Time Nick Message 00:01 dcook Damn, this is addictive... 00:24 mtompset Greetings, #koha. 00:30 mtompset Okay... the OPAC XSLT stuff grabs the MARCXML field, parses and formats it, and then pumps it out to the opac-detail page. 00:31 mtompset If I want to HIDE fields, is there a nice way to do it without converting it to a marc record so I can fliter it against the marc_subfield_structure table for the biblio's frameworkcode? 00:32 eythian probably not 00:33 mtompset Okay, yuck. :) 00:33 mtompset Let the ugliness begin. 00:58 mtompset Progress, but not quite. 00:58 eythian mtompset: what are you doing? because it sounds like something we will need soon also. 00:59 mtompset The opac-MARCdetail.pl properly hides things based on the hidden value in marc_subfield_structure 00:59 mtompset opac-detail.pl does not. 01:00 eythian ah 01:00 eythian yeah 01:00 mtompset I'm fixing it so opac-detail.pl will match opac-MARCdetail.pl 01:00 eythian if you make a nice generic function to filter marc, I'll be able to use that too. 01:00 mtompset So if you do crazy things like hide 100$a, it will. 01:00 eythian I'll be needing something so that logged in users have different rules to anon users. 01:00 mtompset different?! 01:01 mtompset That actually sounds like something WE could use. ;) 01:01 wizzyrea If you're gonna do it, should DO IT. 01:01 eythian basically, they want a public catalogue, but they want notes to be hidden unless you're logged in, as they are potentially a little politically inflammatory. 01:02 mtompset In our case, we want to hide authors which would link the book to an organization that some countries would be unhappy about. :) 01:03 mtompset or publishers. not everything, just not liked parts. 01:03 eythian ah yep 01:03 mtompset So, the hiding functionality needs to work properly across views. 01:03 eythian yeah, same here 01:04 mtompset If you give me a week, I'll have it all nicely up on bugzilla for you, eythian, so you can extend it as you want. 01:04 eythian that'd be cool 01:10 mtompset just my luck, the template toolkit website is timing out on me. 01:16 eythian doh 01:16 eythian might have docs in /usr/share/docs 01:16 eythian (I'm not sure) 01:19 dcook Yeah, the TT manual is on the system. 01:19 dcook Or at least some are... 01:19 * dcook was looking at them the other day 01:20 dcook -> /usr/lib/perl5/.../Template/Manual 01:21 dcook mtompset: ^ 01:21 dcook At least, that's on my OpenSUSE 01:21 mtompset for future reference. 01:28 eythian so routing list notes are terribly broken. 01:29 eythian it looks like someone assumed they worked one way, when they in fact work a different way, and now they don't work at all. 01:45 mtompset oh joy! 01:45 mtompset does anyone know how to DELETE a tag from a MARC::Record? 01:47 dcook MARC::Field documentation would tell you that, I believe 01:48 wizzyrea @quote add A sad story: "it looks like someone assumed they worked one way, when they in fact work a different way, and now they don't work at all." 01:48 huginn wizzyrea: The operation succeeded. Quote #291 added. 01:49 wizzyrea an all too common story :(( 01:50 mtompset That triggered this musical flashback 01:50 mtompset http://youtu.be/lenFPDV76uM 02:02 eythian I'm not convinced the routing list notes have ever worked 02:02 eythian they seem to have been introduced in '06, and the queries look broken then. 02:04 wizzyrea ^ this is what I thought when I was looking at it, but didn't feel qualified to make any pronouncements :/ 02:07 eythian Oh, I just make pronouncements and if I'm wrong, it's safe to assume that everyone has forgotten about them before long anyway. 02:07 dcook I think I've looked at the routing list notes, but only barely 02:19 * mtompset grins. 02:19 mtompset Question... if I decide to hide 260$a,b,c but not the other subfields of 260. 02:20 mtompset Should the label "Publisher:" show if nothing else is printed? 