Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-10-11

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

All times shown according to UTC.

Time Nick Message
00:14 metacpan joined #metacpan
00:14 metacpan [cpan-api] oalders pushed 1 new commit to oalders/cpan: http://git.io/hgQUog
00:14 metacpan cpan-api/oalders/cpan 743b412 Olaf Alders: --force-confdef when installing Elasticsearch via Travis.
00:14 metacpan left #metacpan
00:17 dipsy [ --force-confdef when installing Elasticsearch via Travis. · 743b412 · CPAN-API/cpan-api · GitHub ]
00:44 metacpan joined #metacpan
00:44 metacpan [cpan-api] oalders pushed 1 new commit to oalders/cpan: http://git.io/XkGXZA
00:44 metacpan cpan-api/oalders/cpan a59eae7 Olaf Alders: Moves custom Moose types into MetaCPAN::Types::Internal.
00:44 metacpan left #metacpan
00:45 vroom joined #metacpan
00:46 dipsy [ Moves custom Moose types into MetaCPAN::Types::Internal. · a59eae7 · CPAN-API/cpan-api · GitHub ]
00:53 vroom joined #metacpan
00:58 vroom joined #metacpan
01:29 klapperl joined #metacpan
01:34 metacpan joined #metacpan
01:34 metacpan [cpan-api] oalders pushed 1 new commit to oalders/cpan: http://git.io/46yQNg
01:34 metacpan cpan-api/oalders/cpan 4698081 Olaf Alders: Tidy.
01:34 metacpan left #metacpan
01:36 dipsy [ Tidy. · 4698081 · CPAN-API/cpan-api · GitHub ]
01:43 vroom joined #metacpan
02:01 vroom joined #metacpan
02:30 metacpan joined #metacpan
02:30 metacpan [cpan-api] oalders pushed 3 new commits to oalders/cpan: http://git.io/BWDcEQ
02:30 metacpan cpan-api/oalders/cpan 66117d4 Olaf Alders: Formatting changes in Backup.pm
02:30 metacpan cpan-api/oalders/cpan 699729c Olaf Alders: Use MooseX::Getopt::Dashes in Backup.pm
02:30 metacpan cpan-api/oalders/cpan 967502d Olaf Alders: Add more debugging to backup restore logic.
02:30 metacpan left #metacpan
02:32 dipsy [ Comparing 469808114619...967502dfe51d · CPAN-API/cpan-api · GitHub ]
03:18 vroom joined #metacpan
03:19 metacpan joined #metacpan
03:19 metacpan [metacpan-web] oalders pushed 1 new commit to master: http://git.io/5ikW4w
03:19 metacpan metacpan-web/master 726cc52 Olaf Alders: Add content to coding conventions.
03:19 metacpan left #metacpan
03:23 dipsy [ Add content to coding conventions. · 726cc52 · CPAN-API/metacpan-web · GitHub ]
03:58 vroom joined #metacpan
04:04 vroom joined #metacpan
04:26 vroom joined #metacpan
04:52 descartes joined #metacpan
05:40 Mso150 joined #metacpan
06:49 trs a bit funky: http://tsibley.net/paste/2014-10-10ytcOUanF-Selection-001.png
06:52 FROGGS joined #metacpan
07:08 neilb joined #metacpan
07:11 neilb_ joined #metacpan
07:29 kentnl haarg: How exactly does one build your CPAN-Changes repo? The usual perl Makefile.PL workflow gives me a dud dist
07:30 kentnl oh. I see. have to "make manifest". I always forget that shit.
07:30 haarg it doesn't really need to be built.  you can just add the lib dir to @INC somehow.
07:31 * kentnl usually just goes "fuck it, its easier to install things than to fuck with inc"
07:34 kentnl and now my test suite I have is of course bailing because it requires a specific version of CPAN::Changes, and CPAN::Changes in lib/ has no version.
07:37 * kentnl patches around that and now has a weird thing where entries in ->changes() on a group don't seem to go anywhere.
