IRC log for #koha, 2008-07-15

← Previous day | Today | Next day → | Search | Index

All times shown according to UTC.

Time Nick Message
13:20 nengard owen did you make an major changes here: http://git.koha.org/cgi-bin/gi[…]a0ef90142caca3afb - anything I need to re-document?
13:21 owen No changes in functionality, but the entry form is better-formatted. You'll probably need to update your screenshot
13:24 paul hi nengard&owen
13:24 nengard hiya paul
13:24 owen when you say the 'entry form' do you mean every tab? or something else?
13:24 paul owen, a librarian asked me a good question: why isn't here a form on top of additem.pl as there is on top of other screens ? (the quick queries tab)
13:25 I had no good answer, and think it's just a lacking form
13:25 am I right ?
13:25 owen Our reasoning (kados and I) was that addbiblio and additem are very focused operations that are less likely to be interrupted with day-to-day operations like check-in and check-out
13:26 ...and that those screens needed the screen real estate more than others do
13:27 paul yes, but when you are doing bulkmodifs on items (like changing callnumber), it would be convenient to be able to search catalogue when on additem.pl (search cn1 => edit item => search cn2 => Edit item => search cn3 =>Edit item ...)
13:27 atm, you have 1 more click :
13:27 search cn => edit item => Search => search cn2 => ...
13:29 kados quick poll: would you rather I release Koha 3.0 with current 'blocker' serials bugs, and with no support for item-level holds, or wait two weeks until serials and holds are fixed?
13:29 (hi)
13:29 paul hi kados.
13:29 + with a major problem in CGI::Session cpan package
13:29 kados *nod*
13:30 owen Man, nothing's easy is it?
13:30 kados (thout that has been noted in installation and there is a workaround)
13:30 paul you'll be surprised : i would vote "No". We are in july, no french libraries pressure to have something released for 1 month ;-)
13:31 + that would let more time to translators to do their job
13:32 nicomo : kados is doing a quick poll :
13:32 [15:29] <kados> quick poll: would you rather I release Koha 3.0 with current 'blocker' serials bugs, and with no support for item-level holds, or wait two weeks until serials and holds are fixed?
13:32 I answer:
13:32 [15:30] <paul> you'll be surprised : i would vote "No". We are in july, no french libraries pressure to have something released for 1 month ;-)
13:32 [15:31] <paul> + that would let more time to translators to do their job
13:32 your opinion ?
13:32 nicomo I agree with paul
13:32 better have something "cleaner" even if it means waiting a bbit longer
13:33 paul kados : note I  consider you'll be really strict & accept only patches that are true patches.
13:33 & we should create a new branch immediatly
13:34 kados OK, but Galen is out this week
13:34 so new branch shoudl wait until he gets back at least
13:34 paul no prob, just don't validate patches that are not bugfixes...
13:35 kados not have been ... good :-)
13:35 paul: glad to hear that
13:36 some patches have been submitted that I haven't accepted for instance, some from Michael, about adding support for multiple IP ranges for branch identification
13:36 which was neat, but not a 3.0 feature IMO
13:37 paul did you accept Barcodes - OO replacements, extensible modules, tests from joe ?
13:37 (jul 3rd)
13:38 kados hmmm
13:38 atz paul: that doesn't actually do anything yet
13:38 as in, none of the other koha code references it
13:38 kados paul: yes, i believe I did accept those patches, bt they don't have any functional purpose yet IIRC
13:38 atz: snap
13:39 paul ok, I see
13:39 kados paul: I've just pushed up a patch for 2184 ... can you confirm it works for French language too?
13:41 it should show: Native Description ( Current Language Description )
13:41 advsearch -> More Options
13:41 (in staff interface)
13:42 paul hey, yes, it's OK
13:42 something that is not OK though : the language http_accept chooser
13:42 (tested on ff & opera)
13:42 it's still in english by default :(
13:43 kados paul: can you re-open the bug
13:44 paul can you remember which one is it ?