02:20 * wizzyrea strives to be more like eythian 02:20 mtompset -- this seems to be a pain to do in the XSLT stuffs. 02:20 eythian mtompset: if there are no other subfields, there should be no 260 in the resulting record. 02:21 eythian and so it should behave exactly as though there was no 260 to start with. 02:22 mtompset Hmm... the XSLT is still putting out Publisher, even if the 260 field is all hidden, which I suspect is not there. 02:22 mtompset (I delete fields when there are no subfields) 02:24 eythian mtompset: in that case, just leave it alone. It's not your problem, at least as part of this. 02:24 mtompset So, what is "<xsl:if test="marc:datafield[@tag=260]">" actually looking at? 02:24 eythian that ought to catch it. 02:24 mtompset It should. 02:25 mtompset OOOOO! 02:26 mtompset I have an idea... the delete_subfield fails on the last subfield. 02:26 eythian ahh 02:26 mtompset now to test that idea. 02:26 eythian I don't like that behaviour myself. 02:27 eythian I'd rather it either deleted it, and the field too, or issued an exception so you know you have to handle it. 02:27 mtompset I haven't tested the idea, but that's the only thing I can think of right now. 02:28 mtompset bingo! 02:29 mtompset Oh this is awesome. I'm really close now. 02:29 eythian awesome :D 02:29 mtompset Basically, opac/opac-detail.pl grabs GetMarcBiblio. 02:29 mtompset This is a Marc::Record 02:30 mtompset I think delete fields/subfields based on hidden values from marc_subfield_structure 02:30 mtompset ^think^then^ 02:31 eythian sounds like a good way to do it. 02:31 mtompset I then wrote a function which attempts to takes a partial kohafield and relate it back to the MARC fields/subfields. 02:32 mtompset And create a hash of whether it is hidden or not. 02:32 mtompset I did have to tweak opac-MARCdetail.pl, because it wasn't hiding title properly. 02:33 mtompset And had to tweak C4/XSLT.pm to delete out fields which have no subfields. 02:33 mtompset but it doesn't actually check hidden at all. 02:33 mtompset it's just a fix. 02:34 mtompset oh, and GetMarcSubjects... [ { '', undef } ] is useless to push. 02:34 mtompset so I added a check, so the Subjects hides correctly. :) 02:35 mtompset I suppose I should hide/unhide subtitle correctly in the template. 02:35 mtompset But it looks weird to hide the title and NOT the subtitle. 02:36 eythian mtompset: don't make semantic decisions 02:36 mtompset exactly. 02:36 eythian if people hide title and not subtitle, then that's their own problem. 02:36 mtompset exactly. 02:36 eythian and maybe they have a reason for it anyway 02:36 wizzyrea @quote get 123 02:36 wahanui http://xkcd.com/1172/ 02:36 huginn wizzyrea: Quote #123: "rangi: #thingsihavelearnt if there is a mad scheme a library somewhere will be doing it ... except madder" (added by wizzyrea at 09:20 PM, March 30, 2011) 02:36 mtompset I can't think of a reason to hide the title and not the subtitle. 02:36 wizzyrea ^ 02:36 eythian mtompset: I can. 02:36 eythian BRAIN WORMS! 02:36 mtompset Really?! 02:37 mtompset That doesn't count as a GOOD reason. :P 02:37 eythian you didn't ask for a good one 02:37 wizzyrea Things I've learned: it doesn't have to be a good reason, it just has to be a reason. 02:37 wizzyrea good reasons are better, of course :P 02:46 mtompset but brain worms?! That's not even realistic. :P 02:54 eythian I beg to differ. 03:02 mtompset URL to a scientific study. :P 03:06 * wizzyrea recalls the study where toxoplasmosis makes rates run towards cats, instead of away 03:06 eythian http://starwars.wikia.com/wiki/Brain_worm 03:06 wizzyrea rats* 03:07 wizzyrea not quite a worm, but a parasite. :P 03:07 eythian toxoplasmosis is the disease 03:07 eythian Toxoplasma gondii is the parasite 03:08 wizzyrea just like you are nauseated, instead of nauseous. 03:08 mtompset oh shoot! This has side effects. 