← Previous day | Today | Next day → | Search | Index
All times shown according to UTC.
| Time | Nick | Message |
|---|---|---|
| 00:00 | dcook | Not that I don't think he could, but he's so nice! |
| 00:00 | Dyrcona joined #koha | |
| 00:01 | eythian | <pianohacker> gmcharlt: He grumbled but said he could cut a separate 3.16 package for the overdrive dependencies |
| 00:01 | pianohacker: I don't think that's what I said, was it? | |
| 00:03 | pianohacker | eythian: I thought so, let me check. |
| 00:03 | eythian | I think it was more like "if we set it up right, it should just work" or something like that. |
| 00:03 | pianohacker | eythian: http://irc.koha-community.org/[…]4-06-11#i_1521767 |
| 00:04 | eythian | ohh |
| 00:04 | that part of it | |
| 00:04 | Dyrcona | rangi: got a couple minutes to talk about NCIPServer? |
| 00:04 | eythian | yeah, can stick it in as a recommends, or do the koha-overdrive-perl-deps thing |
| 00:05 | pianohacker | apologies if I didn't communicate that, don't want to volunteer you for work without yer knowing |
| 00:05 | But I see what you meant now | |
| 00:05 | eythian | Dyrcona: he probably doesn't, he's away until monday |
| 00:05 | pianohacker | eythian: I thought you meant throwing the overdrive-perl-deps in as a recommends |
| 00:05 | Dyrcona | eythian: Thanks. I didn't know that. |
| 00:05 | I'm on "vacation," but its a working vacation. | |
| 00:06 | pianohacker | either way works, though, I'd say. The specific approach I'd leave up to you and galen |
| 00:06 | eythian | nah, he's working on a client site for a couple of days, so I assume IRCing isn't so permitted :) |
| 00:06 | Dyrcona | OK. It can wait. |
| 00:07 | I can work on the Evergreen side of things without having the discussion. | |
| 00:07 | I just had some questions and ideas about the generic part. | |
| 00:07 | bag | gmcharlt: you going to ALA? |
| 00:07 | eythian | pianohacker: yeah. recommends means people will be told that they should consider it. And then we can bump it up to required on the next release. |
| 00:07 | *major release | |
| 00:07 | it's quite similar to have a -deps package | |
| 00:08 | bag | well I'm going to guess that you are going Galen - so see you soon :) |
| 00:08 | eythian | we could do both, but making an extra package go away when it's no longer needed might be hard. |
| 00:08 | pianohacker | makes sense |
| 00:12 | eythian: does aptitude auto-install recommends? | |
| 00:12 | eythian | I don't think so, but it definitely tells you that they're an option. |
| 00:12 | well | |
| 00:12 | if you're doing an install, yes | |
| 00:12 | if you're doing an update, no | |
| 00:13 | that's the standard I think, anyway. | |
| 00:14 | pianohacker | ah, okay, that makes sense. That's actually perfect imo |
| 00:20 | eythian | so, when I make a 3.16 package, I'll just regenerate the depends (which I don't normally do) and manually move the new entry down into recommends. |
| 00:20 | That'll be easy enough. | |
| 00:40 | tcohen | bye #koha |
| 00:48 | dcook | Always handy to re-read the coding guidelines... |
| 00:48 | guidelines? | |
| 00:48 | guidelines is http://wiki.koha-community.org[…]Coding_Guidelines | |
| 01:04 | kmlussier joined #koha | |
| 01:44 | a_m0d joined #koha | |
| 01:52 | a_m0d | Hi all - is there anyone who can help me get Koha running on a VPS running CPanel? |
| 01:54 | eythian | a_m0d: what distro is it running, do you have shell access, do you have root access? |
| 01:56 | a_m0d | eythian: CentOS 6.5, yes, and yes |
| 01:56 | eythian | a_m0d: it probably won't be the most fun ride, but there's some centos docs here: |
| 01:56 | http://wiki.koha-community.org[…]umentation#Centos | |
| 01:57 | I'm not sure if cpanel will interfere, I don't know enough about it. | |
| 01:57 | a_m0d | I've seen those, but I haven't been able to follow them 100% because of CPanel. |
| 01:58 | I couldn't install to the root, since there are multiple sites on here and I wanted to keep everything under the one site's home directory | |
| 01:58 | Also, those docs specify to go to http://localhost:8008 to run the web installation, but I can't do that. | |
| 01:59 | I did make a change in the koha-httpd.conf file to make the `intranet` site have its own unique host name. | |
| 01:59 | eythian | you can't really do that, it expects to be at the root from the point of view of the URL (i.e. you can virtual host it just fine, but it can't really share the same URL with something else unless that other thing can handle not being at the root.) |
| 01:59 | a_m0d | But when I hit that host name all I see is a directory listing. |
| 01:59 | eythian | well, you'd change localhost to be the relevant hostname. |
| 01:59 | hmm | |
| 02:00 | that sounds like an apache config issue alright | |
| 02:00 | like, it's not obeying the aliases or something | |
| 02:00 | a_m0d | That's kind of what I figured, but I'm kind of stuck when it comes to configuring apache. |
| 02:00 | eythian | that's when you want to see your logs to ensure the right virtualhost is being hit |
| 02:00 | that's an easy thing to have not happen. | |
| 02:01 | a_m0d | I can't even find an error log for the 500 errors I generate |
| 02:02 | e.g. hitting ***/index.html generates a 500 error for me, but I can't find where that is logged. | |
| 02:02 | eythian | hmm |
| 02:02 | is the log directive specified in the apache config? | |
| 02:03 | a_m0d | Yes, actually, just found that directive. Trying to load the error file now - hopefully it's in there since the CPanel view of the error log shows absolutely nothing. |
| 02:03 | carmenh joined #koha | |
| 02:04 | carmenh left #koha | |
| 02:12 | gmcharlt | bag: Yep |
| 02:12 | bag | cool |
| 02:13 | gmcharlt | eythian: if you need a patch or two in 3.16.x to deal with packagnig for the OverDrive change more easily, bug it and shoot it over to me in the next 8 hours |
| 02:14 | eythian | gmcharlt: I might do that. |
| 02:15 | gmcharlt | great |
| 02:16 | bag: you? | |
| 02:17 | bag | yes I will be arriving tomorrow |
| 02:19 | a_m0d | eythian: I can't really find any errors in any of the apache error logs that relate to this, except for 1 error when I tried to access kohalib.pl |
| 02:19 | That error just says: End of script before headers: kohalib.pl | |
| 02:20 | eythian | don't try to access kohalib.pl |
| 02:20 | it's not really a thing | |
| 02:20 | try mainpage.pl | |
| 02:22 | a_m0d | Okay, that's more like it - now I see an error message (in the logs) saying: |
| 02:22 | Can't locate Modern/Perl.pm in @INC | |
| 02:22 | and then a bunch more about that error | |
| 02:23 | eythian | that's a start |
| 02:23 | now you're just going to have to chase up aalllll the dependencies :) | |
| 02:23 | a_m0d | Okay, looks like I need to adjust the @INC to include another directory |
| 02:24 | eythian | well, do you have Modern::Perl installed? |
| 02:24 | that's the first thing to check. | |
| 02:26 | a_m0d | Yup, but its in a sub-folder of the home directory |
| 02:26 | e.g. ~/perl5/lib/perl5/Modern/ | |
| 02:30 | btw, I found a small issue in the build process | |
| 02:31 | * dcook | perks up a bit |
| 02:31 | dcook | issue with the build process? |
| 02:31 | a_m0d | If the database password contains a & symbol, it is put directly into the hoha-conf.xml file and than nothing can read that file anymore ... |
| 02:31 | Not a huge issue, but since I was prompted for that password as part of the build I expected it would handle it a bit better | |
| 02:32 | eythian | hmm yeah. that's probably worth filing a bug on. |
| 02:32 | a_m0d: to modify the INC you want to change PERL5LIB from within the apache config | |
| 02:32 | probably something like SetEnv PERL5LIB /home/whatever/perl5/lib/perl5 | |
| 02:33 | dcook | Hmm apparently having an ampersand in your MySQL password will cause problems with mysqldump as well |
