| Time |
Nick |
Message |
| 00:47 |
pinesol_green |
Incoming from qatests: Test Failure - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 06:47 |
|
remingtron_ joined #evergreen |
| 06:47 |
|
jboyer_isl joined #evergreen |
| 06:47 |
|
sseng_ joined #evergreen |
| 06:48 |
|
fparks joined #evergreen |
| 06:48 |
|
BigRig_ joined #evergreen |
| 06:48 |
|
tfaile joined #evergreen |
| 06:58 |
phasefx |
that test failure came from live_t/03-overdue_circ.t, expected 13 bills, got 12. probably bad logic in the test surrounding certain dates and not in Evergreen. Will look later, unless someone gets the itch :) |
| 07:07 |
* phasefx |
had to double-take, but today's xkcd: http://xkcd.com/1340/ apropos dates |
| 07:31 |
csharp |
wow - so we'll really be kicking ourselves in 8000 years for not starting to use the five-digit year now, right? ;-) |
| 07:40 |
|
rjackson-isl joined #evergreen |
| 08:00 |
|
collum joined #evergreen |
| 08:09 |
|
akilsdonk joined #evergreen |
| 08:22 |
|
mrpeters joined #evergreen |
| 08:33 |
|
Shae joined #evergreen |
| 08:33 |
|
finnx joined #evergreen |
| 08:58 |
|
ericar joined #evergreen |
| 09:05 |
|
rfrasur joined #evergreen |
| 09:10 |
|
Dyrcona joined #evergreen |
| 09:12 |
|
timf joined #evergreen |
| 09:20 |
csharp |
arg - I'm trying to learn xpath on the fly... my mission is to get the xpath to extract position 6 from the leader in our MARC records (item type) |
| 09:20 |
csharp |
I'm looking for examples in Evergreen, but none are exactly what I'm after :-/ |
| 09:20 |
csharp |
would anyone be willing to share? |
| 09:21 |
|
rfrasur_ joined #evergreen |
| 09:22 |
jboyer-isl |
csharp: I don't know if it's currently used in Evergreen, but it appears there's a substring function in XPath. If you can apply it to the content of the leader, this is the signature: substring(string, offset, length) |
| 09:23 |
jboyer-isl |
so something like substring(<LDR>,7,1) though I'm not up on XPath enough to tell you how to specify the leader in that example. |
| 09:24 |
jboyer-isl |
Or is that where you're getting stuck also? |
| 09:27 |
jboyer-isl |
I also appear to have made an assumption re: the offset param. It's 1-based, not 0. |
| 09:34 |
jeff |
others with more xpath + postgresql chops may correct me or give you better advice, but this at least gives a slight indicator of working: select xpath('substring(//marc:record/marc:leader/text(), 7,1)', marc::xml, ARRAY[ARRAY['marc', 'http://www.loc.gov/MARC21/slim']]) from biblio.record_entry limit 10; |
| 09:34 |
|
timlaptop joined #evergreen |
| 09:34 |
Dyrcona |
csharp: http://www.w3schools.com/xpath/default.asp |
| 09:35 |
|
yboston joined #evergreen |
| 09:36 |
Dyrcona |
jeff: returns empty arrays in my development database. |
| 09:36 |
|
kmlussier joined #evergreen |
| 09:37 |
jeff |
I suspect i'm not accounting for other possible namespaces, or an empty namespace. |
| 09:39 |
eeevil |
csharp: any reason you don't want to just use the value stored in metabib.record_attr? |
| 09:40 |
Dyrcona |
Get empty even if I do it on a record that has the namespace. |
| 09:40 |
Dyrcona |
eeevil++ |
| 09:40 |
jeff |
Dyrcona: interesting. i wonder what i'm doing wrong/different. |
| 09:42 |
jeff |
but yeah, what eeevil said: select id, attrs->'item_type' as item_type from metabib.record_attr limit 10; |
| 09:42 |
Dyrcona |
It works if I remove the substring call. I can get the whole leader no problem. |
| 09:43 |
jeff |
that seems almost more odd. |
| 09:43 |
jeff |
bigshrug |
| 09:44 |
jeff |
meanwhile, running that over a larger set of records on our reporting db turned up invalid xml. :-) |
| 09:45 |
jeff |
not surprisingly, the problem record doesn't seem to be in mrfr. |
| 09:46 |
csharp |
jboyer-isl: Dyrcona: eeevil: - thanks! |
| 09:46 |
csharp |
just as I asked I got pulled into something else |
| 09:46 |
csharp |
multitasking-- |
| 09:46 |
eeevil |
fwiw, it doesn't work because the xpath() function in PG returns a nodeset. substring() in xpath does not return a node or nodeset, but text ... some might call that splitting hairs ("then what does text() return?" ... the answer is "a nodeset of cdata nodes", btw) , but the result is that you can't use a non-set-returning function as the outermost part of an xpath expression in PG's builtin xpath() |
| 09:48 |
jeff |
eeevil: any idea why it works for me? |
| 09:48 |
jeff |
pg 9.2 vs 9.3? |
| 09:48 |
jeff |
dumb luck on my dataset? |
| 09:49 |
|
mrpeters joined #evergreen |
| 09:49 |
eeevil |
you could, conceivably, use it at the end, like: //*[local-name() == "leader"]/substring(.,7,1) ... but I haven't tested it. |
| 09:49 |
eeevil |
jeff: I'm surprised it works, TBH |
| 09:50 |
eeevil |
it's been a pain point for me since we first started thinking about in-db indexing |
| 09:50 |
Dyrcona |
jeff: If I add the name space for xpath-functions, I get the following: DETAIL: xmlXPathCompOpEval: function substring not found |
| 09:54 |
* jeff |
runs a "just how many marc values are invalid xml" query in the reporting db |
| 09:56 |
dbwells |
jeff: Which version of pg are you running? |
| 09:57 |
jeff |
two! two invalid xml documents... ah, ah ah... |
| 09:58 |
jeff |
dbwells: 9.2.6 |
| 09:58 |
eeevil |
interestingly, the 9.2 release notes say: Allow xpath() to return a single-element XML array when supplied a scalar value (Florian Pflug) ... Previously, it returned an empty array. This change will also cause xpath_exists() to return true, not false, for such expressions. |
| 09:58 |
eeevil |
so, that's when it started working |
| 09:59 |
jeff |
eeevil++ |
| 10:00 |