07:44 kentnl haarg: this test code passes on current CPAN::Changes, fails on your version: http://paste.scsys.co.uk/429209
07:44 dipsy [ magnet_web paste from Someone at 217.168.150.38... ]
07:45 haarg kentnl: serialization has been adjusted slightly
07:45 kentnl in that, none of the content of the group is emitted anywhere. It just gives a release without any content.
07:47 kentnl oh. I see. Releases' are just headers of some kind ( ie: events )?
07:47 kentnl Hum.
07:48 haarg Group is now just a proxy object for backcompat
07:50 haarg releases aren't particularly different, but they contain entries rather than groups.
07:54 haarg trying to think how i can make this work reasonably for your case
07:55 kentnl I was just going to test for object serialization instead of release serialization instead, but that has hit some bumps due to me implementing those as lazy-accessors
07:55 kentnl for some reason its behaving as if "->new( name => '' )" was passed.
07:57 haarg because groups don't have names
07:57 haarg a group dispatches to an entry that has a name
07:58 haarg so i can probably partly fix this for what you are doing
07:58 kentnl is the data still hierarchial? /me just work up and code reading is at 25%
07:59 haarg but things won't entirely work because the group objects are entirely ephemeral
07:59 kentnl namely because I do a lot of "Blow away this group and generate it from scratch" logic.
07:59 haarg changelog->release[]->entries[]->entries[]->entries[]
08:00 kentnl kk.
08:00 haarg rather than changelog->release{}->groups{}->changes[]
08:01 haarg it should be pretty easy to adapt your code to work with the new api.  but it would be nice if that was less necessary.
08:01 kentnl I could always make the code /forward/ compatible somehow.
08:01 kentnl Just yeh, the old "Is anyone in darkpan doing what I'm doing" argument :)
08:01 haarg if you are the only person subclassing it i don't mind breaking it
08:02 haarg thankfully changelogs are less likely to be an issue in darkpan code
08:02 kentnl that being true is kinda depressing.
08:02 neilb joined #metacpan
08:03 kentnl I think xdg was doing something there of some sorts though. He was asking me about my code I use to do exactly this.
08:03 haarg i'd also appreciate if you could do some testing against the main CPAN-Changes repo
08:04 haarg i've made some small changes that i think are sensible, but could easily break things doing strict testing.
08:04 haarg so feedback would be welcome.
08:06 * haarg is currently working on re-rewriting the parser to hopefully allow round tripping
08:06 kentnl http://paste.scsys.co.uk/429210 # Indentation change is all I've spotted so far there.
08:08 dipsy [ magnet_web paste from Someone at 217.168.150.38... ]
08:08 haarg which is expected.  i think the new indentation scheme is better.  it also makes it easier to match up with a recursive structure like i have planned.
08:08 kentnl RIP MetaCPAN diffs >:)
08:10 kentnl https://github.com/CPAN-API/CPAN-Changes/commit/087f3b7a5f340a0731811cbd511ab01db6649448 # Also, can you not?
08:10 kentnl I already have that localised myself because 78 columns is rather small for my liking and I've localised it myself to 120
08:11 haarg Text::Wrap is not part of the CPAN::Changes api
08:11 haarg if you want an api for it, request it
08:11 dipsy [ wrap at 78 columns rather than 76 · 087f3b7 · CPAN-API/CPAN-Changes · GitHub ]
08:12 haarg also keep in mind that i likely won't ship CPAN::Changes until i have the rewrite working
08:14 haarg which i probably should use Text::Wrap for, but i'll still be localizing its vars.
08:14 neilb_ joined #metacpan
08:16 haarg and my current rewrite doesn't split long words, and that will continue to be the case.  i would add that to CPAN::Changes if i didn't intent to replace it entirely.
08:18 haarg basically, my changes to CPAN::Changes master are to bring it more in line with how i want things to work, but keeping its tests working.  so that i can take its test suite unmodified and use it against my rewrite.
08:20 kentnl 5 files changed, 4492 insertions(+), 4420 deletions(-)  # This is the sort of delta I'm looking at by simply renormalizing one of my dists with current CPAN::Changes
08:21 * kentnl is not exactly complaining, more complaining on the behalf of others who might complian that my changes jump too much
08:21 haarg i'm going to assume that's pretty much entirely indentation changes
08:22 kentnl Yeh. Literally every line of those files is changed except for the release lines.
