IRC log for #koha, 2008-06-21

All times shown according to UTC.

Time S Nick Message
12:04 hdl salut eric
12:06 mc kados, atz: what about C4::Error ?
12:49 kados paul: I wrote you about 'display english news by default'
12:49 kados paul: and 'bugfixing scan index' didn't patch for me
12:49 kados paul: needs a rebase I think
12:49 kados mc: I haven't seen a patch for C4::Error, did you send one?
12:50 mc kados, no: there was no reaction about it
12:51 mc i wonder what all of you think about that
12:51 mc and about try_to_use
12:51 kados mc: I'm not the one to ask, atz, gmcharlt and acmoore would be better
12:52 mc sure.... what's the key binding to bug everybody on the chan ? ;)
12:52 mc well... thoughts from paul and hdl are welcomed too
12:53 acmoore I'll go back and re-read the emails. I can't recall if it was supposed to be an exception class or what...
12:53 mc perhaps i have to reactivate the feed on the mailing list ?
12:56 acmoore Oh, I see C4::Context::Error was just intended to be a variable. I can see that (or a class method) being a reasonable way to see if we have a valid context.
12:57 acmoore I'm not sure what atz meant by C4::Error, if that was just supposed to be a variable or what.
12:57 acmoore Since we don't have a or a plan C4 namespace, I'm not sure where he intended for that functionality to go.
12:57 acmoore mc, How do you picture $C4::Context::Error working?
12:59 kados paul: scan isn't working at all for me
12:59 kados paul: before or after your patch :/
13:00 paul strange, as it used to work previously
13:00 mc acmoore, $C4::Error would be very basic string
13:00 kados paul: working for you?
13:00 mc just set the $C4::Error string instead of warn it
13:00 mc as simple as a warn
13:01 mc but can be caught by the caller
13:01 paul kados : yes : http://i15.bureau.paulpoulain.[…]sort_by=relevance
13:01 paul (abel / abel)
13:02 kados :(
13:03 kados paul: looks like you haven't updated since yesterday, there was an improvement made to the languages footer
13:03 paul I rebased 10mn ago !
13:03 kados paul: but I don't think it causes scan to stop working :-)
13:03 kados oops, maybe that is the improvement
13:03 kados hmmm, maybe it's only on the OPAC
13:05 paul seems yes, as I seen the improvement on ;-)
13:05 acmoore mc, If it's set by the C4::Context class, it seems that it should be C4::Context::Error to me. All of the other interfaces into C4::Context appear to be class methods to me, though, so maybe doing that would make more sense. We have C4::Context->config(), C4::Context->preference(), C4::Context->dbh(), C4::Context->Zconn,  C4::Context->stopwords already. Perhaps C4::Context->error or somesuch would make more sense.
13:05 acmoore mc, but, maybe I'm missing the point.
13:07 kados paul: we have one final improvement to make to the staff and opac advanced search pages
13:07 kados paul: owen sent it to me yesterday and I'll be testing and pushing up today
13:08 paul kados : yesterday, on a fresh git install, I think i've found some problems with MARC editor, when adding multiple subjects/authors.
13:08 kados paul: it will break adv search into two types: 1 adv search (no boolean or extra stuff); 2. power search (boolean + other advanced parameters)
13:08 paul we tried to add something like 4 or 5, and, some were lost.
13:08 kados paul: this is to solve the problem some libraries have about what should display on adv search page
13:08 mc acmoore, in fact: the discution wasn't on the point to know how to call it but do we do it
13:08 kados paul: yikes ... can you reproduce the problem?
13:08 paul I could not investigate, but would be a pain to release with such a problem.
13:09 mc but your lines make sense: the better would be C4
13:09 mc C4::Context->error
13:09 kados paul: IMO, we should abandon the marc editor in favor of Biblios
13:09 paul was quite easy : UNIMARC install from scratch hide_marc=OFF. Add biblio, use + to repeat a field.
13:09 kados paul: for 3.2 perhaps
13:09 paul kados: maybe/probably, but later ;-)
13:09 mc hmm ...
13:09 mc no!
13:09 mc can it work ?
13:10 mc i mean: is it a way for C4::Context->error to be loaded even if $KOHA_CONF is missing ?
13:11 kados paul: OK, please submit a bug and mark it blocker
13:11 paul + we could not add a supplier. This one seems to be fixed now. did someone work on that yesterday ?
13:11 mc missing $KOHA_CONF was the key of the problem
13:11 kados paul: don't think so
13:11 paul (Perl error like "could not find Main::AddSupplier..."
13:11 paul )
13:11 kados paul: there are a few blocker acqu bugs that need to be fixed ... ryan should fix them today
13:13 acmoore mc, well, if you can figure out how to make $C4::Context::Error work, I think that writing sub error { return $C4::Context::Error } would work.