dbwells |
Ah, there you go. Given the other issues we've seen in 9.2, it seemed pretty likely. |
| 10:01 |
eeevil |
however, that doesn't mean that everyone is actually using the builtins, even on 9.2. if one started out on, say, 8.3, one might still be using (for EG purposes) the plperlu implementation, which might be bug-compatible with pre-9.2 xpath() ... I have not checked |
| 10:03 |
* Dyrcona |
would think that the plperl xpath would depend on your Perl module and module versions, since there are more than one xpath module for Perl. |
| 10:03 |
csharp |
eeevil: two reasons I wasn't going for metabib.record_attr: 1) I really didn't understand it - thanks for the tip! and 2) I was asked by our acq testers to add "item format/type" to acq.lineitem_attr_definition, which is looking for XPath |
| 10:07 |
jeff |
eeevil: fwiw, in my example above i was using the builtin xpath function. |
| 10:07 |
eeevil |
csharp: gotcha. both fine reasons. looks like it might be worthwhile to add a new "acq.lineitem_indexed_attr_definition" that can point at a row in config.record_attr_definition |
| 10:16 |
csharp |
oh forgot to increment jeff :-/ |
| 10:16 |
csharp |
jeff++ |
| 10:16 |
csharp |
@karma jeff |
| 10:16 |
pinesol_green |
csharp: Karma for "jeff" has been increased 409 times and decreased 2 times for a total karma of 407. |
| 10:20 |
|
mllewellyn joined #evergreen |
| 10:33 |
|
dluch joined #evergreen |
| 10:46 |
jeff |
<leader></leader>adatafield tag="901" ind1=" " ind2=" "> |
| 10:50 |
eeevil |
jeff: that is, indeed, invalid ;) |
| 10:50 |
jeff |
two records similarly mangled. :-) |
| 10:51 |
|
rfrasur_ joined #evergreen |
| 10:54 |
jeff |
both records appear to date back to 2009 and have no recent (which means since end of december in this case) auditor entries. |
| 10:54 |
jeff |
thus, i am not overly concerned. |
| 10:59 |
|
Wyuli joined #evergreen |
| 10:59 |
|
dac joined #evergreen |
| 11:00 |
csharp |
@eightball is there a statute of limitations for cataloging errors? |
| 11:00 |
pinesol_green |
csharp: Maybe... |
| 11:04 |
|
dave_bn joined #evergreen |
| 11:05 |
dave_bn |
hi everyone! |
| 11:05 |
dave_bn |
I am looking for a quick way to test evergreen |
| 11:06 |
dave_bn |
so I need to install it on a test server. Is there an automated installation script that comes with a documentation? |
| 11:06 |
dave_bn |
or perhaps a virtual image I could deploy |
| 11:06 |
dave_bn |
or more or less easy installation instructions. |
| 11:07 |
dave_bn |
I have spent several hours with the official installation guides and also with sitka and paxed scripts, but have not been able to configure them properly due to lack of documentation. |
| 11:08 |
jeff |
what documentation did you find lacking? |
| 11:08 |
jeff |
i don't know that there is a current VM image for download, but I do know of at least one public "attempt to install everything automatically" script. let me see if i can find the url before someone else. |
| 11:09 |
dave_bn |
jeff: thank you for your help |
| 11:09 |
dave_bn |
I have spend many hours trying to install with official documentation and then with sitka's and paxed's scripts |
| 11:10 |
jeff |
you specifically mentioned that you found a lack of documentation for something. can you give me further details on that? i'd like to see what can be done to resolve the issue. |
| 11:11 |
yboston |
dave_bn: related to jeff's question, can you post the link to the documentation you are using? or some info on it. i ask because often those that have issues with installation are following outdated installation instructions |
| 11:11 |
jeff |
i have used the installer here with reasonable success recently for installing a test/dev system: http://git.evergreen-ils.org/?p=working/random.git;a=shortlog;h=refs/heads/collab/phasefx/wheezy_installer |
| 11:12 |
paxed |
i doubt my script actually works anymore |
| 11:12 |
jeff |
a more direct link, if you're not using git to clone the random repo: http://git.evergreen-ils.org/?p=working/random.git;a=tree;f=installer/wheezy;h=92dd7d1d5d0f5acfe0b49793eb5e65af78d68455;hb=refs/heads/collab/phasefx/wheezy_installer |
| 11:19 |
dave_bn |
paxed: it did do quite well, stopped at database creation, I think. |
| 11:19 |
dave_bn |
jeff: thank you for the link |
| 11:19 |
csharp |
dave_bn: if you're running an ubuntu 12.04 server, you could try out our debs: http://archive.georgialibraries.org/ |
| 11:19 |
dave_bn |
csharp: yes, I am running a fresh 12.04 |
| 11:20 |
csharp |
dave_bn: they follow the default install instructions as closely as possible, so you'd be on stock |
| 11:20 |
csharp |
however - hmm, I'm seeing that it's on 2.4.6... |
| 11:20 |
csharp |
you probably want 2.5.current |
| 11:21 |
jeff |
dave_bn: you're welcome for the link. can you go into any more detail on your comment "have not been able to configure them properly due to lack of documentation"? |
| 11:22 |
dave_bn |
jeff: I first used official documentation from http://evergreen-ils.org/documentation/install/README_2_4.html |
| 11:22 |
dave_bn |
and then http://evergreen-ils.org/documentation/install/README_2_5.html |
| 11:22 |
dave_bn |
as well as OpenSRF http://evergreen-ils.org/documentation/install/OpenSRF/README_2_2_2.html |
| 11:23 |
jeff |
so far so good. |
| 11:23 |
dave_bn |
but found some discrepencie from the actual 12.04 system and it took quite some hours to guess |
| 11:23 |
jeff |
where did you find discrepancies? |
| 11:23 |
csharp |
dave_bn: if you end up being interested in the APT repo approach, our repo is being updated to 2.5.X right now |
| 11:23 |