03:08 eythian wizzyrea: no, I'm nauseous, you're nauseated :) 03:08 wizzyrea well I meant it how I said it, but you're right now. ;) 03:09 mtompset GetCOinSBiblio -- triggers warnings when there is no 245 field. :) 03:09 wizzyrea you, a person, feel nauseated. Something else makes a person feel nauseous. 03:10 eythian No, something that is nauseous makes someone feel nauseated. 03:10 wizzyrea ^ that's saying the same thing, I think. 03:10 eythian I don't think so. 03:10 mtompset so is the former an adjective and the other a verb? 03:10 eythian you can't feel nauseous. 03:10 eythian mtompset: I think they're both adjectives 03:10 wizzyrea ugh, yes I know! 03:11 wizzyrea we're saying the same thing only I'm saying it poorly 03:11 eythian the nauseous cheese, the nauseated person. 03:11 eythian ah right 03:11 wizzyrea ^.^ 03:13 mtompset Actually, I think you can feel nauseous, but wouldn't it be like "I think I'm feeling like I cause others to be nauseated"? ;) 03:13 eythian yes, that's true :) 03:16 mtompset What is GetCOinSBiblio for? 03:17 mtompset # COinS format FIXME: for books Only 03:18 mtompset <span class="Z3988" title="[% ocoins %]"></span> 03:19 mtompset Okay... some other technology which I must not touch. 03:21 eythian yeah, coins is a thing people use 03:22 eythian I think it lets browser plugins scrape a page for biblio daa 03:22 eythian *data 03:22 eythian so, e.g., citation managers. 03:22 eythian I think that's that one, anyway. 03:30 mtompset Well, if people do crazy things like hide 254$a, that's going to get floody in the error log. 03:31 eythian hmm, it'd be nice if it didn't, if it's an easy fix, I guess. 03:31 eythian if it's not an easy fix, then I'd say it's safe to ignore. 03:53 eythian do authorities always use DOM? 03:56 dcook eythian: I think there's an option but that DOM might've been the default for a while? 03:57 eythian I'm trying to figure it out. 03:58 eythian ahh 03:59 eythian I was looking in the template, but not realising that it had variables in there, so my searches weren't working. 03:59 eythian silly me 04:10 mtompset okay... if I can figure out why XSLT vs. non-XSLT differs in hiding the subtitle, I'll post up my patch. 04:24 eythian you know what we haven't had in a while: test failures when building packages. 04:24 eythian I like this. 04:35 mtompset okay... dumb question. 04:35 wahanui there are no dumb questions. 04:35 mtompset What is the point of fieldmapping vs. marc_subfield_structure? 04:39 mtompset Does anyone actually have anything in their fieldmapping table? 04:40 eythian I tend to grep through the code and see where it's being used, in those cases. 04:41 eythian I just checked on db and it was empty 04:47 dcook mtompset: We use it on our 3.8, but that's for some local stuff that I want to eliminate 04:47 dcook Maybe 04:47 dcook I'm probably lying 04:47 mtompset which is part of my confusion. 04:47 mtompset I also noticed I don't have a 254$b in my BKS framework. 04:47 dcook Mmm, I think it should have subtitle in it at the very least 04:48 mtompset in marc_subfield_structure. 04:48 dcook 254 or 245? 04:48 wahanui 255 04:48 dcook Love you too, wahanui 04:48 wahanui dcook: i'm not following you... 04:48 mtompset I'm pretty sure 254. 04:48 mtompset because 254$a is title, and 254$b is subtitle. 04:49 mtompset well, remainder of title. 04:49 dcook http://www.loc.gov/marc/bibliographic/concise/bd254.html 04:49 dcook $a - Musical presentation statement 04:49 dcook http://www.loc.gov/marc/bibliographic/concise/bd245.html 04:49 dcook $a - Title (NR) 04:49 dcook $b - Remainder of title (NR) 04:49 dcook :) 04:49 mtompset okay, okay... 04:49 mtompset I'm lysdexic. :P 04:49 dcook That said, they should both be there anyway :p 04:59 mtompset oh... is anyone else getting a C4::Context modulus error on line 1250? 05:16 cait morning magnuse 05:41 Bahruz good morning 05:52 mtompset Anyone want a quick sign off? :) 06:05 mtompset Could someone else confirm this problem. 