Time Nick Message
23:43 mtompset There... attached.
23:37 kidclamp kk
23:36 mtompset You'll understand when you see my scaffolding and code.
23:35 mtompset So I purposefully try to make it undef, rather than try to make it work, but the test really should be the latter, not the former.
23:34 mtompset And this last case requires setting up something, somehow, but because it isn't, it is returning undef.
23:34 kidclamp if you did I would be worried about you
23:34 mtompset Because I don't really understand the relationships between all the acquisition tables. :)
23:34 kidclamp thnaks for the assist
23:33 kidclamp a few other things first, will tryto do tonight or in am
23:33 mtompset If I attach a kind of working, but broken test to your bug, can you fix the last case? :)
23:32 kidclamp hi mtompset
23:32 mtompset Greetings, kidclamp
22:18 barton bag, you need JesseM's lightning fast kick reflexes. the man's a 7th degree black belt in TaeKwonIRC.
22:09 cait good night all :)
22:08 talljoy hi cait!
22:02 cait :)
22:02 * cait waves to talljoy
22:02 cait hm?
22:00 Margaret so, who wants to complain to "SUPERNETS STAFF"?
21:59 talljoy o/
21:59 talljoy HI!!!!
21:59 * Margaret waves to talljoy
21:59 * mtompset waves to talljoy to make her feel better about the situation.
21:58 talljoy or indranil....or cait....or anyone
21:58 * talljoy was hoping pianohacker was waving to her
21:58 Margaret and on a Russian irc channel to boot.
21:58 bag yeah pretty sad message
21:58 rocio yikes...
21:58 andrew drive-by hatespeech
21:57 andrew wow, that’s a first for me
21:57 bag darn I’m too slow to get a chance to kick rjsalts
21:57 mtompset What in the world was that?!
21:57 rjsalts887 lukeG alexbuckley JesseM TGoat andreashm cait Scott-CSPL deb_CSPL roch kidclamp_away rocio barton jbeno andrew talljoy mtompset Margaret Gue lari phasefx_ bshum gmcharlt huginn logbot dbs bag rhamby rangi CallBackQueen wahanui jcamins_ sen jeff JesseM_away dj_shiz pastebot irma_ papa ibeardslee m
21:57 rjsalts887 ARE YOU TIRED OF LOW IQ MAXINE WATERS AND THOSE OTHER PORCH NIGGERS PISSING ON TRUMP?? EMERGENCY KKK MEETING @ iÑÑ.sцреÑиетs.оÑg Ñни #superbowl IN 5 MINUTES!! (THE MEETINGS HAVE MOVED FROM FREENODE BECAUSE OF COMPLAINTS DO NOT COMPLAIN TO SUPERNETS STAFF THANKS!!)
19:56 huginn cait: The operation succeeded.
19:56 cait @later tell oleonard you are awesome :)
19:56 cait oleonard++
19:51 andreashm khall++
19:50 * andreashm waves
19:42 LeeJ so would it become 1.5?
19:40 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=16330 new feature, P5 - low, ---, benjamin.rokseth, Pushed to Master , Add routes to add, update and delete patrons
19:39 LeeJ was going to document Bug 16330 and noticed it
19:39 LeeJ glad I noticed lol
19:39 LeeJ okay
19:39 Joubu nope, I guess it will change
19:35 LeeJ Joubu: what version is the REST API going to be in for 18.05? still 1.4?
19:33 LeeJ hi mtompset
19:33 mtompset Greetings, LeeJ cait
19:32 * mtompset puts on his chauffeur hat.
19:26 * LeeJ waves to cait
19:11 * cait waves
19:07 LeeJ okay thanks!
19:07 Joubu yes, I would add them anyway
19:06 LeeJ Joubu: for any entries with the keyword "release-notes-needed" should I still add it into docs now? or wait for specific release notes? :) last question of the day I promise. haha
18:59 LeeJ if that makes sense
18:59 LeeJ meaning that now that I think of it, it's probably just a matter of updating the text for the docs entry for each indicated enhancement
18:58 oleonard See you later #koha
18:52 Joubu include them where?
18:52 Joubu u
18:51 LeeJ Joubu: need to figure out a good way to include the enhancements...perhaps just modifying the docs entries? I *think* that's all there is to do..