dave_bn |
and in the end I probably messed up the jabber configuration |
| 11:24 |
csharp |
and... it's back: http://archive.georgialibraries.org/ |
| 11:24 |
dave_bn |
csharp: thank you. I think it will be the fastest route to use your repo. I will try it now and also take a look at the automated script |
| 11:25 |
csharp |
dave_bn: happy to help |
| 11:25 |
csharp |
I can vouch for it working |
| 11:25 |
dave_bn |
jeff: regarding the lack of documentaiton, I meant the lack of documenatiaon for the sitka and paxed automated scripts |
| 11:25 |
csharp |
I really only install from source now when testing community stuff |
| 11:26 |
jeff |
dave_bn: ah. thanks! that clears that part up. now, how about the discrepancies you found? |
| 11:29 |
Dyrcona |
dave_bn: My suggestion for installation is ignore any documentation on the web, and use the README files inside the tarballs. |
| 11:30 |
Dyrcona |
dave_bn: My next suggestion for installation for installation is don't use the tarballs and install from a git clone. |
| 11:32 |
dave_bn |
jeff: the discrepancies I encountered would probably be insignificant to experienced users. e.g. I did not even know I needed to have installed build essential :) |
| 11:32 |
dave_bn |
Dyrcona: thanks for the advice. is there a guide for the git clone installation? |
| 11:32 |
Dyrcona |
dave_bn: git clone the master repo, read the README. |
| 11:32 |
dave_bn |
jeff: other than that, there were small things like extra slash in with echo "export PATH=\$PATH:/openils/bin" >> /home/opensrf/.bashrc |
| 11:33 |
jeff |
dave_bn: you found you needed to install build-essential, not just make followed by the things installed by Makefile.install? |
| 11:33 |
dave_bn |
dyrcona: thank you |
| 11:33 |
bshum |
Sounds like desktop vs server discrepancies to me. |
| 11:33 |
Dyrcona |
bshum: VM installs don't usually include half the stuff you need. |
| 11:33 |
bshum |
Oh that. |
| 11:34 |
bshum |
*too |
| 11:34 |
dave_bn |
jeff: and when I installed opensrf, I needed to put all the password in configuration, which took some time, only to find out that the configuration files would be overwritten by evergreen installation guide later :) and I would need to put the passwords again |
| 11:35 |
jeff |
dave_bn: slightly annoying, but it is important to test opensrf before continuing on to evergreen installation. :-) |
| 11:35 |
dave_bn |
dyrcona: I see. but the APT method is reliable, yes? |
| 11:35 |
Dyrcona |
I always make sure that the following are installed before proceeding: acpid, vim, openssh-server, screen, wget, git-core, build-essential, zip, nsis, man-db, autoconf, automake, libtool, aptitude, tmux |
| 11:35 |
Dyrcona |
dave_bn: Dunno. I only ever install from git checkouts. |
| 11:36 |
dave_bn |
jeff: I may be mistaken about build essential package requirement. when something went wrong, this was the package I blamed. |
| 11:36 |
dave_bn |
jeff: In documentation I also had trouble guessing whether I should have run certain commands as root or normal user or opensrf user |
| 11:37 |
jeff |
dave_bn: any specific commands where the docs seemed ambiguous? i know there's been some effort to clarify, but perhaps you encountered some new places? |
| 11:37 |
Dyrcona |
dave_bn: The README usually tells you when to run something as root. Been a while since I actually read it, but last time I did it was accurate. |
| 11:39 |
dave_bn |
jeff: I also spent quite some time figuring out why the cpan packages would not install and turned out I only needed to change the repository, as the server of pacages did not work :) |
| 11:39 |
jeff |
oh, meaning that the CPAN mirror auto-selected by the cpan setup was bad? |
| 11:40 |
jeff |
that's... annoying. a bit outside of Evergreen's control, but annoying nonetheless. :-) |
| 11:40 |
dave_bn |
jeff: yes :) |
| 11:40 |
dave_bn |
jeff: yes, definitely outside of EG control :) only took me a while to find out |
| 11:40 |
Dyrcona |
dave_bn: Do you have much experience with GNU/Linux already? |
| 11:41 |
dave_bn |
jeff: I would have found helpful if opensrf installation guide mentioned that it was no need to put all the passwords in the configuration file, if I were to install evergreen afterwards, as it would overwrite my configuration files |
| 11:41 |
dave_bn |
jeff: but I'm sure more experienced users would have guessed that |
| 11:42 |
jeff |
dave_bn: was the point of confusion that you didn't realize that you needed to enter the passwords twice, so you skipped that step after installing evergreen? |
| 11:42 |
dave_bn |
Dyrcona: no :) this is the reason I have not been able to follow the instructions properly obviously |
| 11:43 |
dave_bn |
jeff: no, I just had to look throught the xml configuration file twice, and I guess the first time it was not really required, as following the EG guide overwrites the opensrf configuration anyway |
| 11:46 |
Dyrcona |
dave_bn: If you want to test OpenSRF before installing Evergreen, it is necessary to put the passwords in the file. |
| 11:47 |
Dyrcona |
I always test OpenSRF by doing the math add 2,2 call before installing Evergreen when I set up a new VM. |
| 11:47 |
|
mtate joined #evergreen |
| 11:51 |
|
kayals_ joined #evergreen |
| 11:58 |
dave_bn |
Dyrcona: I see. It also helped me understand the process, though was a bit time consuming at first. I guess I will be able to to it faster the next time. |
| 12:05 |
dave_bn |
csharp: I followed the instructions at http://archive.georgialibraries.org/ and added the repo to the list, but pt-get install evergreen-ils brings 'Unable to locate package evergreen-ils' error |