3.14.x Koha/List/Patron.pm uses IsSuperLibrarian, but C4/Context.pm does not have that function. 06:08 mtompset @later tell gmcharlt Please confirm 3.14.x Koha/List/Patron.pm use of IsSuperLibrarian against C4/Context.pm's lack of that function. bug 10565? Not sure. 06:08 huginn mtompset: The operation succeeded. 06:26 mtompset Have a great day (24 hour period), #koha. 06:38 cait @wunder Konstanz 06:38 huginn cait: The current temperature in Taegerwilen, Taegerwilen, Germany is 3.1°C (7:35 AM CET on January 21, 2014). Conditions: Light Rain. Humidity: 94%. Dew Point: 2.0°C. Windchill: 3.0°C. Pressure: 29.83 in 1010 hPa (Rising). 07:17 * magnuse waves 07:17 cait @wunder Konstanz 07:17 huginn cait: The current temperature in Konstanz, Germany is 3.0°C (8:00 AM CET on January 21, 2014). Conditions: Light Rain. Humidity: 88%. Dew Point: 2.0°C. Pressure: 29.88 in 1012 hPa (Rising). 07:18 magnuse @wunder boo 07:19 huginn magnuse: The current temperature in Bodo, Norway is -2.0°C (7:50 AM CET on January 21, 2014). Conditions: Clear. Humidity: 43%. Dew Point: -13.0°C. Windchill: -11.0°C. Pressure: 30.12 in 1020 hPa (Steady). 07:19 magnuse windchill-- 07:19 magnuse and hiya cait 07:23 mram3 hi everybody 07:24 mram3 it's normal that the script link bibs to authorities runs ok with dom indexing, but not with grs1? 07:25 mram3 bareword "geographicsubdiv" it's indexed with dom but not with grs1? , I'm right? 07:25 jcamins Yes. 07:25 jcamins That is normal. 07:25 cait jcamins: go to sleep. 07:25 mram3 thnk you ¡ 07:26 jcamins GRS-1 indexing can't handle headings. 07:26 dcook Great Scot! 07:26 dcook jcamins: 2:26am? 07:26 jcamins cait: I'm running link_bibs_to_authorities... 07:26 mram3 it was driving me crazy... 07:26 mram3 ok 07:26 jcamins And, errr, finishing my book. 07:28 cait sleep? :) 07:29 jcamins 12k records to go. I have some other scripts I want to kick off when this is done. 07:34 cait bbiab 07:34 cait go to sleep jared! 07:37 reiveune hello 07:37 wahanui niihau, reiveune 07:38 * magnuse waves to jcamins and reiveune 07:38 * dcook should go home but is reading about GPG 07:38 * dcook waves to reiveune and magnuse\ 07:39 * reiveune just say "hello" to dcook magnuse rangi kathryn 07:49 magnuse bonojur paul_p 07:50 magnuse s/bonojur/bonjour/ 07:54 dcook Yay! All the way through the GPG user manual! 07:54 dcook Outdated but mostly good! 07:54 magnuse dcook: yay! now you just have to start using it? :-) 07:55 dcook Yarp! 07:55 dcook I'm not entirely how best to do that at this point 07:55 dcook I mean...I have a work computer and a home computer which would be two different identities... 07:55 alex_a bonjour 07:55 wahanui hi, alex_a 07:55 dcook Well, not necessarily two identities 07:57 paxed is it just me, or is the sample frequency nro 7 wrong? ('1/month', 'month', 1, 1, 7) but [% IF ( subscription_LOO.periodicity7 ) %]3 months 08:04 dcook Mmm, I could just have a subkey on my work computer.. 08:05 paul_p good morning #koha 08:06 dcook bon matin, paul_p 08:08 paul_p hello dcook (in french, we don't say "bon matin", we only have "bonjour" ("good day") and "bonsoir" ("good evening") -in one word- ) 08:09 paul_p and we usually switch from "bonjour" to "bonsoir" when the sun goes down, so the hour depends on the season 08:09 dcook (Interesting! Thanks for the tip :D) 08:10 dcook Is "bonne nuit" just for when you're about to go to sleep? 08:11 Joubu hi 08:11 wahanui privet, Joubu 08:21 gaetan_B hello 08:43 Joubu thank you cait for passing qa on 7825! :) 08:46 cait np 08:46 cait hope it's ok and i didn't miss something :) 09:02 magnuse bug 7825 09:02 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=7825 enhancement, P5 - low, ---, jonathan.druart, Passed QA , Change OPACItemHolds syspref to be of type Choice, choices being "no", "yes", "force" 09:10 MrAgent075 Evening all, who have we got tonight? 