18:50 LeeJ Joubu: thank you :)
18:42 Joubu LeeJ: It would be good to document some of the enhancements, but not all of them
18:42 Joubu LeeJ: No need to document bugfixes (unless it's flagged in bugzilla, but I do not think we have)
18:42 Joubu LeeJ: let start with new feature and then we will see if the doc team has more time :)
18:41 * Joubu wonders if what he wants to say is clearer or more confused...
18:40 Joubu *unless* the interface explicitly tells/shows it's not a OR
18:39 Joubu the only thing is: if "Use a file" if selected and the textarea for "use list of records" is filled, the content of the textarea must not been used
18:37 Joubu your suggestion is ok, I was just asking why not tabs instead (it makes it clear that the fieldset will change when the tab's header will be clicked)
18:35 Joubu oleonard: from my comments (which are not clear, as usual), I understand that the interface suggests to use a file OR a list OR a textarea, but actually will use all of them
18:34 Joubu oleonard: I do not remember how it behaves
18:34 LeeJ Joubu: when you get a sec (no rush) I'd like your opinion...last release's documentation didn't get around to documenting the enhancements or bugfixes...and I was wonding what would be your thoughts about how/where to go about including them? :)
18:33 oleonard But if that simplification is unwanted, that's fine.
18:31 oleonard I'm suggesting that enforcing the selection of only one choice would simplify the interface.
18:30 Joubu so I suggested to tell the user there was something wrong
18:29 Joubu it's bad because the interface does not tell it will use both
18:29 oleonard I thought your comment about the state of the bug implied that you thought it was bad.
18:29 Joubu currently it's "do that OR that OR that", I guess users read that as a XOR :)
18:28 Joubu oleonard: I'd say it depends on what the interface suggests
18:26 roch Currently working on a patch that has more than 8 attachements and it got pretty messy. I just squashed everything on my kohaclone and it's working and pretty clean, How should I name this new attachement If i plan to obsolete everything else? Should I include the old commits name and description?
18:25 oleonard If I paste numbers into the textarea AND select a file of biblionumbers to upload.... is that bad?
18:19 Joubu I do not see the difference with your suggestion
18:16 oleonard Joubu: You commented that it wasn't okay that input was submitted from all form fields if someone chose more than one
18:15 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12759 enhancement, P5 - low, ---, aleisha, Failed QA , Add ability to pass list contents to batch record modification/deletion tools
18:15 Joubu context bug 12759
18:15 Joubu oleonard: why do you want the textarea to be refreshed when switching tabs?
18:14 LeeJ hi #koha
18:14 * LeeJ waves
17:56 khall Joubu: ok, thanks!
17:54 Joubu yes
17:52 khall Joubu: was that in reference to my reply?
17:41 Joubu hum... I will try again later this week
17:38 khall in Koha, any number of holds on a given record can all be set to 'lowest priority'
17:37 khall that is correct behavior
17:37 khall Joubu: you have both those holds set to "lowest priorty" which is why the both have that icon
17:22 Joubu I was referring to the "bottom_lowest" (or whatever it's called) icon you can see on priority 8 and 9 on this screenshot: https://screenshots.firefox.com/TX15YNJl7t9SF3LD/pro.kohadev.org
17:21 Joubu my first point: "1. Last lines of the 2 blocks have the "lowest priority" icon, I would expect only the first one (with priority=9)"
17:21 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19469 enhancement, P5 - low, ---, kyle, Passed QA , Add ability to split view of holds view on record by pickup library and/or itemtype
17:21 Joubu khall: I am not sure you understand what I meant on bug 19469
17:16 Joubu ashimema: you should stop creating new bug report from existing ones
16:35 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20505 minor, P5 - low, ---, mtompset, Needs Signoff , Selenium tests fail if server not running
16:35 mtompset I'm getting around to your bug... care to look at mine? bug 20505? :)
16:35 kidclamp_ ish mtompset
16:34 mtompset kidclamp_, you there?