| 02:33 | a_m0d | eythian: thanks, I didn't realise I could do that. I actually modified mainpage.pl to use the lines |
| 02:33 | use lib qw(/home/.....); | |
| 02:33 | eythian | a_m0d: that won't scale as you'll have to modify dozens of files :) |
| 02:33 | a_m0d | but your suggestion is probably better :D |
| 02:33 | dcook | ^ |
| 02:34 | a_m0d | So now I've run into more issues with the apache config again - since it doesn't seem to have picked up the ~/koha/etc/koha-httpd.conf file, it doesn't know where `/cgi-bin/koha/installer/install.pl` is |
| 02:34 | * dcook | probably wouldn't use an ampersand in a MySQL password as Google seems full of horror stories but definitely an issue nonetheless.. |
| 02:35 | a_m0d | dcook: thanks for the advice - I'll change it then. This was randomly generated by cpanel. |
| 02:35 | jcamins | a_m0d: cPanel's EasyApache substantially changes how Apache behaves. |
| 02:35 | You're going to need to do some major modification to make it work, most likely. | |
| 02:35 | And none of the paths in the instructions are going to be correct for your system. | |
| 02:36 | a_m0d | jcamins: I've found the vhost configurations for my 2 sites (standard one and the 'intranet' one), and I'm wondering now if I could perhaps just remove them from the main apache config file and then replace them with the koha-conf.httpd ones instead ... ? |
| 02:39 | jcamins | a_m0d: it's worth a try. I refuse to work with Koha on CentOS/RHEL/whatever, though, so I can't give you any advice beyond noting that cPanel substantially changes your environment. |
| 02:41 | a_m0d | jcamins: I hate working with cPanel, but unfortunately that's what I have here. I must say, though, that the CentOS instructions on the wiki were actually very useful to get the system built. They just stopped being quite so relevant after that point. |
| 02:44 | jcamins | I used a system that had cPanel once. It was so painful I just installed VirtualBox and used a VM with Debian. |
| 02:52 | * dcook | seems to recall rangi working on Koha::Email |
| 02:53 | thinks a Koha::Notices would be fantastic as well... | |
| 02:58 | dcook | I suppose Koha::Email would include that anyway most likely... |
| 02:59 | @later tell rangi Is Koha::Email going to be responsible for notice generation? (e.g. advanced notices, overdues, etc) | |
| 02:59 | huginn | dcook: The operation succeeded. |
| 02:59 | eythian | well, maybe. I'm not sure, but I'd think notices would sit on top of Koha::Email, and other things like SMS and such. |
| 02:59 | they shouldn't really be tied anyway. | |
| 02:59 | dcook | Agreed |
| 03:00 | But I suppose implementing Koha::Email would probably require other changes to be made as well | |
| 03:00 | I suppose not necessarily... | |
| 03:00 | * eythian | is now wondering how I'm going to design and build this browse interface. |
| 03:00 | eythian | It's something of a fiddly problem. |
| 03:00 | dcook | Seems like it :/ |
| 03:01 | * dcook | isn't sure he quite understands the idea |
| 03:01 | eythian | https://groups.google.com/foru[…]earch/h496-uirfNo <-- like this |
| 03:06 | The problem is that ES is a search engine, not really a sequential index system like a database can be. So at the moment it feels a bit like a square peg into a round hole. | |
| 03:07 | bag: you about? | |
| 03:08 | dcook | Hmm |
| 03:08 | The whole search vs browse thing is something we actually talked a lot about in library school (:p to Jorg Prante) | |
| 03:09 | eythian | as an end user, I can't really see the use case. |
| 03:09 | with exceptions, browsing by call number or similar, sure. | |
| 03:10 | dcook | Author can be handy if it's a quite common name |
| 03:10 | The author browse is basically an authority browse in my mind | |
| 03:10 | Or if you can't remember the name but you know one of the names | |
| 03:10 | eythian | so, that's just an author search sorted by author |
| 03:10 | dcook | A browse can be more precise than a search |
| 03:10 | eythian | the difference is you're not getting records back, you're getting author names back. |
| 03:10 | dcook | Hehe. |
| 03:11 | Yeah, pretty much | |
| 03:11 | I suppose it depends on whether you're doing an authority search or a facet search like you said | |
| 03:11 | eythian | So if you seach for "Smith" you only get "Smith" authors back. |
| 03:11 | dcook | That's the downside |
| 03:11 | Because maybe you want "Smithe" | |
| 03:11 | Or "Smitt" | |
| 03:11 | eythian | Oh, that's not a problem |
| 03:11 | well, smitt is | |
| 03:11 | dcook | I suppose you can have fuzzy type searches at that point |
| 03:12 | Librarians like browse because it's a bit more hands on | |
| 03:12 | You can sort of "feel" the data | |
| 03:12 | eythian | Which is why I'm thinking of a different approach: You search for Smith, and it does that but also searches for "Smit". If that doesn't get many more results, it searches for "Smi", and so on. |
| 03:12 | and then merges all the results. | |
| 03:12 | dcook | I was thinking about that a bit |
| 03:12 | eythian | (All these searches being startswith) |
| 03:12 | dcook | Although you still want to be able to browse back and forth |
| 03:13 | eythian | yeah, it'd allow you to do that. |
| 03:13 | dcook | How's that? |
| 03:13 | eythian | let me draw up some ascii art |
| 03:13 | dcook | \o/ |
| 03:14 | (Btw I've desired a browsing interface in Koha for ages) | |
| 03:14 | I imagine it'll mostly be used by librarians or academics | |
| 03:14 | Err researchers* | |
| 03:14 | Not so much by typical users | |
| 03:15 | * dcook | is reading more of the list now |
| 03:18 | dcook | I suppose what you'd want is to return a list of all authors (with their occurence count), and then just search that sublist for an index to start... |
| 03:18 | eythian | http://paste.koha-community.org/63 <-- dcook |
| 03:18 | something like that | |
| 03:19 | so you search for smith, and it starts "deep" and works its way out as it needs to to get adjacent results until it has enough for fill a page. | |
| 03:19 | dcook | Hmm |
| 03:20 | eythian | If you page back, it'll do the same with an earlier value. |
| 03:21 | chrisvella94_ joined #koha | |
| 03:21 | eythian | A real authority browser would be nice, but not everything is authorities. |
| 03:22 | dcook: opinions welcome, I'm trying to get a firm idea in my head before I really start working on it. | |
| 03:23 | I've had no real specs for this, so am mostly making it up to fit what it seems like people want. | |
| 03:23 | dcook | Yeah, fair enough |
| 03:23 | An authority browser would be nice, but not sure that's what they want :/ | |
| 03:23 | Although in theory you could probably adapt it to either case | |
| 03:23 | Too bad "scan" doesn't sort as well | |
| 03:24 | eythian | what even is scan |
| 03:24 | dcook | I'm not sure, haha |
| 03:24 | I want to say the Zebra scan is a bit like browse.. | |
| 03:24 | * dcook | takes a look |
| 03:24 | dcook | Ah, nah, I guess not |
| 03:24 | Well.. | |
| 03:25 | Sort of | |
| 03:25 | If I scan for "test" in Koha (w/ Zebra), it shows me "test" and the next 19 term/phrases following it in the keyword index | |
| 03:25 | Err "any" index | |
| 03:25 | Also gives a count | |
| 03:26 | eythian | how does it order them? |
| 03:26 | or just, order they're in in the index? | |
| 03:27 | dcook | It looks alphabetic |
| 03:27 | ES docs: | |
| 03:27 | If we want to support accurate ranking, we would need to first gather the term frequencies from all shards to calculate global term frequencies, then execute the query on each shard using these globale frequencies. | |
| 03:27 | Also, because of the need to sort the results, getting back a large document set, or even scrolling it, while maintaing the correct sorting behavior can be a very expensive operation. For large result set scrolling without sorting, the scan search type (explained below) is also available. | |
