Log of the #blacklight channel on chat.freenode.net

Using timezone: GMT-05:00
* rduplain joins09:40
* rduplain leaves09:50
* rduplain joins
* erikhatcher joins10:13
* jamieorc joins10:32
* jkeck joins12:25
* ndushay joins13:15
* bess joins13:41
* bess leaves13:57
* ndushay leaves14:06
* bess joins14:09
* ndushay joins14:10
<bess>hey, does everyone in here know about libdevconX ? 14:13
rduplain, rsinger, jrochkind, jamieorc, erikhatcher, cbeer? 14:14
oh, and MrDys_weekend although the name suggests he might not actually be in here
<rduplain>bess, libdevconX? no.
<bess>rduplain: Stanford is hosting a conference March 23 − 25 around blacklight and repositories14:15
<rsinger>hmm
<bess>rduplain: If you'd like to visit CA, I'm sure I can arrange for an invite for you. No travel funding from us, but it should be pretty fun.
<rsinger>has this been announced anywhere?
<bess>rsinger: no, it's invite only
rsinger: but this is me inviting you
<rsinger>heh :)
<rduplain>bess, that would explain 0 hits on Google
make that 1 when these logs go online14:16
<rsinger>well, actually the timing on that is a little bad for me, unfortunately :(
<rduplain>I actually have those days open... need to be back in Cville Thursday night, though
well, Friday morning really
<bess>rduplain: come on, you know you want to see California. And you could take the red eye back and be in C'ville Friday morning14:17
<rduplain>bess, I've been to Palo Alto, albeit only passing through
<bess>rduplain: see? Time to spend more time here. 14:18
<rduplain>bess, really tempting
<bess>rduplain: well, let me know if you want more info or an official invite to be sent to your employer
<rduplain>bess, shouldn't be too hard to get the funds, esp since I'm not going to PyCon this year14:19
bess, shoot me an invite rduplain@nrao.edu
<erikhatcher>libdevconX++ ... i'm pondering it now myself... but really shouldn't tangent14:20
<bess>erikhatcher: I'm having lunch w/ your boss today. Want me to talk it up how important it would be to have you there? ;)
<erikhatcher>please14:23
my boss?
David?
Taly?
Eric?
too many bosses!
i'm due for a CA trip14:24
implore them to bring me out, i might just do it
<rduplain>"I have 8 bosses, Bob."14:25
bess, I take it you are doing well at Stanford? :-)14:40
<bess>rduplain: Yes, I am. Thanks for asking. I looove it here. :)14:41
erikhatcher: lunch w/ David
<rduplain>bess, that's awesome. sure to be a good fit
cville is cold, windy, and covered in snow
in case you were wondering
<bess>rduplain: Palo Alto is a little overcast, but mostly warm and balmy, and the flowers are starting to bloom. Also, there are oranges and lemons on the trees. 14:47
<rduplain>bess, again congrats on the job :-P
<bess>rduplain: thanks! :) I'm pretty happy about the whole thing myself. 14:48
<rduplain>bess, how much of your time are you spending on blacklight?14:50
<bess>rduplain: not much right now, but that's going to change. One of the reasons I took this job was b/c I'll be able to spend a LOT more time on it14:52
rduplain: mostly on getting it integrated as a front end for data repositories. Which is one of the reasons it would be great to get a closer alignment w/ NRAO14:53
<rduplain>bess, I advocated integrating multiple data repositories under a single blacklight front-end
bess, people just want search apps14:54
bess, so I'm deploying multiple blacklight instances, one for each app
<bess>okay cool
that works too. :)
<rduplain>weird I know, but it's still easier than rolling my own
<bess>rduplain: not weird at all. :)14:55
<rduplain>in fact, it's extremely easy to build search apps this way, customizations are almost entirely in the config files
<ndushay>hey all - libdevconx official invites are coming - stalled with overextended dept head at the moment14:56
<rduplain>I'll send links with each new deployment... been slow deploying. Had to fight for Ruby on Rails. *sigh*
<ndushay>rduplain: we want to have a unified front end for diff content sources
not clear if one big index or federated back end search
also not clear if multiple tabs in UI or what.
<rduplain>ndushay, that might come some day, but at the moment just having fun with heavy reuse of amazing search code14:57
<ndushay>rduplain: cool.
rduplain: coming to code4lib? solr blackbelt?
* bess leaves14:58
<ndushay>"amazing search code" meaning solr in general?
<rduplain>ndushay, currently light on travel plans
so no14:59
<ndushay>rduplain: i'm largely responsible for specifics of solr in bl-demo; come to libdevconx and talk to me!
<rduplain>by "amazing search code" I mean that I can demo a complete search app in hours to a week using blacklight and Solr
<ndushay>yeah! isn't that awesome!
yay team!