13:14 mc hmm ... w8: i test
13:14 kados paul: strange
13:19 paul kados : bug confirmed on a fresh git setup.
13:21 mc acmoore, it's ok C4::Context::Error is loaded as every functions
13:21 mc and returns a C4::Context::context->{error}
13:21 mc that is set during the module init
13:21 mc this can be the patch
13:22 mc oopps
13:22 mc returns a $C4::Context::context->{error}
13:22 mc sorry about the missing sigil
13:23 acmoore OK. Whatever you recommend.
13:24 paul kados :[…]w_bug.cgi?id=2265
13:28 mc acmoore, i'll report the discution on koha-devel. i'll patch if nonone is against
13:29 kados paul: thx
14:36 gmcharlt paul: patch submitted for bug 2265
14:36 paul great ! really !
14:41 kados gmcharlt, paul: pushed
14:41 kados gmcharlt++
14:41 gmcharlt gmcharlt-- # I added the bug in the first place
14:41 paul & kados ++ for pushing so quickly. & me++ for detecting it before RC1 ;-)
14:42 kados gmcharlt: you just did that to make the rest of us feel OK about our programming skills, right :-)
14:42 gmcharlt heh
14:53 ricardo Hi everyone!  :)
14:56 ricardo Can anyone tell me if "we" are still planning to release RC1 of Koha 3 today? (I'm asking because, if RC1 will be delayed, I would submit the INSTALL.opensuse file during the weekend)
14:57 acmoore I think we're intending to. kados?
14:57 kados ricardo: yes, we are
14:59 ricardo kados: OK. Thanks for the feedback. I guess that means I'll go home early today, eheh (I have already set up git in a Virtual Machine at home to submit the patch after copying my documentation about installation in the Wiki to a text file)
15:04 ricardo Paul / Galen: Will this bug be fixed for RC1 (not asking /demanding, just inquiring):
15:04 paul ricardo: which bug ?
15:05 ricardo[…]w_bug.cgi?id=2178
15:06 ricardo paul: This one ^^^  :)
15:07 paul gmcharlt will confirm, but I think that no. Note that a KNOWN ISSUE paragraph should be added to the Release Notes, and explain how to solve this install problem (which is quite easy to do manually)
15:08 gmcharlt ricardo: confirmed - just won't have time to get to it
15:08 paul (& it's simple to fix that manually, but very long to fix that in the installer)
15:08 ricardo paul: OK, thanks  :)
15:09 paul as Zebra/NoZebra is the last question, asked when all other setups are already done. there is a dirty easy way to fix it, but I think it's better not to use it
15:09 ricardo gmcharlt: OK, thanks for the info
15:10 ricardo paul: right... An additional problem will probably be to notify the user to update NoZebraIndexes variable if/when setting NoZebra = ON
15:35 dkg when i do a "find guarantor" search on a koha installation (from git, up-to-date a week ago), i can't seem to get it to return anything.
15:36 paul time to leave for me. happy hacking & releasing.
15:36 dkg i'm in the "modify child patron" screen, and i click on the "Find Guarantor" button.
15:37 dkg a dialog box pops up, titled "Search for guarantor", i put in the name of a known patron, hit search, and it comes up "0 patron(s) found"
15:37 paul_away kados : i've resubmitted 3 patches. I hope they can apply now (+ you'll accept "display english news by default")
15:38 kados paul_away: thanks, have a good evening
15:38 dkg Any ideas what i should look for?  should i file it as a bug?
15:38 paul_away weather very good. beach planned !
15:39 kados dkg: hmmm ... can you reproduce on ?
15:40 dkg kados: is there a child acount on that server?
15:40 dkg ah: i can make one.
15:43 dkg kados: huh.  no, i get a hit when i use
15:45 dkg what version is running?
15:49 dkg on adult patron is "baz fubar", child patron is "qux fubar", i can do a "search for guarantor", and link them cleanly.
15:49 dkg on my install from git, no such luck.
15:51 kados dkg: please submit a blocker bug to bugzilla (
15:52 kados wait
15:52 kados it DOES work at
15:52 kados hmmm
15:52 kados dkg: just a sec, I'll update it
15:52 kados dkg: and you can test again
15:54 kados dkg: OK, I've updated, please test again and let me know if you have trouble finding guarantor
15:55 kados dkg: it may have been a recently introduced bug
16:05 dkg kados: thanks for updating!  It still works for me on (no bug)
16:06 dkg testingn further on my installation, it's actually working for other patrons as well.
16:06 dkg So actually, i should revise this: it appears that there are some patrons who don't show up, but others do.
16:06 dkg does the "find a guarantor" interface refuse to show library staff?
