02:24:51tzt quits [Ping timeout: 265 seconds]
03:06:01HP_Archivist (HP_Archivist) joins
03:58:31tzt (tzt) joins
04:41:19qw3rty__ joins
04:44:53qw3rty_ quits [Ping timeout: 258 seconds]
05:28:12qwertyasdfuiopghjkl quits [Client Quit]
05:30:30HP_Archivist quits [Ping timeout: 258 seconds]
08:09:50sonick quits [Client Quit]
08:35:26AK quits [Quit: AK]
08:37:22AK (AK) joins
13:37:04HP_Archivist (HP_Archivist) joins
13:57:16HP_Archivist quits [Ping timeout: 258 seconds]
14:09:32Arcorann_ quits [Ping timeout: 258 seconds]
14:16:39qwertyasdfuiopghjkl joins
14:32:14HP_Archivist (HP_Archivist) joins
14:38:17HP_Archivist quits [Ping timeout: 258 seconds]
15:08:24]42[ is now known as [42]
16:44:56HP_Archivist (HP_Archivist) joins
17:01:43spirit joins
18:33:27atphoenix quits [Ping timeout: 265 seconds]
18:34:39atphoenix (atphoenix) joins
18:44:00HP_Archivist quits [Ping timeout: 258 seconds]
19:02:52atphoenix_ (atphoenix) joins
19:05:28atphoenix quits [Ping timeout: 258 seconds]
19:20:25<Ryz>Hmm, for the past week or two, there's a notice on the Internet Archive donations when checking archived pages on the Wayback Machine, but it seems to be stuck on $3.5 million raised so far? And sometimes it doesn't appear? Is this normal?
19:50:35HP_Archivist (HP_Archivist) joins
19:56:19qwertyasdfuiopghjkl quits [Client Quit]
20:36:22AlsoHP_Archivist joins
20:40:05HP_Archivist quits [Ping timeout: 265 seconds]
20:51:16AlsoHP_Archivist quits [Ping timeout: 258 seconds]
20:52:04AlsoHP_Archivist joins
20:58:56AlsoHP_Archivist quits [Ping timeout: 258 seconds]
20:59:43AlsoHP_Archivist joins
21:12:21AlsoHP_Archivist quits [Ping timeout: 258 seconds]
21:33:09spirit quits [Client Quit]
21:57:45<@JAA>https://web.archive.org/web/*/https://github.com/jjjake/internetarchive → https://web.archive.org/web/*/internetarchive/*
21:57:48<@JAA>WTF?
22:15:25<OrIdow6>Playing around with this, it looks like (and I do not think this is entirely accurate) if the tokenized path (slashes and periods, at least) has "archive" in it, it changes it to the last path component (as determined by splitting on "/")
22:15:40<OrIdow6>*changes the URL to
22:16:49<OrIdow6>Counterexample: https://en.wikipedia.org/wiki/archive
22:17:01britmob25636477 quits [Quit: britmob25636477]
22:17:34<OrIdow6>Oh, the * on the end also seems to prevent it
22:19:33<OrIdow6>So https://web.archive.org/web/*/https://en.wikipedia.org/wiki/Archive -> https://web.archive.org/web/*/Archive/*
22:22:14<Jake>looks... broken?
22:28:17Cook joins
22:28:26<Cook>is there any way to search all of the Internet Archive by suffix, rather than domain/prefix?
22:30:20<Cook>I'm looking for a series of files that have well-defined names, but were hosted (and archived) on fairly random IP addresses between 2001 and 2003, like here: https://web.archive.org/web/*/66.28.11.52/*
22:31:46<Cook>The most extreme stupid way would be to go through 2^32 IP addresses, but I'm unsure if there's a better index, or an easily accessible list of partially-archived domains, that would make this plausible to look through
22:36:02<@JAA>The index only works by prefix, but the host is indexed in reverse order internally. Through the CDX API, you can query subdomains, and IPs are treated like domains as far as I can tell.
22:36:11<@JAA>So you could do it with 256 queries against the CDX API I suppose.
22:37:04<Cook>Oh, I hadn't considered that IP ranges are themselves sort of prefixes - I'll take a look and see whether that's tractable. thanks!
22:38:02<@JAA>Hmm nope, querying for *.52 doesn't work. But *.11.52 does.
22:38:34<@JAA>So 64 Ki queries. Still an improvement over 2^32.
22:39:10<@JAA>Note that the CDX API has awkward rate limiting which redirects you to an HTML page that's served with HTTP 200.
22:39:55<Cook>Interesting that it works on the last octets rather than first ones
22:40:04<Cook>But this is way better than what I was considering, so thanks
22:40:06Arcorann (Arcorann) joins
22:40:16<@JAA>As I said, hosts are indexed in reverse order of the labels.
22:40:28<@JAA>Otherwise, it wouldn't be possible to do subdomain searches.
22:40:58<@JAA>https://example.org/foo/bar internally becomes org,example)/foo/bar I believe.
22:41:02<Cook>Yeah, it makes sense for domains but it's still kind of funny since it's sort of inverting the order of how IP blocks work
22:41:23<@JAA>I think they simply didn't bother implementing something special for IPs.
22:52:58TheTechRobo joins
22:55:58<Arcorann>Is anyone else having issues with accessing sites on the Wayback Machine?
22:56:32<@JAA>Yeah, it was discussed in here briefly an hour ago.
22:56:51<TheTechRobo>Can you tl;dr it? I just joined
22:57:01<@JAA>TL;DR: Shit's broken, yo.
22:57:05britmob25636477 joins
22:57:11<TheTechRobo>:thumbsup:
22:57:40<@JAA>Prefix searches with a trailing /* seem to work, everything else is getting rewritten incorrectly.
22:59:14Sluggs quits [Client Quit]
22:59:15<@JAA>Actually, only the calendar thing is broken, snapshots still seem to work.
23:19:18Sluggs joins
23:44:47thetechrobo_ joins
23:48:22TheTechRobo quits [Ping timeout: 258 seconds]