[01:23] http://consumerist.com/2012/05/rock-band-for-iphone-game-will-self-destruct-in-29-days.html [01:30] let's archive it! and break a million laws in the process! [01:33] http://i.imgur.com/IP8pI.jpg [01:46] actually- no it won't: http://www.theverge.com/gaming/2012/5/2/2995299/ea-rock-band-ios [02:48] yaay laws [02:49] Nemo_bis: There are gaps in some of these runs of magazines - that's known, right [03:05] AWWWWWWW shit [03:05] I got the scanner running. [03:49] http://archive.org/details/gamesmachine-034 [06:06] SketchCow, what gaps? I sent you a list of all the stuff I was shipping you, with numbers included; that's all I know [06:10] hm or maybe I didn't [06:11] the main part of it was [06:11] *Pc Action: 93 magazines, 46 floppy disks, 55 CD-ROMs/DVDs (1-7, 9-94) [06:11] *Pc World: 160 CD-ROMs/DVDs, 3 booklets (1996-2010) [06:11] *The Games Machine: 94 magazines (87-180), 197 CD-ROMs/DVDs [06:11] Content: 187 magazines, 412 CD-ROMs/DVDs, 106 floppy disks, 6 books: [06:12] I don't remember what I checked and I added some more stuff later. [06:32] No worries [06:32] I just wanted to make sure that when I was digitizing shizzle and then I was missing here and thee, I wasn't, you know, losing stuff. [06:46] http://archive.org/details/gamesmachine-034 [06:46] Tah dah!!! [06:52] :D [06:52] The file listing is empty [06:53] Yeah, the new machine isn't loving ISO life. [06:53] Anyway, minor setback. [06:56] Internet Explorer 4.0 [09:46] err [09:46] how long ago was the "migrate videos from Google video to youtube" thing? [09:50] as I *just* got notifcation of it and switched mine. [11:23] SmileyG: Looong time [11:23] Over 6 months ago [11:23] probably like 9 [11:25] heh weird. [11:25] tho my videos were uploaded.... [11:25] * SmileyG can't rememder the date now, nov 2009 I think.,... [13:42] i need wget help. i would like to mirror http://forum.jamendo.com/ but exclude any URLs that contain /comment/ /profile/ or /entry/ [13:42] those are NOT directories, but arguments to index.php, eg http://forum.jamendo.com/index.php?p=/profile/331639/509 [13:44] I thought the "--reject" argument took care of that. Can use it to get rid of links that have things like "&action=print," and aren't really different content. [13:45] i thought that would only look at the extension (as in end of the url string) [13:45] i totally asked this before i think [13:45] I can test it right quick. Thought it could match patterns as well. [13:45] yeah [13:45] Note that if any of the wildcard characters, *, ?, [ or ], appear in an element of acclist or rejlist, it will be treated as a pattern, rather than a suffix. [13:45] sweet [13:54] can't figure it out. will resume later [16:06] fileplanet is shutting down [16:07] http://www.fileplanet.com/ [16:07] "Note: FilePlanet is no longer being updated and is in the process of being archived." [16:07] ha ha ha [16:07] fuck [16:07] "archived" [16:07] fuckfuckfuck [16:07] it's IGN [16:07] should be fairly easy to grab, you can wget urls you get from the site without cookies [16:07] iirc even the timestamps are correct [16:09] the terrible thing is that many files are from the old hosted planet* sites and probably not listed on the fileplanet pages [16:09] fileplanet isn't the one that requires an obnoxious downloader and has enforces browser useragent strings, is it? [16:09] * shaqfu tries [16:10] No, seems to work fine [16:11] i am logged in, so it might be different but example: [16:11] http://www.fileplanet.com/224345/220000/fileinfo/Elder-Scrolls-V:-Skyrim---Macho-Dragons-v1.0 [16:11] click download [16:11] you get http://www.fileplanet.com/224345/download/Elder-Scrolls-V:-Skyrim---Macho-Dragons-v1.0 [16:11] Click here [16:11] you can wget that just fine [16:11] Phew [16:12] Now I'm curious how far back their downloads go, given how old the site is [16:12] they have a "Customer Support Fax" [16:12] shaqfu: 1996 at least [16:12] Yeah, I saw Half-Life 1 patches [16:12] it started with planetquake and they migrated its downloads to fileplanet afaik [16:12] Yeah, they're def. p. oldskool. I don't think they ever pruned stuff. [16:13] "Customer Support Fax", wow. [16:14] http://www.fileplanet.com/174/0/section/Gaming looks like a good starting point [16:14] We should ask for a fax of their game demos... [16:15] hm, 1999 for the oldest quake dm map [16:15] Get a screenshot faxed back [16:15] Adrian Chen - Gawker [16:15] http://achewood.com/index.php?date=11222006 [16:15] thanks for keeping this place gawker free [16:15] Trying to get an interview? [16:16] Or just hoping something interesting happens to write up a clickmagnet headline? [16:16] I assume same. [16:17] I also assume he'll do the Sekrit Maneuver of coming back in here under an "assumed name". [16:19] heh, wget -m --spider http://www.fileplanet.com/ [16:20] http://www.fileplanet.com/index.html works [16:20] "Jason Scott Bans Me From #archiveteam, Potentially Involved With Chickens" [16:21] or at least that's the caliber of writing Gawker seems to trade in [16:21] lol! [16:22] FPOps@IGN.com seems like a high contact [16:22] Hey, some of you might know this. [16:22] So, jsmess.textfiles.com no longer shows the javascript [16:22] I know the big change - new apache server. [16:22] But what's the setting to get it back? [16:23] Do you get a 'download this javascript' file instead of the javascript? [16:23] Or nothingness? [16:24] Nothiness. Check it out. [16:25] SketchCow: it looks like the Content-Encoding header on the javascript is missing [16:25] I am SURE it's a handler I don't have enabled. [16:25] OK, where would I enable that in apache2.2 [16:25] I went from 1.3 something to 2.2 something so it was a fairly nasty jump. [16:25] in an appropriate configuration file (httpd.conf, virtualhost config, etc): [16:25] [16:25] Header set Content-Encoding gzip [16:25] Header set Content-Type "text/javascript; charset=utf-8" [16:25] [16:25] I *think* that'llw ork [16:27] SketchCow: you can set that in a or context, so scope to whichever one is most specific, etc [16:27] oh, fuck, I forgot that that needs mod_header loaded [16:28] well, maybe it's already loaded, we'll see [16:30] "You have already activated rake 0.9.2.2, but your Gemfile requires rake 0.9.2. Using bundle exec may solve this." [16:30] motherfucking christ I hate Ruby environments sometimes [16:31] yipdw: Did it, no load [16:31] LoadModule headers_module libexec/apache22/mod_headers.so [16:31] hmm [16:31] it's definitely loading it. [16:32] hmm, ok [16:32] well, let's pull that FilesMatch thing out then [16:33] I enabled compress [16:33] already pulled [16:33] ok [16:33] Still not working, but enabled compress. [16:33] the error Chrome is throwing is "Uncaught SyntaxError: Unexpected token ILLEGAL", so it's interpreting the gzipped javascript without decompressing it [16:33] I'm trying to figure out what combination of headers it needs to go "oh, this is gzipped" [16:34] yay, we can go with http://www.fileplanet.com/NUMERICID/download/ [16:34] I thought Content-Encoding was it, and it probably is [16:35] pretty slow though [16:35] but there's probably something else wrong in there [16:35] one sec [16:35] it is both content-encoding and content-type [16:36] devesine: yeah, but I'm not sure what was wrong with either [16:36] [16:36] [16:36] ForceType text/javascript [16:36] Header set Content-Encoding: gzip [16:36] what does ForceType do that Header set Content-Type doesn't? [16:36] heck if i know [16:36] uhh [16:36] yeah, nice. the /download/ HTML includes both the direct download link AND the location of the file, eg "Home / Gaming / RPG / Massively Multiplayer / Gas Guzzlers: Combat Carnage / Game Clients" [16:37] looks like it causes all internal apache bits to treat it as that type [16:37] wtf [16:37] though i don't know what aside from the content-type header cares [16:37] wtf apache [16:37] well, ok [16:37] I guess that can be tried [16:38] I'm not sure why the default MIME type detection would screw it up [16:38] mail to ign sent [16:39] possibly the mime type Content-Type header gets set late in the process? [16:39] devesine: I dunno -> https://gist.github.com/14cb242afb08c4c6714a [16:39] (it's been a blissfully long time since i had to know details about apache internals that fine-grained - these days i'm all about nginx) [16:39] that's what I got with the previous ForceMatch in place, which I thought would work [16:39] but evidently it didn't [16:39] and I'm not sure why [16:40] huh, that looks reasonable to me [16:40] yeah, me too [16:40] unless the charset thing is screwing it up [16:41] http://www.fileplanet.com/024884/download/ is the same as http://www.fileplanet.com/24884/download/ and they both work [16:41] where by "it" I mean Chrome [16:42] https://gist.github.com/14cb242afb08c4c6714a#gistcomment-298537 is my server's response [16:42] it looks like Apache is doing the right things there, unless it's mangling the response body [16:42] yeah, i'm using chrome too [16:42] hm [16:42] i wonder what happens with no charset set [16:43] huh, i'm getting Content-Encoding: x-gzip back from jsmess.textfiles.com [16:43] (in curl) [16:44] the configuration might have changed [16:47] chrome thinks the headers I got back include [16:47] Content-Type: application/x-gzip [16:47] fuckers return 302 for non existing links... [16:50] gah. can i not tell wget not to follow a 302 to a directory? [16:50] "wget -X error http://www.fileplanet.com/1/download/" still downloads "http://www.fileplanet.com/error/error.shtml?aspxerrorpath=/autodownload.aspx" [16:50] hm, chrome wasn't believing that the cache was cleared, it looks like [16:51] clearing the cache, quitting chrome, starting it up again, clearing the cache, quitting chrome, starting it up again, and /then/ going to jsmess.textfiles.com loaded it fine [16:53] http://blog.jamendo.com/2012/05/03/response-to-the-community/ [17:03] i wonder how soon fileplanet links expire [17:12] so, we need to download http://www.fileplanet.com/NUMERICID/download/ , grep that file for the dl link. and download the file (to the same dir i guess, sorting can happen later) [17:12] i love bash [17:14] and jedit [17:16] hm, the actual download links are interesting too. eg /ftp1/fpnew/action/quake/levels/dm/armory2.zip [17:17] not sure if i should rather save the files in that part or in www.fileplanet.com/NUMERICID/download/ [17:17] thoughts? [17:19] nah [17:19] the url is saved so if needed one could do that later [17:21] script is ready \o/ [17:22] hm, any suggestion how i can make a loop like: for i in {$1..$2}? [17:22] i would like to pass the starting and last id to download from the commandline [17:22] doing this i get "{1466..1470}" [17:22] for $i itself [17:23] "for i in $(seq $1 $2)" seems to work [17:24] maybe for i in {$1..