lizmat JRaspass: I've been uploading about 20 modules today to the zef ecosystem 16:36
I've verified they can all be installed using zef 16:37
yet none of them appear on
is there something I need to do still ?
tonyo there's a job that runs on the hour and then they should get indexed 16:44
lizmat so then they should show up on, right? 16:47
in the recent dists list, right? is now at 0.0.7 and uploaded today (more than 6 hours ago) 16:48
and is a 404
16:51 lizmat_ joined
tonyo i wonder if the ingestor can't find the META in that tar/gz 16:54
i think it should, though because the cpan stuff finds it
16:55 lizmat left 16:56 lizmat_ left, lizmat joined
lizmat tonyo: so basically the same thing that fez is complaining about 16:57
but which is ok for zef or the indexer
tonyo the indexer? 17:06
that's a guess, i don't know for sure
lizmat well, if it wouldn't be indexed, how could I install a module with zef then? 18:18
tonyo indexed by what? 18:30
lizmat the zef indexer? 18:32
sorry, lost context
I mean, zef, or the indexer that uses
tonyo well, fez uses a different indexer than rl 18:39
i'd have to look at the ingestor for rl to know if it's just an incompatibility between it and rl
lizmat looks like knows about the modules I uploaded: 18:52
compared to:
now I *do* wonder about the double zef: in that URL
and it apparently cannot find the version either 18:53
however, the version is clearly in the META: 18:54
ok, I guess I'll halt moving modules to fez for now 19:02
wouldn't mind redoing the 25+ that I've done so far
don't want to redo all of them :-)
tonyo it's not an issue with fez or zef, it's an issue with the consumers 19:04
the fez indexer clearly shows 0.0.7 for the version 19:05
Array::Sorted::Util is indexed for fez @ (it's item 277 in that blob) 19:06
JRaspass oh i think i know why RL is broken, is a key needed to read the json file now tonyo? 20:38
i get an xml error fetching
ignore me, i'm not reading my code correctly 20:39
carry on, but yeah i'm debugging your issue liz :-P
lizmat I've worked it out, your tarballs have a nested directory that others lack: bfda98b8b8eb1125c89ad125303871c2a7edcb65/Array-Sorted-Util-0.0.7/META6.json vs 49f69ac09a574065d3824151f8f8ba41cca9af0c/META6.json for example 20:56
I have logic to deal with this with cpan tarballs but not zef, time i made it generic then 20:58
tonyo mi6 packages the modules that way ^ 21:10
JRaspass ah gotcha, i best support it then :-P 21:15
i just used tar ¯\_(ツ)_/¯
I guess I should consider the route wherever the meta file resides, one of the cpan dists has two folders which really confuses my logic 21:16
not two nested, two adjacent 21:17
tonyo excellent 21:18
lizmat JRaspass++ so I guess that means I can continue moving my modules to the zef ecosystem 21:55
JRaspass sure, i should have a working patch to handle directories in tar tonight 21:57
ah this one is funky: "releases/eigenstates-0.0.2/META6.json" 22:08
lizmat ah?
that is a bad tar file
just ignore it
I should probably just remove it from zef
JRaspass oh, i was just going to handle any depth
lizmat no, please don't:: that releases dir is an artefact 22:09
I just keep all previous releases in there
it should not have been part of the tar file
JRaspass oh poo, i'm extracting the first meta6.json i see, this complicates things
ah yes, ill make it a depth search, ta for the heads up 22:10
Seq $metas = $(("releases/eigenstates-0.0.2/META6.json", "releases/eigenstates-0.0.4/META6.json", "releases/eigenstates-0.0.3/META6.json", "releases/eigenstates-0.0.1/META6.json", "META6.json").Seq)
22:39 TempIRCLogger joined
JRaspass lizmat all the dists! fixed via 22:50
lizmat JRaspass++ 22:52
will continue with migration tomorrow, only about 100 to go :-) 22:54
JRaspass only 100 lol 23:01
those should bring the zef %age right up on 23:05