13:45 kados no :-)
13:45 paul ok, i'll dig
13:46 kados : if it's not too long for you, maybe releasing a 3.0RC2 with some notes about serials & item-level holds not working + installing CGI::Session would be the best
13:46 (as there are some improvements btw RC1 & now + lot of questions about instsalling CGI::Session)
13:52 kados paul: yep, I can do that for sure
13:52 paul so I don't vote yes or no, I vote RC2 ;-)
13:53 kados hehe
13:53 sounds good
13:53 OK, RC2 for today then
13:53 paul kados: +++
13:57 kados : today is our national day. You can say something about that if you release today ;-)
13:58 mc something like ""biblibre also works during national day" ? ;-)
13:58 paul lol
13:58 kados hehe
13:58 mc good bye all: it's familly time
13:59 paul bye mc
13:59 kados cya mc
13:59 paul tomorrow, i'll be in Lyon
13:59 & wed = start of Aix-Marseille BU project (1st meeting)
13:59 atz paul: happy bastille day
14:00 mc ho ... will you be reachable ? by cellular ?
14:00 paul mc : yep
14:00 mc cool
14:20 acmoore there's a new "gitcast" on git diff that I found intersting and useful: http://www.gitcasts.com/posts/git-diff.html
16:24 tim I just pulled an upgrade from git and get an error on make test.
16:25 t/Barcodes....................1/126 No max barcode (C4::Barcodes::annual=HASH(0x87ee100)->autoBarcode format) found.  Using initial value. at /home/tmcmahon/kohaclone/blib/​PERL_MODULE_DIR/C4/Barcodes.pm line 126.
16:31 atz tim: do you have any data loaded?
16:31 it appears not
16:31 that's not an error though, just a warning
16:32 you don't have any barcodes loaded, so the test uses the "initial" value.
16:35 acmoore there are tons of errors during the test suite. I don't thknk we're considering them failures yet.
16:47 tim atz: I have over 18000 items in there
16:48 atz what barcode format are you using?
16:48 tim The current Koha version is 3.00.00.094 and it seems to be working fine.
16:52 Our barcode format is just eight digits and increments for each item.
16:53 If that's just a warning, I need to find out what's stopping the test.
16:55 How about just below that?
16:55 Failed test '(annual      ) db_max()         : Database Empty or No Matches'  at t/Barcodes.t line 27.
16:56 atz that makes sense since none of your barcodes will match the "annual" format
16:57 acmoore atz, are these tests in t/Barcodoes.t that are actually dependent on data in the database? Should they be re-written to not depend on that, or moved to t/lib/KohaTest/... so that they can be sure to have a database under them?
16:59 atz acmoore: yes, the main point of db_max is to get the max barcode value from the database
17:05 the next test (max) encapsulates that one, and returns either db_max or initial
17:06 i can modify db_max line to be just a diag, but you need the feeback from that to know whether max is working correctly
17:08 Fallor so what's this serials-bug in rc1?
17:09 atz Fallor: i think there are a couple, one is with handling irregularities
17:10 Fallor such as?
17:11 atz That one is Bug 2114
17:11 the rest: http://bugs.koha.org/cgi-bin/b[…]ail2=&bugidtype=i
17:11 nclude&bug_id=&votes=&chfieldfrom=&chfi​eldto=Now&chfieldvalue=&cmdtype=doit&or​der=Reuse+same+sort+as+last+time&field0​-0-0=noop&type0-0-0=noop&value0-0-0=
17:11 http://bugs.koha.org/cgi-bin/b[…]dtype=include&bug
17:11 _id=&votes=&chfieldfrom=&chfieldto=Now&chfieldva​lue=&cmdtype=doit&order=Reuse+same+sort+as+last+​time&field0-0-0=noop&type0-0-0=noop&value0-0-0=
17:11 damn... gets auto split
17:12 Fallor whoa
17:12 long url
17:12 atz yeah, bugzilla advanced search query
17:12 i think you can cut off all that crap at the end though
17:12 everything after REOPENED
17:13 Fallor i'll just fix them
17:13 (the url's, not the bugs unfortunately ;))
17:20 lots of serials bugs...
17:20 so rc1 won't work for serials for now?
17:21 atz Fallor: yeah, too many blockers right now for my tastes
17:22 Fallor: another way to say it is "it *might* work, but it can't be considered stable"
17:27 Fallor yep
17:28 that's gonna be a change of plans for us then
17:28 we had been planning to start with serials management with Koha3
17:28 atz BibLibre has several ppl who are working on this (though perhaps not on Bastille Day)
17:29 it's rather important to some of our LL clients also.
17:31 Fallor when would you estimate that serials will be usable?
