Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-03-12

| Channels | #metacpan index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 trs neato
00:01 trs it'd be cool if you added that diff calculation to metacpan itself :)
00:06 punter The calculation is like less than half a page of code.
00:06 punter Why don't I give you the code, and let you add it to metacpan, since you know metacpan better?
00:11 trs the benefit of prodding you to do it is that we'll bootstrap up another contributor ;)
00:12 trs but if you don't want, would you submit an issue to https://github.com/CPAN-API/cpan-api with the code?
00:12 dipsy [ CPAN-API/cpan-api · GitHub ]
00:12 trs once it's in metacpan, you can use it to reduce your rss generating code even further :)
00:15 punter there's a problem
00:15 punter I never delete a version from my database ( I do'nt have to)
00:15 punter but metacpan does
00:16 punter so what if someone deletes version 0.17 before uploading 0.18? The changediff will be between 0.16 and 0.18 then
00:16 punter i.e. it will show both 0.17 and 0.18
00:16 punter in that one diff
00:17 punter so your output will not be 100% reliable
00:17 punter maybe we want it to contain two versions in one diff, on the other hand
00:18 punter but maybe you don't want metacpan to do anything that's not 100% correct all of the time
00:19 punter confusing
00:22 trs metacpan doesn't delete anything.
00:22 trs and metacpan does plenty of things that aren't 100% correct.
00:23 trs so does most of the Perl toolchain. :)
00:24 trs I guess realistically, metacpan should just expose the "changes in this release" it already has via the api
00:24 trs i.e. the bottom of https://metacpan.org/release/Moose
00:24 dipsy [ Moose-2.1204 - A postmodern object system for Perl 5 - metacpan.org ]
00:29 trs which would mostly mean moving the changelog parser and last_version filtering to the api instead of the web frontend.
00:30 trs it takes a parsing route rather than a pure diff route, which looks like what you do?
00:30 punter I'm doing a plain diff
00:30 punter pure diff
00:31 punter So you shouldn't use my code, after all, right?
00:32 punter The parser does a better job in some ways
00:34 trs and a worse job in others.
00:35 trs there are lots of things it is too strict about.
00:38 punter the API user could choose which of the two he wants, with an option
00:54 trs yup
00:55 trs either a url part or query param flag, or an Accept: text/x-diff header
00:55 trs the default should probably be a textual representation of the parsed variant (i.e. the one we use on the web)
00:56 punter What uses can you find for diffs?
00:56 punter one is tracking your favorite modules (what I'm doing)
00:56 punter I wonder who's use it
00:57 punter (the API, I mean)
01:03 punter g'night
01:39 omega we wanted to have the diff as a fallback for the release page where it shows CPAN::Changes parsed stuff now, if something doesn't parse well with CPAN::Changes, or fails to parse
02:08 klapperl joined #metacpan
02:27 oiami joined #metacpan
02:29 trs omega: that would be awesome.
02:31 trs esp as most of my dists' changelogs aren't parsed by CPAN::Changes
06:32 [Sno] joined #metacpan
06:37 nbezzala joined #metacpan
07:41 dpetrov_ joined #metacpan
07:49 neilb joined #metacpan
08:39 oiami left #metacpan
08:55 mo joined #metacpan
09:32 rwstaune` joined #metacpan
10:11 kurahaupo_ joined #metacpan
13:07 gugod joined #metacpan
13:23 oiami joined #metacpan
14:03 rwstauner left #metacpan
14:03 rwstauner joined #metacpan
14:51 ether joined #metacpan
14:53 ether_ joined #metacpan
15:04 rwstaune` joined #metacpan
15:09 rwstauner joined #metacpan
15:58 oiami joined #metacpan
16:23 nbezzala joined #metacpan
17:15 cooper joined #metacpan
17:23 ether joined #metacpan
17:28 ether_ joined #metacpan
17:53 ether_ joined #metacpan
17:56 ether joined #metacpan
17:59 ether_ joined #metacpan
18:14 ether joined #metacpan
18:14 ether_ joined #metacpan
18:58 punter joined #metacpan
21:02 trs_testing joined #metacpan
21:07 ether joined #metacpan
21:07 ether_ joined #metacpan
23:14 metacpan joined #metacpan
23:14 metacpan [cpan-api] oalders created oalders/code-cleanup (+3 new commits): http://git.io/QDLimA
23:14 metacpan cpan-api/oalders/code-cleanup 63ffc87 Olaf Alders: Code cleanup....
23:14 metacpan cpan-api/oalders/code-cleanup 66fc9ad Olaf Alders: Explicitly stringify DateTime object in MetaCPAN::Script::Release.
23:14 metacpan cpan-api/oalders/code-cleanup 6bb9132 Olaf Alders: Try::Tiny is not used in MetaCPAN::Util.
23:14 metacpan left #metacpan
23:14 dipsy [ Comparing 63ffc87eb7c7^...6bb913201c28 · CPAN-API/cpan-api · GitHub ]
23:14 metacpan joined #metacpan
23:14 metacpan [cpan-api] oalders created oalders/minor-cleanup (+1 new commit): http://git.io/klA5Zw
23:14 metacpan cpan-api/oalders/minor-cleanup 999b447 Olaf Alders: app.psgi is no longer a symlink to MetaCPAN::Server
23:14 metacpan left #metacpan
23:14 dipsy [ app.psgi is no longer a symlink to MetaCPAN::Server · 999b447 · CPAN-API/cpan-api · GitHub ]
23:15 metacpan joined #metacpan
23:15 metacpan [cpan-api] oalders created oalders/psgi from oalders/minor-cleanup (+0 new commits): http://git.io/I0cNxg
23:15 metacpan left #metacpan
23:15 dipsy [ Commits · CPAN-API/cpan-api · GitHub ]
23:16 oalders some of those builds will be broken
23:16 oalders rwstauner: if you can have a look at oalders/code-cleanup at some point, i wouldn't mind some pointers on how to fix some of those failing tests
23:17 oalders they were failing on my VM, anyway
23:31 punter Is metacpan down?
23:32 grantm slow
23:32 punter ok

| Channels | #metacpan index | Today | | Search | Google Search | Plain-Text | summary