08:23 kentnl If I reduce it just to re-wraps its not so bad
08:23 kentnl 5 files changed, 185 insertions(+), 113 deletions(-)
08:26 kentnl Hm, running my dep changes tool w/ CPAN-Changelog I got this : Can't call method "text" on an undefined value at /home/kent/perl/forks/CPAN-Changelog/lib/CPAN/Changes/Release.pm line 132.
08:30 haarg hmm
08:30 haarg how are things being constructed with that tool?
08:31 kentnl Oh, I probably need to blow away my object cache. Sec.
08:32 kentnl yeh. Seems it doesn't work so good if you resurrect a {} created with an old version and expect it to work with a new version :)
08:32 haarg given that the new code and old code have pretty much nothing in common... yeah
08:34 kentnl So far so good though, CPAN::Changelog version *mostly* works for remunging my Changes file, its just the more complex CPAN::Changes::Dependencies subclass that is printing empty files
08:34 kentnl but that's more to be expected
08:35 kentnl https://metacpan.org/requires/distribution/CPAN-Changes?sort=[[2,1]] # Wow, very useful </sarcasm>
08:37 dipsy [ Distributions which depend on CPAN-Changes - metacpan.org ]
08:37 neilb joined #metacpan
08:38 kentnl why it fails is of course entirely obvious if you read the source.
08:41 haarg as for handling subclasses of ::Group, i should be able to make it work to serialize things that override ->changes
08:42 haarg what won't work is getting back out the group object you put in
09:08 neilb joined #metacpan
09:27 castaway joined #metacpan
12:06 neilb joined #metacpan
12:11 neilb_ joined #metacpan
12:19 vroom joined #metacpan
12:40 vroom joined #metacpan
12:49 vroom joined #metacpan
13:08 kentnl wow. That's an awesome fail: https://metacpan.org/source/BOWTIE/Debug-Client-0.29/eg/debugger.pl
13:08 kentnl I don't even.
13:08 dipsy [ eg/debugger.pl - metacpan.org ]
13:11 kentnl Source view in firefox shows that there's unbalanced HTML
13:19 kentnl Something is fucked today, github is doing that old "hurp durp, can't let you file an issue!" thing.
13:21 kentnl https://github.com/CPAN-API/metacpan-web/issues/1386
13:23 dipsy [ Unbalanced HTML leads to epic fail in source view. · Issue #1386 · CPAN-API/metacpan-web · GitHub ]
13:48 bob_spoonmongler joined #metacpan
14:05 kentnl Good. Coveralls appear to have just broken their image linking service.
14:15 bowtie kentnl: do you have some more info for me to take a look at :(
14:16 kentnl Its not your fault bowtie, I'm pretty sure its metacpan-web breaking
14:17 kentnl Because nobodies source code should be able to break the HTML. That's special kinds of bad.
14:17 bowtie kentnl: ok, thanks for the heads up :)
14:20 kentnl I don't even know what's happening there, there's a "Documentation view" link rendered on the left, and I can't find any other page that has it.
14:21 metacpan joined #metacpan
14:21 metacpan [metacpan-web] haarg pushed 1 new commit to master: http://git.io/lGMKzw
14:21 metacpan metacpan-web/master a29be3d Graham Knop: fix markup on source page
14:21 metacpan left #metacpan
14:22 haarg i could have sworn i had already fixed that issue
14:23 dipsy [ fix markup on source page · a29be3d · CPAN-API/metacpan-web · GitHub ]
14:24 kentnl haarg++ # fixed
14:25 kentnl "reopened"?
