| Time |
Nick |
Message |
| 00:44 |
|
StomproJ joined #evergreen |
| 01:46 |
|
remingtron_ joined #evergreen |
| 01:56 |
|
jeff_ joined #evergreen |
| 02:45 |
|
Stompro joined #evergreen |
| 03:26 |
|
artunit_ joined #evergreen |
| 03:30 |
|
artunit_ joined #evergreen |
| 05:04 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:09 |
|
rjackson_isl joined #evergreen |
| 07:12 |
|
JBoyer joined #evergreen |
| 07:32 |
|
agoben joined #evergreen |
| 07:46 |
csharp_ |
hmm - seeing a weird issue trying to update the location for a particular set of copies - I'll issue the statement 'update asset.copy set location = loc2 where location = loc1 and circ_lib = blah' but when I look at them afterwards, the location is still set to 'loc1' :-/ |
| 07:47 |
csharp_ |
and it appears to only be happening to copies being set to a specific location (in this case, moving from a branch-owned "New Books" location to a system-owned "New Books" location) |
| 07:48 |
csharp_ |
other copies moving to other locations are acting as expected |
| 07:49 |
miker |
csharp_: yep. we disable the trigger that does that. no way to configure it. we are not a fan of that code |
| 07:49 |
csharp_ |
ahh |
| 07:50 |
csharp_ |
miker: acp_location_fixer_trig? |
| 07:50 |
miker |
makes consolidation a huge pain |
| 07:51 |
miker |
yep, that'd be the one |
| 07:51 |
csharp_ |
ooohhhh |
| 07:52 |
csharp_ |
and I'm somewhat responsible for that |
| 07:52 |
csharp_ |
(or it originated as a PINES library problem) |
| 07:53 |
csharp_ |
unintended_consequences-- |
| 07:55 |
csharp_ |
I'll think on how to improve that :-/ |
| 07:55 |
|
csharp joined #evergreen |
| 08:01 |
csharp |
6e6540aa |
| 08:01 |
pinesol_green |
csharp: [evergreen|Thomas Berezansky] LP#778989: Attempt to find "Correct" copy location - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6e6540a> |
| 08:02 |
csharp |
for the logs, this is the bug that originated the change: https://bugs.launchpad.net/evergreen/+bug/778989 |
| 08:02 |
pinesol_green |
Launchpad bug 778989 in Evergreen "Owning lib of asset.copy_location not visible in Item Attributes Editor UI" [Medium,Fix released] |
| 08:04 |
csharp |
the whole thing would be unnecessary with a UI fix that showed the current location's owning lib ( I haven't checked the web client for that ) |
| 08:12 |
JBoyer |
Quick i18n Q: Is this page still current? https://wiki.evergreen-ils.org/doku.php?id=evergreen-admin:customizations:i18n |
| 08:13 |
JBoyer |
We're hoping to try out the improved Spanish translations and that's all I could find so far. |
| 08:24 |
|
rlefaive joined #evergreen |
| 08:31 |
|
kmlussier joined #evergreen |
| 08:38 |
|
mmorgan joined #evergreen |
| 08:40 |
csharp |
JBoyer: we copied the file from /build/i18n/po/opac.dtd/es-ES.po into /openils/var/data/locale/opac, added relevant lines to /etc/apache2/eg_vhost.conf (inline examples are there), inserted the locale into the database as instructed on that page, restarted OpenSRF and Apache and it was there |
| 08:41 |
csharp |
(Terran may have done something else on the tt2 level that I'm not aware of, but that did it for us |
| 08:41 |
csharp |
) |
| 08:42 |
csharp |
oh, actually one of the db upgrade scripts inserts the locale for you |
| 08:43 |
|
collum joined #evergreen |
| 08:46 |
JBoyer |
ls |
| 08:58 |
tsbere |
snow-- |
| 08:58 |
tsbere |
traffic-- |
| 09:00 |
csharp |
@weather 30345 |
| 09:00 |
pinesol_green |
csharp: Atlanta, GA :: Overcast :: 45F/7C | Monday: Considerable cloudiness. Occasional rain showers this afternoon. High 58F. Winds light and variable. Chance of rain 70%. Monday Night: Rain this evening with thunderstorms developing overnight. Low 52F. Winds E at 10 to 15 mph. Chance of rain 100%. 1 to 2 inches of rain expected. |
| 09:05 |
kmlussier |
@weather |
| 09:05 |
pinesol_green |
kmlussier: Seekonk, MA :: Light Rain :: 32F/0C | Wind Chill: 26F/-3C | Monday: Cloudy with light rain this morning...then becoming partly cloudy. High around 45F. Winds WNW at 5 to 10 mph. Chance of rain 70%. Monday Night: A clear sky. Low 33F. Winds WNW at 5 to 10 mph. |
| 09:06 |
tsbere |
@weather |
| 09:06 |
pinesol_green |
tsbere: North Andover, MA :: Light Snow :: 29F/-2C | Wind Chill: 25F/-4C | Monday: Snow showers in the morning will give way to a mixture of rain and snow for the afternoon. High 38F. Winds SSW at 5 to 10 mph. Chance of precip 90%. Snow accumulations less than one inch. Monday Night: Mostly clear. Low around 30F. Winds W at 5 to 10 mph. |
| 09:06 |
kmlussier |
Huh, light rain leaves white powder on the ground? |
| 09:09 |
kmlussier |
@coffee [someone] |
| 09:09 |
* pinesol_green |
brews and pours a cup of Colombia Finca Cerro Azul, and sends it sliding down the bar to jeff |
| 09:09 |
kmlussier |
@tea [someone] |
| 09:09 |
* pinesol_green |
brews and pours a pot of Bao Zhong Oolong Tea, and sends it sliding down the bar to dbwells (http://ratetea.com/tea/rishi/bao-zhong-oolong/768/) |
| 09:10 |
mmorgan |
kmlussier: It's a dry rain. |
| 09:11 |
kmlussier |
That explains it! |
| 09:15 |
JBoyer |
mmorgan++ |
| 09:22 |
rjackson_isl |
A couple of days ago our weather service was hinting at 0 wind chill come Thursday night. winter-- |
| 09:25 |
|
yboston joined #evergreen |
| 09:28 |
tsbere |
rjackson_isl: The way you wrote that I choose to interpret it as "the wind won't be chilling anything". If only because other interpretations are depressing. ;) |
| 09:29 |
rjackson_isl |
tsbere++ for some positive news! |
| 09:30 |
csharp |
@who is chillin'? |
| 09:30 |
pinesol_green |
RBecker is chillin. |
| 09:30 |
gmcharlt |
tossing this out as an experiment - an RSS feed of updates to Perl modules that Evergreen depends on: https://www.perlmodules.net/feed/personal/Mag7ld0xXxF9sSAhsPgaPDiXF1hrfaVlnA7lJApPPMQbrrde82 |
| 09:34 |
|
maryj joined #evergreen |
| 09:38 |
kmlussier |
gmcharlt++ |
| 10:26 |
|
Dyrcona joined #evergreen |
| 10:29 |
bshum |
csharp: Wait, opac.dtd? That doesn't sound right... it should be "tpac" for the catalog |
| 10:29 |
bshum |
opac.dtd is for JSPAC |
| 10:35 |
tsbere |
Gotta hate cryptic messages like "Maintenance required" with no details and no apparent way to *get* details. |
| 10:35 |
bshum |
JBoyer: That wiki page is somewhat accurate, if one were building the locales from scratch via git install. Otherwise, from a tarball, copying as csharp indicates should do the trick just fine. |
| 10:36 |
bshum |
Cause the locale files should already be all created. |
| 10:36 |
bshum |
I'm hoping to fine-tune those instructions from that wiki page to make it easier for future i18n setup for git installers (like me) |
| 10:52 |
|
Christineb joined #evergreen |
| 10:53 |
csharp |
bshum: oh, maybe I'm wrong :-/ |
| 10:53 |
* csharp |
doesn't mean to mislead nobody |
| 10:54 |
|
bos20k joined #evergreen |
| 10:58 |
bshum |
csharp: Many strings would likely be similar between tpac and opac.dtd. But tpac is definitely the file that's mapped to the tt2 strings from that directory |
| 10:58 |
bshum |
I didn't get to try installing from a tarball, but I wonder why we need to manually copy over the file. It used to copy that stuff in automagically. |
| 10:58 |
bshum |
Maybe we're missing a step in the makefile somewhere |
| 11:09 |
csharp |
well, in my case, the target directory /openils/var/data/locale/opac doesn't exist by default (installed from a deb created by 2.11.1 tarball) |
| 11:10 |
csharp |
s/by 2.11.1 tarball/with PINES-ified 2.11.1 tarball/ |
| 11:18 |
JBoyer |
Since I'm also a git-only installer I just added the commands on that page to my playbook, and now all of the files are where they belong. A quick change to eg_vhost.conf and things are rolling. Thanks! |
| 11:19 |
JBoyer |
csharp++ |
| 11:19 |
JBoyer |
bshum++ |
| 11:21 |
bshum |
JBoyer++ # neato :) |
| 11:21 |
JBoyer |
Though I will say it's quite difficult to see what |
| 11:21 |
bshum |
JBoyer: My plan for 2.next is to get an i18n make target that'll have all those pre-reqs on the side. |
| 11:22 |
bshum |
I started up a working branch, but need to tidy it up and then get an LP opened for it |
| 11:22 |
JBoyer |
s been translated currently. :/ dev.evergreen.lib.in.us lets me change to Espanol, but it looks like only the DB translations made it through. |
| 11:22 |
bshum |
They're all already part of the "packager" target. But let's say I didn't want to wait for asciidoc to download and install if I just wanted to play with translations. |
| 11:23 |
JBoyer |
I ran the i18n stuff pre-build, I expected them to be copied over with the english versions, was I assuming too much? |
| 11:23 |
bshum |
JBoyer: It's possible that there's something getting missed and the actual installed files aren't getting put into place, even though there's stuff for source |
| 11:24 |
JBoyer |
I'll check around rather than noise up the channel. It's also possible I'm missing some other apache settings. |
| 11:26 |
bshum |
JBoyer: Verify that you've got files in that installed place, like /openils/var/web/locale/opac/* (or wherever you told eg_vhost.conf to look for locale files) |
| 11:57 |
|
jihpringle joined #evergreen |
| 12:06 |
|
jihpringle joined #evergreen |
| 12:13 |
|
mmorgan joined #evergreen |
| 12:26 |
|
jvwoolf joined #evergreen |
| 13:00 |
csharp |
berick: FYI, I'm finally testing your new hold targeter on our test server - so far so good |
| 13:00 |
csharp |
I like --verbose mode |
| 13:02 |
|
brahmina joined #evergreen |
| 13:02 |
|
kmlussier joined #evergreen |
| 13:10 |
kmlussier |
@coffee |
| 13:10 |
* pinesol_green |
brews and pours a cup of Panama La Esmeralda Especial Mario San Jose, and sends it sliding down the bar to kmlussier |
| 13:13 |
berick |
csharp++ awesome! |
| 13:13 |
berick |
csharp: just holler if you find any issues. i'll patch 'em up. |
| 13:27 |
csharp |
with 3 parallel processes, my 2-core server with 6GB of RAM is totally happy - RAM usage and system load are nice and low |
| 13:36 |
* Dyrcona |
needs to find a faster way to search for matching records. |
| 13:42 |
JBoyer |
re: my es-ES mystery, the examples in Open-ILS/examples/apache_24 don't have complete paths for the translations. Need to stick an opac/ in there before the es-ES.po. Almost there, now it looks like there's been some message id drift that I get to fight with... D: |
| 13:42 |
berick |
csharp: are you targeting a full pines data set? |
| 13:45 |
bshum |
JBoyer: Oh, interesting. Maybe we should add a patch to fix the sample paths for the translations for apache 24. |
| 13:45 |
bshum |
Also, when is Wheezy going away? And then we can get rid of old apache and move all the apache_24 stuff in there. :D |
| 13:46 |
bshum |
"Wheezy also benefits from Long Term Support (LTS) until the end of May 2018." Okay... guess it'll be awhile yet. At least till Debian ends. |
| 13:46 |
JBoyer |
That would explain Terran needing to move the file from one place to another before it would work. |
| 13:47 |
JBoyer |
Did the translation layout change significantly at some point? (i.e. going from one big .po to a bunch of them spread around?) |
| 13:47 |
csharp |
berick: yes |
| 13:47 |
csharp |
it's a couple of months old |
| 13:48 |
bshum |
JBoyer: There's always been a bunch of PO files for translation as applied to different areas. |
| 13:48 |
bshum |
There's just more of them now since I started adding more folders of files for translation in more recent releases. |
| 13:48 |
bshum |
But for the catalog, they're all under the "tpac" PO file for all of it |
| 13:49 |
|
mmorgan joined #evergreen |
| 13:49 |
berick |
csharp: great. should be a good test, then. |
| 13:51 |
bshum |
Interesting... |
| 13:51 |
JBoyer |
Still called opac, unless there are two. I'm pointing the es_es local at /openils/var/data/locale/opac/es-ES.po |
| 13:51 |
bshum |
Well yes, it's relabeled from tpac to opac by the makefile job |
| 13:52 |
bshum |
During the install |
| 13:52 |
bshum |
Question, do you have anything in /openils/var/templates/locale/ > |
| 13:52 |
bshum |
I think the Makefile has been putting all the tt2 translation files over there instead |
| 13:54 |
|
mmorgan1 joined #evergreen |
| 13:54 |
JBoyer |
Ah. I do, all of the loose .po files (including tpac.es-ES.po) I'll redirect that variable and see if that clears up the weird labeling issue we're seeing. |
| 13:56 |
bshum |
That might be part of it then |
| 13:56 |
bshum |
And maybe we need to suggest alternative places for the installed files for the apache configs |
| 13:57 |
bshum |
See, but then that's confusing |
| 13:57 |
bshum |
TT2_OPAC_DATA_DIR=../../Open-ILS/src/data/locale/opac |
| 13:57 |
bshum |
So it should end a copy over there too |
| 13:57 |
bshum |
mkdir -p $(TT2_OPAC_DATA_DIR) |
| 13:57 |
bshum |
cp $(PO)/tpac/$(LOCALE).po $(TT2_OPAC_DATA_DIR)/. |
| 13:58 |
bshum |
So it makes that spot, and then it copies the PO file for tpac over to it |
| 13:59 |
bshum |
Something must be wonky with how I'm reading things |
| 13:59 |
* dbs |
is tempted to look at that makefile again |
| 14:01 |
dbs |
run from an account that didn't have write permission? |
| 14:03 |
dbs |
(we install from git, on ubuntu 14.04 so apache 2.4 as well, and I don't recall having to do much to make our fr-CA locale work) |
| 14:06 |
JBoyer |
I have both /openils/var/data/locale/opac/es-ES.po and /openils/var/templates/locale/tpac.es-ES.po and diff tells me they're identical. There should probably be 1, a single location; and 2, an update to the apache config at the same time. |
| 14:07 |
dbs |
all of ours are only in /openils/var/data/locale/opac/ and Open-ILS/examples/apache_24/eg_vhost.conf.in includes lines like #PerlAddVar OILSWebLocale "@localstatedir@/data/locale/opac/en-CA.po" |
| 14:07 |
bshum |
It could be leftovers from when we build the PO files from scratch. And they're getting stored in templates/locale |
| 14:07 |
bshum |
And not getting removed, before install happens |
| 14:08 |
bshum |
Copying it over |
| 14:08 |
dbs |
thanks to 36fa29fe |
| 14:08 |
pinesol_green |
dbs: [evergreen|Ben Shum] LP#1544606: Change apache examples for TPAC to more specific locale/opac location - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=36fa29f> |
| 14:08 |
bshum |
I would expect it only to be for /openils/var/data/locale/opac/* |
| 14:08 |
bshum |
Oh, okay, I thought that looked familiar... |
| 14:08 |
dbs |
bshum++ |
| 14:08 |
bshum |
:) |
| 14:08 |
bshum |
I thought /opac/ was in there somewhere, haha |
| 14:09 |
dbs |
we're on 2.10, so maybe a regression? haven't looked forward in time |
| 14:09 |
bshum |
I only focused on master, I don't know what if any got backported. |
| 14:09 |
bshum |
So, rel_2_9 |
| 14:10 |
bshum |
according to the LP |
| 14:10 |
bshum |
And forward |
| 14:10 |
bshum |
So it should be fine |
| 14:23 |
|
jeffdavi1 joined #evergreen |
| 14:28 |
|
jeffdavis joined #evergreen |
| 14:50 |
Dyrcona |
Are there any example SQL queries that use index_vector on metabib.real_full_rec and tsquery? |
| 14:50 |
miker |
Dyrcona: you should be able to provoke one for logging by using the "expert" search |
| 14:51 |
miker |
IIRC, it uses plain_to_tsquery (or however that's spelled) |
| 14:51 |
Dyrcona |
miker: Thanks! I've been trying using the Pg docs and some examples and so far all I get is syntax error in tsquery |
| 14:53 |
|
buddytheelf joined #evergreen |
| 14:54 |
bshum |
"I really can't stay..." |
| 14:54 |
Dyrcona |
Well, that's different. I'm getting notices about it not containing lexemes "". |
| 14:55 |
Dyrcona |
So I guess something is sort of working. :) |
| 14:57 |
Dyrcona |
Or not, but something different is "progress." |
| 14:58 |
dbs |
Dyrcona: https://stuff.coffeecode.net/2012/pgopen_fulltext/ holds my gentle intro to full-text search in PostgreSQL |
| 15:00 |
dbs |
in the PDF, section 22/23 covers tsvectors |
| 15:00 |
Dyrcona |
dbs: Thanks, but so far I'm not seeing anything new. I'm just having issues building my query string correctly, I think. |
| 15:02 |
dbs |
e.g. SELECT value FROM metabib.full_rec WHERE to_tsquery('Apache & Derby') @@ index_vector; |
| 15:02 |
dbs |
Dyrcona: okay, so help us help you by showing us what you're doing? |
| 15:03 |
Dyrcona |
Well, for one thing, I'm not editing my code correctly. :) |
| 15:05 |
Dyrcona |
Here's the problem: I want to quickly search metabib.real_full_rec for records mathing 035 and/or 020, and I need records that match either or both. I also need to match on item type and form in mravl. |
| 15:05 |
Dyrcona |
I've got a query using the value field that works, but it is sloooowww. |
| 15:06 |
Dyrcona |
So, I figured I try tsquery, but that's turning out to be trickier than I thought. |
| 15:06 |
Dyrcona |
As for showing what I've got. I'm not ready to do that at this moment. |
| 15:08 |
Dyrcona |
I think my biggest problem is that I have a variable number of 035 and 020 fields in my input, anywhere from 0 to I-don't-know-without-actually-counting, but usually 2 035s and not more than 2 to 4 020s, if present. |
| 15:09 |
Dyrcona |
So, before I had value = any(?) in my where and passed in the arraryref to the bind_param statement. |
| 15:10 |
Dyrcona |
What I'm finding is trying to do an equivalent with tsquery isn't so simple.... |
| 15:13 |
Dyrcona |
Hmm.. I didn't try passing an arrayref of tsquery to any after @@. I doubt that will work, but what's the worst that will happen? ;) |
| 15:14 |
tsbere |
Dyrcona: index_vector @@ to_tsquery(array_to_string(?,' | ')) |
| 15:15 |
Dyrcona |
tsbere: Maybe, but that will match too much. I really want 'blah & something' || 'notblah & something_else' |
| 15:16 |
tsbere |
Dyrcona: Then I think you need to build your string before calling your query. Also, I think tsquery strings support '(blah & something) | (notblah & something_else)' type syntax |
| 15:17 |
Dyrcona |
Yes, I started with with something almost like that. |
| 15:17 |
Dyrcona |
When I started down this path. |
| 15:18 |
Dyrcona |
I was casting each 'str1 & str2' pair to tsquery and joining with ||, but I'm sure I did it wrong. :) |
| 15:22 |
Dyrcona |
So, I'm getting NOTICE: text-search query doesn't contain lexemes: "" again, which I think happens when I end up with an empty tsquery. |
| 15:34 |
miker |
Dyrcona: yes (to the last statement) it is |
| 15:55 |
|
mmorgan1 joined #evergreen |
| 16:08 |
|
Enjabain joined #evergreen |
| 16:11 |
Enjabain |
Hello, I am wondering if there is an API or feed from evergreen that I can use to automatically list new additions to our library on our own site (Not the catalog site). Not sure if this is the place to ask. |
| 16:12 |
Dyrcona |
Enjabain: There is a feed. |
| 16:13 |
Dyrcona |
Enjabain: https://wiki.evergreen-ils.org/doku.php?id=backend-devel:supercat:examples&s[]=freshmeat#return_a_feed_of_recently_edited_or_created_records |
| 16:15 |
Dyrcona |
The example URL from GA PINES returns an atom feed. You can get other formats. |
| 16:18 |
Enjabain |
This is great, is there a limit to how many records you can get? |
| 16:18 |
Dyrcona |
Well, yes, you don't want to try to get the entire database. ;) |
| 16:19 |
Dyrcona |
It really depends on your hardware, network, and load, etc. |
| 16:19 |
Enjabain |
OK just making sure it wasn't 10 the wording was a little confusing |
| 16:20 |
Dyrcona |
Twenty to thirty should be OK. |
| 16:21 |
Dyrcona |
More than that will likely work, but I wouldn't go over 100 or so. :) |
| 16:30 |
|
kmlussier joined #evergreen |
| 17:00 |
|
mmorgan1 left #evergreen |
| 17:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 17:17 |
|
jvwoolf left #evergreen |
| 17:45 |
csharp |
85a470ef |
| 17:45 |
pinesol_green |
csharp: [evergreen|Dan Pearl] LP#1501781 - Make patron name search diacritic/space insensitive. - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=85a470e> |
| 17:52 |
csharp |
miker++ # the usrname index did the trick |
| 19:51 |
|
StomproJ joined #evergreen |
| 23:19 |
|
Stompro joined #evergreen |