16:30 mtompset Well, medium easy... have to know how to install missing requirements manually. :)
16:30 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20505 minor, P5 - low, ---, mtompset, Needs Signoff , Selenium tests fail if server not running
16:30 mtompset Anyone care for an easy sign off? bug 20505
16:01 Shane-S okay well 3.0X or something and stopping at 3.20
16:01 Shane-S considering I only did it maybe a handful of time running 3.20 :) for 4 years
16:01 Shane-S the file does use /tmp as default, maybe I should try it...because I don't reboot this server often
15:59 mtompset Perhaps /run/koha/{instance}/template_cache could be another reasonable location.
15:58 mtompset Though, I vaguely recall some /tmp permissions issue thing with /tmp recently.
15:58 mtompset I think that's because when you reboot, the /tmp might go bye, bye. :)
15:57 mtompset Well rewrite-config.PL defaults it to /tmp/koha
15:57 mtompset No problem...
15:56 Shane-S mtompset: sorry I assumed having plack on the modules page meant it was on...I should have ask you last time you suggested that
15:56 mtompset Mmm....
15:55 Shane-S any suggestion on the what dir is preferred for the cache?
15:54 Shane-S I could count to 3 sometimes before it loaded
15:54 mtompset Because with more memory, you are less likely to swap out to disk which is painfully slow by comparison.
15:54 Shane-S well the page response didn't even take a second
15:54 mtompset Remember. *grunt grunt grunt* More memory! *grunt grunt grunt*
15:53 mtompset Well, no it's significantly faster, but not 100x. Your increase in CPU and RAM also brought a significant change.
15:53 Shane-S now to check the resource usage
15:52 Shane-S Plack is like 100x faster
15:52 * wahanui blesses crap
15:52 Shane-S holy crap....
15:52 mtompset You've upgraded from an older version, right?
15:52 mtompset I say go for it.
15:51 Shane-S should I also specify that cache dir?
15:51 Shane-S I will check it out and see if it helped
15:51 Shane-S just reports "koha" which I just enabled and is muy default install
15:50 mtompset You should notice a significant speed boost once plack is enabled.
15:49 mtompset That will tell you the instances that are running plack.
15:49 mtompset sudo koha-list --plack
15:48 mtompset I was thinking more of a list placks.
15:48 mtompset I believe that is the enable plack, yes.
15:48 Shane-S mtompset: sudo koha-plack --enable library ?
15:47 Shane-S mtompset: would this help too? Warning You are missing the <template_cache_dir> entry in your koha-conf.xml file. That will bring a performance boost to enable it.
15:46 mtompset There's a debian command for packages... let's see...
15:46 mtompset That doesn't mean you are running Koha in plack.
15:46 Shane-S yes 17.11 sorry
15:46 Shane-S Plack::Middleware::ReverseProxy 0.15
15:46 mtompset I'm also assuming you typo'd and meant 17.11 :)
15:45 mtompset Because non-plack versions will run slower.
15:45 Shane-S let me check, but I think so yes
15:45 mtompset Shane-S, are you running plack?
15:43 Shane-S I even gave the new 7.11 VM 3 CPUs, 6Gb ram, and 60Gb space, old 3.20.02 VM was 1 CPU, 2Gb, and 30Gb space
15:42 Shane-S anyone have a suggestion on what I can check for why 7.11 seems to take longer to process then 3.20?
15:29 kidclamp there'd be a lot of trouble if we didn't excuse typos :-)
15:29 roch thing* excuse the typo
15:29 roch Kinda new to Perl, so I still had the kubernetes's definition of pods in mind. Definitely not the same think lol
15:27 roch Alright thanks
15:26 kidclamp https://perldoc.perl.org/perlpod.html
15:26 kidclamp documentation of the functions roch
15:26 roch In the QA test tool; What does "pods coverage" mean?
15:10 Shane-S My only guess is this one has memcache...3.20 did not, is there anyway to speed up the check-in or allow for a windows to "batch" scan books. I tried it with 4 barcodes, and pausing a second between then, I still would scan while it was "waiting" for the server to process
15:10 reiveune bye
15:09 Shane-S so...I have been trying to determine why circulation is slower on 7.11 from 3.20. It appears I am just getting a longer "waiting for response from <IP here>" in the browser status bar. All that is different is the server..same client laptop and browser (chrome), same network, same VM Host, Ip address is just 1 digit higher (so I could load the old one).