| 12:09 |
bshum |
Assuming you ran apt-get update before the install, then that sounds like a repo weirdness. |
| 12:09 |
csharp |
dave_bn: did you do 'sudo apt-get update'? |
| 12:09 |
dave_bn |
csharp: yes, and also added the key |
| 12:10 |
dave_bn |
csharp: i'm on fresh Precise 12.04 x32 |
| 12:10 |
|
kayals joined #evergreen |
| 12:10 |
csharp |
dave_bn: would you mind doing 'sudo apt-get install pastebinit' then 'pastebinit /etc/apt/sources.list' and share the link? |
| 12:10 |
dave_bn |
csharp: does it matter if I add the repo at the end or start of the sources.list? |
| 12:10 |
csharp |
dave_bn: it shouldn't matter |
| 12:11 |
dave_bn |
sharp: thanks for helping. I did not know pastebin worked from shell too. its http://paste.ubuntu.com/7068418/ |
| 12:13 |
csharp |
dave_bn: what about 'apt-cache search evergreen-ils'? does that return anything? |
| 12:14 |
csharp |
yeah pastebinit is nice (but also dangerous when pasting config files that might have passwords) ;-) |
| 12:14 |
dave_bn |
csharp: no, nothing comes up |
| 12:15 |
dave_bn |
csharp: should I donwload the dpkg file directly? |
| 12:15 |
csharp |
dave_bn: no - don't do that yet |
| 12:17 |
csharp |
dave_bn: just to humor me, could you do 'sudo apt-get update' again? |
| 12:19 |
dave_bn |
csharp: I did. update went good, but the error remains about evergreen-ils package not being found |
| 12:19 |
csharp |
dave_bn: ok - try 'sudo apt-get clean && sudo apt-get update && sudo apt-get install evergreen-ils' |
| 12:20 |
dave_bn |
csharp: no luck. Unable to locate package evergreen-ils error again |
| 12:21 |
csharp |
dave_bn: it's working for me on a fresh 12.04 server installation |
| 12:21 |
csharp |
dave_bn: are there any errors when you do 'sudo apt-get update'? |
| 12:22 |
jeff |
(like key errors?) |
| 12:22 |
csharp |
right |
| 12:23 |
dave_bn |
csharp: no, all goes ok. it says Hit http://archive.georgialibraries.org precise Release.gpg then Hit http://archive.georgialibraries.org precise Release then Hit http://archive.georgialibraries.org precise/main i386 Packages and moves on to the next repos |
| 12:23 |
dave_bn |
csharp: if it is working on your end, perhaps I should try on a new server just to make sure. |
| 12:23 |
csharp |
dave_bn: ah - so this is an i386 server? |
| 12:24 |
dave_bn |
csharp: yes, virtual x32 |
| 12:24 |
csharp |
dave_bn: okay - that's probably it :-( |
| 12:24 |
csharp |
do you have the means to set up a 64-bit server? |
| 12:24 |
dave_bn |
csharp: should I try on x64 server? :) |
| 12:24 |
bshum |
http://www.youtube.com/watch?v=M6KOEMJKdEI |
| 12:24 |
csharp |
yeah, definitely |
| 12:24 |
dave_bn |
csharp: yes |
| 12:25 |
bshum |
That's my reaction to 32-bit :( |
| 12:25 |
csharp |
bshum++ |
| 12:30 |
eeevil |
grabbing 0871 |
| 12:32 |
|
jihpringle joined #evergreen |
| 12:33 |
bshum |
eeevil: re cleanup script for the deleted metarecords, my paste from that night worked pretty quickly to clean up our dead entries, but please let me know if you have better suggestions for inclusion with the branch. |
| 12:33 |
bshum |
http://pastie.org/8897557 (so that we don't have to scroll so far back) |
| 12:34 |
eeevil |
heh, thanks. I was about to ctl-f for pastie ;) |
| 12:34 |
csharp |
@weather 02142 |
| 12:34 |
pinesol_green |
csharp: The current temperature in MIT (Green Bld Roof), Cambridge, Massachusetts is 39.4°F (12:30 PM EDT on March 10, 2014). Conditions: Overcast. Humidity: 59%. Dew Point: 26.6°F. Windchill: 37.4°F. Pressure: 29.80 in 1009 hPa (Falling). |
| 12:35 |
pinesol_green |
Showing latest 5 of 17 commits to Evergreen... |
| 12:35 |
pinesol_green |
[evergreen|Bill Erickson] LP#1284864 MR display avoid fetching master - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3bd56ef> |
| 12:35 |
bshum |
eeevil: I'm sure there's more optimal ways of doing what I did, but I just wanted to get a more solid thing in place before we start committing fixes. |
| 12:35 |
pinesol_green |
[evergreen|Bill Erickson] LP#1284864 MR opac-visible constituent records - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=097d661> |
| 12:35 |
pinesol_green |
[evergreen|Bill Erickson] LP#1284864 repair MR hold possibilty depth calculation - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c675041> |
| 12:35 |
pinesol_green |
[evergreen|Bill Erickson] LP#1284864 correct TPAC metarecord list pagination - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=03c8b33> |
| 12:35 |
pinesol_green |
[evergreen|Mike Rylander] Stamping upgrade script for TPAC metarecord / formats repairs and usability additions - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d470c2d> |
| 12:35 |
eeevil |
bshum: the second query looks like just the thing we'll need. want me to toss that on the end of the upgrade script in my branch? |
| 12:36 |
bshum |
eeevil: That'd be perfect I think. |
| 12:36 |
kmlussier |
csharp: I checked the 10-day forecast, and next Wednesday isn't looking as great as I had been hoping. |
| 12:36 |
csharp |
kmlussier: it'll be awesome no matter what! |
| 12:36 |
kmlussier |
@eightball Did you steer me wrong when you said we would have spring-like weather for the conference? |
| 12:36 |
pinesol_green |
kmlussier: The answer is certainly yes. |
| 12:36 |
kmlussier |
Curse you eightball! |
| 12:36 |
bshum |
kmlussier: As long as Tuesday is fine, since that's when I will have to drive up. With equipment and stuff in tow. |
| 12:37 |
* csharp |
plans to arrive Tuesday early afternoon, drop his bags somewhere and take a train to fair Hahvahd |
| 12:37 |
eeevil |
