| Time |
Nick |
Message |
| 04:01 |
|
abowling1 joined #evergreen |
| 06:30 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:23 |
|
rjackson_isl joined #evergreen |
| 07:28 |
|
agoben joined #evergreen |
| 08:25 |
|
collum joined #evergreen |
| 08:28 |
|
jvwoolf joined #evergreen |
| 08:32 |
|
rlefaive joined #evergreen |
| 08:34 |
|
remingtron joined #evergreen |
| 08:39 |
|
kmlussier joined #evergreen |
| 08:52 |
|
mmorgan joined #evergreen |
| 08:52 |
|
Dyrcona joined #evergreen |
| 08:54 |
|
bos20k joined #evergreen |
| 09:21 |
|
yboston joined #evergreen |
| 09:21 |
Bmagic |
@coffee jeff |
| 09:21 |
* pinesol_green |
brews and pours a cup of Panama Gesha 2011, and sends it sliding down the bar to jeff |
| 09:21 |
Bmagic |
@coffee [someone] |
| 09:21 |
* pinesol_green |
brews and pours a cup of Kenya AA, 2003 Crop, and sends it sliding down the bar to abneiman |
| 09:21 |
Bmagic |
@coffee |
| 09:21 |
* pinesol_green |
brews and pours a cup of La Esperanza Colombia Huila, and sends it sliding down the bar to Bmagic |
| 09:21 |
Bmagic |
On me yall |
| 09:22 |
abneiman |
bmagic++ man, I NEVER get coffee from pinesol! :) |
| 09:22 |
Bmagic |
haha! Today is your lucky day! Time to play the lotto |
| 09:23 |
abneiman |
nope, clearly I used up all my luck on virtual coffee |
| 09:24 |
Bmagic |
:) |
| 09:33 |
Bmagic |
Just got hit with an email about a company wanting to register our domain name in China. They wanted to know if we had business partners in China. I was fooled into replying but I should have googled it |
| 09:33 |
Bmagic |
https://www.chinalawblog.com/2017/08/the-china-domain-name-scam-redux.html |
| 09:42 |
Dyrcona |
I had someone ask to buy my domain a few years ago. |
| 09:55 |
|
yboston joined #evergreen |
| 10:02 |
|
rlefaive joined #evergreen |
| 10:06 |
|
littlet joined #evergreen |
| 10:11 |
|
rlefaive joined #evergreen |
| 10:11 |
|
stephengwills joined #evergreen |
| 10:33 |
|
rlefaive joined #evergreen |
| 10:40 |
|
Jaswinder joined #evergreen |
| 10:41 |
Jaswinder |
Hi Guys, I want to test Ebook API methods form srfsh. I tried to send the request it is throwing 500 error, Basically, I am trying to call this Ebook API method: open-ils.ebook_api.title.details |
| 10:42 |
Jaswinder |
Example: request open-ils.ebook_api open-ils.ebook_api.title.details {"sessionid":"1522161337.001863.152216133624029","title_id":"a"} |
| 10:42 |
Jaswinder |
Any thoughts, how do I test it? |
| 10:48 |
|
kmlussier joined #evergreen |
| 10:52 |
* kmlussier |
plans to buy lottery ticket since she hasn't used up her luck yet on virtual coffee. |
| 10:55 |
Jaswinder |
Can anyone help here? |
| 10:56 |
dbs |
sorry Jaswinder I don't know anything about the ebook api |
| 11:01 |
Jaswinder |
Anyone else? |
| 11:03 |
JBoyer |
jeffdavis would know the most about it, but I'll mention some low-hanging fruit: You already have the setup you need on the Overdrive or OneClick side? This service can't do anything without some configuration on their side. (and they usually point you at a test system, etc.) |
| 11:05 |
|
mmorgan1 joined #evergreen |
| 11:07 |
dbwells |
Jaswinder: it looks like that method is not expecting an argument hash, but rather just two unnamed arguments, something like: request open-ils.ebook_api open-ils.ebook_api.title.details "1522161337.001863.152216133624029", "a" |
| 11:07 |
|
stephengwills joined #evergreen |
| 11:08 |
dbwells |
(using your example data, which may or may not do something of value) |
| 11:10 |
Jaswinder |
JBoyer - Yes, i do have setup |
| 11:12 |
JBoyer |
Ok. Did dbwells suggestion help? I've not called these services manually so I have no idea if that should work. The title id seems a little unusual, does it need a real UUID or something, or can you not even connect to the remote service to make the call? |