14:48 oleonard http://zivotdesign.com/p/#1/15239764835671
14:40 oleonard Would it make sense to completely reset the form if someone selects a different input method (file/list/numbers) ?
14:40 oleonard (and anyone else's)
14:39 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12759 enhancement, P5 - low, ---, aleisha, Failed QA , Add ability to pass list contents to batch record modification/deletion tools
14:39 oleonard Joubu: I'm looking to see if I can revive Bug 12759 and wanted to get your opinion on something
14:37 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19469 enhancement, P5 - low, ---, kyle, Passed QA , Add ability to split view of holds view on record by pickup library and/or itemtype
14:37 khall andreashm: Bug 19469 now has a followup to address Jonathan's points!
13:51 andreashm (slow reply, forgot to set away)
13:51 andreashm khall: np
13:50 Freddy_Enrique Good morning everyone
13:48 mtompset Hence, my work laptop is single boot Ubuntu 16.04 LTS.
13:47 mtompset I, however, no longer support Windows folks. :)
13:47 mtompset You have to use the OS of the people you support.
13:47 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20594 trivial, P5 - low, ---, mtompset, Needs Signoff , Revert Perl version to 5.10
13:47 mtompset Still, I submitted a Perl version reversion patch bug 20594
13:46 * tcohen recalls him using Windows XP too
13:46 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20549 enhancement, P5 - low, ---, koha-bugs, NEW , Avoid indexer race condition when moving item
13:46 mtompset Still, I submitted a Perl version reversion patch bug 20549
13:45 mtompset I vaguely recall him using Arch Linux as well.
13:44 mtompset He's using a lot of non-Debian's actually.
13:44 tcohen mtompset: he's using OpenSuSE
13:42 mtompset I figured, but sometimes one keeps strange hours.
13:42 mtompset Greetings, tcohen.
13:40 huginn calire: oleonard was last seen in #koha 37 seconds ago: <oleonard> @seen tcohen
13:40 calire @seen oleonard
13:39 huginn oleonard: tcohen was last seen in #koha 17 seconds ago: <tcohen> @seen mtompset
13:39 oleonard @seen tcohen
13:39 oleonard Middle of the night for dcook, mtompset
13:39 huginn tcohen: mtompset was last seen in #koha 3 minutes and 38 seconds ago: <mtompset> @seen dcook
13:39 tcohen @seen mtompset
13:35 huginn mtompset: dcook was last seen in #koha 9 hours, 51 minutes, and 44 seconds ago: <dcook> Oh also if someone wants a super easy pretty trivial sign off: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20592
13:35 mtompset @seen dcook
13:35 mtompset Greetings, #koha.
13:07 wahanui okay, Joubu.
13:07 Joubu no wahanui, feature freeze is April 20th (soft) and April 27th (hard)
13:06 wahanui well, feature freeze is Oct 27th (soft) and Nov 3 (hard)
13:06 tcohen feature freeze?
13:05 clrh oups sorry
13:05 clrh §3
13:00 khall andreashm: thanks! I had not
12:30 * andreashm waves
12:29 calire hi nlegrand
12:27 nlegrand o/ severine_q
12:27 severine_q hi everyone !
12:26 oleonard \o
12:26 nlegrand o/ calire, oleonard
11:49 AndrewIsh OK, thanks cait
11:47 cait AndrewIsh: you don't need to provide translations, just copy the english sql into the translated sql files, translators will take over
11:38 wahanui hi olé onard
11:38 calire hi oleonard
11:37 oleonard Hi #koha
10:20 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19469 enhancement, P5 - low, ---, kyle, Failed QA , Add ability to split view of holds view on record by pickup library and/or itemtype
10:20 andreashm khall: I just wanted to alert you to Joubu's comments on bug 19469, in case you haven't already seem them?
10:14 khall mornin! not yet
10:13 andreashm khall: did you see my later?
10:10 andreashm hi khall
10:02 nlegrand andreashm: nice !
10:01 andreashm nlegrand:
10:01 andreashm Unless you have already done so, I would recommend contacting Gothenburg Univ. Library. They went live, using ES, a few weeks ago.
10:00 andreashm haha
10:00 nlegrand We had the most critical colleague testing ES. They screamed screams of joy. It was an unusual sight.