17:36 atz i'll defer to kados on that one (i haven't been following serials code that tightly)
17:37 kados Fallor: stay tuned to koha-devel for an email from me about that
17:38 Fallor ok :)
17:46 atz tim: just submitted a patch to improve the Barcodes.t test script
17:46 should avoid tripping up the harness after that
18:39 how is it we have a "category" EXPR sneaking back into the codebase?
18:39 i could've sworn we'd killed those off.
18:48 owen atz, they come back unless you sever their heads completely
19:05 tim Maybe I messed up with a setting, but I can't find anything about barcode formats except autoBarcode.
19:05 atz tim: yeah, that's pretty much it
19:06 the rest is just whatever data you put in
19:06 tim I still can't get the test to run.
19:07 atz i sent my patch, but it hasn't (afaik) been pushed up yet
19:07 kados tim: I'll ping you when it's pushed
19:08 tim Ah.  I did another pull and there were some changes.  I figured that was it..
19:11 kados tim: OK, rebase now and let us know
19:13 pianohacker What is a "category" EXPR? I'd like to know so I can avoid them myself
19:13 owen <!-- TMPL_IF EXPR=" " -->
19:14 pianohacker Yeah, I'm familiar with EXPRs
19:14 owen I don't think atz was referring to one particular bad EXPR, but the use of them in general
19:14 pianohacker Ah, joy
19:14 tim It worked!  Thanks kados and atz.
19:14 owen yeah :(
19:15 pianohacker So we'll have lots of template vars like variable_is_equal_to_other_variable
19:15 tim brb
19:15 pianohacker Instead of just using EXPR="variable eq other_variable" ?
19:15 atz pianohacker: almost never can you use EXPR for that
19:16 pianohacker Okay
19:16 Are we trying to avoid EXPRs completely, or can we use them with discretion?
19:17 atz the part that messes it up is that it jumps out of execution flow
19:17 or rather, you'd think that it gets evaluated wherever it is (in a loop for example)
19:19 but it doesn't, it gets evaluated at the beginning, every time and gives errors if both vars aren't defined
19:19 pianohacker Wow
19:19 That's... pathological
19:20 atz yeah, useless "feature"
19:21 owen atz, have you heard anything about that being considered a bug that might be fixed?
19:22 atz owen: no, i'm fairly sure it won't.  the point of H:T:P is that it is single pass compile/execute.  i.e., fast.
19:22 owen I see
19:22 atz it doesn't want to add an interpretation layer
19:24 pianohacker Is this when we start fervently wishing we could use something like Genshi that actually did this right?
19:30 owen I don't think anyone wishes they "could" re-write Koha from scratch :|
19:31 atz sure, this weekend i'm rewriting koha in PHP  :)
19:32 pianohacker I did say could
19:59 eric Is translate.koha.org valid for koha 3?
20:00 paul eric : not for french
20:00 eric ok.  tx paul
20:04 su-erin what linux distro seems to be the quickest to install koha on?
20:05 pianohacker Ubuntu and Fedora seem to be the list favorites
20:06 acmoore debian seems to be working pretty well for people, but opensuse seems to be popular too. I think they and fedora all have specific install instructions available for them.
20:07 su-erin thanks all
20:07 just wanted to quickly install koha for library staff members to start customizing
20:07 acmoore I think debian and ubuntu are considered the same in these instances. ;)
20:08 pianohacker Yeah
20:08 I think ubuntu might be a tad easier to install, but for servers, it's not a huge difference
20:10 su-erin i have the ubuntu 8.04 installed already, so i might try it on there if not i'll just use 7.10
20:27 pianohacker su-erin: I have 2.2 and 3.0 with Zebra running nicely on Ubuntu, though I had to compile YAZ and Zebra myself
20:31 eric Anybody is aware of a Best Practices page on Koha.org?  Exemple: best tools/utilities/setup for programmers, translators, etc?
20:31 pianohacker eric: There is a coding style guide at http://wiki.koha.org/doku.php?[…]:codingguidelines
20:34 acmoore and there is now a perltidyrc file in the distribution
20:34 eric acmoore: perltidyrc?
20:35 acmoore there's a tool called "perltidy" that will take perl code and clean it up a bit to match things like indentation preferences. You configure it with a .perltidyrc file. there's one in the xt directory.
