IRC log for #koha, 2012-08-16

All times shown according to UTC.

Time S Nick Message
00:00 libsysguy joined #koha
00:10 melia left #koha
00:13 Judit joined #koha
00:13 Judit monrning
00:19 Judit1 joined #koha
00:28 Judit joined #koha
00:32 Judit1 joined #koha
01:03 Judit joined #koha
01:03 Irma joined #koha
02:02 ala_3vo joined #koha
02:04 ala_3vo Hello #koha
02:05 ala_3vo After installing koha-common, mysql and koha-create, I encountered an error message
02:05 ala_3vo Syntax error on line 69 of /etc/koha/apache-shared.conf:
02:05 ala_3vo Invalid command 'RewriteEngine', perhaps misspelled or defined by a
02:06 ala_3vo module not included in the server configuration
02:06 ala_3vo Action 'configtest' failed.
02:06 ala_3vo The Apache error log may have more information.
02:20 libsysguy ala_3vo you may need to enable mod_rewrite
02:21 libsysguy $ a2enmod rewrite
02:26 ala_3vo libsysguy: thanks that worked
02:26 libsysguy sweet, good luck
02:33 ala_3vo if I koha-create --create-db instancename another, I still have that same error.
02:33 ala_3vo why is that?
02:36 libsysguy I'm not sure ala_3vo
03:25 janPasi joined #koha
03:53 bartsimpson joined #koha
04:10 cait joined #koha
05:16 cait hi #koha
05:21 Judit1 joined #koha
05:24 * mtj waves to cait
05:29 BobB joined #koha
05:33 cait hi mtj :)
06:27 mtj woahh - jcamins' c4::search test looks impressive
06:28 mtj i cant remember seeing a patch thats too big for bugzilla, ever?
06:28 kf joined #koha
06:29 Judit1 left #koha
06:29 mtj jcamins++
06:33 kf hi mtj
06:34 kf jcamins++ :)
06:36 ala_3vo joined #koha
06:37 mtj ->
06:37 mtj ... twas the test, i was talking about
06:38 julian_m joined #koha
06:42 julian_m hello
06:42 wahanui niihau, julian_m
06:53 francharb joined #koha
06:54 francharb Good morning
06:58 kf hi francharb
06:59 drojf hi #koha
07:00 kf hi #koha
07:00 drojf moin kf
07:04 hdl joined #koha
07:06 drojf uk--
07:11 francharb joined #koha
07:21 hdl joined #koha
07:26 ala_3vo Hello #koha
07:27 ala_3vo After installation of koha-common, and mysql successfully there was a syntax error when I run koha-create
07:28 ala_3vo cat: /etc/koha/passwd: No such file or directory
07:28 ala_3vo Koha instance is empty, no staff user created.
07:28 wahanui i already had it that way, ala_3vo.
07:29 ala_3vo okay, what was the problem wahanui?
07:29 eythian joined #koha
07:30 ala_3vo I have tried about 3 times on test environment and it came out the same
07:32 ala_3vo but when I upgrade from previous version it was okay...
08:03 rangi n/names
08:41 WaqarAzeem joined #koha
08:45 ala_3vo joined #koha
08:59 WaqarAzeem Hello! how can i find... the version of Koha running on my server?
08:59 eythian WaqarAzeem: you can go to the 'About' page in the staff client
09:00 WaqarAzeem thanks
09:02 koyauni joined #koha
09:03 koyauni hi guys, anyone good with Debian dedicated servers
09:03 eythian koyauni: it depends, what's your question?
09:04 koyauni I got this unmanaged server with and I can not run php pages,
09:05 koyauni I say this sort of thing are so basic that a server should do these thing out of box but they say this is it take it or leave it :(
09:05 koyauni here is a sample page
09:05 koyauni
09:05 koyauni it is Debian 6, with Apache 2, Php 5.x
09:06 koyauni and I have gone though hell putting koha on it I do not want to juet leave it
09:06 eythian "More information about this error may be available in the server error log." <-- what does that information tell you?
09:07 koyauni in /var/log/apache/
09:07 koyauni last line says
09:07 koyauni [Thu Aug 16 10:42:25 2012] [notice] Apache/2.2.16 (Debian) DAV/2 mod_fcgid/2.3.6 mod_python/3.3.1 Python/2.6.6 mod_ssl/2.2.16 OpenSSL/0.9.8o mod_perl/2.0.4 Perl/v5.10.1 configured -- resuming normal operations
09:08 eythian that's not the important line. The one with the error is the one you need to be looking at.
09:08 koyauni I do not see PHP resumin after restart
09:08 eythian I don't really know much about PHP
09:08 koyauni that is in the error.log
09:09 eythian Yes, but it's not an error.
09:09 eythian It's a notice.
09:09 koyauni I see this line as well
09:09 koyauni [Thu Aug 16 10:42:25 2012] [error] python_init: Python path being used '/usr/lib/python2.6/:/usr/lib/python2.6/plat-l​inux2:/usr/lib/python2.6/lib-tk:/usr/lib/pytho​n2.6/lib-old:/usr/lib/python2.6/lib-dynload'.
09:10 eythian you should try the request again and see what gets added to the logs
09:10 eythian all these look like startup information.
09:10 eythian also, if this is a koha server, why do you have php and such installed on it anyway?
09:11 koyauni where is this log file
09:11 koyauni may be I look in the wrong log
09:11 eythian I don't know
09:11 eythian somewhere in /var/log/apache2 most likely
09:11 eythian do the request and see what files change
09:11 koyauni I have my page for uni with information about the library
09:12 koyauni how do you make sure in SSH command that php is runing
09:12 koyauni I an in there and I am looking at error.log
09:12 eythian I don't know, I don't use PHP much, and when I have had to, it just works.