| 11:13 |
Jaswinder |
dbwells - I have my own ebook version of API setup locally which I am calling using the srfsh utility. I tried what you sent but it gave no response. |
| 11:13 |
|
rlefaive joined #evergreen |
| 11:14 |
kmlussier |
dbwells++ |
| 11:14 |
Jaswinder |
This is an example command: request open-ils.ebook_api open-ils.ebook_api.get_patron_id "ASDABS-ASSAS-2323", "1212133", "BAC GMAIL.COM" |
| 11:15 |
kmlussier |
gmcharlt / phasefx: I was thinking of upgrading the MassLNC community demo server to 3.1. Would you be okay with keeping the Equinox demo server at 3.0 so that folks have a choice of which release they try out? |
| 11:20 |
dbwells |
Jaswinder: also, your session ID doesn't look like what I would expect for an Ebook API session ID. It looks like an OpenSRF session ID, which isn't what you want, I think. |
| 11:21 |
dbwells |
Jaswinder: in my basic testing, my session ID looks like "ebook_api.ses.{HASH_VALUE}" |
| 11:21 |
|
Christineb joined #evergreen |
| 11:21 |
dbwells |
This is after starting a "test" session using: request open-ils.ebook_api open-ils.ebook_api.start_session "ebook_test", "1" |
| 11:24 |
|
rlefaive joined #evergreen |
| 11:35 |
Jaswinder |
dbwells - You are correct. I tried with new session id and I got no response. |
| 11:36 |
gmcharlt |
kmlussier: in the short term, but not indefinitely |
| 11:36 |
Jaswinder |
I tried the following: request open-ils.ebook_api open-ils.ebook_api.get_patron_id "B225D41C-CDFS-4F60-9FF5-AASSWD11", "ebook_api.ses.a8f6bb29081a44b1ff83f7e542dcca6d", "abcgmail.com" |
| 11:37 |
Jaswinder |
The perl code looks for 3 arguments: token, session_id, and an email |
| 11:38 |
kmlussier |
gmcharlt: OK, thanks. I'll upgrade it for now then. I can always move it back to 3.0 when the other server moves to 3.1. |
| 11:38 |
bshum |
systemd-- # cries to himself as he keeps poking at Ubuntu 18.04 |
| 11:39 |
JBoyer |
gmcharlt, while you're around, can you look at the last comment of bug 1758347 |
| 11:39 |
pinesol_green |
Launchpad bug 1758347 in Evergreen 3.0 "wrong version of asset.merge_record_assets() stored function" [High,Fix committed] https://launchpad.net/bugs/1758347 |
| 11:39 |
JBoyer |
? |
| 11:39 |
pastebot |
"Jaswinder" at 64.57.241.14 pasted "srfsh logs" (21 lines) at http://paste.evergreen-ils.org/1049 |
| 11:39 |
gmcharlt |
JBoyer: crap. thanks for the heads-up |
| 11:40 |
Jaswinder |
I have pasted the logs here |
| 11:40 |
JBoyer |
gmcharlt, it happens. I just wanted to make sure it was seen before 3.0.6 is cut. Thanks! |
| 11:41 |
kmlussier |
JBoyer++ |
| 11:50 |
Jaswinder |
dbwells - any comments/suggestions based on the logs? |
| 11:58 |
dbwells |
Jaswinder: sorry, stepped away for a few minutes. |
| 11:58 |
Jaswinder |
np |
| 12:02 |
dbwells |
Jaswinder: I don't see anything jumping out in that log, sorry. |
| 12:02 |
Jaswinder |
hmm - Is there any place that I can check? |
| 12:03 |
dbwells |
Most likely place to start would be the osrfsys.log |
| 12:05 |
|
khuckins joined #evergreen |
| 12:06 |
Jaswinder |
let me check that as well |
| 12:07 |
dbwells |
Jaswinder: So, 'get_patron_id' is something you have added? I do not see it in the code. |
| 12:08 |
* berick |
tries to devise the incantation that makes an opensf 2.5 api call behave like an opensrf 2.4 api call WRT bundling/chunking. |
| 12:08 |
Jaswinder |
Yes, I did as a part of addition API vendor |
| 12:10 |
Jaswinder |
So, basically, I added an entry to EbookAPI.pm inside the %vendors% block as : 'ebooks' => 'OpenILS::Application::EbookAPI::EBooks' |
| 12:11 |
Jaswinder |
and then created a perl class similar to oneclickdigital.pm as Ebooks.pm inside /Application/EbookAPI/ |
| 12:11 |
|
jihpringle joined #evergreen |
| 12:11 |
Jaswinder |