bshum: a minor modification (to allow for existing config) and I'll push the change soon |
| 12:37 |
kmlussier |
bshum: Well, there aren't any snowflake pictures in the Tuesday forecast. |
| 12:37 |
bshum |
eeevil++ berick++ # thanks! |
| 12:42 |
ktomita |
ldwhalen: I am taking a look at the two bugs you emailed the list about. Is there anything else I should lookout for other than what was outlined in launchpad? |
| 12:42 |
dave_bn |
csharp: this time it worked all the way to the end, and only failed after issuing 'Starting OpenSRF C' and 'Starting web server apache2 ', which says [fail] - The Apache error log may have more information. and hangs |
| 12:44 |
csharp |
dave_bn: this is upon rebooting after installation? |
| 12:44 |
csharp |
or during installation? |
| 12:45 |
eeevil |
bshum: working/collab/miker/metarecord-deleted-constituents updated |
| 12:45 |
dave_bn |
csharp: this is during installation. should I reboot manually now? |
| 12:45 |
csharp |
dave_bn: you shouldn't have to |
| 12:45 |
csharp |
dave_bn: give me a minute to try something on my test server |
| 12:45 |
ldwhalen |
ktomita: I think the hardest part to test is the warning message appearing when it should not appear. From my testing it works as it should, but there may be cases where the variables I am using are persisting when they should not, so the message might appear elsewhere. I do not expect this to happen, but that is a concern. |
| 12:47 |
ktomita |
ldwhalen: I will look out for that. |
| 12:47 |
ldwhalen |
ktomita: thank you for looking at the branches |
| 12:47 |
bshum |
eeevil: Cool, I'll check that in, once I figure out what that internal flag is all about. I didn't remember that one before. |
| 12:48 |
eeevil |
bshum: it keeps the record source map entry around so that you can search for deleted records with the #deleted modifier |
| 12:50 |
bshum |
eeevil: That's what I figured, I just couldn't find it in my upgraded DB. But it looks like it might just be a quirk of our system |
| 12:51 |
eeevil |
it could be. that's why I'm using an unadorned from-clause entry and coalesce |
| 12:51 |
bshum |
Oh, internal_flag, not global_flag |
| 12:51 |
bshum |
Duh :( |
| 12:51 |
eeevil |
in case you don't have it, it defaults to "you don't want that, it seems" ;) |
| 12:51 |
eeevil |
and, yes, it's internal only ... not something you want to switch on and off |
| 12:51 |
bshum |
Yeah |
| 12:52 |
bshum |
Looks good to me. |
| 12:52 |
|
akilsdonk_ joined #evergreen |
| 12:52 |
bshum |
Calling 0872 |
| 12:53 |
|
dave_bn joined #evergreen |
| 12:53 |
csharp |
dave_bn: almost done with my test... |
| 12:53 |
dave_bn |
csharp: thanks for your help |
| 12:54 |
csharp |
dave_bn: no problem - I want our APT repo to be a viable option for installing stock test servers ;-) |
| 12:55 |
eeevil |
How would you like comments that apply to the OmniTI suggestions you put up on the wiki? (re: http://wiki.evergreen-ils.org/doku.php?id=dev:testing:performance_report ) ... would you like them in-line on that page? |
| 12:57 |
pinesol_green |
[evergreen|Mike Rylander] LP#1284864: Don't leak deleted constituent records - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5f03068> |
| 12:57 |
|
akilsdonk joined #evergreen |
| 12:57 |
pinesol_green |
[evergreen|Mike Rylander] LP#1284864: Forcibly update deleted MR masters - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=f4d5813> |
| 12:57 |
pinesol_green |
[evergreen|Ben Shum] LP1284864 - stamping upgrade for deleted metarecord function fixing - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b9f8c5b> |
| 12:57 |
pastebot |
"csharp" at 64.57.241.14 pasted "for dave_bn: the last lines from my installation on a fresh Ubuntu 12.04 VM" (99 lines) at http://paste.evergreen-ils.org/31 |
| 12:58 |
csharp |
dave_bn: so yours errored out at the end of starting OpenSRF services? |
| 13:03 |
dave_bn |
csharp: mined errored out much earlier, the first time it tried to restart apache2 |
| 13:04 |
dave_bn |
csharp: I can try again on a fresh 12.04 x64 |
| 13:05 |
dave_bn |
csharp: my fresh 12.04 comes with root activated, with no other user in place. and i issue the apt-get install from root directly |
| 13:05 |
csharp |
dave_bn: are you able to pastebin the output from when yours errored out? (with as many details around the error as possible) |
| 13:05 |
csharp |
dave_bn: that shouldn't be an issue since you have to install as sudo/root anyway |
| 13:07 |
bshum |
eeevil: Now that I pushed your changes, was there anything else in that bug 1284864 before I mark it as "fix committed"? |
| 13:07 |
pinesol_green |
Launchpad bug 1284864 in Evergreen "TPAC metarecord / formats repairs and usability additions" (affected: 1, heat: 6) [Undecided,Confirmed] https://launchpad.net/bugs/1284864 |
| 13:08 |
bshum |
Do we plan to tack on the whole reingest stuff or I guess maybe that's to be a new bug while we get the final plan for that done? |
| 13:08 |
|
hbrennan joined #evergreen |
| 13:08 |
|
mcooper joined #evergreen |
| 13:09 |
dave_bn |
chsarp: I have just restarted the process on a fresh server and will pastebin the results |
| 13:10 |
csharp |
dave_bn: okay - thanks |
| 13:10 |
eeevil |
bshum: 1) nothing else on that bug and 2) I think a reingest (at least for attrs) should be separately planned |
| 13:12 |
bshum |
eeevil: Sounds good to me. |
| 13:14 |
bshum |
phasefx: Do you have any final ideas about bug 1249480 ? I'd like to commit it, but can't think of any way of explaining about the localstore.rdf file other than maybe as a release note or something? |
| 13:15 |
pinesol_green |