09:13 koyauni me too, I have never seen something like this
09:13 eythian and is error.log the one that changes when you do the request that causes the error.
09:13 koyauni and this bloody server admin say this it we do not give support :(
09:13 eythian Well, that's what an "unmanaged" server means.
09:16 koyauni this is odd, nothing is add to, when I reload or reaccess the page
09:16 koyauni that you have to manage it yourself after purchase
09:17 koyauni but I just got this, it is like buying a TV and take it home, and it does not switch on
09:17 eythian are you sure you're hitting the right server, or looking in the right place for your log files? I'd have a look through your apache configuration and see if it's logging somewhere else.
09:17 koyauni so they keep telling you that is find out why
09:17 koyauni :(
09:18 koyauni these are two domains two domains I have activated on that server for now, and
09:18 koyauni I can see html pages
09:19 koyauni but no script pages
09:19 mtj koyauni, you have a php problem - not a koha problem
09:19 koyauni I am in the root of the server
09:19 eythian Well, if nothing is being logged at all, then you are either looking in the wrong place for your log files, or you're accessing the wrong server (or something more unusual, but those are the common situations.)
09:20 koyauni I have koha at and it is installed, do not know if it is working yet
09:20 mtj (there are not many php experts in this #irc channel)
09:20 koyauni I need to get the server right before I point to it
09:21 koyauni I do not think this is PHP problem rather Debian server issue to run any script
09:21 eythian if you can't see the logs, you need to find out what's going on. To do that you need to look through your configuration. Things are pretty much always logged.
09:24 mtj koyauni, well.... whatever your problem is... its not a koha problem, yes?
09:26 mtj do you have php installed, even?
09:27 mtj ... via your systems parallels-panel admin-gui?
09:27 rangi   <-- what page do you see here? I see something about parallels
09:28 mtj rangi, yep, me too
09:28 koyauni that is normal HTML page server defualt
09:28 rangi thats not debian
09:29 koyauni I checked yes, php5 is on Debian
09:29 eythian yeah, that is not the normal server default for debian at all.
09:29 eythian I think you are accessing the wrong server.
09:30 koyauni no this is Debian 6 dedicated server with Plesk that is Plesk default page
09:30 mtj also, shows parellels page, not koha
09:30 mtj koyauni, no not plesk - parellels
09:30 eythian then maybe you need to be looking into your plesk configuration.
09:31 mtj ahh, parellels == plesk, sorry...
09:32 koyauni I found the bloody things
09:32 mtj koyauni, so your plesk did not install php properly
09:33 koyauni they have a setting with run PHP as FastCGI thing
09:33 koyauni I changed that to Apache mode and it worked
09:33 mtj yes, you dont have fast-cgi installed...
09:33 mtj ->
09:33 koyauni sorry guys to take you time
09:34 koyauni[…]ted_on_plesk.html
09:34 koyauni that page saved me,
09:35 koyauni so they have set up my server and put Plesk on Fastcgi without having FastCGI
09:36 koyauni do I even need this
09:36 koyauni[…]ebian-6-0-squeeze
09:37 eythian Not for koha you don't
09:37 eythian beyond that, I don't know
09:38 mtj koyauni, koha uses perl, not php
09:40 mtj dear #koha, is anyone using a snazzy carousel thingy on their opac?
09:41 koyauni I have installed that, I want to make sure that the server is working for all my projects before I point my domain to this new NS
09:41 mtj ... a book carousel, that is
09:44 * mtj sends the biblibre folk an email re: carousel
10:00 mtj ... but i would like to see some other examples, for improvements & ideas
10:04 eythian When you click left, it moves to the right which is confusing initially
10:04 eythian also should be mouse draggable :)
10:08 mtj yeah, i agree re: left/right
10:09 mtj ... mouse stuff in version 2 :)
10:11 mtj this carousel is quite nice too ->[…]Flow/download.php
10:46 drojf joined #koha
11:02 francharb mtj, I emailed you back on the carousel thing! ;)
11:02 francharb_lunch now, time for lunch!
11:08 mib_pzwneg joined #koha
11:09 mib_pzwneg Has anyone noticed LDAP auth seems to stop when upgraded from 3.6.x to 3.8? Or is there a subtle change in the config I am missing!?
11:10 nengard joined #koha
11:13 jcamins mib_pzwneg: I don't really know anything about LDAP, but did you look at the blog post on ByWater's blog about configuring LDAP?
11:15 mib_pzwneg jcamins: No, I will take a look. I had it configured fine on 3.6 though, so I thought if there was some substantial changes it would have been in the release notes for 3.8 and there is nothing.
11:16 jcamins I am not aware of any changes, but I use 3.6, and don't use LDAP.
11:17 jcamins Hm. There don't appear to be any further changes to the LDAP code.
11:18 jwagner joined #koha
11:19 eythian I've done a 3.4 to 3.8 upgrade, and LDAP didn't break.
11:19 mib_pzwneg Thanks jcamins. The blog doesn't offer me anything I didn't know - If I switch on debug in the Net::LDAP class I can see the binding correctly, and even the user's (trying to login) attributes get logged fine, just fails to login to Koha!
11:20 jcamins What version of 3.8?
11:21 mib_pzwneg I just pulled the latest stable from koha-community - one second I will get the exact version
11:22 mib_pzwneg 3.08.03
11:22 jcamins Okay, that's a good version to be using.
11:22 jcamins eythian: how's Konstanz?
11:23 eythian It's good. Today is getting better now that the rain stopped, too.
11:25 mtj heya francharb_lunch , thanks for the reply
11:25 mtj my carousel is using ->[…]site/carousel.htm
11:26 mtj (which is the same as, afaik)
11:27 nengard joined #koha
11:45 clrh dear develppers, in wich debian version do you program ?