If you want, I can share a piece of code to your email or another location |
| 12:13 |
dbwells |
Jaswinder: From the log you posted, it seems like your request "worked", it just didn't return anything. I might try making my 'get_patron_id' method return "TEST" right away or something like that just to make sure everything is wired up correctly. I need to head out for a bit, but if you want to email, it is dbw2calvin.edu. I'll be back in around an hour or so. |
| 12:14 |
Jaswinder |
Thanks and I appreciate it! I will try it out first and then email you if I need help. |
| 12:16 |
|
terran joined #evergreen |
| 12:17 |
|
ngf42 joined #evergreen |
| 12:38 |
|
rlefaive joined #evergreen |
| 12:41 |
mmorgan1 |
Anyone willing to run a quick query? I'm wondering if anyone gets significant numbers from the query in comment #1 of bug 1758975 |
| 12:41 |
pinesol_green |
Launchpad bug 1758975 in Evergreen "Item status can be set to Lost and Paid inappropriately when closing a transaction for a Long Overdue or Lost returned item" [Undecided,New] https://launchpad.net/bugs/1758975 |
| 12:43 |
|
yboston joined #evergreen |
| 12:44 |
jeffdavis |
mmorgan1: that query gives 7 hits with null stop_fines and 26 with stop_fines = LOST on our system |
| 12:47 |
Dyrcona |
Gives me 1 with null stop_fines and 1,938 with LOST. |
| 12:50 |
mmorgan |
jeffdavis: Dyrcona: Interesting. I don't find any with NULL stop_fines, but not too long ago we closed a large number of transactions that had been erroneously left open. |
| 12:53 |
mmorgan |
Our circs go to LONGOVERDUE at 42 days. Dyrcona: I seem to remember your timetable for going to LOST was similar? Or maybe shorter? |
| 13:05 |
Dyrcona |
Looks like like we have two a/ts for lost at 14 days and 28 days. |
| 13:06 |
Dyrcona |
We don't use longoverdue. |
| 13:12 |
collum |
"No rows found" |
| 13:12 |
collum |
Lost at 60 days and we don't use longoverdue. |
| 13:14 |
Dyrcona |
The 14 days lost is limited to 3 org units by a filter. I suspect 3 banches of the same system, but I didn't look them up. |
| 13:14 |
Dyrcona |
Everyone else uses 28 days. |
| 13:15 |
mmorgan |
We also have the ou settings to restore overdues on long overdue/lost item return set to TRUE, which likely plays into this. |
| 13:24 |
mmorgan |
jeffdavis++ |
| 13:24 |
mmorgan |
Dyrcona++ |
| 13:24 |
mmorgan |
collum++ |
| 13:24 |
* mmorgan |
runs off to fix status in 84 items |
| 13:28 |
Jaswinder |
dbwells - I have sent you an email with the source code. |
| 13:47 |
dbwells |
Jaswinder: well, I can answer at least one question. In Evergreen, an "authtoken" is a hash value you receive when you first login, and is then sent on all subsequest request to make sure you are a valid user with a valid session. The easiest way to get one is via the srfsh "login" command, e.g.: login userName password123 |
| 13:49 |
dbwells |
That should give you a response of "Login Session: c35674c2d671..." The token is that long hexidecimal string. |
| 13:52 |
dbwells |
So, the full call, something like: request open-ils.ebook_api open-ils.ebook_api.get_patron_id "<authtoken goes here>", "ebook_api.ses.<some session value>", "abcgmail.com" |
| 13:54 |
Jaswinder |
Thanks, I did try new token and session id for ebooks and ran the request. The logs says that the authentication failed. |
| 13:56 |
Jaswinder |
I also noticed this in the log: No user in the cache exists with key oils_auth_ebooks |
| 13:57 |
Jaswinder |
Does this have any role in authenticating the user? |
| 14:01 |
dbwells |
Jaswinder: can you paste up your exact error log? (with truncated session values, if needed) |
| 14:04 |
dbwells |
That last error comes from the Evergreen auth service (oils_auth.c), but I don't know where that key would come from. |
| 14:06 |
dbwells |