09:11 cait hm? :) 09:11 cait hi MrAgent075 09:12 MrAgent075 hi cait, how's it goin'? 09:13 cait tuesday morning, long to do list :) 09:14 cait how are you doing? 09:26 MrAgent075 I'm doin' alright. Might get back into the work tomorrow - resetting routers a thousand times is not the best thing to sit through... :/ 11:22 cait magnuse++ 11:34 magnuse cait: huh? 11:35 cait email to dev list 11:38 magnuse ah, we get pluses for asking dumb questions? ;-) 12:22 cait pluses for communicating 12:32 mtompset Greetings, #koha. 12:58 francharb good morning 13:00 magnuse khall++ for answering 13:08 tcohen_ morning #koha 13:19 magnuse hola tcohen_ 13:19 tcohen_ hi magnuse ! 13:26 oleonard Hi #koha 13:30 magnuse wheee 13:32 magnuse sophie_m: could be :-) 13:40 tcohen morning? 13:40 wahanui morning is, like, a state of cat 14:00 tcohen fridolin? 14:02 fridolin hie all 14:03 * cait waves 14:05 * druthb waves 14:05 tcohen hi cait druthb 14:05 tcohen druthb: did my email reach koha-translate? 14:06 cait the one about arabic? 14:06 cait i saw something form you 14:07 druthb tcohen: yup! 14:08 tcohen good 14:08 tcohen its just that I expected being flooded by answers heh 14:09 druthb The Arabic community for Koha is not large, partly because of issues like that one. Give it a little time. 14:09 druthb a hunt through the archives for Karam might get his address, and you could write him directly. 14:22 magnuse i think i found a problem at the intersection of DBIC and koha-gitify: https://github.com/mkfifo/koha-gitify/issues/7 14:22 magnuse no time for proper testing at the moment, though 14:40 tcohen cait: there are 32 strings to translate in german 3.12 14:41 cait that's not much :) 14:41 cait i amon it 14:44 druthb super-cait! 14:52 tcohen portuguese translator? 15:36 mtompset Karam? He's trying to link to me on LinkedIn. ;) 16:02 tcohen Joubu: steps to reproduce? 16:02 wahanui steps to reproduce are on comment 1 16:03 Joubu tcohen: I applied yours patches, launch a search and... that's all 16:03 tcohen UNIMARC grs-1? 16:03 tcohen or DOM? 16:03 wahanui DOM is the answer... 16:03 Joubu tcohen: grs1 16:04 Joubu unimarc 16:05 tcohen http://snag.gy/LdI8f.jpg 16:09 cait translations are all done :) 16:13 druthb cait++ 16:13 druthb hm. 16:13 druthb @most increased 16:13 huginn druthb: "gmcharlt": 18, "cait": 11, "oleonard": 9, "rangi": 8, "khall": 5, "magnuse": 5, "jcamins": 5, "tcohen": 5, "eythian": 5, "Joubu": 4, "petter": 3, "francesca": 3, "community": 3, "fridolin": 3, "davidnind": 3, "wizzyrea": 3, "druthb": 3, "ashimema": 3, "paul_p": 2, "ebegin": 2, "MrAgent075": 2, "pianohacker": 2, "jatara": 2, "*+++++++++": 1, and "davidtennant": 1 16:19 fridolin see you 16:20 mtompset YAY! I'm not on the hit list of blame. ;) 16:24 mtompset ARG! Just when I thought I was good... [ view plain ] has to come along! 16:42 mtompset How do I turn the contents of marcxml into a MARC::Record? 16:42 mtompset (and vice versa) 16:55 reiveune bye 16:59 Joubu mtompset: MARC::Record->new_from_xml ? 16:59 JesseM Hello 16:59 Bahruz hi all 17:00 Joubu hi / bye :) 17:07 gaetan_B bye ! 18:07 * magnuse waves 18:56 * mtompset struggles to stay away. 18:56 mtompset stay AWAKE. 18:57 druthb lolz 18:57 mtompset Just caught the tail portion of a meeting online. 19:39 mtompset oh come on! I hide 0,3,5, and 8! 19:40 * mtompset yells at his VM. 19:41 * oleonard shakes his walking stick menacingly at mtompset's VM 19:41 mtompset Try something crazy... uncheck the visibility for 0,3,5, and 8. 19:42 mtompset Then go look at opac-MARCdetail for that framework. 19:42 mtompset click [ view plain] 19:42 mtompset Still shows! 19:44 oleonard Is this surprising because you simply expect things to work or because the script appears to be written such that it should work? 19:45 mtompset I've been fixing this whole hiding mess. 19:45 mtompset I hid everything for this one record I'm looking at. 19:46 mtompset except 999$c, 999$d, and 245$b. 19:46 mtompset I suppose the conversion from MARC::Record back to MARCXML probably creates those. 