Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2016-06-19

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

All times shown according to UTC.

Time Nick Message
04:30 cjm joined #metacpan
05:13 cjm joined #metacpan
05:30 metacpan joined #metacpan
05:30 metacpan [metacpan-web] haarg force-pushed topic/permalinks from 9b24564 to e48aa49: https://git.io/voaBp
05:30 metacpan metacpan-web/topic/permalinks e48aa49 Graham Knop: use release specific pages based on url, not release status...
05:30 metacpan left #metacpan
06:08 Tempesta joined #metacpan
06:52 oiami joined #metacpan
07:47 neilb joined #metacpan
09:11 skaji Hi,
09:12 skaji I'm currently trying to move over cpanminus MetaCPAN client to v1.
09:12 skaji Then I found some issues.
09:12 skaji https://api-v1.metacpan.org/download_url/Net::LDAP::Server::Test?version=%3D%3D%200.14 (here the version query is "== 0.14") returns "status cpan" but I think it should be "status backpan".
09:13 skaji http://www.cpan.org/authors/id/K/KA/KARMAN/Net-LDAP-Server-Test-0.14.tar.gz returns 404 Not Found
09:13 skaji http://backpan.perl.org/authors/id/K/KA/KARMAN/Net-LDAP-Server-Test-0.14.tar.gz returns 200 OK
09:39 castaway skaji: you may want to add a github issue so that this doesn't get lost..
09:42 skaji castaway: Thank you for your advice! I'll open an issue on github.
10:05 dp_ joined #metacpan
11:03 ranguard skaji: thanks for the ticket update, I'm just trying to build some test cases for that end point now
11:04 ranguard ok, being an idiot, rather than faffing, how do I get a list of all versions of a dist (I want the download URL) from our api?
11:05 ranguard /release/ endpoint?
11:07 * haarg would have to look it up in the codebase
11:08 haarg really want to find time to fix metacpan explorer, especially given v1.  but i'm a bit out of date with js packaging garbage.
11:09 ranguard so many yacks!
11:09 haarg aye
11:09 ranguard isnt' the issue that we just need a backend (we now have postgres) to save the queires to?
11:09 haarg well
11:10 haarg if you want to get to the raw issue
11:10 haarg we need the api key to not be in a public repo
11:10 haarg doesn't matter how the front end gets it, as long as it's not in a public github repo things work
11:11 haarg this is github protecting us from ourselves
11:11 haarg but as long as all that gives access to is gists, i don't really care
11:12 haarg a more permanent solution is to have a different backing data store than gists
11:12 ranguard which is better, gists or other backend?
11:12 haarg better vs faster
11:13 haarg other backend is probably a better long term solution
11:13 ranguard well, either way means a server controller, but I guess gist then does't need fancy editor as you get one built in
11:13 haarg but currently explorer has to private backend at all
11:13 haarg it's just a js webpage
11:13 * ranguard nods
11:13 haarg changing the key doesn't require a server controller
11:14 haarg it just means we need to serve the static files differently
11:14 ranguard if you want to hide the key from the public is would?
11:14 ranguard s/is/it
11:14 haarg because currently explorer is a pile of static js files
11:14 haarg we don't need to hide the key from the public
11:15 haarg we just need it to not show up in a raw text search of public github repos
11:15 haarg if we stored it in a private repo but served it publicly, it would work as it used to
11:16 ranguard oh, I can do that easily then through puppet (we have the private repo for secrets that I can then deploy from)
11:16 haarg yeah
11:16 haarg thus the comment about figuring out js packaging
11:17 haarg having an api key available "publicly" isn't idea, but all it can do is interact with gists
11:17 haarg *ideal
11:17 ranguard ok, I'll stick this on my list after working out building better test data for the download_url end point
11:17 haarg the thing that broke explorer is that github scans their own public repos for api keys
11:18 haarg if it's not in a public repo, that stops being an issue
11:18 * ranguard nods
11:18 haarg more work is needed, but i'm comfortable with the compromise in the mean time
11:18 ranguard k
11:22 metacpan joined #metacpan
11:22 metacpan [metacpan-examples] ranguard created leo/all_releases from master (+0 new commits): https://git.io/voaV5
11:22 metacpan left #metacpan
11:45 ranguard skaji: thanks for the bug report btw, we'll update the ticket when we work out what's happening
11:47 ranguard sawyer: should MC::Client feature/v1 branch be passing tests?
11:47 ranguard oh, LWP::Protocol::https not installed, ignore me, will try later
11:48 ranguard all passing
11:52 ranguard could we release a RC for MC::Client that points to api-v1? - just a thought, then people (incuding me) can use that for testing :)
12:01 haarg mickey: ^
12:36 skaji randguard: Thanks! I hope wee see progress during YAPC::NA.
12:44 skaji randguard: Sorry for bothering you, but I found another issue related backpan.
12:52 skaji commented https://github.com/CPAN-API/cpan-api/issues/483
14:53 Lee_ joined #metacpan
14:54 Lee_ joined #metacpan
14:55 Lee_ joined #metacpan
14:57 leejo joined #metacpan
14:57 leejo left #metacpan
16:24 hui_ joined #metacpan
19:08 metacpan joined #metacpan
19:08 metacpan [metacpan-examples] ranguard pushed 1 new commit to leo/all_releases: https://git.io/voajV
19:08 metacpan metacpan-examples/leo/all_releases c45eeb0 Leo Lapworth: Create 2-all-versions-of-dist.pl...
19:08 metacpan left #metacpan
19:49 neilb joined #metacpan
21:21 * ranguard is close to script to take dist name and build configs for all versions to feed to api's CPAN::Faker setup
21:21 ranguard but sleep needed
21:30 alh joined #metacpan

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