Launchpad bug 1249480 in Evergreen "staff client localstore.rdf can grow too large" (affected: 2, heat: 10) [Medium,Triaged] https://launchpad.net/bugs/1249480 - Assigned to Ben Shum (bshum) |
| 13:17 |
|
ericar_ joined #evergreen |
| 13:20 |
|
ericar__ joined #evergreen |
| 13:21 |
jeff |
phasefx: related followup question: in a current client, is any information that a user would care about lost by removing localstore.rdf? |
| 13:21 |
dave_bn |
csharp: now it managed to start apache, but stopped a bit later. http://pastebin.com/BCjnMP6p |
| 13:22 |
|
ericar__ joined #evergreen |
| 13:41 |
|
serflog joined #evergreen |
| 13:41 |
|
Topic for #evergreen is now Register to vote in the upcoming EOB election: https://www.surveymonkey.com/s/WM5ZVWG | Welcome to the #evergreen library system channel! | We are publicly logged. | Large pastes at http://paste.evergreen-ils.org |
| 13:42 |
ktomita |
ldwhalen: I don't think the change is necessary, I was just curious on the difference. From your explanation it seems they are two different message because of two different actions that can occur on the same screen. |
| 13:43 |
ldwhalen |
ktomita: I think in this case the TCN search was given a different message because it would prompt for a lose of data from a dialog as opposed to the UI where the change was made. |
| 13:43 |
bshum |
Yeah, it's definitely broken. z3950 searches with the local catalog just spin on searching till it times out with a network error. |
| 13:44 |
bshum |
Sigh. And this isn't the AUTOLOAD thing that jeff saw cause we don't have that patch yet. |
| 13:44 |
ldwhalen |
bshum: is is it broken when you do not use a pubdate in the search? |
| 13:44 |
bshum |
ldwhalen: I was just doing a title search and it dies horribly. I could try other types of searches too. |
| 13:45 |
Dyrcona |
ldwhalen: I verified it doing only a title. |
| 13:45 |
ldwhalen |
bshum: I have a fix for a similar bug when using a pubdate. It has been tested. I will push it to a working branch. There is an issue when the source does not return results. |
| 13:46 |
bshum |
Given that the source is the native catalog (aka our own system) that's not very comforting either. |
| 13:46 |
bshum |
But I'll be curious to see what you saw with pubdate |
| 13:48 |
bshum |
And sorry phasefx, just got distracted by this z3950 bug. But I plan to take a closer look at that bug with localstore.rdf. Seems like something we should try fixing as a performance type of issue. |
| 13:49 |
jeff |
bah. there is no url param to change the number of items displayed in a list in myopac/lists -- only a user pref or the default of 10. |
| 13:51 |
ldwhalen |
bshum: http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/ldw/LP1271559_z3950_timeout |
| 13:52 |
csharp |
dst-- |
| 13:57 |
bshum |
Well it works in our production. So I guess it's something that's changed in the past 3 months. :\ |
| 14:01 |
Dyrcona |
ldwhalen: Just the one commit on that branch, right? |
| 14:01 |
ldwhalen |
Dyrcona: that is correct |
| 14:02 |
bshum |
Oh |
| 14:02 |
bshum |
There's a bug already |
| 14:02 |
bshum |
https://bugs.launchpad.net/evergreen/+bug/1271559 |
| 14:02 |
pinesol_green |
Launchpad bug 1271559 in Evergreen "Z39.50 Import errors in staff client" (affected: 2, heat: 12) [Medium,In progress] - Assigned to Liam Whalen (whalen-ld) |
| 14:03 |
Dyrcona |
ldwhalen: OK. I'm giving it a whirl, waiting a restart-all, 'cause just restarting search didn't do it. |
| 14:03 |
kmlussier |
@karma dst |
| 14:03 |
pinesol_green |
kmlussier: Karma for "dst" has been increased 0 times and decreased 4 times for a total karma of -4. |
| 14:03 |
kmlussier |
dst-- |
| 14:03 |
|
finnx joined #evergreen |
| 14:04 |
ldwhalen |
Dyrcona: I am not sure if this bug is relaated to the new issues. It was reported for 2.4 |
| 14:04 |
bshum |
I'm going to really miss osrf_control |
| 14:04 |
bshum |
Err, osrf_ctl.sh rather |
| 14:04 |
bshum |
:D |
| 14:05 |
|
dMiller joined #evergreen |
| 14:05 |
bshum |
Or maybe I'm used to it already! |
| 14:05 |
Dyrcona |
ldwhalen: I don't think it is either. Your patch makes no difference for me. |
| 14:05 |
bshum |
I think it'll help with non remote targets |
| 14:05 |
bshum |
But this is a local search problem |
| 14:05 |
Dyrcona |
osrf_ctl.sh? What's that, again? :p |
| 14:06 |
Dyrcona |
@karma DST |
| 14:06 |
pinesol_green |
Dyrcona: Karma for "DST" has been increased 0 times and decreased 5 times for a total karma of -5. |
| 14:06 |
Dyrcona |
Guess it isn't case sensitive. |
| 14:06 |
bshum |
It isn't. |
| 14:07 |
bshum |
Or at least, it hasn't been. |
| 14:07 |
bshum |
We switched to a new karma module couple years ago that doesn't pay as close attention to casing, etc. |
| 14:07 |
Dyrcona |
bshum: Do you want to bug this z39.50 thing? |
| 14:07 |
bshum |
Dyrcona: Yeah I suppose I should. It must be something newly introduced during the beta period. |
| 14:08 |
Dyrcona |
I know I've done some z39.50 searches more recently than the last three months and they worked. |
| 14:08 |
Dyrcona |
I just wish I could remember dates. |
| 14:12 |
eeevil |
kmlussier: I don't want the conversation to get lost, so the mailing list would make sense, but I also don't want multiple sources of authoritative info ... on the other hand, the specifics are of a nature that benefits from as much direct context as possible, which makes me want to put it on the wiki where all the info is in one place at a glance... so, I'm torn ;) |
| 14:12 |
phasefx2_ |
bshum: re bug, didnt i fix it? |
| 14:13 |
bshum |