19:47 magnuse sounds like something gmcharlt would know :-) 19:50 mtompset oh... perhaps I have to code the @ trick. 19:54 magnuse mtompset scores an 'at-trick! 19:54 * druthb rolls her eyes. 19:56 * magnuse watches druthb's eyes rolling past his desk 19:57 druthb lol 20:00 mtompset Sorry... only a partial @ trick. 20:00 mtompset Still working on the LEADER field. ;) 20:00 mtompset But that was it. :) 20:46 mtompset I'm going to give up on trying to hide the leader record. 20:46 mtompset I got everything else workings, as far as I know. 20:46 mtompset I mean leader field. 20:52 rhcl @seen bag 20:52 rhcl huggin? 20:52 wahanui i think huggin is a bot 20:52 rhcl yea, well, where is (s)he? 20:53 rhcl @wunder 64507 20:53 rhcl d* bot is off on a bender somewhere 21:14 tcohen @seen Joubu 21:15 wizzyrea you will note that huginn is not in the user list at the moment. 21:15 wizzyrea ping gmcharlt, when you have a moment, huginn needs a kick :) 21:18 tcohen oh, thanks wizzyrea 21:22 pablito hi all 21:23 pablito my koha 3.14 doesn't send overdue messages to patrons 21:23 pablito the triggers are setup 21:23 pablito but the process table is not populated 21:40 wizzyrea Does anyone out there have a (hopefully nice) comment about the work the academy kids did? 21:40 wizzyrea hehe. 21:49 wizzyrea there he is! 21:49 rhcl bum 21:49 gmcharlt @quote random 21:49 huginn gmcharlt: Quote #15: "<|Lupin|>ricardo: you are like the Portuguese Hitchcock!!" (added by wizzyrea at 02:58 PM, July 10, 2009) 21:49 rhcl @seen bag 21:49 huginn rhcl: bag was last seen in #koha 1 week, 4 days, 22 hours, 34 minutes, and 19 seconds ago: * bag makes a loud banging noise 21:49 wizzyrea pablito: I'm going to guess you have the cron jobs set up? 21:50 pablito the overdue.pl is already setup by default 21:51 pablito wizzyrea: now, I try to run that script manually and it gives me this error: unable to locate Koha configuration file koha-conf.xml at /usr/share/koha/lib/C4/Context.pm line 372. 21:51 pablito Can't call method "config" on unblessed reference at /usr/share/koha/lib/C4/Context.pm line 797. 21:52 eythian pablito: how have you installed Koha? 21:52 wizzyrea also wb gmcharlt 21:53 pablito eythian: I did apt-get install koha-common in Ubuntu 21:53 eythian OK 21:53 eythian do you have email configured on your server. 21:53 eythian ? 21:53 pablito that's the thing, I remember ever setting up the email server for Koha 21:53 pablito I DON'T remember ... 21:54 pablito it didn't ask me to setup the email server 21:58 wizzyrea why oh why do we still allow the DB user to log in to the regular staff client. Why. WHY. 21:59 pablito wizzyrea: ??? 21:59 wizzyrea it's not related to you ;) 22:00 wizzyrea unless you are loggin in as the database user. In which case, don't do that. 22:00 eythian wizzyrea: because if you didn't, you couldn't create a user. 22:00 wizzyrea logging* 22:00 wizzyrea then there needs to be another mechanism for that. 22:00 eythian it's also very handy for nipping in and tweaking a syspref 22:01 wizzyrea I agree that it's handy for power users. 22:01 eythian pablito: have a look at a user that should have got an email, see if it's in there and marked as sent. 22:01 wizzyrea for people who know better than to circ with that account. 22:02 eythian wizzyrea: There is a big screamy warning now, don't know if it works though. 22:02 wizzyrea it doesn't 22:02 wizzyrea bug 11590 22:02 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11590 critical, P5 - low, ---, kyle, Needs Signoff , Librarian able to checkout item without setting branch first 22:03 wizzyrea GIANT WARNING ... does it anyway. 22:03 eythian pablito: if it says it has sent, then you probably have a misconfigured mail server on your system. 