Nevermind, I see, somewhere you are passing an authtoken of "ebooks". That obviously isn't right. |
| 14:06 |
pastebot |
"Jaswinder" at 64.57.241.14 pasted "osrfsys.log" (8 lines) at http://paste.evergreen-ils.org/1058 |
| 14:08 |
dbwells |
Jaswinder: In that example, you have an extra argument "recordedbooks". It is not needed. |
| 14:09 |
Jaswinder |
Actually, it is ebooks |
| 14:10 |
Jaswinder |
I needed to specify a vendor name. so the call can go to correct vendor file. In my case, it is Ebooks.pm file |
| 14:10 |
dbwells |
Well, in any case, the first argument must be the authtoken, then the session, then the identifier (email, in this case). |
| 14:10 |
dbwells |
No, the vendor is "baked in" to the ebook_api session at that point. |
| 14:11 |
dbwells |
Only the intial "start_session" needs the vendor key. |
| 14:11 |
|
yboston joined #evergreen |
| 14:13 |
dbwells |
Jaswinder: Does that make sense? |
| 14:13 |
* dbwells |
goes to check his assumptions... |
| 14:13 |
Jaswinder |
Yes |
| 14:15 |
Jaswinder |
One more question |
| 14:16 |
Jaswinder |
I have called the request without the vendor name. I see authentication failed again. I see this in the log: EbookAPI: getting patron id for patron abcgmail.com How do I make sure that this user exist in EG? |
| 14:21 |
Dyrcona |
Jaswinder: That code is normally used from the OPAC, after a patron has logged in. It generally sends the patron barcode to the vendor. |
| 14:21 |
Dyrcona |
Jaswinder: What, exactly, are you trying to do? |
| 14:23 |
_bott_ |
Does anyone have a good reference for the color codes for the rectangle around a user's name in the XUL client? |
| 14:25 |
dbwells |
Jaswinder: You will see the "getting patron id for patron..." message every time if you have "info" or lower level logging on. Are you also getting the "authentication failed" message still? If so, your authtoken is bad. |
| 14:27 |
dbwells |
Jaswinder: Other than that, there is no inherent connection between Evergreen users and the token you are passing along to your provider. I think the assumption here is you will do everything logged in as yourself, so having an authtoken already guarentees you are in Evergreen. |
| 14:27 |
kmlussier |
_bott_: I've seen good documentation somewhere on it. I can't remember if it was in the eg docs or somewhere else. |
| 14:28 |
Jaswinder |
Yes, I am getting that. I want to get the patron id from external API call after the user is authenticated by EG. |
| 14:28 |
_bott_ |
kmlussier: I know I've seen the same, but I haven't tracked it down |
| 14:28 |
kmlussier |
_bott_: http://docs.sitka.bclibraries.ca/Sitka/current/html/patron-status-colour.html |
| 14:28 |
kmlussier |
When in doubt, always start with the Sitka documentation. |
| 14:28 |
_bott_ |
sitka++ |
| 14:29 |
kmlussier |
_bott_: I don't know if those are standard colors or if they have been customized. |
| 14:29 |
_bott_ |
It's a start! |
| 14:30 |
dbwells |
Jaswinder: Then I would say you do not need to further check for the user existence based on their email, they have already logged in at that point? Does that help? |
| 14:31 |
dbwells |
Jaswinder: It looks like you are on the right track. |
| 14:32 |
Jaswinder |
dbwells: so basically, I can remove these lines: # verify that user is authenticated in EG my $e = new_editor(authtoken => $auth); if (!$e->checkauth) { $logger->error("EbookAPI: authentication failed: " . $e->die_event); return; } |
| 14:32 |
dbwells |
Jaswinder: No, that is what checks the token. |
| 14:33 |
dbwells |
Jaswinder: if *that* is still failing, for some reason you don't have a valid authtoken. Try logging in again? |
| 14:33 |
Jaswinder |
The call is failing and I am getting :EbookAPI: authentication failed message |
| 14:33 |
Jaswinder |
yes, I can try that |
| 14:34 |
dbwells |