| 03:27 | I wonder if "scrolling" might be what you want in ES.. | |
| 03:28 | eythian | ohh |
| 03:28 | dcook | Ah, maybe not.. |
| 03:30 | eythian | yeah, looks like no |
| 03:31 | I was thinking you were talking about the scan option to Search.pm | |
| 03:31 | (initially) | |
| 03:31 | dcook | I was initially |
| 03:31 | eythian | oh right :) |
| 03:34 | dcook | Ideally it would be great if you could retrieve all authors (via search), sort them, store them, then re-query to get the offset of the one you want...then work from there. |
| 03:35 | Damn, this is a hard one :p | |
| 03:35 | eythian | yeah |
| 03:36 | it almost seems to me that a set of tables in the DB that are periodically regenerated might be the best approach | |
| 03:36 | dcook | "dfs_query_then_fetch" seems useful.. |
| 03:37 | Hmm | |
| 03:37 | Actually, it might be worthwhile looking at what Dspace does | |
| 03:37 | They used to do search via a home-cooked Lucene search engine and then browse using DB tables like you're thinking | |
| 03:37 | eythian | alternately, load authorities in and just use that. |
| 03:37 | dcook | Nowadays, they do both search and browse through Solr |
| 03:40 | eythian | it looks like the hollis example pointed to is just an authority browser. |
| 03:40 | That would be easier. | |
| 03:40 | Though I don't look forward to implementing authority linking. | |
| 03:40 | dcook | Would it? |
| 03:41 | eythian | Well. |
| 03:41 | not harder. | |
| 03:41 | dcook | Wouldn't you still have the problem of getting to the right place in the list? |
| 03:41 | eythian | that's true |
| 03:41 | I guess I didn't think about it hard enough :) | |
| 03:43 | actually | |
| 03:43 | I can run two queries, and sort them (one asc, one desc.) | |
| 03:44 | I don't actually need to know _where_ in the list I am, I just need to know how to page back and forward. | |
| 03:44 | dcook | Interesting.. |
| 03:44 | wahanui | somebody said interesting was sometimes good and sometimes bad |
| 03:45 | eythian | All I'd need is an index containing only author names (and one for every other thing to search, but that's no big deal.) |
| 03:46 | then for each author name it can do a search on the main index to get the counts. | |
| 03:46 | deleting records would be mildly tricky (i.e. detecting that there are none left) but solvable. | |
| 03:47 | dcook | Wouldn't you still run into the problem of paging though? |
| 03:47 | Or knowing what results should come before/after? | |
| 03:47 | jenkins_koha joined #koha | |
| 03:48 | eythian | no, because if my page size is ~20 I go 30 back (to work out the middle of the prev page) and 30 forward (to work out the middle of the next page) but only display the 10 forward and 10 back. |
| 03:49 | dcook | You'd still need to be browsing though, right? |
| 03:50 | That is... | |
| 03:50 | eythian | yeah sure. So I have an index of all authors. I can page back and forward through them by knowing what search term to generate the prev/next page from. |
| 03:50 | dcook | Using that method you described before in that pasted explanation? |
| 03:51 | eythian | No, that's a different idea :) |
| 03:51 | dcook | Hmm, then I'm not sure I understand |
| 03:51 | So you have an index of all authors...I'd assume you'd need to retrieve the whole thing, sort it, then look for your search term | |
| 03:51 | eythian | oh, but it still would only return the one you typed in. |
| 03:51 | damn. | |
| 03:52 | I always need to loosen the search term, or have an ordering value. | |
| 03:52 | dcook | Yeah, the ordering value would be nice |
| 03:52 | eythian | I don't like an ordering value as it makes insertions really hard. |
| 03:52 | dcook | I assume ES must use an ordering value but must be done internally |
| 03:52 | On the fly | |
| 03:52 | eythian | I doubt it does. |
| 03:52 | because there's no such thing as a native order. | |
| 03:53 | dcook | Right, I suppose it could just sort it into a response and use that for ordering |
| 03:53 | I don't know. I don't know enough about sorting :) | |
| 03:55 | eythian | the problem then is it would be relative to your search |
| 03:55 | so I can search for "smith" and page through all the "Smith*" results, easy. | |
| 03:55 | but I can't go backwards, or to things that aren't "Smith*" | |
| 03:56 | dcook | Which is a huge problem |
| 03:58 | eythian | yeah |
| 03:58 | dcook | Hmm seems like sorting in ES might be interesting as well.. |
| 03:58 | eythian | so I end up back at having the code chop a letter off and looking for "Smit*" on your behalf. |
| 03:58 | dcook | http://www.elasticsearch.org/g[…]multi-fields.html |
| 03:59 | eythian | yeah, I'm doing something like that as it is |
| 04:00 | ooh | |
| 04:00 | quakey | |
| 04:00 | ah, just a baby one, that's good | |
| 04:02 | http://www.geonet.org.nz/quake[…]aland/2014p476125 <-- I missed the one this morning, so I guess it's only fair | |
| 04:16 | dcook | hit.sortValues |
| 04:16 | But I think that's internal.. | |
| 04:16 | Glad that you're all right, eythian :) | |
| 04:18 | eythian | I think that's for scrolling again |
| 04:18 | dcook | Might be |
| 04:19 | It's certainly used in scroll | |
| 04:19 | Love how folks don't document things | |
| 04:20 | I think they use sortValues on InternalSearchHit but that doesn't get surfaced in SearchResponse | |
| 04:21 | * dcook | is just skimming code though |
| 04:21 | eythian | yeah, I've not seen it anywhere. |
| 04:22 | dcook | According to the ElasticSearch cookbook on Google Books: |
| 04:23 | SortValues(): This is the value/values used to sort the recored. It's only available if sort is specified during search phase. | |
| 04:23 | I thought that might be the other case... :/ | |
| 04:24 | Dunno. Definitely have a few problems here. | |
| 04:25 | jburds__ joined #koha | |
| 04:27 | dcook | If you could access "score" that would probably do it.. |
| 04:27 | But then still the problem of finding your term in the first place in that list | |
| 04:28 | eythian | yeah |
| 04:28 | I might have to give the relaxing of search terms thing a try. | |
| 04:29 | It just seems a bit ugly | |
| 04:29 | dcook | Definitely a bit ugly :/ |
| 04:29 | eythian | The other way would be to do it in mysql |
| 04:29 | chrisvella94 joined #koha | |
| 04:30 | dcook | That also seems a bit suboptimal :/ |
| 04:30 | Could you create a temporary index on the fly in ES? | |
| 04:31 | eythian | Not at any sort of speed, it'd have to reindex everything. |
| 04:31 | dcook | Do a search, throw it into the index (where the ids would be sequential) then search that index and return your one result and do things based on id.. |
| 04:31 | Hmm | |
| 04:31 | But if it's indexing in memory rather than on disk, it would be faster, yeah? | |
| 04:32 | I suppose you have an arbitrary number of hits though so that wouldn't scale even if that seemed doable | |
| 04:32 | eythian | It's still taking /all/ the documents and putting parts of them into a table. |
| 04:32 | There might be millions | |
| 04:32 | dcook | Yeah :/ |
| 04:32 | eythian | well, taking all of the authors, anyway |
| 04:32 | dcook | I suppose in that case an authority browser would be easier |
| 04:33 | Possibly | |
| 04:38 | chrisvella94 joined #koha | |
| 04:55 | dcook | Hmm what about this? |
| 04:55 | http://www.elasticsearch.org/g[…]aggregations.html | |
| 05:00 | wizzyrea | @marc 700 |
| 05:00 | huginn | wizzyrea: An added entry in which the entry element is a personal name. (Repeatable) [a,b,c,d,e,f,g,h,j,k,l,m,n,o,p,q,r,s,t,u,x,3,4,5,6,8] |
| 05:00 | wahanui | i already had it that way, huginn. |
| 05:01 | rocio left #koha | |
| 05:04 | cait joined #koha | |