16:09 kados dkg: possibly
16:10 gmcharlt dkg, kados: just checked - it looks like it restricts itself to patrons whose category's type is A (Adult)
16:12 dkg gmcharlt: ah, thanks.  that explains it!  the patron i was searching for was myself (type "S") :P
16:13 dkg maybe that could be documented in the UI someplace?  Any reason staff couldn't be a guarantor?
16:13 hdl acmoore: sorry to bother you.
16:13 acmoore hdl, hi.
16:14 hdl for your sms stuff to work, we should add SMS::Send to dependencies should we not ?
16:15 acmoore hdl, yes. Did I fail to do that? I think so. I'll fix that.
16:15 acmoore Thanks for pointing that out.
16:17 acmoore hdl, offhand, do you know if I need to edit more than Makefile.PL?
16:18 hdl too
16:19 hdl maybe gmcharlt knows better.
16:20 gmcharlt hdl, acmoore: those two will suffice
16:21 ricardo gmcharlt: : About Makefile.PL and Algorithm::CheckDigits - is the comment in the code about "awaiting package maintainer's use of $VERSION" still current / valid?
16:23 gmcharlt ricardo: it appears that as of version 0.50 of the package, the maintainer has put VERSION back in
16:24 ricardo ricardo: So, should that line in Makefile.PL be un-commented?
16:24 ricardo (above line meant to gmcharlt and not to myself, obviously!  ;-)
16:27 dkg gmcharlt, kados: i just added a note to[…]w_bug.cgi?id=2221 about documenting the guarantor search limitations.  hope that's OK.
16:27 gmcharlt ricardo: probably - Algorithm::CheckDigits can be 0.50; however, Algorithm::CheckDigits::M43_001 does not set $VERSION
16:27 gmcharlt dkg++
16:29 ricardo gmcharlt: I think that you NO longer need to install Algorithm::CheckDigits::M43_001 by itself anymore (installing Algotihm::CheckDigits now "works" - does not return an error message anymore)
16:30 gmcharlt ok, as long as it picks up M43_001 during a CPAN install, great
16:31 ricardo gmcharlt: Yes, I think that id did pick it up in an installation I did in openSUSE 11 RC11 last week or so...
16:31 ricardo (via CPAN, yes)
16:32 ricardo s/RC11/RC1  )openSUSE did NOT have that many Release Candidates! ;-)
16:32 gmcharlt lol
16:33 ricardo Eheheh
16:56 mc i would like to add a default value for the session driver string
16:57 mc to share it between the *Auth*.pm
16:57 mc someone can tell me the best way to do it ?
16:58 mc (i think that it has to be context independant but i'm not sure)
17:02 gmcharlt mc: is this regarding using the same session between and during a database upgrade?
17:13 gmcharlt paul, hdl, or mc: could you give me a translation for "Allow hold requests to be placed on items that are not on loan' (or 'that are on the shelf')
17:14 gmcharlt it's for a new syspref, AllowOnShelfHolds, to specify whether hold requests can be placed on any available item or only those items that are currently on loan
17:28 acmoore hdl, now that slef pointed it out, I guess SMS::Send is an optional dependency. It fails to send SMS messages without it, but in a pretty controlled way. So, I guess that additionb should not be made. What are your thoughts?
17:30 gmcharlt acmoore: IMO leave it in - the installer does not have a good way to tell an inexperienced user when something is optional
17:30 gmcharlt and if they decide to use a feature later on, it is not yet documented in any way comprehensible to a non-programmer which Perl modules are needed for which optional features
17:33 gmcharlt re extensive indentation cleanup, probably better to separate whitespace-only patches from those that would have a functional effect
17:35 acmoore I would, but that would violate the policy about refactoring just for the sake of refactoring. I figured since it was ONE STATEMENT that it would be reasonable to adjust the whitepace. I honestly try to make this stuff reasonable, rational, and considerate to other developers, but I feel like I'm danged if I do and danged if I don't.
17:46 ricardo OK. Time to go home to edit that INSTALL.opensuse file. Take care everyone!  :)
20:03 mc re
20:04 mc gmcharlt, not the same session
20:04 mc just the session driver string
20:04 hdl gmcharlt: translaton :
20:04 mc that is use to initialize the session persistance
20:06 hdl gmcharlt: Permet de faire des réservation sur des exemplaires qui sont 'exclus du prêt'  (ou qui sont 'en rayon')
20:06 gmcharlt hdl: thanks - please email it though, the diacriics are not coming across in my IRC client
20:07 mc hdl: loan ? shelf ?
20:07 mc i was unable to translate, in fact :)
20:08 gmcharlt hdl: and while you're at it, could you translate the following for the other new syspref, AllowHoldsOnDamagedItems: 'Allow hold requests to be placed on damaged items'
20:08 acmoore kados, thanks for pushing in my "Adding an automated test for POD correctness." patch. I'm working on some tests that go in the "xt" directory that do some basic sanity checks that should be run before we check in code.