20:36 perltidy is at http://perltidy.sourceforge.net/
20:37 eric oh, I see.  that's cool.  So i guess that every files go through are perltidified before a release
20:38 kados eric: that's the goal, yes, I'm sure a few miss that step
20:39 su-erin pianohacker, what version of ubuntu?
20:39 pianohacker The latest
20:39 8.04
20:41 kados: Patch prepared for bug 2282
20:47 kados pianohacker++
20:48 pianohacker I have about 15 more minutes before I knock off for the day; anything I should start on?
20:48 kados pianohacker: i sent you and galen a list
20:48 pianohacker: maybe something from there if any of those are small enough
20:49 pianohacker 2348 looks nice and bite-sized
20:53 eric perltidy looks very powerfull.  I used to place my braces on the line following (as example) my condition.  Perltidy takes care of bringing the braces at the end of the line.
20:55 thanks acmoore!
20:55 acmoore yep. FYI, I don't think of our code passes it yet, but maybe we'll get there eventually.
20:56 pianohacker acmoore: On that subject, we use 4-wide spaces, not tabs, right?
20:56 eric i guess that something like perltidy -R koha once a while would do the job...
20:57 acmoore pianohacker, I think so, though I see a lot of tabs around. Most of it is 4 spaces, though.
20:57 pianohacker Should I correct those file-wide whenever I have a good chance during some other work?
20:57 acmoore eric, I just perltidy stuff around where I'm working for now. It's a little less intrusive and such.
20:58 pianohacker, I don't really know. I try to stick to just the area that I'm working on. Though I'm always tempted to untabify the whole file ;)
20:58 pianohacker Heh
20:59 acmoore I personally don't think it's worth monkeying with the code that I'm not trying to edit, and it's probably distracting to others to modify stuff unnecessarily
20:59 eric acmoore: I guess that everybody is working like this (about being tempted to untabify files)
21:00 chris yeah, its nicer if you just tidy the code you are editing, else looking back through diffs is hard
21:02 eric christ, IIRC, merging tools can do abstraction of whitespaces
21:03 chris yep, they can, human eyes cant that well
21:03 eric :)
21:03 chris if i want to look at what changed between this commit and 12 commits ago, its hard if one of those commits was a big perltidy on the whole file
21:04 pianohacker Next time LibLime gets a new dev, we can have them perltidy -R the whole tree, then whenever a weird problem pops up, we say, "Blame So-and-So!" ;)
21:04 chris heh
21:10 eric Can't git support  pre-checkin scripts?  We could add a perltidy there.  That would ensure that every check-in respects the coding standards
21:11 kados eric: I think it can, so mostlikely we could enforce that as part of the QA process ... it'd be good for 3.2 and beyond
21:12 chris yep we just have to pick a point and do it, and the diffs from behind that would be annoying, so when we branch for 3.2 it could be enforced from then on
21:13 eric I second chris (i was typing it actually) :)
21:13 chris if we do it now, kados is gonna have a heck of a job reading patches
21:13 kados yea
23:51 atz cnighs: looks like reintroduced EXPR may come from your 6/20 patch
00:42 ricardo Hi all!
00:51 Is there a "git" expert here?
00:55 Never mind. I solved it!  :)
00:56 kados: If you're reading this, I just updated the INSTALL.opensuse file to add a missing step (install yaz "itself" - instructions about installing "libyaz" and "libyaz-devel" were already there).
00:56 Patch submitted.
00:56 Take care everyone!
07:16 chris evening
07:32 slef hi all
07:35 mc hi slef
07:36 evening, chris
08:33 masonj evening gentlemen
08:34 js hello masonj, what you do for women ;)
08:34 masonj anything they want....
08:35 js hé hé ;)
11:47 Fallor hi koha :)
11:47 +#
11:50 this is turning out to be a hellish project
11:51 first there was incompatibility with the router
11:51 it wanted netbios name for the computer or it wouldn't let it network
11:51 so put samba up
11:52 then we were able to connect to the network, but the router still kept dropping us
11:52 so we switched the router
11:52 then there was the problem with CGI::Session
11:53 and now we have a faulty disk in raid-array and it doesn't want to be mirrored
11:53 :P
11:53 not to mention the incredibly slow internet connection...
11:54 this is getting ridiculous

← Previous day | Today | Next day → | Search | Index

koha1