09:58 AndrewIsh I've checked the wiki and not found anything
09:57 nlegrand Nice!
09:57 gaetan_B we should answer this afternoon, once i'm done and we've tested a few things
09:57 AndrewIsh Hey cait :) Following on from the help you gave me yesterday. Is there any documentation on the steps required for me to provide translations for the new Authorised Values Category I've created?
09:57 gaetan_B actually i'm going through all this to answer Séverine's email nlegrand
09:57 andreashm pfff
09:57 andreashm =)
09:56 gaetan_B i know! can't they use unimarc like everyone?
09:56 nlegrand gaetan_B: ha! Those MARC21 people.
09:34 nlegrand o/ cait gaetan_B
09:34 nlegrand andreashm: o/ andreashm
09:26 gaetan_B also there is a specific group of marc21 indexes for personal-name authorities but not for other types, what is the purpose of these?
09:20 gaetan_B looking at the ES conf files, can someone explain what the match index is for in authorities ?
09:04 cait totally agreed
09:04 andreashm cait: yeah, it would need much work. but would love to see that feature.
08:57 cait but yep, I'd like that one a lot
08:56 cait it's all in C4 module sstill
08:56 cait andreashm: the write protect needs major work before it could go in i think
08:48 andreashm hey nlegrand
08:45 nlegrand Salut Koha
08:44 huginn Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14957 enhancement, P5 - low, ---, eivin, In Discussion , Write protecting MARC fields based on source of import
08:44 andreashm together with bug 14957
08:43 andreashm that would be a nice feature.
08:42 andreashm cait: great!
08:41 cait :)
08:41 cait I sent him the link yesterday, remembered it from when I was in Sweden
08:41 andreashm Ah. He had already commented on that one....
08:40 huginn Bug 14367: enhancement, P5 - low, ---, xarragon, ASSIGNED , History for MARC records. Roll back changes on a timeline or per field.
08:40 huginn andreashm: The operation succeeded.
08:40 andreashm @later tell rangi I think this is the bug you were looking for regarding saving record history: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14367
08:34 * andreashm waves
08:10 cait no issue
08:08 jajm cait, sorry about that, i should have `git fetch`ed before the rebase :/
08:07 cait thx jajm
07:18 wahanui privet, fridolin
07:18 fridolin hi there
07:06 gaetan_B hello
06:36 cait bbiab :)
06:33 huginn cait: The operation succeeded.
06:33 cait @later tell tcohen can you take another look at bug 20400 please?
06:31 reiveune hello
06:29 alex_a bonjour
06:11 huginn cait: Konstanz, Germany :: Clear :: 48F/9C | Wind Chill: 48F/9C | Tuesday: Sunny. High 69F. Winds NE at 5 to 10 mph. Tuesday Night: A clear sky. Low 48F. Winds NE at 5 to 10 mph.
06:11 cait @wunder Konstanz
06:09 cait trying to fix my own patches for a little bit this morning :)
06:08 josef_moravec cait++
06:08 cait josef_moravec++
03:55 rangi ahh gotcha
03:50 mtj rangi: i think RHSC is more like deb backports, rather than the universe repo
03:43 huginn Bug 20592: minor, P5 - low, ---, koha-bugs, Needs Signoff , updateitem.pl causes database errors when empty non-public item notes updated
03:43 dcook Oh also if someone wants a super easy pretty trivial sign off: https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20592
03:43 dcook Ok. Food time. No starving.
03:43 dcook (And at that point I complained because I was getting all these warnings about deprecated Perl features being used in Koha >_<)
03:42 dcook (Also funny is that I know I've used Perl 5.20 on a different version of Suse before...)
03:42 dcook Interesting that Eric was asking for 5.20 in July 2017.
03:41 mtj "Red Hat Software Collections were introduced to provide capabilities and development libraries at a different cadence to the core applications as there was a demand for this (primarily from developers it seems), especially in later stages of major releases. The primary difference with RHSC versions is that they have shorter support lifecycles"
03:41 mtj "Red Hat will update major versions and major releases of RHEL. In rare cases they will do a 'rebase' at a minor RHEL release, but this is less likely for core components. This is a feature not a bug of RHEL as it provides a standard platform for enterprises to develop applications against for an extended period of time (one of the primary reasons enterprises pay for Red Hat support)"
03:41 dcook "The default versions of Perl or PostgreSQL, for example, remain those provided by the base Red Hat Enterprise Linux system."