20:09 kados acmoore: sounds good
20:09 mc permet de réserver les examplaires endommagés
20:09 owen I'm looking at the new patron messaging settings pages and wondering what goes in that table...does that get populated only if you have notices defined?
20:09 acmoore kados, the full test suite is unweildy to run repeatedly during development, so I'm hopeful that in addition to the tests that touch the particular part of the application that is being changed, the xt tests can be run each time.
20:10 kados acmoore: sounds very sensible
20:10 acmoore kados, then, perhaps a buildbot will come in behind and run the full test suite on whatever gets pushed in. But that's a pipe dream for the not too distant future.
20:10 mc oops ... gmcharlt: the french sentence was your translation
20:10 kados buildbot++
20:11 kados owen: did you update before or after the updatedatabase patch I just pushed up?
20:11 mc is it possible to use buildbot with git ?
20:11 acmoore mc, I think so. It appears to have a connector (or whatever they're called) that can be run after a git-pull or whatever.
20:11 kados owen: you'll also have to change your version in sysprefs to 070 to trigger the update again
20:12 kados oops, 090 i think
20:12 mc cool
20:12 acmoore mc, or if not, then I'll "just" write a cronjob or something to git fetch, run the test suite, and report somehow, like to smolder or IRC or email or something.
20:12 mc ok
20:13 acmoore mc, have you used buildbot before?
20:13 ricardo kados: Hi! I'm almost finishing the INSTALL.opensuse file  :)
20:13 owen kados: I updated, but I'm not seeing anything different in the interface. still shows a table with only a header ("days in advance, sms," etc) but no contents
20:14 mc no but the  java dev team at university was fan of it
20:14 acmoore mc, because I tried to install it the other evening and was overwhelmed at the complexity.
20:14 mc they also was svn fans
20:14 acmoore mc, Oh. I think it has lots of fans. I'd like to be one, too.
20:14 mc so perhaps they aren't references :)
20:14 mc hehe
20:14 acmoore ha! well, I don't think svn is that bad.
20:15 kados acmoore: should I have the ability to do more than add/edit my SMS number on the opac?
20:15 hdl is there  a buildbot for PERL ?
20:15 mc i don't know
20:15 kados acmoore: looks like there's a table with just a header, but no contents (just like Owen mentioned)
20:15 hdl seems there is for python.
20:15 kados acmoore: do we have to install the sample data or something?
20:16 mc just remember that buildbot drive its sysop crazy :)
20:16 acmoore kados, yeah, the example data needs to be installed for it to make much sense. What's the procedure for that?
20:16 acmoore hdl, they're python scripts, but you can use it to run a perl test suite.
20:17 kados acmoore: i don't see anything in admin or tools, is there a way to generate the sample data somehow on the staff side?
20:17 kados acmoore: we don't have a mechanism other than updatdatabase with which to add sample data
20:17 acmoore kados, there's installer/data/mysql/en/optional/s​ample_notices_message_attributes.sql and installer/data/mysql/en/optional/s​ample_notices_message_transports.sql which must be run in alphabetical order.
20:18 kados actually, you can run through the installer again from scratch
20:18 acmoore kados, oh, and installer/data/mysql/en/optional/sample_notices.sql
20:18 kados nope ... doesn't take me to the sample data step
20:18 acmoore I don't really understand how the mandatory ones get applied, but installer/data/mysql/en/manda​tory/message_transport_types.sql ought to be.
20:19 acmoore kados, these are all from[…]5ae396b4016877ec2
20:19 kados acmoore: *nod* ... I see them
20:19 kados just puzzling over how we get them into an already installed systm
20:20 kados hdl: do you have an idea?
20:20 hdl Indeed, adding data samples to already installed system is a pain now.
20:21 gmcharlt practice usually has been to dupe the SQL inserts in, similar to the way the sysprefs are added both to the SQL scripts and
20:21 acmoore anyone else getting "Can't use an undefined value as an ARRAY reference at /home/acm/koha/dev/C4/ line 150" in their staff interface now?
20:21 gmcharlt admittedly that's not ideal
20:21 kados acmoore: yes, and cnighs is working on that I think
20:21 acmoore gmcharlt, I can do that if that's what is recommended. Or, is there something in C4::Installer that I can call with the filename?
20:21 hdl If we could get to installer step 3 on an installed system, it could be OK.
20:21 acmoore kados, OK good.