22:03 pablito eythian: where do I check for the message or the status "sent"? 22:04 tcohen_ pablito: apt-cache policy postfix 22:05 eythian http://manual.koha-community.org/3.14/en/patroninformation.html <-- pablito, section 9.7 22:06 * eythian wishes you could link to sections 22:06 pablito tcohen_: Here's what I got after running that command: 22:06 pablito postfix: Installed: (none) Candidate: 2.9.6-1~12.04.1 Version table: 2.9.6-1~12.04.1 0 500 http://il.archive.ubuntu.com/ubuntu/ precise-updates/main i386 Packages 2.9.1-4 0 500 http://il.archive.ubuntu.com/ubuntu/ precise/main i386 Packages 22:10 eythian wizzyrea: that's also part of the reason we never give out the database passwords, so end users _shouldn't_ be able to do that. Also, giving out database passwords strikes me as a terrible idea. 22:11 pablito eythian: hmm, sect 9.7 it doesn't really tell me how to get to the message list 22:11 wizzyrea having the login be the db password seems like a bad idea to me :/ 22:11 wizzyrea at all, ever. 22:11 wizzyrea except maybe for upgrades. 22:11 eythian wizzyrea: it's how you configure things 22:12 eythian pablito: it's the notices tab on the left hand side when you're viewing a user. 22:12 wizzyrea yea, I'm thinking of a mechanism by which you add a user and define your first library at install time 22:12 wizzyrea add a user, define your first library, and your first circ rule. 22:12 eythian wizzyrea: it'd make life a lot harder for us 22:12 pablito eythian: ah, I saw that there was no message. 22:12 eythian wizzyrea: because now we have no sure way of logging in ever 22:13 wizzyrea right, I thought of that as well - which could lead to things like a "vendor mode" which has a pre-defined login added as part of the koha-sites or something 22:13 eythian and then someone deletes it or something. 22:14 eythian dollars to doughnuts it'll just cause people to get stuck, and piles of questions asking how to log in the first time. 22:14 wizzyrea yep, it would have to be undeletable 22:14 wizzyrea ok, for arguments sake 22:14 wizzyrea wordpress requires a user be defined by the person installing at install time 22:15 wizzyrea it comes with a default username, highly suggesting you change it 22:15 wizzyrea allows the user to set the password then 22:15 wizzyrea it's an ok workflow, I think 22:16 eythian yeah, but we need a separation between "users we have" and "users they have" so that they can't lock us out. 22:16 wizzyrea right, I think that falls into what I might think of as "vendor mode" 22:17 wizzyrea rather like the wordpress multisites, a user of a site can't delete the administrator of the overall site 22:17 gmcharlt spitballing -- is "you must give us shell access on request in order to support you" a solution for you? 22:18 eythian gmcharlt: no, because we have shell access :) but often need to log in to flip a syspref or something. The easy way of doing that is to look up the password in koha-conf and log in using that to do it. 22:18 gmcharlt especially since it might solve the following problem from the user's POV: "my vendor has somehow gotten their internal database of credentials compromised, and I want to change all the passwords until they sort it out" 22:19 gmcharlt eythian: indeed, that's the easy way ;) 22:20 gmcharlt possibly an obvious step forward: even if we keep a koha-conf.xml-defined superlibrarian account, there is zero reason why it *has* to be based on the DB credentials 22:21 gmcharlt it can just be a username and password 22:21 eythian There is an issue with having a default database user like some install guides show, as it's hard to change and easy for someone to figure out. 22:21 gmcharlt although -- convenience aside -- it really ought to be hashed 22:21 wizzyrea ^ is kind of what I was thinking, a user added by koha-sites, and possibly flagged as undeletable 22:21 wizzyrea password randomly generated 22:22 wizzyrea different from the db user 22:22 eythian a similar option would be to have a user/pw in there that maps to a specially flagged user in the db that can't be deleted. 