phasefx2_: You have a patch, it hasn't been committed yet. And your notes on the bug mention that we need to explain how to deal with the localstore.rdf files that people may have bloated on their local machines. |
| 14:13 |
phasefx2_ |
ah |
| 14:14 |
bshum |
I can commit it, but I figure we should have some sort of PSA about dealing with those files. Though I can't think of any smart ways of handling it yet. |
| 14:14 |
bshum |
If we tell people to delete it, they'll have to reset their columns, etc. right? |
| 14:14 |
bshum |
That may be more hasslesome than I'd care to admit to |
| 14:14 |
kmlussier |
eeevil: Heh. Well, let's just say, in either case, it's worthwhile to put something on the list so that it gets more attention. That was my original intent, but I also intended to post it back in January. |
| 14:15 |
kmlussier |
And then maybe we can link to the discussion thread in the wiki comments? |
| 14:15 |
eeevil |
sure |
| 14:15 |
* kmlussier |
will send something to the list today, tomorrow morning at the latest. |
| 14:15 |
eeevil |
kmlussier++ |
| 14:15 |
bshum |
Dyrcona: Going backwards, Z3950.pm seems to point at some call to open-ils.search.biblio.zstyle.staff which in turn leads me to Biblio.pm and trying to decipher what's in that z_style thing |
| 14:15 |
bshum |
For native-evergreen-catalog that is |
| 14:16 |
bshum |
(may be juggling too much) |
| 14:17 |
Dyrcona |
bshum: you may be, but zstyle is definitely where things go. |
| 14:19 |
ktomita |
ldwhalen: I am seeing an issue when I am getting the prompt when pressing Shift-F3 on a new tab. |
| 14:19 |
ktomita |
ldwhalen: I will outline my steps in launchpad. |
| 14:19 |
ldwhalen |
ktomita: thank you for finding that |
| 14:20 |
ldwhalen |
I will take a look once you have posted your findings |
| 14:21 |
|
kayals joined #evergreen |
| 14:24 |
|
dave_bn joined #evergreen |
| 14:29 |
ktomita |
ldwhalen: I have posted two of them to launchpad ticket, https://bugs.launchpad.net/evergreen/+bug/1282286 |
| 14:29 |
pinesol_green |
Launchpad bug 1282286 in Evergreen "Pressing Shift-F3 in the MARC editor results in invalid lose data warning" (affected: 1, heat: 8) [Undecided,New] - Assigned to Liam Whalen (whalen-ld) |
| 14:30 |
ldwhalen |
ktomita: thank you. I am looking at them now |
| 14:31 |
ktomita |
ldwhalen: let me know if they don't make sense |
| 14:32 |
dave_bn |
csharp, Dyrcona, dbs: I have increased my machines ram to 2GB and apt-get method worked fine :) Thanks for your help! |
| 14:33 |
csharp |
dave_bn: excellent! |
| 14:33 |
dave_bn |
csharp: this explains why I have not been able to install with the regulat method :) |
| 14:33 |
Dyrcona |
dave_bn: Great! Glad to hear it worked. |
| 14:34 |
dave_bn |
Dyrcona: later I will try the git method too |
| 14:34 |
csharp |
dave_bn: a word of caution... now that you have Evergreen installed, I would recommend commenting out the apt/sources.list entry |
| 14:35 |
csharp |
otherwise you may risk inadvertently "upgrading" when our folks put in newer versions of Evergreen/OpenSRF |
| 14:35 |
dave_bn |
csharp: thank you, will do that. |
| 14:35 |
csharp |
we haven't tested upgrading via APT yet |
| 14:36 |
dave_bn |
csharp: i see :) which staff client would be compatible with the current version I have installed? |
| 14:36 |
jeff |
or risk getting owned when they turn evil ;-) |
| 14:36 |
csharp |
jeff: exactly! |
| 14:36 |
csharp |
muhahaha |
| 14:37 |
csharp |
dave_bn: http://evergreen-ils.org/downloads/evergreen-setup-2.5.2.exe |
| 14:38 |
* csharp |
notes that that link needs to be added to the "past staff clients" page |
| 14:38 |
bshum |
Dyrcona: a very boring and generic https://bugs.launchpad.net/evergreen/+bug/1290496 filed while we dig deeper. |
| 14:38 |
pinesol_green |
Launchpad bug 1290496 in Evergreen "Import z3950 - using local catalog search option results in network error" (affected: 1, heat: 6) [Undecided,New] |
| 14:38 |
dave_bn |
csharp: thaks :) I wonder how many hours I have wasted during the past years trying to install eg on machines with 512 and 1GB ram |
| 14:38 |
csharp |
dave_bn: understandable ;-) I did that when I first started too |
| 14:39 |
* csharp |
sits in his rocking chair and shoos kids from his lawn |
| 14:40 |
dave_bn |
csharp: another question i always had is about integrating electornic database and external sources, like archive.org or JStor, ebsco, isi web of knowledge etc. |
| 14:40 |
csharp |
way back in ought-seven before the autotools days |
| 14:40 |
dave_bn |
csharp: is it possible to integrate those sources without importing the marc records from them? |
| 14:41 |
csharp |
dave_bn: I'll defer to others on that question |
| 14:41 |
csharp |
dave_bn: but you can include URLs to external resources in the MARC record - ($856) - those work |
| 14:42 |
Dyrcona |
dave_bn: You have to import MARC records. |
| 14:42 |
ldwhalen |
ktomita: I can reproduce both issues you have found. I will fix them, and update the commit by tongiht. |
| 14:42 |
ldwhalen |
ktomita++ |
| 14:43 |
dave_bn |
Dyrcona: if I import records, is there a chance to keep them updated? what if a book is removed from the external collection |
| 14:44 |
ktomita |
ldwhalen: I will try to take a look at the update today or tomorrow. |
| 14:44 |
dave_bn |
Dyrcona: and how would I know which marc records have changed or are new, e.g. from archive.org |
| 14:44 |
Dyrcona |
dave_bn: That depends on the vendor. Generally, I get update files that tell me which bibs to delete and will have any new bibs to add. |
| 14:44 |
Dyrcona |
dave_bn: Another useful thing is to create a MARC source for each vendor. |