20:21 kados acmoore: it only happens when you have a db update to do in my experience
20:22 gmcharlt acmoore: yeah, there's a method in C4::Installer you could call to do that
20:23 gmcharlt acmoore: C4::Installer->load_sql, after using  sample_data_sql_list() and parsing the results to get the correct path
20:23 atz btw, just got an error on rebase:
20:23 atz DBD::mysql::db do failed: Duplicate entry 'OPACShelfBrowser' for key 1 at /home/atz/koha/production/koha/inst​aller/data/mysql/ line 1616.
20:24 gmcharlt you'll know language ('en' as default, section, and filename)
20:24 kados atz: that's harmless
20:25 acmoore gmcharlt, I'm following you. I'll see what I can whip up. thanks.
20:25 gmcharlt acmoore: feel free to add a method to C4::Installer to locate a script, given a known name
20:26 acmoore gmcharlt, sql_file_list and sample_data_sql_list seem very similar. What's the difference?
20:26 acmoore besides the structure of the returned data
20:28 gmcharlt sql_file_list returns all scripts, including MARC frameworks and sample data scripts
20:28 gmcharlt sample_data_sql_list is just a subset
20:28 acmoore ah. OK.
20:35 ricardo Darn... Sorry people. Newbie git question.
20:35 ricardo I already read[…]lopment:git_usage
20:35 ricardo Created a branch
20:36 ricardo Created an INSTALL.opensuse file inside the directory in that branch
20:36 ricardo wrote "git-add INSTALL.opensuse" (no error message)
20:36 kados so far so good
20:37 ricardo kados: And now, should I run git-commit ?
20:37 kados ricardo: yep
20:37 ricardo Ah, this time it seems to be working. Strange. Thanks!  :)
20:37 kados then git format-patch origin
20:37 kados and finally git send-email
20:38 kados mc: ?
20:38 mc do you think it's a good name for an env varaible ?
20:38 kados mc: what does it do?
20:38 mc (i'm writtibng ma patch)
20:38 kados who what where? :-)
20:38 mc $session = new CGI::Session("driver:MySQL;serializer:yaml;id:md5", ...
20:39 mc ;serializer:yaml;id:md5
20:39 mc with a default i will hardcode in C4::Context
20:39 ricardo kados: Done git format-patch origin. Now I should mail this. Right?
20:40 kados ricardo: yes, and use git send-email
20:40 mc the idea is that this string
20:40 atz ricardo: git send-email
20:40 kados atz++
20:40 ricardo atz: Thanks for the tip!  :)
20:40 mc - appears twice in installer/
20:40 mc - appears twice in
20:40 kados mc: are you hoping to get this change in before RC1?
20:41 kados mc: otherwise, I'm afraid I'll have to ask you to wait until after RC1 to ask about it ;-)
20:41 mc - aren't the same for the moment
20:41 mc - could me averwritten by sysop
20:41 ricardo "Who should the emails be sent to?". Do I just press ENTER here?
20:41 kados ricardo: send to
20:42 mc hmm ... how to write a blocking bug ?
20:42 ricardo kados: Oops. Right. That info is in the wiki. Sorry about that.
20:42 mc this problem is responsible to not install correctly on debian
20:42 mc can hdl confirm that ?
20:42 ricardo "Message-ID to be used as In-Reply-To for the first email?"  ?!
20:42 kados ricardo: just hit enter Ithink
20:43 ricardo kados: Right. Done!  :)
20:43 mc kados, the fact is that i can't run the web installer because of a problem if session
20:43 mc everything is installed afail
20:43 mc afaik
20:44 kados ricardo: haven't received yet, hang on until it comes through so we can verify ti worked
20:44 ricardo kados: Right!  :)
20:44 mc kados, i can write a stupid quick patch
20:45 mc just applying the same value everywhere
20:45 mc but this is at least
20:45 mc and before the RC1 i think
20:45 hdl kados : mc is right.
20:45 ricardo kados: You'll probably NOT get it... I see that I have errors in /var/log/mail
20:46 hdl Wefound that CGI::Session just changed the way it was used.
20:46 hdl And that serialize:yaml;id:md5 was necessary on a fresh install
20:46 kados hdl: 'just changed' ... you mean today?
20:47 kados hdl: or some time in the recent past?
20:47 hdl some time in the recent past.
20:47 kados and I think gmcharlt is the one who added the yaml stuff, but I could be wrong
20:47 ricardo kados: I'm using a Virtual Machine with the domain set up as an internal one (koha.test)
20:48 hdl So that we usually donot have to add those since they were defaulted.
20:48 ricardo I have this in /var/log/mail :
20:48 ricardo Jun 20 22:43:33 tygra postfix/smtp[5712]: 9DA62532A6: to=<>,[]:25, delay=31, delays=0.46/0.14/16/14, dsn=4.0.0, status=deferred (host[] said: 450 <root@tygra.koha.test>: Sender address rejected: Domain not found (in reply to RCPT TO command))
20:48 gmcharlt kados: I added the YAML stuff months ago
20:48 gmcharlt kados: one issue is that the Debian package for CGI::Session dropped the yaml serializer
20:48 kados ricardo: hmmm ...