| 05:05 | eythian | So, I'm thinking of taking everything that we want to extract and shoving it into a db table and using that. |
| 05:05 | With some smartly crafted indices, it could be speedy, and then query ES to get counts. | |
| 05:06 | hi cait | |
| 05:06 | cait | hi eythian :) |
| 05:15 | dcook | Just re-read Jorg Prante's earlier post...that must be what DSpace does as well.. |
| 05:16 | The adding a sort value before indexing | |
| 05:16 | * dcook | apparently can't stop thinking about this |
| 05:16 | dcook | hey cait :) |
| 05:16 | eythian | the problem with that is adding an author in the middle is hard. |
| 05:16 | cait | alphabetic indexes? |
| 05:16 | dcook | eythian: Yeah, that's the thing that gets me too :/ |
| 05:16 | Adding them anywhere except the end really | |
| 05:17 | cait: yeah trying to do an author browse | |
| 05:17 | eythian | cait: they're more or less uncountably infinite, so saying "give me the ten previous" is tricky. |
| 05:18 | cait | hmm |
| 05:18 | wonder how others do it | |
| 05:19 | eythian | adding a special index with a pre-sorted index is probably the cleanest way, but if you add an author you have to reindex the whole lot, which is a problem. |
| 05:19 | dcook | Agreed |
| 05:19 | cait: I think that's how others do it ^ | |
| 05:19 | At least that's how it looks like it's done in DSpace and by another person using ES | |
| 05:20 | eythian | my plan of backing off the search precision (i.e. manually stemming) is ugly but wouldn't require reindexing. |
| 05:20 | dcook | What sort of data do you get in a result set, eythian? |
| 05:20 | cait | dcook: hm not sure |
| 05:20 | eythian | my other plan of throwing them all in a database table and doing sorted/limited queries might be the best way. |
| 05:20 | dcook: you get the document that you put in. | |
| 05:20 | dcook | That's it? |
| 05:21 | No ES metadata about the document? | |
| 05:21 | Hmm | |
| 05:21 | eythian | "_index" : "koha_robin_biblios", |
| 05:21 | "_type" : "data", | |
| 05:21 | "_id" : "122", | |
| 05:21 | "_score" : 1.0, "_source" | |
| 05:21 | where _source contains your data | |
| 05:22 | _id is the biblio number, and the score is how well it matches. | |
| 05:22 | dcook | Mmm |
| 05:22 | eythian | cait: yeah, that's the sort of thing I want |
| 05:23 | cait | eythian: really hard it's going to be when people are asking for callnumber ... |
| 05:23 | from my experience | |
| 05:23 | eythian | we already support call number in the shelf browser. |
| 05:23 | cait | because most callnumbers can not really be sorted easily |
| 05:24 | yeah, using cn_sort i think :) | |
| 05:24 | eythian | I was thinking of using an approach similar to how we do that. |
| 05:24 | yeah | |
| 05:24 | using cn_sort is no problem. | |
| 05:24 | cait | probably the only way |
| 05:24 | creating a sortable form | |
| 05:24 | eythian | I'm planning on allowing collations so weirdo languages like German can be handled too :) |
| 05:24 | cait | and use our currnet plugin system to help people adapt it to their custom schemes if necessary... |
| 05:24 | lol thx | |
| 05:24 | they use diacritics here in callnumbers | |
| 05:25 | and ([,:-° ... | |
| 05:25 | * dcook | shudders |
| 05:26 | eythian | heh baka |
| 05:26 | cait | eythian: stop thinking that :) |
| 05:27 | it's an abbreviation | |
| 05:27 | dcook | That Horizon one cheats a bit I think |
| 05:27 | eythian | cait: it's Japanese for "idiot" |
| 05:27 | cait | i know |
| 05:27 | dcook | Actually it's really bad.. |
| 05:27 | cait | it's not really |
| 05:27 | why do you think that? | |
| 05:27 | did you went backwards? | |
| 05:27 | dcook | Because I don't read German ;) |
| 05:27 | cait | heh |
| 05:27 | dcook | Not so bad after a few clicks |
| 05:27 | cait | i think it takes stop words into account |
| 05:28 | i looked up a and went back, now there are all records with numbers | |
| 05:28 | 'a' | |
| 05:28 | dcook | Yeah, I did "e" |
| 05:28 | eythian | the first link is odd, it changes styling when you hit previous the first time |
| 05:28 | * dcook | always searches "e" |
| 05:28 | dcook | or "pickles" |
| 05:28 | Looks like an authority browser | |
| 05:28 | cait | eythian: i am not going to correct htat now - it#s the last one that is left :) |
| 05:29 | eythian | well, I was thinking it'd be possible to tie authorities in too |
| 05:29 | but maybe another time | |
| 05:29 | ah right | |
| 05:29 | cait | hm |
| 05:30 | looking for another example i am thinking about | |
| 05:30 | icould ask how the second example did it... coworkers should know | |
| 05:30 | eythian | So my plan is to throw every field that you might want to do this sort of search on into a database table, and then treat it like we do with the shelf browser. |
| 05:30 | then use ES to get counts. | |
| 05:33 | cait | eythian: can you get the capitalization right? :) most catalogs seem not to :) |
| 05:33 | eythian | cait: well that'd be part of the collation I expect. |
| 05:33 | cait | how does zebra do it? |
| 05:34 | our implementation there is not really good, but it's gotten nicer since tcohen fixed the display | |
| 05:34 | * dcook | might send the dspace folk an email anyway |
| 05:34 | eythian | I have no idea how zebra does it. |
| 05:35 | cait | k |
| 05:38 | i think at leat in our feature for that you can't page backwards | |
| 05:38 | but not sure if that was us ...or a limitation of it | |
| 05:46 | * dcook | grumbles a bit about creating notices |
| 05:46 | JesseM_a_ joined #koha | |
| 05:49 | * eythian | gets to have a little snark about supporting XSLX on the mailing list. |
| 05:49 | cait | eythian: we do support odf now :) |
| 05:49 | since 3.16 | |
| 05:49 | eythian | oh cool |
| 05:50 | cait | looking for the bug |
| 05:50 | bug 11679 | |
| 05:50 | huginn | 04Bug http://bugs.koha-community.org[…]_bug.cgi?id=11679 enhancement, P5 - low, ---, jonathan.druart, Pushed to Master , Add an ods export choice for reports |
| 05:51 | dcook | Yay |
| 05:51 | CSV wasn't really cutting it | |
| 05:54 | This C4::Letter stuff is a bit bonkers.. | |
| 05:54 | eythian | it really is |
| 05:57 | OK, tomorrow I'm going to have a go at this database table plan. I think there's not much point twisting ES into doing something it's not designed for. | |
| 05:58 | * dcook | shrugs |
| 05:58 | dcook | If you go the database route, you could use other search engines as well I suppose |
| 05:59 | eythian | that too |
| 05:59 | dcook | It's not a bug, it's a feature ;) |
| 05:59 | Oh dear God. My English profs are rolling in their... | |
| 05:59 | Well most of them are still alive. | |
| 05:59 | It's not a bug; it's a feature. | |
| 05:59 | There we go.. | |
| 05:59 | eythian | I just wish the facet stuff was more powerful, it's so close to doing what I want. |
| 05:59 | dcook | Totes |
| 05:59 | eythian | they're rolling in their tenure. |
| 06:00 | dcook | I think there is something called ElasticSearch Rails which might have some support for this...but :S |
| 06:00 | Probably reducing complexity is good | |
| 06:01 | * dcook | is curious about Zebra's scan now.. |
| 06:01 | eythian | yeah. |
| 06:03 | dcook | Looks kinda cool actually |
| 06:03 | http://www.indexdata.com/yaz/doc/yaz-client.html | |
| 06:04 | "Set step-size for scan. This value is used in next scan sent to the target. By default step-size is 0." | |
| 06:04 | Yaz-client: "[205] Only zero step size supported for Scan -- v2 addinfo ''" | |
| 06:04 | * dcook | facepalm |
| 06:04 | wahanui | http://buikitty.files.wordpres[…]120202-094434.jpg |
| 06:05 | eythian | heh |
| 06:05 | dcook | scanpos |
| 06:05 | That's a bit neat.. | |
| 06:06 | You can set it to be in the middle of a range... | |
| 06:06 | Then use your top and bottom figures to determine your next range I guess | |
| 06:06 | cait | eythian: is the alphabetic indexing going to be a separate bug? |