11:45 clrh wheezy ?
11:45 wahanui hmmm... wheezy is testing
11:45 mtj ... ldap problem looked like a problem worthy of a bug report :/
11:45 clrh squeeze?
11:45 wahanui squeeze is 3.2.x
11:45 clrh squeeze is 6.0
11:46 mtj clrh - the stable version of debian :)
11:46 mtj ... always
11:50 eythian clrh: mostly squeeze, but I tend to build against unstable so I have advanced warning of things changing.
11:50 eythian no wahanui, squeeze is 6.0
11:50 wahanui okay, eythian.
11:51 clrh ok eythian thanks
11:51 clrh I hope your last too wekks at Marseille were cool too :)
11:51 clrh I didn't tell you goodbye!
11:52 eythian Yeah, it was good :)
11:52 eythian No, I think I you got back the work day after I finished.
11:52 jcamins I use Squeeze.
11:58 mtj but, its great to know people like eythian++ are testing koha on debian-testing, for the rest of us :)
12:04 mtj woah, awesome icons ->[…]Awesome/#examples
12:05 mtj yes please :)
12:05 nengard joined #koha
12:06 mtj hmm, is the (CC BY 3.0) license compatible with koha's license?
12:07 mtj 'The Font Awesome webfont, CSS, and LESS files are licensed under CC BY 3.0'
12:07 mtj
12:09 jcamins CCSA2.5 is.
12:09 jcamins CC BY 2.5, rather.
12:17 francharb mtj, you're welcome! ;)
12:25 mtj jcamins, aah ok - sounds like we need to bump Koha to gpl3+ to use those fonts/icons...
12:28 mtj ftw, ive recently discovered some very interesting svg/vector icons too ->
12:28 mtj ... first ive ever seen, i think?  (mit license too :) )
12:31 drojf ecuador++
12:42 mbalmer joined #koha
12:49 chris_n` joined #koha
12:51 Callender joined #koha
12:53 tcohen joined #koha
12:59 ago43 joined #koha
13:05 Callender joined #koha
13:11 talljoy joined #koha
13:20 maximep joined #koha
13:21 oleonard joined #koha
13:26 libsysguy joined #koha
13:27 oleonard Hi #koha
13:36 mtj hiya oleonard
13:38 druthb joined #koha
13:41 datadoctor joined #koha
13:41 datadoctor joined #koha
13:43 datadoctor I'm looking at the Tools > Notices editor - looks like it is much easier to edit the quick slip and print slip.
13:45 datadoctor You can add your branch information, biblio and biblioitem information (like MARC info), borrower, news and issue information to the quick slip, print slip, etc.
13:49 datadoctor I am wondering how easy it would be to add renewal information and item replacement price, as they are not in the list. Are we limited to the values that appear in the list?
13:50 datadoctor Tools > Notices > Edit Quickslip (Modify Notice)
13:54 mtj you might need to modify the sql to pull that info from other tables ( like 'reserves' table )
13:55 datadoctor Thanks. I will check that. Reserves might give me current holds.
13:55 mtj rep-price should be currently available, not sure about renewal info
13:56 datadoctor Price looks easier. We currently total the replacement prices on our quickslip, and inform the patron about how much they have saved by using the library.
13:56 datadoctor Hoping to implement that on 3.8 as well.
13:58 datadoctor For renewals, we can compare date issued with current date, and cross-check with remaining renewals. I think that will isolate today's renewals.
13:59 datadoctor If I can't deliver it through the Notices editor for the quickslip, I can probably pull the information from mysql and deliver it through the template.
13:59 datadoctor I'm thinking that's or
14:00 datadoctor Although a lot of the information comes from
14:11 pastebot "nengard" at pasted "looking for report help - is this possible? how would I get just the numbers from the isbn field?" (4 lines) at
14:23 eythian[…]ng-functions.html <-- nengard, there may be something here that helps
14:24 eythian (sorry, that answer is a bit RTFM-y, but it's all I got right now :)
14:24 nengard I love the mysql manual :)
14:24 nengard and did look there, but I think this needs regex and no manual will ever help me understand that :) hehe
14:26 eythian oh, I understand regexes. If you just want the part up-to the space: s/^([^ ]*) .*$/$1/
14:27 oleonard Hey, watch your language!
14:28 eythian I don't think mysql regexes can do this
14:29 eythian they seem to be for truthiness of matching, or used as LIKE
14:29 datadoctor We are using regex to parse the last renewed date and the current date for comparison, converting to a date in the format: 20120816
14:29 eythian datadoctor: in SQL?
14:29 datadoctor In PERL
14:29 datadoctor Sorry, mixing my metaphors
14:29 eythian Yeah, this case needs SQL.
14:30 eythian nengard: I think SUBSTRING_INDEX may be helpful
14:31 nengard okay i'll take a look
14:31 eythian it'll let you grab everything before the space, anyway.
14:32 eythian as for applying the dates, I'm not sure. Probably group by the sanitised ISBN and take the max of the dates. Then test that.
14:32 eythian That might just work.
14:32 nengard thanks! I'll see what I cam come up with - the dates part I can do
14:32 nengard she wants the date added
14:32 nengard that's another field
14:34 eythian yeah, so something like select [fix up isbn] as isbn, count(*) as count, max(date_added) as max_date where max_date > <<limit>> group by isbn;
14:34 eythian with the appropriate from and joins and stuff
14:34 eythian Not sure that'll work, but it might.
14:35 datadoctor Would SELECT SUBSTRING be useful?