20:48 gmcharlt somebody (dkg?) filed a bug with Debian to fix that
20:48 acmoore ricardo, it looks like your machine can't send email very well. kados would it be OK for him to attach that patch file to a regular email and send it with his regular mailer?
20:49 hdl But on fresh installs from CPAN, opening a Session is quite tricky.
20:49 kados acmoore: good idea
20:49 kados ricardo: can you zip the file and email it ?
20:49 kados as an attachment
20:50 ricardo kados: Sure!  :)  Let met try a thing first
20:50 ricardo acmoore: Thanks for the tip, also!  :)
20:51 acmoore sure. In fact, I'm not really sure if you have to zip it aas long as you make it an attachment and not just pasted into your message.
20:52 kados acmoore: i've seen oddness with unzipped patch attachments in the past
20:52 kados acmoore: mail servers do weird things with stuff sometimes ...
20:52 acmoore kados, oh very well.
20:53 acmoore actually, this is an entirely new file, right? He could send the file itself to one of us to add, commit, and email a patch. although, that's not too great of a solution.
20:53 kados acmoore: yea, that'd work too, but I'd like to get his credits into Git
20:53 acmoore aye.
20:54 kados cnighs seems to have dropped off
20:54 kados I wonder if that means he's done for the day
20:55 kados acmoore: you have anything pressing atm? if not, can you fix that error that happens on the opac ?
20:55 gmcharlt kados: got a good news/sort of thing for you
20:55 gmcharlt kados: re bug 2248 (displaying in transit status in search results), I got a fix just about ready to go
20:55 kados acmoore: to reproduce, artificially increment and load the OPAC page
20:56 kados gmcharlt: wow, that _is_ good news
20:56 gmcharlt kados: the bad news is that because the intransit status is not mapped to the MARC record ...
20:56 kados *cough*
20:56 gmcharlt my current fix involves querying the branchtransfers table for each item in teh seach results
20:56 acmoore kados, I'm going to add something to load up the messaging sample data on upgrade so that table looks better and it works. OK?
20:57 kados acmoore: yes, please do
20:57 ricardo kados: I think I did it!  :)  I edited my  /etc/postfix/  file, restarted postfix and ran the  git send-email  command again. Could you check it please?
20:57 kados gmcharlt: hmmm ...
20:57 kados gmcharlt: this is on the search results page?
20:57 gmcharlt yes
20:57 gmcharlt I've got a separate fix for hte OPAC bib details page, but that doesn't have any performance implications
20:58 acmoore ricardo, I don't see it yet. It should show up at[…] if you have sent it to
20:58 kados ricardo: not yet, but I'll keep my eyes peeled
20:59 ricardo acmoore: OK. Many thanks for that info!
21:02 ricardo Well.... At least, I am receiving the Cc: of the Patch to my e-mail address (that is NOT managed by the Postfix in the Virtual Machine)
21:02 acmoore I don't think you need to be subscribed to patches@ to post there, but I could be wrong.
21:06 kados ricardo: got your patch!
21:06 ricardo kados: Great!  :)
21:07 ricardo Basically, I had to lie a lot in my /etc/   ;-)
21:07 ricardo I mean  /etc/postfix/
21:07 atz what's the interactive git command?
21:08 ricardo acmoore: You're right. It is NOT necessary to be subscribed to patches@ to post
21:08 kados ricardo: actually, hdl had to manually approve your email ;-)
21:08 acmoore atz, there's "git rebase --interactive" and "git add --interactive"
21:08 kados ricardo:[…]?p=Koha;a=summary
21:08 kados ricardo: your patch has entered the Koha project, congratulations :-)
21:09 ricardo kados: That's great!  Thank you very much!  :D
21:09 kados ricardo: no, thank you!
21:09 ricardo :)
21:10 ricardo kados: So, did hdl had to approve /add me to the list of subscribers?
21:10 kados ricardo: if you sigh up to the mailing list he won't have to in the future
21:10 hdl yes*
21:10 kados ricardo:
21:10 ricardo hdl: OK. Thanks. And sorry about that!
21:11 hdl ricardo: you are not subscribed.
21:11 hdl But you are allowed to send patches.
21:11 kados ricardo: you could also add a note to our wiki explaining that :-)
21:11 ricardo hdl: Right. Not to that one. I may subscribe to it and DISABLE to e-mail delivery so I can send patches but NOT get flooded with commits, right?
21:11 kados ricardo: if you're in a contributing mood :-)
21:12 ricardo kados: Why not? :)  I'll do it in a while (I need to get dinner first - it's 22H12 here in Portugal!)