| 06:06 | dcook | scansize 30 |
| 06:06 | scanpos 15 | |
| 06:07 | eythian | cait: define separate |
| 06:07 | of | |
| 06:07 | separate bug | |
| 06:07 | *oh | |
| 06:07 | probably not | |
| 06:07 | cait | ah |
| 06:07 | dcook | scan @attr 1=4 water |
| 06:07 | cait | because the elastic search will get quite big to test :) |
| 06:07 | dcook | Zebra could totally do this |
| 06:07 | eythian | I do plan to split all the patches up eventually |
| 06:08 | cait | it's just a question of how to organize best i guess |
| 06:08 | eythian | yeah |
| 06:08 | cait | not sure what's better really |
| 06:08 | just thought that might be separate enoughwith database tables that it might make sense | |
| 06:08 | but on the other hand... esier for early testers if it's a one bug thing maybe | |
| 06:09 | do you think we could do a demo opac somewhere? like we had for bootstrap? | |
| 06:10 | eythian | that's probably a good idea. |
| 06:11 | I'll see if I can put one on our demo server or something. | |
| 06:12 | cait | cool :) |
| 06:14 | dcook | eythian: You could always use Zebra for it :p |
| 06:14 | eythian | heh |
| 06:14 | wash your mouth out ;) | |
| 06:14 | dcook | Really, "scansize 30", "scanpos 15", "scan @attr 1=1003 test" |
| 06:15 | Done and home in time for dinner ;) | |
| 06:15 | eythian | shush |
| 06:15 | cait | heh |
| 06:15 | eythian | will it scan backwards |
| 06:15 | pastebot | "dcook" at 127.0.0.1 pasted "zebra scan output" (30 lines) at http://paste.koha-community.org/64 |
| 06:15 | eythian | ? |
| 06:15 | dcook | Yep |
| 06:15 | But you'd need to use the value rather than an id | |
| 06:16 | In this example, to scan backwards you'd issue a "scan @attr 1=1003 ed" query to jump backwards | |
| 06:16 | You'd need to play with the values a bit and what you wanted to show, but I think you could do it | |
| 06:16 | eythian | ah yep |
| 06:17 | * dcook | looks at zoom |
| 06:17 | cait | poor understimated zebra |
| 06:17 | underestimated... | |
| 06:17 | dcook | http://search.cpan.org/~mirk/N[…]lib/ZOOM.pod#scan()_/_scan_pqf() |
| 06:19 | cait | i htink position looks like it would do it |
| 06:19 | dcook | and number |
| 06:19 | * dcook | apparently likes to make eythian's life more complicated |
| 06:20 | cait | me too :) |
| 06:20 | heh | |
| 06:20 | eythian | meanies |
| 06:20 | With that, I'm leaving! | |
| 06:20 | cait | hehe |
| 06:20 | finally! | |
| 06:20 | eythian | have a good (evening|morning) all. |
| 06:20 | cait | have a nice evening :) |
| 06:20 | eythian | later |
| 06:21 | cait | dcook: someone should really impelment those facets with zebra... ;) |
| 06:21 | dcook | hehe, night, eythian |
| 06:21 | cait: Yeah, it's sort of on the list | |
| 06:21 | cait | no pressure |
| 06:22 | dcook | I think my first step would be to decompose the existing code |
| 06:22 | cait | it just seems for everytime we said: zebra can't do that... it was just something we got wrong |
| 06:22 | dcook | So that we'd use the same code but in a where we could use a syspref to switch to native facets |
| 06:22 | cait | yeah ... like that :) |
| 06:22 | dcook | cait: Yeah, there's definitely something to be said for Zebra just being set up wrong |
| 06:22 | Painfully wrong | |
| 06:23 | cait | yeah:( |
| 06:23 | dcook | Well... |
| 06:23 | Maybe not set up wrong | |
| 06:23 | But the query building is horrible | |
| 06:23 | The facets just weren't done (or really understood until now) | |
| 06:23 | cait | yeah |
| 06:24 | and the alphabetic index was hardly implemented at all and the gui is buggy | |
| 06:24 | dcook | A lot of the indexing isn't bad though |
| 06:24 | Alphabetic index? | |
| 06:24 | Gui? | |
| 06:24 | cait | @later tell eythian - in case you didn't know - index search with zebra is a bit ugly and hiding on the staff advanced search page |
| 06:24 | huginn | cait: The operation succeeded. |
| 06:24 | cait | heh, see my comment |
| 06:25 | dcook | Where? |
| 06:25 | cait | add a word to a title search, check the scan index checkbox |
| 06:25 | in staff advanced search | |
| 06:25 | dcook | Yeah, I was playing with that earlier |
| 06:25 | I had tried it out ages ago and forgot about it until today | |
| 06:25 | cait | ah right |
| 06:25 | so waht was your question then? | |
| 06:25 | *confused* | |
| 06:25 | dcook | Hence the idea to try out yaz-client ;) |
| 06:25 | cait | aah |
| 06:25 | dcook | I'm not sure anymore haha |
| 06:25 | cait | i thought i had triggered that with the question how zebra did it heh |
| 06:26 | * dcook | should concentrate for a bit though |
| 06:26 | dcook | eythian and his derailing me :p |
| 06:26 | Today is my Friday as I"m heading to a wedding tomorrow.. | |
| 06:26 | cait | yeah he is good at that |
| 06:26 | mine too | |
| 06:26 | concert | |
| 06:30 | liw joined #koha | |
| 06:33 | dcook | Concert? |
| 06:35 | cait | i am going to a concert on friday - not a wedding :) |
| 06:39 | ok bbiab | |
| 06:43 | alex_a joined #koha | |
| 06:44 | alex_a | bonjour |
| 06:56 | yohann joined #koha | |
| 07:03 | reiveune joined #koha | |
| 07:03 | reiveune | hello |
| 07:05 | irma | Hi all in #koha |
| 07:05 | Hi dcook | |
| 07:05 | dcook | heya irma :) |
| 07:05 | how goes it? | |
| 07:06 | irma | dcook ; I got your koha-oz email and we should meet after (when we return from Melbourne) for a catch-up |
| 07:07 | dcook a Friday is probably the best day for me but let's email and see what suits ...ok? | |
| 07:08 | dcook | Sure, that sounds good to me :) |
| 07:08 | Might be a bit tight for the next couple weeks but I think it would be doable in July | |
| 07:09 | irma | July or before mid August would be good. After that I am taking a week off |
| 07:10 | cheers for now dcook ... | |
| 07:10 | have a nice evening/day all :-) | |
| 07:10 | dcook | I'm heading to Europe in late September so before that sounds good |
| 07:11 | Busy busy folks.. | |
| 07:11 | night irma :) | |
| 07:12 | irma | night dcook ^..^ |
| 07:13 | magnuse joined #koha | |
| 07:15 | magnuse | whoa http://www.indexdata.com/news/[…]oha-smart-widgets |
| 07:15 | gaetan_B joined #koha | |
| 07:15 | gaetan_B | hello |
| 07:16 | magnuse | bonjour gaetan_B |
| 07:17 | dcook | magnuse: Yeah, it looks like it could be neat! |
| 07:20 | Got to love that moment when you're kicking yourself wondering where your code isn't working... | |
| 07:20 | Then you remember that you're setting the value of a key in a hashref all wrong.. | |
| 07:25 | * dcook | thinks parse_letter is a stupid sub :p |
| 07:26 | dcook | Well...I guess maybe it does have a purpose sometimes... |
| 07:26 | Would be better if it were centralized.. | |
| 07:26 | * dcook | looks at the clock |
| 07:26 | dcook | But this guy should go home |
| 07:26 | (Well, to the vet, then home) | |
| 07:27 | * dcook | apparently has no concept of English grammar today. |
| 07:27 | waves goodbye | |
| 07:32 | magnuse | have fun dcook |
| 08:01 | irma_ joined #koha | |
| 08:17 | cait joined #koha | |
| 08:17 | magnuse | kia ora cait |
| 08:19 | cait | hi magnuse :) |
| 08:26 | ashimema joined #koha | |
| 08:27 | ashimema | morning #koha |
| 08:28 | cait | good morning ashimema |
| 08:29 | Viktor joined #koha | |
| 08:29 | cait | hm looks like you can only have a search history if you have patrno accunts |
| 08:30 | hm bug? | |
| 08:30 | wahanui | bug is it stores biblionumber not itemnumber |
| 08:32 | * magnuse | waves to ashimema and Viktor |
| 08:32 | * Viktor | waves back |
| 08:32 | * cait | waves at Viktor too |
| 08:32 | Joubu | Hello |
| 08:33 | * Viktor | waves to cait and all the other nice people here :) |
| 08:33 | ashimema | hmm.. I thought there was a local anonymous search history cait? |
| 08:34 | Joubu | cait: I would like to get your opinion on my last note of bug 11872, if you have time... |