14:35 datadoctor for trimming the suffix out
14:35 datadoctor Oh, I see that's covered by the SUBSTRING INDEX
14:39 pastebot "nengard" at pasted "resulting report" (5 lines) at
14:39 nengard eythian, okay so this worked ? but you're right - adding the date is hard
14:39 nengard where does WHERE go when there is a HAVING
14:40 nengard something like WHERE b.datecreated between '2011-01-01' and '2012-01-10'
14:41 gbk77 joined #koha
14:42 eythian hmm, I don't know, I've not used HAVING
14:42 nengard sekjal taught me about HAVING
14:42 gmcharlt nengard: SELECT ... GROUP BY ... HAVING
14:43 nengard and can i put WHERE in there?
14:43 gmcharlt the where clause is optional, but if you need it...
14:43 gmcharlt SELECT ... FROM ... WHERE ... GROUP BY ... HAVING
14:43 nengard oh!
14:43 nengard okay
14:43 eythian that makes some sense
14:44 wajasu at mysql prompt> help select;   for reference
14:44 pastebot "nengard" at pasted "final report! Thank you all!! it's going on the wiki" (6 lines) at
14:46 eythian awesome :)
14:46 nengard eythian++
14:46 nengard gmcharlt++
14:46 nengard :)
14:47 wajasu jcamins_away: i think i got you remote search_t  search unit test setup in a local branch.   what would you like me to run to test. all tests? how?
14:48 mveron joined #koha
14:49 gmcharlt nengard++
14:50 mveron Hi #koha
15:01 datadoctor g'day!
15:07 eythian
15:09 datadoctor :D
15:12 hdl1 joined #koha
15:14 wajasu if i want to run a single unit test, does anyone know how i would do that?
15:16 gmcharlt wahanui: prove t/Foo.t
15:16 wahanui gmcharlt: i'm not following you...
15:16 gmcharlt er,
15:17 gmcharlt wajasu: prove t/Foo.t
15:17 gmcharlt ;)
15:17 wajasu here goes.
15:17 wajasu do i have to run perl Makefile first?
15:18 gmcharlt generally speaking, no
15:19 wajasu trying prove t/db_dependent/Search.t  ...
15:21 gmcharlt wajasu: db_dependent tests require a little more setup
15:21 gmcharlt in particular, KOHA_CONF would have to be set, pointing to a database with the sample data loaded
15:21 gmcharlt and some of the non-DB-dependent tests may want to point to the stock koha-conf.xml template, e.g.
15:22 wajasu i'm trying to test bug 8649
15:22 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
15:22 gmcharlt but just to suppress the "unable to locate Koha configuration file" warning
15:30 wajasu it ran will some failures, i'll fool with it a bit and wait until jcamins shows up
15:42 maximep wow, I guess I really did bug 8645 without thinking about it
15:42 huginn Bug[…]w_bug.cgi?id=8645 normal, P5 - low, ---, pelletiermaxime, ASSIGNED , Fix background-job-progressbar.js for translations
15:43 maximep in 3.6 it was in an .inc file, so it was translatable
15:43 maximep it was moved in a js which isnt
15:43 maximep and I just copied the code -_-
15:43 jcamins Someone looking for me?
15:43 jcamins Ah, wajasu.
15:43 jcamins wajasu++
15:43 jcamins Actually, t/db_dependent/Search.t does not require a working database.
15:44 jcamins It just requires MySQL.
15:44 jcamins wajasu: what failed?
15:51 pastebot "wajasu" at pasted "bug 8649 Search.t test" (203 lines) at
15:51 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
15:51 jcamins Ahh... everything.
15:52 francharb see you!
15:52 jcamins Okay, it looks like it's searching your Zebra server and not the test's.
15:53 wajasu jcamins: the first time i ran it i had less tests, then i only installed DBD::Mock and you see those results in paste
15:53 jcamins Do you have Test::MockModule?
15:54 jcamins Could you paste the output of perl t/db_dependent/Search.t?
15:54 wajasu yes. koha_perl_deps reports 0.05
15:55 wajasu ok i'll do that
15:55 jcamins I suspect I probably missed a file or two.
15:55 jcamins I shall delete my data directory and check it out again.
15:56 jcamins No, it's not that. All the files exist.
15:56 wajasu you saw my git commands.   do i need to do a git pull or fetch/merge?
15:56 jcamins No, that looked right to me.
15:59 pastebot "wajasu" at pasted "bug 8649 Search.t test > perl t/db_dependent/Search.t" (199 lines) at
15:59 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
16:00 jcamins Hrm.
16:02 jcamins So why would the test be connecting to the wrong Zebra server?
16:02 jcamins Do you have anything in t/db_dependent/data/zebra.log?
16:03 wajasu empty
16:04 jcamins Okay, so set_context isn't working.
16:04 jcamins Oh.
16:04 jcamins Geez.
16:04 jcamins This is embarrassing.
16:04 gmcharlt jcamins: I suspect that using set_context() and mocking the methods needs to be done before use_ok('C4::Seaarch')
16:04 gmcharlt plus what you just noticed ;)
16:04 jcamins gmcharlt: nah, I think the problem is the whole "jcamins" in the path name.
16:06 gmcharlt jcamins: out of curiosity, what's your platform? I'm not entirely sure that the compiled zebra indexes are portable, at least to the box I've been using
16:06 jcamins wajasu: okay, do a git pull and try again.
16:06 jcamins gmcharlt: Debian...
16:06 wahanui debian is what most people use and there are good manuals and even packages for that
16:07 gmcharlt jcamins: right, but what architecture?
16:07 jcamins I'm trying to figure that out.
16:07 gmcharlt uname -a
16:07 jcamins Thanks.
16:07 jcamins i686
16:08 jcamins So I need to do the indexing, too.