21:12 ricardo One thing that seems to be missing in the wiki is the  git-add command (necessary to submit NEW files, I think... like in my case)
21:13 hdl ricardo: yes you can also do that.
21:14 ricardo hdl: eheh
21:17 ricardo OK. Dinner time. kados  what time is it there?
21:17 kados ricardo: in ohio it's 5:18
21:17 kados pm
21:18 ricardo kados: So, I only submitted the file about 10 minutes late (5pm of your time). Not bad!  ;-)
21:18 kados hehe
21:18 ricardo :)
21:18 ricardo OK. Dinner time. See you later!  Thanks for all the useful tips for this Git Newbie!
21:31 acmoore gmcharlt, ping?
21:31 kados acmoore: he just stepped out
21:31 acmoore thanks. not urgent.
21:32 acmoore kados, I have one for you, though. If I add some code to to run the .sql files I added with optional and mandatory data in them, shall I make that a new database version, or tuck it back into .091 that has the rest of the enhanced messaging stuff?
21:33 kados acmoore: you can tuck it back into 091
21:33 acmoore very well, thanks.
21:33 kados acmoore: everyone who's updated to 091 is on IRC so you can just tell them to roll back and re-update
21:33 kados :)
21:33 acmoore OK.
21:34 acmoore kados, I figured I'd use the methods to load SQL files present in C4::Installer. But, doesn't use C4::Installer yet. You don't see any reason why it can't do you?
21:35 acmoore kados, I don't think it's a problem, but new installations and upgrades are sort of odd edge cases that I'm likely to miss things on.
21:36 acmoore that's what I was going to ask gmcharlt since he wrote C4::Installer recently. If nothing jums out at you, I'll just assume I can do it this way and see if he votes either way when he gets back.
21:36 kados acmoore: no, should be fine to use Installer
21:36 kados go for it
21:36 kados won't hurt anything, and will actually give us a useful procedure for loading large SQL
21:36 kados without pasting it in directly into updatedatabase
21:50 kados mc: so you are hoping to have a fix for 2216?
22:04 atz gah... TmplTokenizer is some garish code!
22:05 atz it actually has an array of paired \012
22:05 atz it actually has an array of paired \012\201 style values for it's own @latin1_utf8
22:06 atz crazy!
22:08 hdl kados gmcharlt : imho, independentBranches should be splitted into 3 (at least systempreferences)
22:08 gmcharlt back
22:08 mc re
22:08 mc kados, i'm on the work to it
22:09 hdl One For Circ, one for budgets, one for member management, and one for cataloguing
22:09 hdl Would that make sense for you ?
22:09 hdl Not now, but for RC2 or 3.2
22:09 mc i just thought about it
22:10 hdl I could send an RFC, if you like
22:10 kados hdl: gmcharlt has an RFC for that
22:10 mc i'll just provide a minimalistic patch
22:10 kados hdl: it's called 'System Groups'
22:10 kados hdl: can you reply to that with your suggestions, so they can be incorproated?
22:11 mc and propose something better for next koha release
22:11 mc but for now, i want to sleep
22:11 mc 'night koha people
22:15 atz later mc
22:19 ricardo Back
22:20 kados please submit all remaining patches before 7pm EST
22:27 ricardo kados: That will be midnight here in Portugal. So I was actually right when I thought that I could send patches until midnight!  ;-)
22:29 kados ricardo: :-)
22:29 ricardo Eheh
22:29 kados ricardo: yes, but I have to say 5pm to make sure I have time to fit it in before midnight :-)
22:30 ricardo LOL! Ok. Fair enough
22:31 ricardo kados: On a semi-serious note, I'm wondering if this shoud be called "Beta 3" instead of "RC1". When I think of a "RC1", I think that all "major" / "critical" / " blocker" (known) bugs are solved (and I think that's NOT the case here, in spite of a REMARKABLE effort by the developers)
22:32 kados ricardo: I agree, but I am getting pressure from paul, hdl, and others to release RC1
22:32 kados ricardo: in fact, pressure to release 3.0-stable ...
22:33 ricardo kados: Fair enough.
22:33 ricardo kados: Right... Well, I agree with others that one thing that helps there is declaring "Feature Freeze". I believe this has been done during the last month  :)
22:34 hdl kados : not only us but also many customers.
22:35 ricardo hdl: Hey, you're awake!  ;-)  That's true. I'm one of those "customers" (actually, someone that works at a place that has a Library running Koha 2.2.9 - migratedfrom a Proprietary system - and trying now to migrate it to Koha 3)
22:37 ricardo kados: If you know what will be the exact URL of the .tar.gz for RC1, I can send you a patch (the wget line for that URL). It's not critical, obviously.
