Time  Nick     Message
13:21 gmcharlt now to figure out why it randomly died
13:29 owen     Ah, so gmcharlt holds the secret reins to newlogbot! Good to know. Now if we could only get one of our regulars op status.
13:37 slef     owen: gmc is the clue
13:41 slef     the gmc@ bit in newlogbot's id
13:49 ryan     fbcit-away: ping
13:50 owen     By the way slef, thanks for the positive feedback on the changes to opac-topissues. I'm going to be looking for additional cases where that kind of change might be workable.
14:12 fbcit    ryan: can you point me to some docs/examples of loop context vars for TemplatePro?
14:33 gmcharlt hdl or paul: about?
14:33 paul     yep
14:33 paul     hello gm
14:34 paul     hello gmcharlt
14:34 gmcharlt hi paul
14:34 gmcharlt could you take a quick look at bug 2000?  I think ryan's proposal is the way to go, but it depends on people being comfortable with the smart-rules.pl interface
14:35 paul     bug 2000... sounds a nice number. I'll look
14:35 slef     owen: other type of widely-applicable change is jumping through to the details if a search returns only one result
14:38 paul     gmcharlt: i don't understand well the problem...
14:38 gmcharlt paul: issue is this - if you are using the two separate interfaces
14:38 gmcharlt e.g., fines admin & issuing rules admin
14:39 gmcharlt you can define fine rules, but let's say your policies are simply and you're content to just default a default fine for all branches and each item type
14:39 gmcharlt then when you go to add issuing rules in the older interface
14:39 gmcharlt and define a specific one for a branch, itemtype, and patron cat
14:39 paul     older interface : the non smart one, right ?
14:40 gmcharlt (right)
14:40 gmcharlt it adds an empty value for the fine rules for that specific entry
14:40 gmcharlt thus "hiding" the default fines rules you set up initially
14:41 gmcharlt but because the fine entries are blank, going back to the fine rules interface won't show you that this (presumably unintended) override has taken place
14:42 gmcharlt whereas if you use the smart rules interface, it's obvious from the UI that you've defined a more specific fine rule
14:50 paul     ok, I think I understand (although it's a complex situation...)
14:51 paul     Shouldn't we change the older interface behaviour more than do what rch suggest ?
14:51 paul     to save nothing when fine is empty ?
14:52 gmcharlt but having the issuingrules interface save nothing in the fines column is exactly the problem
14:52 paul     so In don't have understood well...
14:53 gmcharlt right now when looking for a fine or issuing rule to apply for a transaction
14:53 gmcharlt it goes from most specific to least specific
14:53 gmcharlt if the most specific entry has null or nothing in the fine columns
14:53 gmcharlt it curently can't tell whether the policy *intentionally* overrides a default fine set at a less specific level
14:54 gmcharlt or whether the override was accidental
14:55 paul     ah, ok, I understand now.
14:55 paul     does "0" mean "no opinion" or "set to 0" ...
14:55 gmcharlt exactly
14:56 paul     I thought, in the old interface, that I had differentiated 0 and "" (empty)
14:56 paul     to avoid this problem
14:56 paul     mmm... I did this for issuing, not for fines...
15:03 ryan     the problem then is you are testing 0 vs '' on issuelength.  Which column do you test on for fines, and then you are using two different rows from the able to specify circ params and fines params.
15:03 ryan     Seems to me that one row in the table should specify all the behaviors
15:03 paul     right. maybe having 1 table for both issuing & fines was a wrong ide
15:03 paul     idea
15:05 ryan     i think it's interface problem more than db problem
15:06 gmcharlt I could see solving it by spliting the rules into two separate tables, but I think changing the UI is better (and certainly quicker, since we already have everything we need)
15:45 fbcit    any ideas on how to say this so that perl likes it: $debug ? (use Data::Dumper);
15:45 atz      fbcit: it's tricky, you have to do it in a BEGIN blick
15:45 atz      *block
15:45 atz      and you can't use use
15:46 atz      see the way C4::Context handles CGI::Carp
15:47 atz      basically it does  "require" and then "import"
15:47 fbcit    so I'll need to require and then import the subs I want to use?
15:48 atz      right, because perl won't let you conditionalize a use statement
15:48 fbcit    k
15:48 fbcit    tnx atz
15:48 atz      np
16:53 paul     gmcharlt ?
16:53 gmcharlt hi paul
16:53 paul     are you 3 patches "granular permissions - ..." related to the mail you've sent to koha-devel ?
16:54 gmcharlt yes (actually 9, plus one I had to send to kados directly)
16:54 paul     OK, 3 more in my mailbox.
16:55 paul     seems the bot deliver them through boat or horse...
16:56 gmcharlt paul: hopefully the general mechanism is small enough for 3.0.  fully fleshing out the permissions for each module I would see as a longer-term project
18:48 owen     fbcit, do you know how the label item search works? Or anyone else? The results for a keyword search from that interface do not match those made from the catalog
18:49 fbcit    owen: it works in a very broken fashion
18:49 fbcit    I started into it a while back to attempt to *fix* it, but gave up due to lack of time
18:50 fbcit    it really needs to be refactored to use the search API
18:53 owen     Okay, so it uses it's very own searching method
18:53 owen     fbcit: Have you seen bug 1870?