14:26 haarg not sure what happened there
14:26 haarg may have mis-clicked
14:34 bob_spoonmongler joined #metacpan
14:36 vroom joined #metacpan
14:44 bob__spoonmongler joined #metacpan
14:53 neilb joined #metacpan
15:00 vroom joined #metacpan
15:07 bob_spoonmongler joined #metacpan
15:22 neilb_ joined #metacpan
15:32 neilb_ joined #metacpan
15:35 vroom joined #metacpan
15:54 kentnl http://github-meets-cpan.com # DNS has expired
15:58 dipsy [ github-meets-cpan.com - github-meets-cpan Resources and Information. This website is for sale! ]
16:17 vroom joined #metacpan
17:18 vroom joined #metacpan
17:19 Mso150 joined #metacpan
18:32 bowtie joined #metacpan
18:36 ranguard oalders: ^^ I can get it on TPF DNS if you can get the transfer details
19:15 Mso150 joined #metacpan
19:42 vroom joined #metacpan
20:03 andreea joined #metacpan
20:05 vroom joined #metacpan
20:14 bowtie joined #metacpan
21:22 Mso150 joined #metacpan
22:05 oalders ranguard: ok. if i'll ask plu about it.  if it's already expired, he'd have to renew it before transfer
22:05 oalders but not sure if it expired last month or one year ago and it's just squatted on now
22:06 oalders i'd also be fine with github-meets-cpan.metacpan.org or something more succinct
22:21 trs whois says that the domain registration was Sept 26 2014.
22:22 trs github-meets.metacpan.org?
22:31 oalders yeah, that's probably even better :)
22:31 oalders unless there's a legal issue with using github in a domain name
22:31 neilb_ joined #metacpan
23:03 Mso150 joined #metacpan
23:03 bowtie joined #metacpan
23:03 andreea joined #metacpan
23:03 bob_spoonmongler joined #metacpan
23:03 FROGGS joined #metacpan
23:03 klapperl joined #metacpan
23:03 bvierra joined #metacpan
23:03 tokuhirom joined #metacpan
23:03 daxim joined #metacpan
23:03 zostay joined #metacpan
23:03 jnbek joined #metacpan
23:03 sivoais joined #metacpan
23:03 bricas joined #metacpan
23:03 n0body joined #metacpan
23:03 Khisanth joined #metacpan
23:03 dipsy joined #metacpan
23:03 SineSwiper joined #metacpan
23:03 vanstyn joined #metacpan
23:03 ilmari joined #metacpan
23:03 ribasushi joined #metacpan
23:03 osfameron joined #metacpan
23:03 GumbyNET7 joined #metacpan
23:03 Ralesk joined #metacpan
23:03 kentnl joined #metacpan
23:03 cjm joined #metacpan
23:03 andrefs joined #metacpan
23:03 grantm joined #metacpan
23:03 skaufman joined #metacpan
23:03 GumbyNET5 joined #metacpan
23:03 ingy joined #metacpan
23:03 jibsheet joined #metacpan
23:03 BooK joined #metacpan
23:03 michael_ joined #metacpan
23:03 tianon joined #metacpan
23:03 alh joined #metacpan
23:03 marcus joined #metacpan
23:03 wreis joined #metacpan
23:03 mattp_ joined #metacpan
23:03 oalders joined #metacpan
23:03 ether joined #metacpan
23:03 mst joined #metacpan
23:03 mdk joined #metacpan
23:03 BinGOs joined #metacpan
23:03 punytan_ joined #metacpan
23:03 ivan joined #metacpan
23:03 ranguard joined #metacpan
23:03 rsrchboy joined #metacpan
23:03 mo joined #metacpan
23:03 rafl joined #metacpan
23:03 priodev joined #metacpan
23:03 celogeek joined #metacpan
23:03 Kovensky joined #metacpan
23:03 Mithaldu joined #metacpan
23:03 avenj joined #metacpan
23:03 berekuk joined #metacpan
23:03 omega joined #metacpan
23:03 mala joined #metacpan
23:03 alnewkirk joined #metacpan
23:03 Seveas joined #metacpan
23:03 charsbar joined #metacpan
23:49 kentnl thought: why target only github
23:49 kentnl in theory you could target >1 source control system or even >1 social coding target
23:50 kentnl and github would simply be the #1th of them
23:51 kentnl maybe coders-of.metacpan.org or something
23:51 kentnl hackers-on.metacpan.org

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