16:08 jcamins Okay, we can do that.
16:08 gmcharlt yeah
16:08 gmcharlt (box I'm using for testing is x86_64)
16:08 pastebot "wajasu" at pasted "bug 8649 Search.t test > perl t/db_dependent/Search.t #2" (35 lines) at
16:08 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
16:08 jcamins Well... that's not an improvement.
16:08 jcamins Are you on x86_64?
16:09 jcamins 'Cause apparently that's going to be an issue.
16:09 wajasu i'm on x86_64 deb
16:09 jcamins Right. Guess I'd better change the patch to reindex, too.
16:09 jcamins This might take a little bit...
16:09 * jcamins sighs.
16:09 jcamins I knew it was too easy.
16:10 gmcharlt jcamins: well, I think you're very close
16:10 wajasu I'm grabbing lunch then later you let me know and I'll pull and try again.
16:11 melia joined #koha
16:11 talljoy1 joined #koha
16:11 jcamins gmcharlt: I think I can probably use rebuild_zebra -k, use the export in the test, and then index that directly with zebidx.
16:11 jcamins I hope.
16:12 jcamins Mocking the data in would be... not fun.
16:12 gmcharlt jcamins: that should be close enough -- besides, it will be handy to have the actual MARC file(s) anyway
16:12 gmcharlt not to be too much of a pain, but are the records CC0 or the like?
16:13 jcamins They're the sample records from wizzyrea.
16:13 gmcharlt ok
16:14 jcamins Maybe we need a "sample data generator," which takes a bunch of records, and adds a bunch of items to them.
16:15 * jcamins does not volunteer to write it.
16:16 gmcharlt jcamins: could you do me a favor, grab the snippet of C from[…]ian-or-big-endian
16:16 gmcharlt then do gcc test.c; ./a.out
16:16 jcamins Sure.
16:17 jcamins little-endian.
16:18 jcamins How could an x86 machine be big-endian?
16:18 gmcharlt my amd64 is also little endian
16:19 wajasu i ran it in my koha debian VM and it said little-endian
16:20 wajasu uname -a => Linux biblio 2.6.32-5-amd64 #1 SMP Sun May 6 04:00:17 UTC 2012 x86_64 GNU/Linux
16:20 jcamins gmcharlt: bet Zebra was implemented using int rather than long.
16:21 jcamins I hate int.
16:22 jcamins Actually, I guess they'd have to use size_t, if they support little endian systems.
16:22 gmcharlt I attached an strace to the zebrasrv child process ... trying a search on my box causes it to die with a floating point exception
16:24 jcamins wajasu: I'm going to have to rebase the branch.
16:25 jcamins Well, I guess I could call it search_t2.
16:26 jcamins Oh, darn.
16:26 jcamins I forgot to do git rm.
16:27 talljoy joined #koha
16:27 jcamins Fear not. regular expressions to the rescue.
16:29 * chris_n bemoans spending ~8hours to write 12 lines of code
16:29 chris_n but loves the results
16:29 kf what are you working on?
16:29 kf :)
16:30 chris_n a webmin interface for pykota
16:30 datadoctor sounds cool
16:30 jcamins Woohoo!
16:30 kf what is pykota? :)
16:30 * chris_n has successfully ridden his network of Win32 servers
16:30 kf python... ?
16:30 chris_n yay!
16:30 jcamins Wait.
16:30 chris_n kf: its written in python
16:30 jcamins I forgot to add the data.
16:30 chris_n kf:
16:30 chris_n its print accounting backend for cups
16:31 chris_n probably very useful to libraries and academic institutions
16:31 kf oh interesting
16:31 wahanui interesting is sometimes good and sometimes bad
16:31 kf good this time :)
16:31 chris_n jcamins: data is always important
16:32 jcamins Also, it leaves .mf files.
16:33 adnc_ joined #koha
16:33 jcamins Wait... this is no good.
16:33 * jcamins moves things between scripts.
16:36 jcamins Wait...
16:36 jcamins Drat.
16:38 jcamins Great...
16:38 jcamins Now the test doesn't work on my system either.
16:39 kf oh no, even 25
16:40 jcamins Yay!
16:40 jcamins (regarding the registrations, not regarding the fact that Zebra hates me.
16:42 jcamins Ah.
16:42 jcamins Here's the problem.
16:46 kf ah, wrong window :)
16:49 eythian joined #koha
16:52 jcamins Yay!
16:52 jcamins New patch attached to the bug.
16:52 jcamins Note that you still require the fix for bug 7648 to use it.
16:52 huginn Bug[…]w_bug.cgi?id=7648 enhancement, P5 - low, ---, kyle.m.hall, NEW , staff can make themselves superlibrarians
16:53 jcamins This should work on any architecture.
16:53 jcamins I hope.
16:53 jcamins Note that it's a bit slower now.
16:57 gmcharlt jcamins: do you have it on a branch as well?
16:57 jcamins gmcharlt: I'll push it, one moment.
16:58 jcamins It's on search_t2.
16:58 jcamins On my github repo.
17:10 wajasu got search_t2
17:10 wajasu ran. but errors.
17:10 jcamins Same ones?
17:10 wajasu yes
17:11 jcamins Hm.
17:11 gmcharlt jcamins: once I finish setting up a new repo, I have a follow up patch for you
17:11 jcamins Same as the first run or same as the second?
17:11 jcamins gmcharlt: ah, I guess you were able to figure out what was wrong?
17:11 gmcharlt jcamins: yeah, I've got the test cases working for me
17:12 jcamins Yay! Progress!
17:13 jcamins So I guess set_context still isn't working for wajasu.