18:55 owen     Those may be issues that should wait until someone tackles the search method problem
18:55 fbcit    how coincidental
18:55 fbcit    I just fixed that very thing
18:55 fbcit    actually it was there
18:56 fbcit    I think andrew added it
18:56 fbcit    but I broke it and then fixed it
18:56 fbcit    owen:did you notice if it appeared to work ok with the files I sent to you?
18:57 fbcit    opps...
18:57 gmcharlt owen: re closing of 1809, I disagree unless it just changed - the horizontal scrollbar was justed moved to within the page
18:57 fbcit    owen: I see, it is the links that are broken
18:57 acmoore  It's better than it was, but it could use a lot of help.
18:57 fbcit    owen: your correct, that should probably wait until the search stuff is corrected
18:57 owen     gmcharlt: moved within the page, and the edit/delete links were moved to the beginning of the table so that you don't have to scroll to see them.
18:58 fbcit    acmoore: yea, the entire search code needs to be thrown out and rewritten
18:58 owen     I interpreted the scrolling to the links to be the core of the issue
18:58 acmoore  fbcit, should it not be using SimpleSearch? Should SimpleSearch even exist?
18:59 acmoore  besides that, even if it's using the right search stuff, there is a bunch of useless code in that script.
18:59 fbcit    acmoore: I looked at SimpleSearch some thinking it might work
18:59 gmcharlt owen: I think a horizontal scrollbar within the page is still less than ideal, but obviously this isn't the most signficant bug
19:00 fbcit    acmoore: I agree
19:01 fbcit    no use re-inventing the wheel here
19:01 owen     gmcharlt: I think the best solution would be to be able to customize which columns show up in that additem table...perhaps by changing the framework? I think we've talked about that before.
19:01 acmoore  I just didn't really feel right ripping out dozens of lines of code that I didn't fully understand, considering that I'm a rather new kid on the block.
19:02 gmcharlt owen: yeah, we did talk about it befoe
19:02 acmoore  fbcit, next time I do something in that file, though, I'm not going to be as reserved.
19:02 fbcit    hehe
19:03 fbcit    you see how much commenting I did to figure out what was going on originally in the results sub
19:03 acmoore  yup. me too.
19:04 acmoore  b
19:04 fbcit    b ?
19:04 acmoore  heh. wrong window.
19:05 acmoore  I thought I was in emacs and hit C-x b <return>. Nothing happened.
19:20 atz      how does "issuingbranch" in issues get set?
19:20 ryan     it doesn't.
19:20 atz      mine all appear to be NULL
19:20 ryan     vestigial column, i think
19:20 ryan     issuingbranch would never be different from branch
19:21 atz      ok, cool... i'll update bor_issues_top
19:21 atz      (still working on that)
19:39 ryan     atz: did you look at catalogue_stats ?
19:40 atz      i don't think so
19:40 atz      do you want me to shift to that one?
19:41 ryan     that and circ stats need the item-level itype fix
19:41 atz      yeah, in this case a main flaw was the separation of issues and old_issues
19:41 atz      and other outdated mappings
19:42 ryan     ah, right
19:44 atz      there may be other reports similarly affected
19:45 gmcharlt alas, those reports were pretty much broken before the issues/old_issues split
19:45 atz      gmcharlt: true
19:46 kados    better get testing :-)
19:51 fbcit    ryan: can you confirm bug 1916 is still an issue?
19:58 ryan     fbcit: you are able to print accented characters with the labels module ?
19:58 ryan     i.e. pdf generation ?
20:07 ryan     fbcit: ok, just tested, and seems ok.
20:07 ryan     i was working with a backport of the labels code on a 2.2 system when I reported it.
20:07 ryan     must be specific to that installation, or have been otherwise fixed
20:08 fbcit    k
20:08 fbcit    I was trying to hunt up some data with accented chars
20:08 fbcit    I'll mark it fixed then.
20:09 ryan     cool, thx
20:19 frapell  hi guys, does anyone knows what happened with the skinning contest ?
20:19 kados    frapell: yes, it's proceeding, there were three submissions
20:19 kados    frapell: darci is back among the living and she should be announcing something very soon
20:19 kados    frapell: did you send in a submission?
20:20 frapell  kados: glad to hear about darci :)
20:20 frapell  kados: yes, i worked in the "Temita" product
11:16 sylvain  Hi all
11:17 sylvain  I have a little question about 2.2.9, when you search via isbn, dashes are stripped away from the query, and they should be too when you input isbn, but it seems it is not the case. is there a special config to enter ?
11:32 fbcit    g'morning #koha
11:37 kados    morning fbcit
11:58 fbcit    kados: why is it that during some db updates I am dumped back to the installer login rather than to the standard koha login page?
11:58 fbcit    perhaps its a cookie thing?
11:59 fbcit    ie. after the update from this last rebase I was sent back to http://biblios.campus.foundations.edu:8080/cgi-bin/koha/installer/install.pl?step=3&op=finished
11:59 fbcit    logging in there results in being sent back to mainpage.pl to login again :\