22:22 eythian if it is deleted, then it falls back into the current partially broken mode to allow you to recover it. 22:23 wizzyrea I guess you can't guarantee that your first user won't be deleted if you say, truncate borrowers. 22:23 wizzyrea but wp would have the same issue. I've never tried it 22:23 eythian yeah exactly 22:24 wizzyrea idk what it does there. 22:24 wizzyrea I suspect it would be massively broken and you'd have to do gymnastics to get out of that situation 22:24 gmcharlt there has to be *some* point where one just sighs, and fires up koha-mysql ;) 22:24 wizzyrea well and if the #1 user doesn't exist 22:24 wizzyrea why wouldn't it just prompt you to create a new one? 22:25 eythian then anyone could just create an admin user 22:25 wizzyrea or "go set a variable in your conf to allow creation of a new user" 22:25 eythian (assuming it got deleted) 22:25 wizzyrea that's kind of like koha-mysql, you can do anything if you have sudo on the server 22:26 wizzyrea honestly it could go on the way it is, it's just... it seems it could be better. clearly the warning isn't being heeded. 22:26 * dcook whilst stumbling upon this conversation : http://www.youtube.com/watch?v=xp0vr1n25u4 22:26 wizzyrea (actually I just a couple of weeks ago had a library that was using the DB user systematically as their only staff account.) 22:27 dcook (O_O) 22:27 wizzyrea it might have worked ok in 3.2, but 3.14? yea, naw, that won't really do. 22:27 eythian wizzyrea: not one of ours I hope? 22:28 wizzyrea not one on our servers, no. 22:28 eythian that's why I'd be quite happy to turn off things like the circ function. And/or having an overlay that you have to click through on the main screen that tells you why you're a bad person for doing this. 22:28 wizzyrea THIS IS BAD AND YOU SHOULD FEEL BAD. 22:28 dcook ^ 22:29 wizzyrea to be fair, most places catch you if you try to do something that say, requires a branch 22:29 wizzyrea like circ. 22:29 wizzyrea this one spot, didn't 22:29 wizzyrea but I feel that's enabling people to do the thing we don't want them to do 22:30 eythian well, that's not hte whole problem. Not having a borrower account can cause problems. 22:30 wizzyrea yea, there's bad all over the place with the DB user. 22:31 wizzyrea probably instead of prompting to select a branch, we should just take them to the add user page. :P With giant red arrows "TICK HERE. AND HERE. AND DO THIS." 22:31 wizzyrea that's probably overkill ;) 22:34 eythian well, it's also good should someone do something with a syspref that locks them out or something 22:34 eythian I'd think blocking access to anything other than admin, borrowers, and so forth would be good. Along with a message saying why they're not available. 22:43 wizzyrea either way, what started this whole thing was someone doing something they were told not to, then marking it critical. :P 22:43 wizzyrea that's like reading the law, breaking it anyway, then being mad at the people who wrote the law. 22:44 wizzyrea because they let you break ti. 22:44 wizzyrea fortunately, we are in the lovely position of rigidly defining the law so that it can't be broken. :P 22:58 cjh magnuse: just replied to your github koha-gitify issue :) 23:08 * cait waves 23:08 rhcl waves 23:08 * eythian calms 23:09 rhcl calms waves? 23:10 eythian well everyone else was waving, but it's not very windy outside. 23:11 rhcl ah, compare - contrast 23:13 cait :) 23:15 eythian gmcharlt: there seems to be a versioning issue with MARC-XML 1.0.2, debian things are seeing it as 1..2. 23:15 eythian I can work around it, and don't know where it's coming from, so just fyi 23:16 gmcharlt noted, thanks 23:21 eythian libmarc-xml-perl packages updated.