17:13 pastebot "wajasu" at pasted "bug 8649 search_t2 > perl t/db_dependent/Search.t #3" (42 lines) at
17:13 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
17:14 gmcharlt yeah, that's what my patch deals with
17:14 * gmcharlt waits not-so-patiently for a push to finish
17:14 jcamins Oh, the set_context is in the wrong place.
17:15 jcamins Also whatever problem you located.
17:16 gmcharlt jcamins:, branch bug8649
17:17 jcamins Fetching now.
17:18 jcamins Ohhh.
17:19 jcamins wajasu: gmcharlt's branch?
17:19 wajasu i see it
17:19 jcamins s/: /: could you try /
17:19 * jcamins sighs.
17:19 wajasu ok
17:19 jcamins English I speaking goodly do.
17:20 jcamins I'm going to blame that on C4::Search.
17:20 mbalmer joined #koha
17:20 druthb lol
17:20 gmcharlt jcamins: another thing I noticed -- if a test fails, the test case hangs waiting for zebrasrv to die ... which never happens
17:20 jcamins Ooh.
17:20 jcamins That'd be a problem.
17:21 mbalmer hi
17:21 wahanui que tal, mbalmer
17:21 wajasu jcamins: you want me to remote add the gmcharlt's koha-test?
17:21 jcamins wajasu: yes please.
17:21 wajasu ok
17:21 jcamins And if it works, you can sign off on those patches, and upload them to the bug.
17:21 jcamins Meanwhile, I shall ponder how to catch a catastrophic failure.
17:21 mbalmer kf?
17:21 wahanui kf is cait or really, really sweet. or <reply>she gives me memory loss or a holds expert.
17:22 jcamins Stick the entire thing in an eval {} ?
17:22 gmcharlt jcamins: sec, I'm trying something
17:24 jcamins gmcharlt: has my monomania rubbed off on you? :)
17:26 gmcharlt jcamins: wajasu: I've pushed a follow-up to my branch
17:27 kf mbalmer? :)
17:27 druthb kf?
17:27 wahanui kf is cait or really, really sweet. or <reply>she gives me memory loss or a holds expert.
17:27 jcamins druthb?
17:27 wahanui i heard druthb was a migration specialist and great friend or crazy as a loon.
17:27 wajasu gmcharlt: just did pull
17:28 druthb :D
17:30 pastebot "wajasu" at pasted "bug 8649 gmcharlt koha-test search_t > perl t/db_dependent/Search.t #4" (47 lines) at
17:30 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
17:30 melia joined #koha
17:30 gmcharlt wajasu: do you have KOHA_CONF set in your environment?
17:31 wajasu yes
17:31 gmcharlt what happens if you unset it?
17:31 wajasu hold on
17:32 wajasu same
17:32 gmcharlt wajasu: probably a silly question, but is idzebra installed on that box?
17:33 wajasu its koha-common  based
17:33 jcamins Oooh... the data directory isn't installed properly?
17:34 jcamins I shall build packages and find out.
17:37 gmcharlt hmm -- me just got it working on a koha-common based box as well
17:37 gmcharlt wajasu: another thing to try -- unsetting PERL5LIB
17:39 wajasu gmcharlt: no diff w/o PERL5LIB
17:39 kf left #koha
17:39 gmcharlt is there anyting in zebra.log?
17:40 wajasu empty
17:40 gmcharlt wahanui: in Search.t, try changing the log option from none,request to all
17:40 wahanui gmcharlt: i'm not following you...
17:40 gmcharlt then running the test again
17:40 gmcharlt second time ...
17:40 gmcharlt wajasu: ^^
17:41 wajasu ok
17:41 gmcharlt if that doesn't work, try deleting t/db_dependent/data/etc/koha-conf.xml before running the test again
17:44 jcamins If nothing shows up in the log, it means the fork isn't working.
17:44 jcamins (or the exec)
17:45 gmcharlt wajasu: you have /usr/bin/zebrasrv and /usr/bin/zebraidx, right?
17:45 wajasu its running
17:46 wajasu yes both
17:47 gmcharlt anything showing up in zebra.log now that the log verbosity is turned up?
17:47 wajasu yes
17:47 wajasu i should be in your branch right?
17:48 gmcharlt yes
17:48 wajasu hold on.
17:50 libsysguy joined #koha
17:50 jcamins wajasu: you don't need to have Zebra running when you start the test.
17:51 wajasu wait pasting  log
17:51 jcamins How did you build the packages? I can't build packages based on master.
17:52 pastebot "wajasu" at pasted "bug 8649 gmcharlt koha-test zebra.log > perl t/db_dependent/Search.t #5" (50 lines) at
17:52 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
17:52 jcamins Ohhh.
17:53 jcamins wajasu: kill all the Zebra processes running.
17:53 jcamins Then try again.
17:53 wajasu ok
17:53 jcamins (you can restart Zebra after you've killed all the processes)
17:55 gmcharlt also, I wonder why its trying to bind to port 9999 -- the koha-conf.xml generated by isn't uncommenting that out
17:55 gmcharlt or in, as the case may be
17:55 wajasu killed zebra.  now zebra log shows starting zebrasrv and entering event loop
17:56 jcamins gmcharlt: that's a good point.
17:56 jcamins wajasu: the log showed that only after you ran the test, right?
17:57 wajasu it must have, what else would have started it.
17:57 jcamins wajasu: nothing should, that's why I was checking that you hadn't manually started Zebra pointing at the log.
17:57 pastebot "wajasu" at pasted "bug 8649 gmcharlt koha-test zebra.log > perl t/db_dependent/Search.t #6" (66 lines) at
17:57 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
17:58 jcamins wajasu: wait... I'm confused.
17:58 jcamins That's the output you're getting?
17:58 jcamins With part of the log in the middle?