22:39 kados ricardo: ahh, right ... good point, yes I do know it
22:40 ricardo kados: fire away!  :)
22:40 kados ricardo: koha-3.00.00-stableRC1.tar.gz
22:40 kados T-MINUS 15 MINUTES TO 3.0 PAKAGING
22:40 kados PACKAGING I MEAN :-)
22:41 ricardo[…]-stableRC1.tar.gz
22:41 ricardo Right?
22:42 kados ricardo: yep
22:45 ricardo kados:  Thanks! Patch submitted :)
22:50 kados T-MINUS 8 MINUTES
22:52 ricardo kados: Is my patch in your mailbox?
22:53 kados ricardo: it's already pushed
22:53 ricardo kados: The one with the change in the wget URL? I can't see it at[…]?p=Koha;a=summary
22:54 kados ricardo: it updates every 5 minutes
22:54 kados ricardo: so wait a bit, it'll show up
22:54 ricardo Ah, OK
22:54 ricardo Didn't know that  :)
22:55 ricardo kados: Will you also approve all the translations waiting for moderation at and add the corresponding .po files to RC1?
22:56 kados ricardo: yep ... once I get all remaining patches from everyone I will:
22:56 kados 1. review all bugs in bugzilla, to validate there are none already fixed
22:56 ricardo kados:  Right. That's great
22:57 kados 2. approve all translation moderation, and update with changes since beta2 (so translators can hash them out before final release)
22:57 kados 3. package 3.0 RC1 and test installation and translations
22:57 kados 4. revise release notes and email the list, upload the package
22:58 kados 5. go home and watch television :-)
22:58 ricardo LOL
23:03 kados down to 365 bugs from 404 this morning
23:04 ricardo OK. 365 bugs? That could gives us a new motto: "1 bug per day"  ;-)
23:05 kados hehe
23:06 ricardo s/gives/give
23:08 kados hey owen
23:08 kados owen: any last minute patches you wanna sneak in before i roll this RC1?
23:12 owen No, I'll just have to hold my horses, since I can't jump in at the moment :)
23:12 kados :)
23:13 kados owen: 40+ bugfixes today
23:13 kados we should do a release every week :-)
23:13 owen I know you don't mean that!
23:13 owen Now just imagine how the Firefox people felt this week
23:14 ricardo owen: LOL! True, true...
23:16 ricardo owen:  And also openSUSE 11 (released yesterday  :)
23:19 kados yea, I can't imagine the pressure of being an RM of such a large project
23:19 kados though, to be fair, they have a fair amount of resources in that project :-)
23:19 kados doesn't compare to our puny dev team on #koha :-)
23:22 ricardo kados: How many developers does Koha have contributing code for it? Ten?
23:24 kados ricardo: actively, maybe 12 or so
23:24 ricardo kados: Right...
23:24 ricardo Thanks for the info
23:24 ricardo kados: Is there a way in http://git.koha to browse files like they were in, say, the "3.0 Beta 1" release?
23:25 kados ricardo: yep
23:25 kados ricardo: scroll to the 'tags' at the bottom
23:26 ricardo Right... I think the problem was that I scrolled to tags and then clicked on "tag". Clicking on "shortlog" seems to give better results (and only THEN clicking on "tree"). Right?
23:45 kados ricardo: yes
23:45 kados OK, all bugs reviewed, and some were able to be closed out
23:45 kados we're don'w to 354 bugs remaining
23:45 kados down I mean
23:45 kados now comes the hard part :/
23:46 ricardo kados: No one said that a Release Manager's life was easy!  ;-)
23:52 kados final sweep of bugzilla, we're down to 350
23:53 ricardo YAY!  :D
23:53 kados translations next
23:53 gmcharlt kados: speaking of bugs, 2254, 2206, 2171 on my list I should be able to close before ALA
23:53 kados gmcharlt++
23:53 ricardo gmcharlt: Hi Galen!  :)
23:54 ricardo ALA is some kind of Conference, right?
23:54 gmcharlt (post RC1, I know, but shouldn't require any string changes)
23:54 kados gmcharlt: *nod*
23:54 kados gmcharlt: that's fine with me
23:54 gmcharlt ALA = AMerican Library Assocation's conference
23:54 ricardo gmcharlt: Right. Thanks for the info!
23:54 kados gmcharlt: we also need to kick off 3.2 at some point soon
23:54 kados gmcharlt: post-ALA I'm sure
23:56 hdl when is ALA ?
23:57 gmcharlt hdl: Anaheim, California (i.e., home of Disneyland)
23:57 gmcharlt 26 June to 2 July
00:10 kados translations approved
00:14 ricardo kados:  Great!  :)
01:34 ricardo Time to go to sleep. Congrats for the great work kados!

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