$2}? [17:24] nope, that was my first try [17:25] Ah [17:25] this works great, yay [17:25] * bayleef` grins [17:34] http://jsmess.textfiles.com/ is back. Thanks for the help, everyone. [17:34] As a bonus - some keys work! [17:36] https://github.com/SpiritQuaddicted/fileplanet-file-download/blob/master/download_pages_and_files_from_fileplanet.sh [17:36] we'll need to assign ranges. starting from 1 and ending somewhere in 200k [17:36] it takes about one second for non-existing IDs so far [17:36] the downloads much be big [17:39] i'll start with 1-2000 just to see what happens [17:39] Looks good [17:40] Is it roughly chronological in terms of file ID? [17:40] yeah, i think very much so [17:40] Gotcha. Should go quicker early, then [17:43] If it does its job, let me know and I'll run it [17:46] sweet :) [17:47] ahaha http://www.kb.cert.org/vuls/id/520827 [17:49] I'll give it a shot now, actually, on 2001-4000 [17:49] thanks [17:50] http://jsdosbox.appspot.com [17:54] <15 minutes per 1k so far [17:54] but only single digits of actual found files ;) [17:55] and i am quake fanatic enough to know all of them [17:55] i'll go 4001-10000 [17:56] Sigh; I really need to get a real OS on this server, and not one with flaky wget that doesn't have --max-redirect [17:57] haha [17:59] Got it this time; doing 10k-15k [17:59] yay [18:00] Seems to be moving a bit slower for me, although those are probably much larger files than Quake maps [18:01] might also be just fileplanet sucking [18:01] That too [18:01] eralier i sometimes had 5-10 seconds per html page download [18:03] hm, i have a bug [18:04] ? [18:05] And it looks like there are ~220k fileIDs... [18:05] ah, haha. i try to download even if there is no link. so files_$1_$2.log gets a lot of "http://: Invalid host name." [18:05] i should make that nicer really [18:05] scared me for a moment [18:05] Ah, phew; nothing critical then [18:06] holy hell this jsdosbox [18:25] Feel inspired. [18:32] doing fileplanet 15k-16k [18:36] Are there any IGN network sites at-risk like this? [18:36] any other* [18:37] i'd consider them all at risk :( [18:38] interesting, the new planet* files are not served from fileplanet anymore, i never noticed [18:38] eg http://planetquake.gamespy.com/View.php?view=Quake4.Detail&id=157 [18:39] I don't see the warnings on planetquake; guess it's not at immediate risk [18:39] haha, no way. they seem to have dropped forumplanet without any notice [18:39] wtf [18:40] jesus [18:40] and i mirrored by pure random intend [18:40] Ouch [19:01] hm, the forums might just have been moved. eg http://www.ign.com/boards/forums/diablo.5218/ exists [19:47] 15k-16k took 70 minutes [19:47] ~300 mb [19:48] bedtime for me. anything above 16k is free [19:48] night [21:33] http://eindbazen.net/2012/05/php-cgi-advisory-cve-2012-1823/ I'll bet this is how the old wiki kept getting hacked [21:33] (current one does not seem vulnerable) [21:36] does your old wiki use php-cgi? [21:37] according to the page, it was a common configuration on dreamhost [21:37] Oh PHP. [21:37] oh [21:41] Oh Dreamhost [21:41] That too... [21:49] Well, today's foray into scripting taught me something: line breaks suck [22:03] "We found that giving the query string ‘?-s’ somehow resulted in the “-s” command line argument being passed to php, resulting in source code disclosure." [22:03] what the fuck [22:03] not "somehow", some guy removed protection against that in like 2004 because it made unit testing more complicated [22:05] unit testing of what [22:05] I guess I should look at the commit history [22:05] probably unit testing that didn't exist [22:06] because the fact that an application query string can affect the command line seems ridiculous on its face [22:06] the query string is commandlineified [22:06] I probably sound like I'm making excuses, don't I? [22:07] oh [22:07] fuck those guys. they couldn't program themselves out of a paper bag. [22:07] "From the regression testing system we [22:07] use -d extensively to override ini settings to make sure our test [22:07] environment is sane." [22:07] that's still a what teh fuck [22:07] why not just put in a "sane" ini file as a precondition [22:07] too complicated [22:46] yipdw: Testing is hard.