17:59 wajasu after the test, there is no zebra process running.  the zebra log doesn't show anything after :  [debug] Entering event loop.
17:59 gmcharlt jcamins: I've seen that before with the log level change I ask him to make -- I think zebrasv is spitting some of that to stderr
17:59 jcamins gmcharlt: oh, okay.
18:00 wajasu he had me change it to "all"
18:00 jcamins Right.
18:00 jcamins I didn't realize that always printed to stderr.
18:01 gmcharlt I think those are just before it opens its logfile
18:02 gmcharlt wajasu: what are the contents of t/db_dependent/data/var/li​b/zebradb/biblios/register on your system? (it should just contain a README file)
18:03 wajasu yes, but its called READMEE  with two EEs
18:04 * gmcharlt blinks
18:04 jcamins Did I do that?
18:04 wajasu no.
18:04 wajasu my typo
18:05 wajasu its there its fine.
18:05 gmcharlt and no other files, right?
18:05 wajasu nothing else
18:06 wajasu all the other zebradb places are the same.
18:06 rambutan joined #koha
18:11 gmcharlt wajasu: could you add the following line after the set_context() line, then rerun?
18:11 gmcharlt diag($context->{'listen'}->{'​biblioserver'}->{'content'});
18:14 jcamins Wait a minute.
18:14 jcamins wajasu: how do you have kohaclone mounted?
18:15 pastebot "wajasu" at pasted "bug 8649 gmcharlt koha-test zebra.log > perl t/db_dependent/Search.t #7" (105 lines) at
18:15 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
18:15 wajasu over sshfs
18:16 * jcamins bets sshfs doesn't support special files.
18:16 wajasu i'll pull it into the VM if you like.   or maybe i need a sshfs flag.
18:17 jcamins Give it a try.
18:17 gmcharlt jcamins: I think you have it -- I can't imagine a file socket working working over sshfs
18:17 jcamins If that's the problem, I'll just modify the paths to use a temporary directory on /tmp
18:17 cait joined #koha
18:18 nengard when using the marc export tool what encoding is it using? is it utf-8 by default?
18:18 jcamins nengard: defaults to utf-8, yes.
18:18 cait hi all
18:18 nengard thank you
18:19 maximep left #koha
18:19 wajasu i'm cloning my sshfs kohaclone mount to an inVM one,
18:21 jcamins I guess this means I need to create the directories on the fly, too.
18:22 wajasu All tests successful.
18:22 jcamins Woohoo!
18:22 gmcharlt yay
18:22 wajasu you guys are tenacious!!!
18:22 maximep joined #koha
18:22 * jcamins isn't quite sure how gmcharlt caught it.
18:23 wajasu it was all wajasu's fault :|
18:23 * jcamins is on a white whale quest to rid the world of C4::Search and bring excellent searching to Koha.
18:23 wajasu please, lets get it under /tmp or /var/tmp.  i like my sshfs
18:24 gmcharlt jcamins: tempdir() from File::Temp would be a good choice, methinks
18:24 pastebot "wajasu" at pasted "bug 8649 gmcharlt koha-test zebra.log > perl t/db_dependent/Search.t #8" (12 lines) at
18:24 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Needs Signoff , C4::Search needs a unit test
18:25 gmcharlt actually, hang on a sec
18:25 jcamins gmcharlt: that was my plan.
18:25 gmcharlt jcamins: cool, I'll leave you to it
18:25 jcamins If you have a better plan, though, I'm happy to change plans.
18:26 gmcharlt jcamins: I have a patch coming in a minute
18:26 jcamins Woohoo!
18:28 wajasu another idea.  jcamins, you had an issue posting a large patch.   what if you tar gzipped it.  would it be smnaller an d work?
18:28 jcamins wajasu: no, we'd still have socket problems.
18:29 wajasu with bugzila?  thats the issue?
18:31 jcamins No, the large patch wouldn't fix the socket issue you ran into.
18:40 wajasu[…]x_allowed_packet/
18:40 jcamins wajasu: turns out that there was no reason for such a large patch anyway.
18:41 wajasu i wondered why.
18:41 jcamins wajasu: I was trying to eliminate the need to reindex every time you run the test suite.
18:41 jcamins No dice.
18:42 wajasu is that the initial delay before output?
18:42 jcamins Yup.
18:43 jcamins Well, there's a delay for installing files, then a much longer delay for indexing.
18:43 gmcharlt jcamins: wajasu: I've pushed a follow up to my branch
18:45 wajasu pulled it. it ran
18:45 gmcharlt and from your sshfs clone, in particular?
18:45 wajasu i see Name "DBI::errstr"  used only once: possible typo at t/db_dependent/Search.t line 37.
18:46 wajasu whats up with that
18:46 jcamins wajasu: I'm not sure, actually.
18:46 gmcharlt just a minor warning
18:48 jcamins Works great.
18:48 jcamins gmcharlt++
18:49 gmcharlt wajasu: jcamins: I've pushed to a new branch called bug8649_signed_off
18:49 gmcharlt that includes a follow-up to quiet that warning
18:49 gmcharlt also includes my signoffs
18:49 gmcharlt wajasu: I'd appreciate it if you used that branch for generating your own signoffs
18:51 jcamins Thanks!
18:52 gmcharlt as far as the issue of dealing with patches that are too big to attach... I suggest just putting in a pull request instead
18:53 jcamins I also do that for the overly complex patches like bug 2060.
18:53 huginn Bug[…]w_bug.cgi?id=2060 enhancement, P3, ---, jcamins, Signed Off , Ability to import Authorities from Staff Client
18:53 gmcharlt we could borrow an idea from the Evergreen project and set up a working Git repo that on that anybody could push to
18:53 jcamins Oooh.
