke [23:43:16] jeremyb: Maybe one day! [23:43:21] I have enough work already. :-)