| 08:34 | huginn | 04Bug http://bugs.koha-community.org[…]_bug.cgi?id=11872 major, P5 - low, ---, kyle, Signed Off , Lost overdue items should not generate fines |
| 08:34 | cait | ashimema: there is -- but it won't display when opacuserlogin is off |
| 08:34 | bug 12485 | |
| 08:34 | huginn | 04Bug http://bugs.koha-community.org[…]_bug.cgi?id=12485 enhancement, P5 - low, ---, oleonard, NEW , Show OPAC search history when opacuserlogin is OFF |
| 08:34 | cait | which means i have to explain to a library now that they can't use the feature... meh. |
| 08:35 | hm I don't have an answer right away Joubu | |
| 08:35 | there are so many prefs and options for lost items now nd our libraries don't use it yet, i have to think more about it | |
| 08:36 | ashimema | oh.. thats interesting. |
| 08:36 | cait | but i think it's a good point - is a lost item overdue. |
| 08:36 | hm. | |
| 08:37 | i think if it doesn't accrue fines, it's not overdue but 'just' lost | |
| 08:38 | but then i am not sure where checkoverdues and GetOverduesForBranch are used - we might want to make sure the lost items on the patron account stll display in OPAC etc. | |
| 08:40 | Joubu | cait: I think they should be displayed too. And the itemlost specific case should be manage in the fines calculation script |
| 08:41 | cait | I tend to say so too. writing up a comment and checking something in one of our databases |
| 08:42 | Joubu: i was wondering if maybe having a parameter for the subs might be better... but it's a more complicated change too | |
| 08:42 | Joubu | cait: yes, a filters parameter could be a good idea |
| 08:46 | cait | I've added a comment |
| 08:51 | jseplae joined #koha | |
| 08:56 | Joubu | cait: thanks |
| 09:07 | magnuse | Viktor: a small thing in Bug 7843... |
| 09:07 | huginn | 04Bug http://bugs.koha-community.org[…]w_bug.cgi?id=7843 enhancement, P5 - low, ---, viktor.sarge, Needs Signoff , Create RSS for OPAC news |
| 09:07 | magnuse | you added the new file opac/opac-news-rss.pl, right? but it says "Parts Copyright (C) 2013 Mark Tompsett" |
| 09:08 | it should probably be copyright you, but also saying "based on <filename> by ..." | |
| 09:48 | Viktor | Ah - thanks for pointing it out magnuse. |
| 09:48 | That sounds about right. | |
| 09:49 | I'm a bit behind with work on this one since I've been away from work due to my son being ill. | |
| 09:49 | But next step would be to add dates. | |
| 09:50 | magnuse | hope the son is/gets better! |
| 09:50 | Viktor | It's easy enough but they were not in the right format to validate. And even though there's a function in Koha to get the right dates I ran into some things with references that I didn't quite get right. |
| 09:50 | * cait | too |
| 09:51 | Viktor | magnuse He is getting better. But had a few days now that were a little more than the usual things you expect. |
| 09:51 | Thanks! | |
| 09:52 | And I'm off for vacation now so even if I highly enjoy coding for Koha patches will be quite a bit slower. | |
| 09:52 | But know that I didn't quit - I just went of into the sun for a few weeks :) | |
| 09:52 | a_m0d joined #koha | |
| 09:55 | cait | Viktor: enjoy your vacation! |
| 10:25 | hm does someone know if anonymous search history gets cleaned up when the session is deleted? | |
| 10:32 | Joubu | cait: when the user log out? |
| 10:32 | it should | |
| 10:32 | cait | Joubu: when the user doesn't log in |
| 10:32 | Joubu | cait: yes, anonyous, sorry... |
| 10:33 | cait | np :) |
| 10:33 | Joubu | when is the session deleted? When the browser is closed I suppose |
| 10:34 | cait | hm or when it's cleaned out of the session table |
| 10:34 | do the entries in the db table get deleted? | |
| 10:35 | Joubu | cait: yes, if the session does not exist anymore in the table, the search history is deleted |
| 10:35 | cait | thx a lot! |
| 10:35 | Joubu++ | |
| 10:49 | nlegrand | hey $koha |
| 10:51 | cait | hi nlegrand |
| 11:30 | a_m0d joined #koha | |
| 11:51 | beni joined #koha | |
| 11:58 | alex_a joined #koha | |
| 11:59 | meliss joined #koha | |
| 12:03 | fridolin joined #koha | |
| 12:03 | fridolin | hie lal |
| 12:10 | magnuse | bonojur fridolin |
| 12:10 | fridolin | bonjour magnuse : I have a new kb, so i make a lot of mistakes |
| 12:10 | ;) | |
| 12:10 | oleonard joined #koha | |
| 12:12 | oleonard | Hi #koha |
| 12:13 | magnuse | fridolin: yeah, i struggle a bit with my new laptop, fortunately i have my old external keyboard at home :-) |
| 12:13 | hiya oleonard | |
| 12:28 | edveal joined #koha | |
| 12:32 | collum joined #koha | |
| 12:44 | indradg joined #koha | |
| 12:46 | oleonard | Joubu or anyone else know if something is going on with the Biblibre sandboxes? |
| 12:48 | I tried to set up Sandbox 13 with Bug 12220 but haven't gotten any email after 30 minutes | |
| 12:48 | huginn | 04Bug http://bugs.koha-community.org[…]_bug.cgi?id=12220 normal, P5 - low, ---, oleonard, Signed Off , bootstrap not responsive on all devices |
| 12:48 | Joubu | oleonard: they don't work ? |
| 12:48 | kmlussier joined #koha | |
| 12:48 | oleonard | I tried the same thing yesterday and the only message I got was "Sandbox not ready!" |
| 12:48 | Joubu | oleonard: yes, since the last hackfest, it seems that all emails are not sent |
| 12:49 | oleonard | Ah, okay I see now that the patch was applied this time. |
| 12:50 | Joubu | oleonard: yes, you can take a look at the news, on the mainpage |
| 12:50 | Sandbox setup by oleonard myacpl.org with database -1 and bug 12220 on Thu Jun 26 14:21:00 2014 |
|
| 12:50 | oleonard: maybe it's because you didn't choose a DB :-/ | |
| 12:52 | oleonard | Anyway I'm glad to confirm that my patch works since I submitted it without doing a real test :) |
| 12:55 | jcamins | Are test plans no longer de rigeur? |
| 13:02 | magnuse | in patches? i'd say they are |
| 13:02 | jcamins | magnuse: yeah, it looks like it was just happenstance that the first patch I looked at when I considered signing something off didn't have one. |
| 13:03 | * magnuse | chants "fail it, fail it, fail it" |
| 13:03 | jcamins | Determining that there was no test plan took all the time I had for testing, though, so I didn't look at any other patches before asking. |
| 13:03 | talljoy joined #koha | |
| 13:05 | kmlussier joined #koha | |
| 13:06 | * jcamins | failed it. |
| 13:07 | * magnuse | sheds a (small) tear |
| 13:08 | magnuse | fridolin++ for 3.14.8 |
| 13:09 | fridolin | thanks |
| 13:27 | nlegrand | fridolin: new keyboard? A squarred stuff like Joubu or Stéphane? |
| 13:29 | fridolin | nlegrand: nop, I'm a fan of curved KB : http://www.microsoft.com/hardw[…]desktop/L5V-00001 |
| 13:30 | indradg_ joined #koha | |
| 13:31 | Joubu | erk |
| 13:32 | fridolin | ;) note it uses radio wireless, not bluethooth |
| 13:33 | nlegrand | :D |
| 13:33 | fridolin | paring via bluethooth is enoying when you whant to use your KB on another pc |
| 13:42 | jenkins_koha | Yippee, build fixed! |
| 13:42 | wahanui | o/ '`'`'`'`'`'`'`'`'` |
| 13:42 | jenkins_koha | Project Koha_3.14.x_D7 build #8: FIXED in 32 min: http://jenkins.koha-community.[…]Koha_3.14.x_D7/8/ |
| 13:42 | * Bernardo Gonzalez Kriegel: Translation updates for 3.14.08 release | |
| 13:42 | * Fridolin Somers: Update release notes for 3.14.8 release | |
| 13:42 | * Fridolin Somers: Increment version for 3.14.8 release | |
| 13:53 | JesseM joined #koha | |
| 13:57 | alex_a_ joined #koha | |
| 14:06 | Joubu | @later tell rangi on the dashbard, the search for "pushed to mm" displayed an error "The search named PushedMM has not been made visible to you.". I'm not sure it's expected. |
| 14:06 | huginn | Joubu: The operation succeeded. |
| 14:25 | rocio joined #koha | |
| 14:25 | Tony joined #koha | |
| 14:26 | Tony | I'm wondering if someone can help me with a search issue in Koha |