okay, off to lunch with Erik's boss.15:00
<rduplain>later
* ndushay leaves
* goodieboy joins15:09
* ndushay joins16:36
* Naomi joins
* ndushay leaves16:38
* ndushay joins
* Guest21049 leaves16:41
* goodieboy leaves
* bess joins16:46
* mattmitchell joins16:49
* bess leaves16:50
<ndushay>erikhatcher: u here?16:57
* rduplain leaves17:04
* bess joins17:29
* mattmitchell leaves17:32
<cbeer>bess++ -- i'd love to hear more about libdevconX .. not sure it can happen for me, but...17:34
<bess>cbeer: I'd like to get my boss to send an official invite to your boss
Do you think you might not make it b/c of money, or b/c of other commitments? 17:35
<cbeer>money might be a problem, but we're launching 2 sites in april, so who knows what march will look like
bess@stanford?17:37
<bess>cbeer: that's me
* bess leaves17:38
* jvenner joins18:13
request for help with 2.4, I just downloaded and installed the demo app, and edited the config/initializers/blacklight_config.rb file with data for my schema18:14
I run script/server
which binds to localhost:3000
however when I attempt t connect from a browser I get a redirect error
* mattmitchell joins18:16
* bess joins18:34
* bess leaves18:35
* bess joins18:42
* bess leaves18:49
* mattmitchell leaves18:50
<erikhatcher>ndushay: i'm here now18:59
<ndushay>erikhatcher: tried to convince David to have you at libdevconX19:01
he totally shot it down19:02
he's thinking some other guy i haven't heard of.
i am leaning hard on Tom Cramer to get the invites out TODAY.
we'll be sure to invite you anyway.
* cbeer leaves19:25
* bess joins19:32
* bess leaves
* rduplain joins19:42
* ndushay leaves20:56
* jkeck leaves21:04
* cbeer joins21:13
* erikhatcher leaves21:22
* jvenner leaves21:26
* jamieorc leaves21:36
* jvenner joins21:59
* ndushay joins22:00
<jvenner>thank you to ndushay I am up and running22:35
<ndushay>i'm so glad to hear it jason
<jvenner>I had a plethora of user generated problems and one actual blacklight/solr integration issue
<ndushay>and you actually solved all your own problems
I just provided moral support
<jvenner>the key issue is that with stock solr 1.4, the empty q= strings passed for the facet queries result in a NullPointerException in the query parser. This should have been fixed but clearly it is not.22:36
What I ended up doing is the following horrible block of ruby code in solr_helper.rb
# puts "Raw input is #{input[:q]}" |Rendering template within layouts/application
if input[:q].nil? |Rendering catalog/index
# puts "Input null #{input[:q]}" |Rendered catalog/_hidden_filters (1.0ms)
mq = "*:*" |Rendered catalog/_search_form (2.8ms)
else |Rendered catalog/_did_you_mean (0.1ms)
# puts "Input valid #{input[:q]}" |Rendered catalog/_constraints (9.7ms)
mq = input[:q] |Rendered catalog/_hidden_filters (0.9ms)
end |Rendered catalog/_hidden_filters (0.9ms)
# puts "mq is #{mq}" |Rendered catalog/_sort_and_per_page (5.7ms)
per_page = per_page.to_i > 100 ? 100 : per_page |Rendered catalog/_bookmark_control (1.1ms)
{ |Rendered catalog/_index_partials/_default (0.5ms)
:qt => qt, |Rendered catalog/_bookmark_control (0.1ms)
:per_page => per_page.to_i, |Rendered catalog/_index_partials/_default (0.4ms)
:q => mq, |Rendered catalog/_bookmark_control (0.1ms)
:phrase_filters => input[:f], |Rendered catalog/_index_partials/_default (0.4ms)
:facets => {:fields=>facet_fields}, |Rendered catalog/_bookmark_control (0.1ms)
where I substitute mq with either *:* or the actual q passed in the input array
oh funny, my past is googy22:37
goofy from my split window text emacs22:38
around line: 53
# puts "Raw input is #{input[:q]}"
if input[:q].nil?
# puts "Input null #{input[:q]}"
mq = "*:*"
else
# puts "Input valid #{input[:q]}"
mq = input[:q]
end
# puts "mq is #{mq}"
per_page = per_page.to_i > 100 ? 100 : per_page
{
:qt => qt,
:per_page => per_page.to_i,
:q => mq,
:phrase_filters => input[:f],
:facets => {:fields=>facet_fields},
<ndushay>jvenner: is your default request handler using dismax?22:39
b/c dismax accepts empty q
<jvenner>Next on my agenda will be figuring out how to allow more facets when the users wants them
<ndushay>but standard does not.
<jvenner>I actually cut and pasted the default handler from the example code
and altered the dismax fields to match my schema
and I still received the NPE's
there may be more user level confusion in place22:40
but empty q should work
and I got the lovely NPE
<ndushay>if you send me your solrconfig.xml i'll take a quick look22:41
<jvenner>do you have a moment to explain how the search sorting fields work?
email?
<ndushay>see private message
sorting fields: just fields that have the correct lexical sort22:42
in marc, 245 fields allow indication of "non-filing" characters, like "the" at the beginning of a title.
so sort field can really be anything that will order results lexically the way you want.22:43
<jvenner>in the config the line looks like this:
config[:sort_fields] << ['relevance', 'score desc, pub_date_sort desc, title_sort asc']
what are the values in the array
<ndushay>ah22:44
the first is the word that will appear in the pull down ("sort results by")
the second is the argument passed to Solr22:45
in that case, by score (descending order), then by pub date (descending order), then by title (ascending order)
where pub_date_sort and title_sort are specific fields in the index designed for sorting. :-)
<jvenner>making good progress, thank you22:51
I suspect my dumb questions make good pointers for additional comments22:52
things are working very nicly23:04
Is there a way to have a more button below a set of facet counts, to get the next page of facet counts?23:05
<ndushay>yes, but it's not yet in projectblacklight23:15
we've done it
uva has done it
ask on bl-dev23:16
cbeer: u around ?
* jkeck joins23:18
<ndushay>jessie!23:24
* jkeck leaves23:55

Generated by Sualtam