03:40 dcook "Red Hat Software Collections does not replace the default system tools provided with Red Hat Enterprise Linux 6 or Red Hat Enterprise Linux 7. Instead, a parallel set of tools is installed in the /opt/ directory and can be optionally enabled per application by the user using the supplied scl utility."
03:40 dcook I mean for Plack... that's all right. Doesn't work for CGI with Apache though
03:40 dcook I think they usually install to /opt/rh/<stuff> I think
03:39 mtj https://access.redhat.com/discussions/3108531
03:39 dcook I looked at bit once at using Red Hat Software Collections but they can be a bit of a pain
03:39 rangi mtj: i wonder if thats like ubuntus universe
03:39 mtj https://access.redhat.com/documentation/en-us/red_hat_software_collections/2/html-single/2.4_release_notes/index
03:39 dcook mtj: Yeah, it's interesting
03:39 rangi hehe
03:39 dcook But I mourn that half-eaten burrito from the other night heh
03:39 mtj hmm, it seems redhat has some sorta 'Red Hat Software Collections' repo, that has newer versions - like perl 5.20/5.24
03:39 dcook But sometimes it's just nah
03:39 rangi heh yep
03:39 dcook Which seems crazy... as you know I can put away the food...
03:39 dcook Mmm burrito. I had part of one the other night but was too excited to finish it.
03:38 rangi :) get a burrito thats the best food package
03:38 dcook But speaking of packaging, I really should go find some food packages
03:37 rangi for the most part, yep
03:36 dcook In the case of Koha, it should be pretty much the same
03:36 dcook Yeah I figure so long as both debian and the specfile put things in places the OS would expect them.. it's all good
03:36 rangi but it puts things in places that the OS would expect them
03:36 dcook https://github.com/indexdata/idzebra/tree/master/debian
03:36 dcook https://github.com/indexdata/idzebra/blob/master/idzebra.spec
03:36 dcook https://github.com/indexdata/idzebra
03:36 dcook Here we go...
03:35 dcook I guess the specfile wraps everything in one file whereas Debian uses a number of files
03:35 dcook I know I've looked into it before but have only used build-git-snapshot and made one mini Debian package by hand
03:35 rangi close
03:35 dcook dh_auto_make or something like that?
03:35 dcook What's the debian version..
03:35 dcook That's the thing with the specfile. It does a source install and then you just modify around that really
03:34 dcook I figure collecting a variety of examples doesn't hurt though
03:34 rangi basically you have a base, then you build from there
03:34 dcook Fair enough :p
03:34 rangi im not sure i'd want to copy their practices ;)
03:34 dcook Yeah must be specfile in the repo and the debian outside it
03:34 dcook Hmm https://github.com/indexdata/yaz4j
03:34 dcook There have to be people who do both. Like Indexdata have..
03:33 rangi yep
03:33 dcook I'm guessing that Debian is doing the Debian packaging?
03:33 rangi yep
03:33 dcook Say in this case: https://packages.debian.org/jessie/nagios3
03:33 dcook Maybe they do have debian/ separate
03:33 dcook But I haven't seen a repo yet that has both
03:33 dcook I often see spec files in places like https://github.com/NagiosEnterprises/nagioscore
03:32 dcook I agree with what you're saying about having the build script do the hard work
03:32 rangi most other projects do all their packaging outside the main repo
03:32 dcook MMm yeah
03:31 dcook I'd want to look at other projects to see if it's bonkers or not
03:31 rangi you're building on that
03:31 rangi debian is the base
03:31 rangi not really
03:31 rangi i defintely wouldn't have them in 2 places, but just make the build script put them in the place the need to be
03:31 dcook A RPM specfile reaching into a debian directory
03:31 dcook rangi: Yeah I thought about that, but it seems a bit hacky
03:31 dcook I think Marcel uses Fedora as well iirc
03:30 rangi the script can copy those to the places the rpm expects them
03:30 dcook Or maybe it would be impetus to unite things
03:30 dcook I just worry it would further fragment things
03:30 dcook I just worry about having Apache config in debian/, in etc/, in...