18:53 jcamins I like that idea.
18:53 gmcharlt if folks would prefer to avoid too much scattering about on github or gitorious
18:54 jcamins I don't mind having a dozen remotes, but it seems to me that would be helpful for developers who are newish, and maybe aren't 100% clear on setting up their own public repos.
18:55 wajasu so you want me to pull from bug8649_signed_off ,  run passing tests, then signoff referring to gmchart's git repo branch?
18:56 gmcharlt wajasu: almost -- add your signoffs to the commits, then push to a branch on one of your repos that the RM can get to
18:56 talljoy1 joined #koha
18:57 jcamins wajasu: you sign off on each commit, then either push it to a public repo or attach the patches to the bugzilla bug.
18:58 wajasu ok.  i'll try to push to the public repo.
18:59 jcamins wajasu: do you have a public repo?
19:02 wajasu test passed without warning. now i'll signoff
19:10 wajasu no public repo.  can i push to you?
19:11 jcamins wajasu: nah, best bet is to just attach to the bug, then.
19:11 jcamins All the patches are small enough now.
19:18 rambutan joined #koha
19:20 mtj wajasu, no public repo? just use github...
19:22 wajasu jcamins: i switched to gmcharlts branch, ran tests fine.  when i git commit --amend  it just mentions  "quiet a variable use warning" so i think its just a followup?
19:22 wajasu my format patch is 29 lines
19:22 jcamins wajasu: you need to use git rebase -i to sign off on each one.
19:24 wajasu here goes.
19:26 jcamins[…]n_a_pushed_branch
19:30 wajasu i did a git log and see 5 commits.  will that mean i'll have 5 patches to attach to the bugzilla?
19:31 edveal joined #koha
19:31 rangi Morning
19:31 jcamins Morning.
19:31 wahanui morning is a state of mind. or whenever the cat wants breakfast.
19:31 wajasu afternoon
19:31 jcamins wajasu: right.
19:31 jcamins Sorry, didn't see the question earlier.
19:35 wajasu I'm going to try[…]-git-format-patch
19:35 wajasu if thats ok.
19:35 jcamins We don't want to squash them all because gmcharlt wrote some of the patches.
19:36 wajasu so you want them able to pick and choose?
19:36 jcamins No, but it's important for history.
19:36 jcamins Oh, you can use this: git format-patch master --stdout > my_new_patch.diff
19:37 rangi Stealing authorship is bad mmmkay
19:38 jcamins Also, squashing patches is one of rangi's pet peeves. :)
19:38 rangi Yes yes it is
19:38 wajasu i'm just trying to help the push master :)
19:39 wajasu master pusher
19:40 rangi Seriously tho history is important, and knowing who authored code too
19:40 rangi For all sorts of legal reasons
19:40 wajasu i won't squash.
19:40 rangi Cool
19:41 rangi Squashing a series from one person is different
19:41 rangi But mixing gets messy
19:56 maximep left #koha
19:56 maximep joined #koha
19:57 adnc_ joined #koha
19:58 rangi My stop
20:03 rangi back
20:13 drojf joined #koha
20:13 drojf good evening
20:13 rangi hi drojf
20:13 drojf hey rangi
20:21 bag morning rangi
20:22 rangi heya bag
20:23 cait hi bag, hi rangi
20:23 bag hey cait
20:25 oleonard See ya #koha
20:25 cait bye oleonard :)
20:37 kathryn joined #koha
20:40 edveal joined #koha
20:41 wajasu rangi: after git amending each with [SIGNED-OFF] prefixing each 1st line, i ran format-patch -s origin/master and only the last one has has SIGNED-OFF in prefix.  is that OK?
20:42 rangi as long as the they have the signoff in the patch the subject line doesnt really matter
20:42 drojf joined #koha
20:42 wajasu it also generated 6 patches instead of 5. so they also have [PATCH 1 of 6]  though i am only uploading 2-6.  is tht ok.
20:43 rangi thats fine
20:43 rangi you can turn that off by -N for future reference
20:43 wajasu good to know.
20:44 adnc_ joined #koha
21:04 wajasu whew. glad to have bug 8649 signed off :)
21:04 huginn Bug[…]w_bug.cgi?id=8649 enhancement, P5 - low, ---, jcamins, Signed Off , C4::Search needs a unit test
21:05 rangi well done
21:06 jcamins Yay!
21:07 wajasu i'm going for a run.  then later will attempt to sign off on a couple that i tried, which folks have updated.
21:08 wajasu jcamins++
21:08 wajasu gmcharlt++
21:08 wajasu and it runs on my sshfs mount too.
21:13 wajasu no if someone could signoff on my bug 8378 patch, so that bug 8607, which i signed off on,  can get ready.
21:13 huginn Bug[…]w_bug.cgi?id=8378 critical, P5 - low, ---, koha-bugs, Needs Signoff , <fine> syntax not working on overdues anymore
21:13 huginn Bug[…]w_bug.cgi?id=8607 critical, P5 - low, ---, jonathan.druart, Signed Off , overdue_notices is broken
21:16 maximep left #koha
22:38 edveal left #koha
22:42 papa joined #koha
22:45 papa joined #koha
22:56 papa joined #koha
22:59 papa left #koha
23:41 wajasu ok bug 8382 signed off.
23:41 huginn Bug[…]w_bug.cgi?id=8382 blocker, P1 - high, ---, christophe.croullebois, Needs Signoff , It is not possible to cancel an order when receiving on a filtered page in
23:43 jcamins Does anyone have any idea what this regular expression does: (?-xism:^\s*)
23:45 jcamins Oh, xism are regexp options.
23:48 BobB_ joined #koha
23:51 Irma joined #koha

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary