| Time |
Nick |
Message |
| 04:50 |
|
lualaba joined #evergreen |
| 04:50 |
lualaba |
hello. i am not able to delete row delete from asset.copy_location where id = 113 please help |
| 04:59 |
lualaba |
select * from asset.copy where location = 113 (zero result) |
| 05:01 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 05:02 |
lualaba |
i need just remove from asset.copy_location one record (deleted = 't') but anyway i see in web staff client record |
| 07:15 |
|
rjackson_isl joined #evergreen |
| 07:22 |
|
JBoyer joined #evergreen |
| 07:41 |
|
dteston joined #evergreen |
| 07:53 |
csharp |
@later tell lualaba if you're able to share the error message(s) you're getting, someone may be able to help |
| 07:53 |
pinesol_green |
csharp: The operation succeeded. |
| 07:56 |
|
berick joined #evergreen |
| 08:12 |
|
agoben joined #evergreen |
| 08:24 |
|
rlefaive joined #evergreen |
| 08:29 |
|
dteston joined #evergreen |
| 08:32 |
* csharp |
learns that setting his Windows VM's password hint to "you know what it is" actually works |
| 08:39 |
tsbere |
csharp: I am personally fond of "why are you asking me?" |
| 08:46 |
csharp |
heh |
| 08:47 |
* csharp |
enables sudo insults just because it's Monday |
| 08:50 |
|
Dyrcona joined #evergreen |
| 08:50 |
csharp |
@insult sudo insults |
| 08:50 |
pinesol_green |
sudo insults: You are nothing but a left-over quart of swag-bellied manuals. |
| 08:55 |
|
bos20k joined #evergreen |
| 08:57 |
tsbere |
csharp: That is a wonderful idea! All the machines I am willing to enable sudo insults on, however, don't require passwords for sudo. |
| 09:02 |
|
maryj joined #evergreen |
| 09:02 |
|
maryj_ joined #evergreen |
| 09:04 |
|
maryj joined #evergreen |
| 09:19 |
|
kmlussier joined #evergreen |
| 09:39 |
|
maryj joined #evergreen |
| 09:49 |
|
yboston joined #evergreen |
| 09:58 |
|
DPearl joined #evergreen |
| 09:59 |
bshum |
csharp: Isn't there a delete flag on copy locations now? I can't recall if there's a delete protect trigger on the table now, so getting rid of the row might not be all that simple. |
| 10:00 |
bshum |
Oh, I re-read luababa's thing and I get it, web client view, hmm |
| 10:00 |
* bshum |
goes back to trying to forget that it's Monday |
| 10:02 |
|
Lualaba joined #evergreen |
| 10:04 |
Lualaba |
I am not getting any err |
| 10:07 |
kmlussier |
bshum: Yes, there is a delete flag on copy locations now. |
| 10:07 |
|
berick joined #evergreen |
| 10:09 |
|
barbara joined #evergreen |
| 10:22 |
|
mmorgan1 joined #evergreen |
| 10:25 |
csharp |
@later tell Lualaba there's probably some sort of error in the syslog (check opensrf and postgres logs) |
| 10:25 |
pinesol_green |
csharp: The operation succeeded. |
| 11:05 |
|
Christineb joined #evergreen |
| 11:05 |
|
sandbergja joined #evergreen |
| 11:15 |
|
abowling joined #evergreen |
| 11:31 |
Stompro |
Good morning Evergreen, I'm trying to setup a synonym dictionary as described at https://wiki.evergreen-ils.org/doku.php?id=scratchpad:brush_up_search#synonym_dictionary. Should the results of the synonym dictionary show up in metabib.title_field_entry? |
| 11:31 |
|
brahmina joined #evergreen |
| 11:39 |
dteston |
Are there any plans to change the password storage from MD5 to SHA, crypt, etc.? I saw bug 1468422 and would like to help implement a new hashing/encryption style if possible |
| 11:39 |
pinesol_green |
Launchpad bug 1468422 in Evergreen "Improve Password Management and Authentication" [Undecided,Fix released] https://launchpad.net/bugs/1468422 |
| 11:41 |
Dyrcona |
dteston: Notice the status of Fix Released on that bug. It is done, and became available in Evergreen 2.10. |
| 11:43 |
Dyrcona |
dteston: I heard some chatter at the Hack-a-way about making more changes to improve it. I think berick and jeff were the voices I overheard. :) |
| 11:44 |
berick |
dteston: there are some md5 remnants in the code for backwards compatibility, but the passwrds are not stored as md5 anymore. |
| 11:45 |
berick |
they're salted and bcrypt'ed |
| 11:48 |
kmlussier |
Stompro: The instructions came from NOBLE, but I followed along with them when I posted them to the wiki. My memory of the whole procedures is fuzzy, but I'm pretty sure the synonyms should show up in the index_vector. |
| 11:48 |
dteston |
berick: thanks, that's just what I was looking for. How does the code differentiate between bcrypt and MD5? |
| 11:49 |
Dyrcona |
Stompro: I was just about to say what kmlussier said, more or less. Step 6 makes me think that they should show up in the index_vector. |
| 11:51 |
Stompro |
kmlussier, thanks, I'll keep poking at it then. I'm trying to get "&" and "and" to be interchangable, but I'm not seeing results after a reingest, but other things like "20th" -> "twentieth" are also not showing up.. so I must have missed something. |
| 11:52 |
berick |
dteston: the MD5 hashing occurs at a layer above the password storage. it's only used for communicating with the authentication API. at the storage/db layer, there is no md5, with the 1 exception of the password migration function. |
| 11:53 |
berick |
well, that and existing passwords are derived from the old md5 hashed passwords |
| 11:53 |
Stompro |
Is it ok for multiple ts_config metabib classes to have the same index_weight? It looks like title, english_nostop will share the same weight as the new synonym ts_config I created. |
| 11:56 |
kmlussier |
Stompro: I don't think that would be a problem. |
| 12:07 |
* kmlussier |
did a quick test and can confirm that the synonym should show up in the index_vector. |
| 12:08 |
|
jihpringle joined #evergreen |
| 12:15 |
|
mmorgan joined #evergreen |
| 12:33 |
mmorgan |
Stompro: We tried to add "&" and "and" to the synonym list also, but were not successful with that particular entry either. |
| 12:34 |
* csharp |
would expect trouble with characters like &, $, and # unless the code anticipates them |
| 12:34 |
Stompro |
mmorgan, "and" is in the english.stop list, I wonder if that is excluding it? |
| 12:35 |
mmorgan |
We suspected there was some special treatment of &, but haven't pursued it further at this point. |
| 12:38 |
mmorgan |
Stompro: We're using english_nostop |
| 12:40 |
kmlussier |
Stompro: Are you using the english.stop list? |
| 12:41 |
Stompro |
mmorgan, I'm also usign english_nostop... I was just reading that as "don't include stop words". |
| 12:43 |
kmlussier |
Stompro: nostop means there are no stopwords. We don't use stopwords by default in Evergreen because they're sometimes necessary to perform a search. |
| 12:43 |
kmlussier |
For example, finding the book 'It'. |
| 12:44 |
miker |
to be or not to be |
| 12:44 |
Stompro |
kmlussier, thanks, that make sense. |
| 12:44 |
mmorgan |
right. We're using the default "no stop words" config. |
| 12:44 |
kmlussier |
the the |
| 12:45 |
kmlussier |
Not that any of our libraries own anything from the the |
| 12:53 |
mmorgan |
kmlussier remembers better than I do, actually. We're not using the default since we disabled stemming. |
| 13:02 |
|
jvwoolf joined #evergreen |
| 13:11 |
|
dteston_ joined #evergreen |
| 13:22 |
|
lualaba joined #evergreen |
| 13:23 |
lualaba |
hello. any changes which i made in DB not affected to web staff client ex (remove org_unit, copy_location) but in staff client everything is ok. i restart everything can someone help? |
| 13:24 |
dbs |
lualaba: autogen.sh -u ? |
| 13:24 |
lualaba |
already done |
| 13:24 |
dbs |
ah yeah, staff client shows the changes, hmm |
| 13:24 |
lualaba |
./autogen.sh |
| 13:24 |
lualaba |
-u is needed? |
| 13:25 |
JBoyer |
-u used to be necessary for older databases. it's not necessary anymore, though it doesn't hurt anything. |
| 13:25 |
lualaba |
version 2.10 |
| 13:25 |
* dbs |
has major muscle memory |
| 13:26 |
dbs |
hell, I'm one of the ones who pushed to make -u automatic IIRC |
| 13:26 |
lualaba |
have you any suggestion? |
| 13:26 |
kmlussier |
lualaba: Did you clear the browser cache? |
| 13:27 |
lualaba |
let me check |
| 13:27 |
JBoyer |
I haven't used the web client to check on some of those things, I'm not certain the version I'm on (2.9) would be helpful to check against. :( |
| 13:29 |
lualaba |
oho cache clean help for org_unit, but old shelving locations is still there |
| 13:29 |
JBoyer |
lualaba, is this a production system or testing? I wonder if restarting the memcache server might help, but that would be a bad idea on a production server in the middle of the day. |
| 13:29 |
lualaba |
production |
| 13:30 |
lualaba |
i already restart memchached |
| 13:33 |
JBoyer |
I see. I don't have enough experience troubleshooting the web client to be much help, sorry about that. |
| 13:33 |
lualaba |
now i have problem with shelvin locations (in asset.copy_location delated = 't') but anyway i see on web |
| 13:33 |
lualaba |
Thank you JBoyer |
| 13:39 |
kmlussier |
I don't know. I just went on to webby and deleted a copy location, which should set the asset.copy_location deleted flag to true. It immediately disappeared without a refresh. |
| 13:40 |
kmlussier |
The admin interfaces were still fairly new in 2.10. Maybe it's not the best release to test this interface in. |
| 13:40 |
lualaba |
possible to remove totaly copy location without change flag? |
| 13:41 |
lualaba |
delete from asset.copy_location where id =113, but this just change this flag |
| 13:43 |
lualaba |
affected row = 0 |
| 13:44 |
Dyrcona |
lualaba: drop the protect_copy_location_delete rule on asset.copy_location. |
| 13:48 |
lualaba |
drop trigger protect_copy_location_delete ? |
| 13:48 |
Dyrcona |
It's a rule, not a trigger. |
| 13:51 |
Dyrcona |
You should be aware that rule does a lot more than just set the deleted flag to true. If you can disable it instead of dropping that might be better. |
| 14:03 |
lualaba |
i don't know how disable/enable rule |
| 14:07 |
Dyrcona |
ALTER TABLE x DISABLE RULE y; |
| 14:07 |
Dyrcona |
IIRC. |
| 14:17 |
csharp |
lualaba: if there were ever any circulations for items that were in a copy location, you'd need to change those to another location too |
| 14:17 |
csharp |
s/a copy location/the copy location you're trying to delete/ |
| 14:18 |
lualaba |
thank you very much i will try |
| 14:19 |
csharp |
if you do '\d asset.copy location' in psql, you'll see the tables you need to check under "Referenced by:" |
| 14:20 |
Dyrcona |
The FK constraints are still there. You can't fully delete a copy location that has circulations against it. |
| 14:21 |
Dyrcona |
So, yeah, you'd need to assign them to another copy location. The location with id 1 makes a nice fall back. |
| 14:46 |
Dyrcona |
JBoyer | miker: After doing a little more research and pondering, I have softened my position on C# as it relates to Hatch. |
| 15:16 |
JBoyer |
Dyrcona, miker: I'm about to disappear into a puff of smoke for a few days, but I'll say that we have to use Java on Windows for the foreseeable future anyway, my plan was to eventually remove Java as a client dependency. |
| 15:16 |
JBoyer |
If I really wanted to get my troll on though, I'd start talking about how Visual Basic .Net has made some real strides in the last few releases. ;) |
| 15:17 |
Dyrcona |
JBoyer: I do not want to see Hatch become a Windows-only thing. |
| 15:17 |
jeff |
meanwhile, I'll continue to ponder options that make it work on mobile and/or ChromeOS without Java... :-) |
| 15:17 |
jeff |
also, yes... why "have to use Java on Windows"? |
| 15:17 |
Dyrcona |
jeff: mono works on Android and iOS. |
| 15:18 |
jeff |
or was that "when on Windows, have to use Java"? |
| 15:18 |
jeff |
silly english language. :P |
| 15:21 |
Dyrcona |
At the risk of putting words into his mouth, I think JBoyer means that we have to use Java for now, because that's what berick has given us. :) |
| 15:24 |
Dyrcona |
Neat! I just got a 419 scam written in very good French, claiming to be from a French person, sent from an Indonesian domain and IP address. |
| 15:25 |
|
alynn26_ joined #evergreen |
| 15:25 |
Dyrcona |
And, it scored 80% on the Bayes filter.... Interesting. |
| 15:25 |
Dyrcona |
I figured it would have scored lower, being in French. |
| 15:29 |
berick |
Hatch will in effect be an API. I'll use Java for the reference implementation. |
| 15:30 |
berick |
API, protocol, tomato, tomato |
| 15:56 |
tsbere |
berick: protocol-driven API? ;) |
| 16:07 |
|
lualaba joined #evergreen |
| 16:35 |
|
jvwoolf left #evergreen |
| 16:35 |
|
jvwoolf joined #evergreen |
| 16:48 |
JBoyer |
For jeff and the logs: Dyrcona is right, I meant we have to use Java Hatch on Windows at first because I can't get anything else done in time for 2.12 and possibly 3.0. It's something I want to see done but don't yet have a time line on. (work piles up around me like flakes of snow, but less flurry than blizzard. :( ) |
| 17:01 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 17:07 |
|
mmorgan left #evergreen |
| 18:05 |
csharp |
looks like the Open-ILS/src/sql/Pg/live_t/purge-user-activity.pg test doesn't account for the actor.usr_activity_transient_trg trigger on actor.usr_activity, which deletes a row of the same activity type if transient without checking whether the time is newer or not |
| 18:15 |
berick |
tfw you realized you launched irssi w/o screen |
| 18:15 |
|
berick joined #evergreen |
| 18:16 |
berick |
quickly remedied |
| 18:56 |
csharp |
hmm - so if I want to fix a test, does that warrant a bug report? feels kinda meta |
| 19:13 |
csharp |
b1f4d599 |
| 19:13 |
pinesol_green |
csharp: [evergreen|Bill Erickson] LP#1570909 User activity transient default - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b1f4d59> |
| 20:00 |
|
dteston joined #evergreen |
| 20:39 |
|
abowling left #evergreen |