03:30 rangi yep
03:30 dcook I mean we'd need to start somewhere right?
03:30 rangi then you could make gitlab-ci file to have it build on push too, that way you'd catch things that break it early
03:30 dcook While everyone else benefits from the same specfile
03:29 dcook So people could take advantage of Fedora being current
03:29 dcook There's no reason Fedora/Redhat/Centos and Suse can't share a specfile as well
03:29 dcook Me and Nicholas talked about it a few years ago
03:29 dcook Yeah I used build-git-snapshot to inspire my own build scripts here
03:27 rangi if you can develop a script like build-git-snapshot but for rpm not deb, that would make it easier to automate testing it
03:24 rangi basically i'd never be super interested because I think those distros move too slowly to be useful. But others may well, if you say there are rpm packages if you want to work with them
03:23 dcook Although maybe it's one of those... build it and they will come things
03:23 dcook And most Koha developers would say they just know how to use it with Debian
03:23 dcook I suppose although if we provided that as a community I think there would be an expectation for supporting RPMs too
03:22 rangi and to package every release in a timely manner etc
03:22 dcook And it would then require coordinating developers to make sure things work for DEB and RPM :/
03:22 rangi it needs someone to set up and maintain a repository, that's the bit that makes the packages useful, not the packaging itself
03:22 dcook Maybe if it could match the Debian layout
03:21 dcook I don't know how usable it would be
03:21 dcook I would be happy to write a RPM spec file for Koha, but...
03:20 dcook Although I don't actually see a spec file anywhere in https://github.com/vanoudt?tab=repositories
03:20 dcook Nicholas van Oudtshoorn*
03:19 dcook A git repo it seems
03:19 dcook I think he has koha itself somewhere
03:19 dcook Yeah
03:19 rangi yeah for the bits around koha, not koha itself (its like the koha-deps package for debian)
03:18 rangi too
03:18 rangi CVE-2018-6797
03:18 dcook So Nicholas van Oudt does provide a RPM repo for Fedora: https://copr.fedorainfracloud.org/coprs/vanoudt/KOHAExtra/
03:16 dcook Curious why an attacker would be able to provide a template to pack() but.. still
03:15 dcook But yeah I don't see anything from Suse about it
03:14 dcook Yeah looking at that one now
03:14 rangi https://rt.perl.org/Public/Bug/Display.html?id=131844
03:13 rangi im not sure why you wouldn't just patch the bug anyway, even if you think you can't exploit in in 64bit versions
03:13 rangi yeah, so does freebsd
03:13 dcook Hmm
03:13 dcook But then Debian say vulernable and fixed... https://security-tracker.debian.org/tracker/CVE-2018-6913
03:11 dcook https://access.redhat.com/security/cve/cve-2018-6913
03:11 dcook "The 64-bit versions of perl have not been found to be affected."
03:10 dcook Reading about it on the Red Hat site now
03:09 rangi (its a heap overflow one, quite nasty)
03:09 rangi its patched on all debian and ubuntu versions
03:08 dcook The idea that they're more secure just boggles my brain
03:08 rangi CVE-2018-6913
03:08 dcook I've worked on RHEL systems that have unpatched kernel vulnerabilities...
03:08 dcook That's the thing I don't get about governments and SLES and RHEL
03:08 dcook That's cool
03:08 dcook Ahh I must be thinking of Jessie
03:08 rangi they patch all the way back to oldstable, there was a perl patch yesterday for jessie
03:07 rangi stable is 5.24
03:07 rangi the debian perl has security patches
03:07 dcook I mean with Plack... we could actually use our own Perl
03:07 dcook By that token, we shouldn't go with the Debian Perl either
03:07 rangi because that actually is a disservice not a service
03:06 rangi and making koha work with insecure eol software should be no ones priority
03:06 dcook Agreed
03:06 dcook Yeah I remember hearing that a few years ago
03:06 rangi they should really stop forcing people to use it
03:06 rangi 5.18 was eol years ago
03:06 rangi yes
03:06 dcook I think 5.20 is EOL, yeah?
03:06 rangi and 5.20 was released in may 2015
03:06 dcook I think it would just create division
03:06 rangi also 5.18 in opensuse has an unpatched security flaw
03:05 dcook I could share it but it wouldn't help the community
03:05 rangi thats worse than not doing it
03:05 dcook But we don't share it
03:05 dcook Me, Nicholas, probably others
03:05 dcook rangi: A few people have
03:05 dcook I'd be more willing to work on creating a RPM packaging model
03:05 rangi so easy that absolutely no one has done it
03:05 dcook As per the email I just sent, if the DEB packages allowed for multiple versions of Koha on the same server..
03:04 dcook Building RPMs also seems a million times easier than DEB packages >_>
03:04 dcook And a number of governments and such only use SLES or RHEL
03:03 dcook I have tried to convince the higher ups to move to Debian in the past, but haven't gotten there yet
03:03 dcook I think we mostly have Koha on openSUSE, SLES, and RHEL
03:03 dcook tcohen: I don't call the shots here :p
03:03 dcook Mmm yeah Fedora is often pretty current
03:02 mtj aah, thanks for updating the bug dcook :)
02:49 tcohen OpenSuSE? Really? :-P
02:48 mtj i didnt realise there was such a big gap btw RHEL and fedora's perl versions
02:45 mtj 5.24.4
02:44 mtj https://koji.fedoraproject.org/koji/buildinfo?buildID=107121 4
02:44 mtj hmm, fedora26 seems to have perl 5.24.1 ?
02:39 * dcook wonders why nothing in life can be perfect :p
02:39 dcook But then you're missing out on all that upstream OS packager maintenance..
02:39 dcook It would be nice to encapsulate an app so that the versioning is less of an issue
02:38 dcook I love packages like RPMs and DEBs but...
02:38 dcook Which makes me understand why people like things like perlenv or whatever
02:38 dcook Either way, yeah Red Hat is a Perl nightmare
02:38 dcook Maybe it was 5.16
02:38 dcook Or maybe not..
02:37 dcook RHEL 7 was using 5.14 last I looked last year
02:37 dcook I'm surprised it's that new
02:37 mtj http://mirror.centos.org/centos/7/os/x86_64/Packages/
02:37 mtj i was surprised to see that centos7 is still using perl.5.16.3
02:37 dcook All of this was actually just so I can check to see if a little bug on a cataloguing page has been fixed or if I need to write a patch.. heh
02:36 dcook Cheers mtj :)
02:34 mtj https://perldoc.perl.org/5.8.8/functions/caller.html
02:34 * dcook thumbs up
02:34 dcook https://perldoc.perl.org/5.8.9/functions/caller.html
02:34 mtj https://perldoc.perl.org/5.18.2/functions/caller.html
02:34 mtj .. yet the sub exists
02:34 dcook Yikes perldoc
02:34 dcook Yeah I just tried that
02:33 mtj https://perldoc.perl.org/5.18.2/search.html?q=caller
02:33 mtj searching gives no results
02:33 dcook I wonder why they would've thought to even check for that one though..
02:33 dcook Feel free to add that to the email thread :D
02:33 dcook Mmm I gotcha
02:32 mtj im guessing that might have caused people to think that it didnt exist for < 5.20.x
02:32 wahanui somebody said more was only a nother display for what can be accessed from global
02:32 dcook Can you tell me more?
02:32 dcook Oh?
02:31 mtj .. re: your caller() email
02:31 mtj hiya dcook, there seems to be some problem with the search results, on the perldoc site
02:30 dcook Just needed to update that Apache configuration...
02:30 dcook And now my Koha is back :D
02:03 wahanui interesting is sometimes good and sometimes bad
02:03 dcook Interesting...
02:02 huginn Bug 12904: enhancement, P5 - low, ---, kyle, RESOLVED FIXED, Force browser to load new JavaScript and CSS files after upgrade
02:02 dcook https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12904
02:01 dcook When did we start doing KOHA_VERSION anyway...
02:01 huginn Bug 20538: enhancement, P5 - low, ---, julian.maurice, Pushed to Master , Remove the need of writing [% KOHA_VERSION %] everywhere
02:01 dcook Right... https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20538
01:50 * dcook assumes he might just be missing some modern process perhaps
01:49 dcook So is master broken for everyone or just me?
01:49 * dcook waves