| 14:28 | When you use the advance search and say your are wanting "Germany" and "England" you will get one set of results and use "Germany" or " England" and you get the same results.. the number of returns should be different.. | |
| 14:32 | pianohacker | Tony: what would you expect in the second case? |
| 14:39 | Tony | Well if you are using and against an or it would appear that the number should be higher on the or than on the and.. and is looking for both combined and or is looking for both as individual.. |
| 14:46 | pianohacker | Tony: Ah, I see what you mean |
| 14:50 | Tony | Thank you for understanding.. This is causing our students at the college a real headache because they can't trust the numbers or results |
| 14:55 | Is there anyone that can help me with this issue.. I don't know if Koha has a fix or not.. I'm sure I'm not the only one with this problem. | |
| 14:59 | yohann left #koha | |
| 15:00 | pianohacker | Tony: It would probably be best to send a post to the users mailing list... a lot of the developers and librarians that would have better knowledge of the search interface would be there rather than IRC |
| 15:00 | a nice way of saying "Sorry, I don't got no ideer" | |
| 15:04 | reiveune | bye |
| 15:04 | reiveune left #koha | |
| 15:04 | oleonard | Tony: You're talking about using the advanced search form with "more options?" |
| 15:10 | Tony | Yes I am.. but I didn't use any of the options when I did this search on the top area with the and, or, not selections |
| 15:13 | pianohacker do you know the URL for me to get to the developers because I thought they would be on here.. | |
| 15:14 | oleonard | Tony: pianohacker is suggesting the Koha users list, not the developers list |
| 15:14 | ...since your question is about Koha usage. | |
| 15:15 | My understanding of Koha "quick search" is that and/or operators are not recognized, but I could be wrong | |
| 15:15 | Using the and/or operators in the advanced search form seemed to work for me though. I got different numbers of results with the two different operators | |
| 15:15 | Tony | That is just what I am seeing as the problem so is there a fix that you are aware of or not |
| 15:16 | Using the example that I gave... hmmm it doesn't do that for us.. so maybe we have a setting incorrect | |
| 15:17 | oleonard | Is there a fix for "quick search" not recognizing and/or operators? It's not that something is broken, it's simply a feature Koha doesn't have. |
| 15:17 | jcamins | oleonard: use QP. |
| 15:17 | Or write ccl using ccl= | |
| 15:18 | Tony | What is QP and where would I write ccl using ccl= |
| 15:18 | jcamins | QP is QueryParser. ccl= is what you have to prefix your queries with if you want to write your queries in CCL. However, you don't want to write your queries in CCL. |
| 15:19 | Tony: but that was in answer to oleonard not you. I have no idea what's going on with your system. | |
| 15:20 | oleonard | jcamins: Basically Tony wants to be able to use "and" and "or" in searches |
| 15:20 | ...and have each give him different results ;) | |
| 15:20 | jcamins | oleonard: I thought the problem was that using and and or didn't work. |
| 15:20 | Or do you mean literally typing "and" and "or"? | |
| 15:20 | Tony | jcamins: that is just what I'm trying to do |
| 15:20 | jcamins | Literally typing "and" and "or"? |
| 15:21 | Tony | jcmins: no, using the option operators in the search block is what I would like to function correctly |
| 15:21 | jcamins | Right, I don't know why they're not working for you. |
| 15:22 | I don't know if they work for other people, though. | |
| 15:22 | Tony | hmmm I don't know either but I really need to figure this out for the students here |
| 15:22 | jcamins | Good luck. |
| 15:22 | Tony | Thank you |
| 15:26 | cait | Tony: just checking... you are using the first and second field? not leaving any blank? |
| 15:29 | Tony | cait: yes that is correct... I'm only using the first two fields and changing the operator on each search |
| 15:29 | cait | what does the link look like that koha generated for your 2 seaches? |
| 15:31 | oleonard | paste? |
| 15:31 | wahanui | I eat paste! It's tasty! http://paste.koha-community.org |
| 15:31 | cait | thx oleonard |
| 15:33 | Tony | cait: /cgi-bin/koha/opac-search.pl?idx=kw&q=germany&op-and&udx=kw&q=Europe&op=and&idx=kw&do=Search&sort_by=relevance |
| 15:34 | cait | hm really udx? |
| 15:34 | Tony | hmmm it looks like the operator never changed |
| 15:34 | sorr typed it wrong i think it is idx | |
| 15:35 | oleonard | Please copy and paste |
| 15:35 | Tony | Okay let me bring it up on this machine |
| 15:35 | nengard joined #koha | |
| 15:36 | nengard left #koha | |
| 15:37 | Tony | here it is : /cgi-bin/koha/opac-search.pl?idx=kw&q=Germany&op=and&idx=kw&q=Europe&op=and&idx=kw&do=Search&sort_by=relevance |
| 15:38 | cait | and with or? |
| 15:44 | Tony | cait: sorry for the dely .. give me a sec |
| 15:45 | cait here you are /cgi-bin/koha/opac-search.pl?idx=kw&q=Germany&op=or&idx=kw&q=Europe&op=and&idx=kw&do=Search&sort_by=relevance | |
| 15:46 | cait | is our opac publicly accessible? |
| 15:46 | which version are you on? | |
| 15:48 | Tony | 3.12.01.000 |
| 15:48 | Yes it is public | |
| 15:49 | oleonard | ... |
| 15:49 | cait | 3.12 is pretty old |
| 15:49 | Tony | Okay so do you suggest that I should update |
| 15:49 | What is the current version | |
| 15:49 | gaetan_B | bye! |
| 15:50 | oleonard | For what it's worth Tony trying your searches as you pasted them above in my 3.16 test system returned different results sets |
| 15:51 | cait | we are running 3.12.9 - works there as well |
| 15:51 | Tony | oleonard: not following |
| 15:52 | Oh okay I get it... sorry so you are thinking that is fixed in a later version... Okay so I need to go through the update process... | |
| 15:52 | cait | Tony: for me it's also working in 3.12 |
| 15:52 | i think it's something in your setup | |
| 15:52 | jcamins | 3.12.9 works fine for me. |
| 15:52 | cait | yep |
| 15:52 | for me too | |
| 15:53 | Tony | Perfect guys... then let me do the upgrades and see what I get at that point... Thank you all for help and suggestions... |
| 15:53 | cait | Tony: if your opac is publicly available... can you give us the url? :) |
| 15:53 | Tony | cat.columbiasc.edu |
| 15:53 | cait | an upgrade to the newest verseion would be pretty huge and should be prepared well |
| 15:55 | Tony: i think there is something wrong with your search | |
| 15:55 | Tony | cait: thank you very much |
| 15:55 | cait | http://cat.columbiasc.edu/cgi-[…]levance&do=Search |
| 15:55 | i tried this and I get a 404 for the books | |
| 15:55 | when i try to access the detail page | |
| 15:55 | i think your index might be not ok | |
| 15:55 | first thing i'd try is a complete reindex | |
| 15:56 | for the first 2 results on the page at least | |
| 15:56 | fridolin | byyyy |
| 15:56 | fridolin left #koha | |
| 15:56 | Tony | I totally agree because that wasn't happening earlier |
| 15:57 | Thanks for finding that | |
| 16:01 | cait | ok... also your simple search is broken |
| 16:01 | i think it's something in your setup | |
| 16:02 | a keyword search for germany from the simple search won't give any results | |
| 16:02 | I've never seen that | |
| 16:07 | cait left #koha | |
| 16:15 | Tony | Okay I'm trying to reindex my zebra but I'm getting a lot of errors.. can someone help me on this: rootcat:/usr/share/koha/bin/migration_tools# Can't locate C4/Context.pm in @INC (@INC contains: /etc/perl /usr/local/lib/perl/5.14.2 /usr/local/share/perl/5.14.2 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.14 /usr/share/perl/5.14 /usr/local/lib/site_perl .) at ./rebuild_zebra.pl line 6. > BEGIN failed--compilation aborted at ./rebuild_zebra.pl line 6. > root |
| 16:27 | gmcharlt | @quote random |
