IRC Logs for #crux-devel Friday, 2017-05-05

*** onodera has quit IRC01:08
*** mavrick61 has quit IRC02:37
*** mavrick61 has joined #crux-devel02:38
*** Romster has quit IRC05:12
*** Romster has joined #crux-devel05:15
*** deus_ex has joined #crux-devel08:05
*** Workster has quit IRC10:49
*** noqugputmwpwodhw has joined #crux-devel10:49
*** Workster has joined #crux-devel10:49
*** onodera has joined #crux-devel11:59
juefrinnst: do you have the time to fix dbus or should I do it?12:53
juejust_fun: I need you help, I do not understand the logic of your prtverify patch ;)
jpb3text/plain; charset=utf-8, 644 bytes ( linked by jue12:55
*** deus_ex has joined #crux-devel12:57
*** jpb3 has quit IRC12:57
*** jpb3 has joined #crux-devel12:57
*** _mavrick61 has joined #crux-devel12:58
just_funI've dropped .md5sum from my repo, which I understand will be done in the official repos too, and prtverify will complain12:59
juesure, but the present of .signature will never be checked, right?13:00
just_funthe page drop .md5sum from the filelist "to be checked" if .signature is present13:00
just_funhow is it checked13:01
just_funfor presence only?13:01
just_funinitialy, I've added .signature to that list13:01
just_funif .signature is not present, it should let .md5sum in the list (to be checked)13:02
*** mavrick61 has quit IRC13:02
just_funif .signature is present and .md5sum is not, don't run filters for .md5sum (the .signature is checked in this condition).13:05
just_funisn't enough?13:06
juesounds good13:07
just_funI wasn't sure if .md5sum was tested more in the awk filters13:07
just_funlike: non-zero file size13:07
jueno, only tested if present13:08
just_funin that case, .signature will have to be checked too13:08
just_funI know this patch is ugly: because, every time I've looked it wasn't obvious.13:11
just_funIt will be clear if both checksum files were added and check one or the other, but was to complicated for me.13:12
juewell, what do we expect in the midterm? That every repo uses signatures? If so we should simply ignore .md5sum13:16
just_funThis is what I was thinking: a temporary patch13:17
frinnstjue: sorry if you have time and energy feel free13:17
juefrinnst: ok, will do13:17
frinnsti've been super busy at work and have had close to 0 energy for crux13:18
frinnstbut things are looking brighter again so i'll be more active again :)13:18
frinnsttaking monday off from work so extra long weekend that I intend to spend by mostly sleeping13:19
juefrinnst: np, if I can do something, just drop me a note ;)13:19
frinnstmy inbox is looking scary - ~90 unread emails13:19
jueguess a new prt-utils release is coming due, anything to fix or add?14:33
just_funWatch: patch?14:34
juewell, at first we should discuss if we want that ;)14:41
juestrictly I'm not against to include something like that in the port. btw, it's an old wish of Romster15:01
juewithout any tests I see two problems with your implementation: performance and the lack of the 'macro' feature we have with ck4up15:03
juethe first can be solved with a standalone program, not using prtverify, the second might be not very important15:04
just_funthanks for your feedback15:05
just_funa standalone program is ck4fun, in the same location15:05
just_funwhich uses multiple "threads", initial in perl, but after awk implementation from prtverify I reuse that15:06
jueI'll play the next days a bit with the prtverify module and report how it goes15:11
just_funabout the macro stuff: I thought is important not only to match what changes, but to match against the version.15:11
just_funThanks for thinking about testing.15:12
just_funI've also thinking about multiple Watch expression, if needed:15:12
just_funNow is Watch: url regex (checked against version)15:13
just_funI was thinking about "Watch: url regex varname"15:13
just_fundefault varname (if missing) is version15:13
just_funAnyway, thanks, and I'm waiting for your feedback.15:14
just_funbtw, my whole repo has these Watch expressions15:15
juewelcome and thanks for your various patches and sharing your ideas15:15
just_funI have more ideas, but I don't think you are ready :))15:16
juemight be :) it was always a good thing for CRUX that we are moving very slowly and deliberately15:25
just_funI agree.15:27
juePer said sometimes in the past: my most important task is to say 'no' in the first run15:28
*** chinarulezzz has quit IRC16:36
*** chinarulezzz has joined #crux-devel16:48
Romsterfeature creep for a small distro would be bad, if it made it complex and required more maintenance.17:04
just_funMe too, but what are the things that can make crux easier to maintain?17:59
*** onodera has quit IRC21:12
*** onodera has joined #crux-devel21:12
*** chinarulezzz has quit IRC23:36

Generated by 2.14.0 by Marius Gedminas - find it at!