| 14:45 |
Dyrcona |
dave_bn: Also note, I don't use the normal Evergreen tools for keeping the records up to date. I generally write my own. |
| 14:46 |
dave_bn |
Dyrcona: I see. Would it not be more convenient if external sources were integrated same way as google books? |
| 14:46 |
bshum |
Absolutely more convenient. |
| 14:46 |
bshum |
:) |
| 14:47 |
Dyrcona |
dave_bn: And, just how your catalog know what records you have from the external source? |
| 14:48 |
bshum |
But even Google books needs local records to match onto / link with |
| 14:49 |
bshum |
It's a nice thing to dream / work towards though. I feel that way about getting Overdrive's API working with Evergreen's catalog. |
| 14:49 |
* bshum |
stares intently in jeff's direction |
| 14:49 |
* bshum |
stares some more |
| 14:49 |
jeff |
hi. |
| 14:49 |
bshum |
:D |
| 14:50 |
dave_bn |
Dyrcona: I see. I thought if library subscribed to certain content, e.g. Emerald, the admin could activate the subscription in settings and search results would include publications from Emerald as well |
| 14:51 |
Dyrcona |
dave_bn: Much harder than it sounds since there is no standard API for searching that every service uses. |
| 14:51 |
dave_bn |
Dyrcona, bshum: i thought all the commercial databases supported harvesting |
| 14:52 |
Dyrcona |
dave_bn: And, just where do you get your unique ids, etc. to harvest? |
| 14:54 |
dave_bn |
Dyrcona: as databases would be all electronic, my guess was to simply augment the search results with results from external sources, directling going to the external resource when clicked on |
| 14:54 |
dave_bn |
Dyrcona: but I may be confusing an ILS with metasearch I guess |
| 14:54 |
Dyrcona |
dave_bn: You do realize that you are dealing with external actors who may have a financial disincentive to allow that kind of access. |
| 14:56 |
dave_bn |
Dyrcona: I see. So if a library subscribed to a certain electronic publication, would they want to restrict acess |
| 14:57 |
dave_bn |
Dyrcona: archive.org and other open sources would be less risky in that case |
| 14:57 |
Dyrcona |
dave_bn: Of course. It can't be available to just anyone who can browse the catalog that may be available from anywhere on the Internet. |
| 14:58 |
Dyrcona |
The usual approach is to download records from the vendor. |
| 14:58 |
Dyrcona |
"Open Access" unfortunately doesn't really mean "open." |
| 14:59 |
dave_bn |
Dyrcona: I see :) Do libraries with Evergreen usually import records from electronic subscription databases? |
| 14:59 |
csharp |
dave_bn: in our case we have a state-wide provider of online reference databases (GALILEO) and we have seamless authentication to their site |
| 14:59 |
csharp |
but it's not integrated into the Evergreen OPAC |
| 14:59 |
dave_bn |
csharp: are patrons able to search from evergreen interface? |
| 14:59 |
csharp |
dave_bn: nope |
| 14:59 |
Dyrcona |
dave_bn: We import such records from Overdrive and Safari Online. |
| 15:00 |
csharp |
dave_bn: but we're in the process of adding PINES holdings to their discovery interfaces, so that would allow integrated searching |
| 15:00 |
jeffdavis |
Sitka imports records like that too. |
| 15:00 |
dave_bn |
csharp, Dyrcona: I see. So you would use something like vufind |
| 15:01 |
Dyrcona |
dave_bn: No, I import the records into Evergreen and they are searchable there. |
| 15:01 |
dave_bn |
Dyrcona: ok. are you also updating the 858 fields during importing |
| 15:02 |
Dyrcona |
dave_bn: Yes, my Safari code modifies the 856 fields on import. I don't know what is done with Overdrive since someone else actually imports those by hand. |
| 15:03 |
dave_bn |
Dyrcona, csharp: could you recommend which scripts to use for importing external database marc records and keeping them up-to-date |
| 15:04 |
Dyrcona |
dave_bn: AFAIK, this is the only one publicly available and it is for Safari: http://git.mvlcstaff.org/?p=jason/safariload.git;a=summary |
| 15:05 |
csharp |
dave_bn: yeah - we're not doing that at the moment :-/ |
| 15:07 |
dave_bn |
Dyrcona, csharp: thank you, I will look. probably it can be modofied to work with other sources |
| 15:07 |
csharp |
dave_bn: happy to help! |
| 15:09 |
csharp |
huh - I just found a broken symlink - /openils/var/web/opac/common/js/date.js -> ../../Open-ILS/xul/staff_client/chrome//content//util/date.js |
| 15:09 |
csharp |
guess it can't matter that much ;-) |
| 15:10 |
csharp |
that's in master as of today |
| 15:10 |
csharp |
(and in 2.5.1, which is where I noticed it) |
| 15:33 |
Dyrcona |
csharp: Looks the same here, your broken symlink. |
| 15:43 |
|
dMiller__ joined #evergreen |
| 15:49 |
|
kazakovk joined #evergreen |
| 16:06 |
|
berick joined #evergreen |
| 16:14 |
|
kbutler joined #evergreen |
| 16:23 |
|
berick_ joined #evergreen |
| 16:36 |
|
zerick joined #evergreen |
| 16:38 |
|
mrpeters joined #evergreen |
| 16:48 |
|
dluch joined #evergreen |
| 17:07 |
jeff |
and the rfid pad paid someone's bills. oops. :-) |
| 17:29 |
|
mrpeters left #evergreen |
| 17:34 |
|
mrpeters joined #evergreen |
| 18:00 |
|
dMiller joined #evergreen |
| 20:38 |
bshum |
@roulette |
| 20:38 |
pinesol_green |
bshum: *click* |
| 20:43 |
hbrennan |
yikes. |
| 21:26 |
phasefx |
@dnd |
| 21:26 |
pinesol_green |
phasefx: strength:9 dexterity:12 constitution:15 intelligence:15 wisdom:10 charisma:13 |
| 21:26 |
phasefx |
I'll take it |
| 22:03 |
|
b_bonner_ joined #evergreen |
| 22:13 |
|
zxiiro_ joined #evergreen |
| 23:21 |
|
ktomita_ joined #evergreen |
| 23:35 |
|
dbwells_ joined #evergreen |
| 23:44 |
|
dluch joined #evergreen |