| 16:27 | huginn | gmcharlt: Quote #145: "Don't ask to ask, just ask!" (added by wizzyrea at 02:37 PM, July 07, 2011) |
| 16:32 | rocio joined #koha | |
| 16:51 | pianohacker | dernit. Anybody use dh-make-perl around here? (And successfully at that?) |
| 16:58 | future note: dh-make-perl depends on $PWD, so you can't run it under sudo (unless you manually pass through the environment variable) | |
| 17:10 | liw | pianohacker, I need to dash to the train, but: dh-make-perl shouldn't require any root access, so running it under sudo should be unnecessary; the #debian-perl channel (this irc network) is usually very friendly and helpful, when anyone's awake |
| 17:24 | cda_brannon joined #koha | |
| 17:25 | Alphaman joined #koha | |
| 17:31 | cda_brannon | \nick |
| 17:34 | cait joined #koha | |
| 17:39 | CBrannon | \notice cait Got a moment? |
| 17:40 | oleonard | CBrannon: You don't have to "/" anything to get cait's attention, just say her name :) |
| 17:40 | cait | here :) |
| 17:42 | CBrannon | I replied to your comment in bug 9303. I understand that prog and ccsr are depreciated, but I'm wondering if since they are still available as options, should preferences indicate what themes will or won't work with certain themes? |
| 17:42 | huginn | 04Bug http://bugs.koha-community.org[…]w_bug.cgi?id=9303 enhancement, P5 - low, ---, kyle, Failed QA , relative's checkouts in the opac |
| 17:42 | cait | CBrannon: all the files will be removed soon I think |
| 17:43 | actually i tihnk we deprecated them a while ago... and now we are about to remove them totally | |
| 17:43 | the next release won't have them | |
| 17:44 | hope i ammaking sense :) | |
| 17:44 | bit distracted by watching germany vs USA :) | |
| 17:45 | CBrannon | What happens if a patch is applied to an older version though? Is there some indication that it won't work? Should the patch be limited to specific versions at that point? |
| 17:45 | rangi | http://wiki.koha-community.org[…]precation_of_the_.27prog.27_and_.27CCSR.27_OPAC_themes |
| 17:46 | cait | CBrannon: patches iwth new features are not backported to older versions |
| 17:46 | for a bugfix it would be discussable if it should be backported, but then best to make a separate patch for each theme | |
| 17:46 | oleonard | Hi rangi |
| 17:46 | CBrannon | Okay. Thanks. Still figuring things out. :) |
| 17:46 | rangi | and yes it should, and yeah, thats a new feature so will only go in master, so actually adding the changes to prog .. would make the patch fail qa |
| 17:46 | hi oleonard | |
| 17:46 | cait | CBrannon: no problem, just ask - and thx for taking the times to write notes about your testing, i really like that :) |
| 17:47 | a_m0d joined #koha | |
| 17:48 | sophie_m left #koha | |
| 17:49 | CBrannon | cait: so was it appropriate that I brought up the themes in the bug, since the feature was showing up in them? |
| 17:49 | cait | yes :) |
| 17:49 | i had missed that | |
| 17:50 | CBrannon | Good. I know I've been making some mistakes, but I am trying to get procedures down. I miss some things too when the ticket gets so long. :P |
| 17:51 | cait | you are doing well, really, don't worry |
| 17:52 | CBrannon | Does anyone know if you can apply more than one patch at a time on a sandbox? Particularly if one patch depends on another? |
| 17:54 | cait | hm not sure |
| 17:55 | Joubu would know, but I think it's too late for him to be around | |
| 17:57 | oleonard | But 12464 is caused by some missing CSS not transferred from the prog theme, but copying the relevant CSS doesn't entirely fix the problem. |
| 17:57 | Bug 12464 I mean. | |
| 17:57 | huginn | 04Bug http://bugs.koha-community.org[…]_bug.cgi?id=12464 normal, P5 - low, ---, oleonard, NEW , Bootstrap XSLT view no longer displaying 505 correctly formatted |
| 17:57 | oleonard | It's weird: http://www.screencast.com/t/YtBR8MBiyHg |
| 17:57 | CBrannon | I'll probably send a message to Jonathan an ask. It would be helpful if it did. Maybe they can figure out a way to enter multiple bug numbers separated by commas, if it doesn't already. |
| 17:57 | cait | :) |
| 17:58 | oleonard | Oh and it gets weirder... The arrows appear correctly if I scale the font size up or down... |
| 17:58 | Must be something about how the font renders at certain sizes... | |
| 17:59 | CBrannon | Are bug updates being posted to the IRC? |
| 18:00 | cait | CBrannon: hmm updates are not... commits i think are still |
| 18:00 | whensomething goes into master that is reported here | |
| 18:00 | and the results of the automated testing | |
| 18:00 | and you can always refer to a bug using bug <number> here | |
| 18:00 | and it will bring up the link | |
| 18:01 | you can subscribe to the bugs mailing list if you want info about updates | |
| 18:01 | CBrannon | Ah...I was trying to figure out what was going on. |
| 18:14 | edveal joined #koha | |
| 18:20 | CBrannon | Ist die USA gehen, um zu gewinnen? |
| 18:28 | cait | CBrannon: hm not sure i understand :) but the result was 0:1 for germany |
| 18:33 | CBrannon | cait: I don't claim to have any skill in speaking German. :) Hooray Germany. :) |
| 18:36 | cait | :) |
| 18:41 | CBrannon | barton: you're a little strange. :) |
| 18:41 | oleonard | He must have asked someone to make him a sandwich |
| 18:42 | barton | CBrannon, you're not the first person who's said that ... |
| 18:42 | CBrannon | oleonard: nice |
| 18:43 | barton | yeah, and talljoy was waving a magic wand around in #bywater ... |
| 18:44 | CBrannon | \nick MasterOfAllThingsFluffy |
| 18:44 | * oleonard | wonders if more recent version of git can git-send-email patches with lines longer than 998 characters |
| 18:52 | pianohacker | oleonard: I think you're one of the few people that sends patches to the mailing list these days, is the thing. |
| 18:53 | oleonard | Okay then I'll stop :P |
| 19:08 | jburds_ joined #koha | |
| 19:11 | tcohen joined #koha | |
| 19:11 | tcohen | hi #koha |
| 19:18 | kmlussier joined #koha | |
| 19:19 | cait | ih tcohen |
| 19:43 | jburds__ joined #koha | |
| 19:43 | oleonard_ joined #koha | |
| 19:58 | jburds_ joined #koha | |
| 20:04 | oleonard_ | Bye #koha |
| 20:05 | cait | oleonard++ # all those lovely patches |
| 20:09 | kathryn | morning :) |
| 20:10 | cait | good morning kathryn :) |
| 20:10 | kathryn | hi cait |
| 20:17 | rhcl joined #koha | |
| 20:18 | rhcl | @seen jcamins |
| 20:18 | huginn | rhcl: jcamins was last seen in #koha 4 hours, 25 minutes, and 41 seconds ago: <jcamins> 3.12.9 works fine for me. |
| 20:30 | jburds__ joined #koha | |
| 22:28 | wizzyrea | hi |
| 22:38 | eythian | hi |
| 22:57 | irma joined #koha | |
| 23:09 | jenkins_koha | Starting build #7 for job Koha_3.16.x_U14 (previous build: STILL UNSTABLE) |
| 23:12 | gmcharlt | Koha 3.16.1 is now available - http://koha-community.org/koha-3-16-1-released/ |
| 23:12 | Topic for #koha is now Koha 3.16.1 is available! Next general meeting is 9 and 10 July 2014 at 22:00 and 15:00 UTC. Welcome to the IRC home of Koha http://koha-community.org. Please use http://paste.koha-community.org for pastes. | |
| 23:15 | eythian | oh bother, I forgot to make that patch |
| 23:15 | never mind, I'll roll it in anyway. | |
| 23:16 | huginn | New commit(s) kohagit: Bug 12143 - hiding alternate address on patron self registration doesn't hide heading <http://git.koha-community.org/[…]912cbfd2b3d37565f> |
| 23:19 | gmcharlt | eythian: yep, I figured you might do that |
| 23:37 | jenkins_koha | Project Koha_3.16.x_U14 build #7: STILL UNSTABLE in 28 min: http://jenkins.koha-community.[…]oha_3.16.x_U14/7/ |
| 23:37 | * Galen Charlton: release notes for Koha 3.16.1 | |
| 23:37 | * Galen Charlton: add DBRev for 3.16.1 release |
← Previous day | Today | Next day → | Search | Index
![http://buikitty.files.wordpres[…]120202-094434.jpg](http://buikitty.files.wordpress.com/2012/02/20120202-094434.jpg){kind=link}