by default, a token expires after one hour. |
| 14:34 |
Dyrcona |
Jaswinder: Have you logged in to Evergreen in srfsh before trying this? |
| 14:34 |
Jaswinder |
yes, using logind egadmin password |
| 14:34 |
|
dwgreen joined #evergreen |
| 14:35 |
Jaswinder |
*correction login egadmin password |
| 14:36 |
Dyrcona |
OK, and you are passing the authtoken that you got back from that as a quoted string in the first argument to your calls? |
| 14:36 |
Jaswinder |
yes |
| 14:36 |
Dyrcona |
Then, you would not see that message from that line of code. |
| 14:39 |
Dyrcona |
If that line that you pasted fails, you're not logged in or you've passed an invalid token. |
| 14:40 |
Jaswinder |
Okay, so I am pass the authentication process. Now, I need to get the user email associated with that username from db and the library id. |
| 14:40 |
Jaswinder |
Any thoughts on this? |
| 14:41 |
dbwells |
Jaswinder: yeah, progress! |
| 14:41 |
Jaswinder |
Yup :) |
| 14:41 |
Jaswinder |
Thank You guys for helping |
| 14:43 |
Dyrcona |
Jaswinder: There's a call to retrieve the patron session...open-ils.auth.session.retrieve. |
| 14:43 |
Dyrcona |
You pass the Evergreen session token, and it returns the user object. |
| 14:44 |
Dyrcona |
That object has an email field with the email address, if any. |
| 14:44 |
Dyrcona |
So, are you adding a new vendor that requires the email address to authenticate the patron? |
| 14:46 |
jeffdavis |
Jaswinder: what ebook vendor are you trying to work with? |
| 14:46 |
Jaswinder |
Dyrcona - I received an error because there is no library id associate with this user. So, the external API call failed. I need to have a library ID |
| 14:47 |
Jaswinder |
I am adding a new vendor. |
| 14:47 |
jeff |
Jaswinder: which vendor? |
| 14:47 |
Dyrcona |
Jaswinder: You copied the Overdrive code? |
| 14:47 |
Jaswinder |
I may need to add an entry to DB |
| 14:47 |
Dyrcona |
You would get your library ID from your vendor, if necessary. |
| 14:47 |
Jaswinder |
No, I copied the OneClickdigital library code |
| 14:48 |
Dyrcona |
OK. It's similar. |
| 14:48 |
jeffdavis |
Jaswinder: I'm asking because your code earlier mentioned "recordedbooks", but Recorded Books and OneClickdigital are the same thing. |
| 14:48 |
Jaswinder |
The requirement is to have library ids and a key in a separate table |
| 14:49 |
Jaswinder |
Jeff - Yes, that is correct |
| 14:49 |
Dyrcona |
Jaswinder: There are settings for that in config.org_unit_setting_type. |
| 14:49 |
Jaswinder |
But, I am writing the code with different set of APIs |
| 14:49 |
Dyrcona |
I'll ask again, what are you trying to do? |
| 14:50 |
Dyrcona |
What's your overall goal? |
| 14:50 |
dbwells |
I think he is adding "ebooks.com" as a provider. |
| 14:51 |
Jaswinder |
Dyrcona: My goal is to add additional vendor that will be used for basic search, advanced search, checkouts, and etc. vendor API will return books or audio data and that will be displayed on the results. |
| 14:52 |
Dyrcona |
Jaswinder: The best way to get the results in search is to add MARC records supplied by the vendor, then you use the oneclick digital settings with Evergreen. |
| 14:52 |
Dyrcona |
If you want to use a search API for recorded books to add to Evergreen search results, a different approach would be in order. |
| 14:53 |
Dyrcona |
I think you'd want to add a module to search that would use the vendor's api and add the results to your evergreen search results. |
| 14:53 |
Dyrcona |
I'm not sure if anyone has ever done that. |
| 14:54 |
Jaswinder |
I am not aware of MARC records. However, I want another vendor just like Overdrive with additional functionality in place |
| 14:54 |
JBoyer |
I gave it serious consideration until I saw the API I was going to have to talk to and then changed my mind. (XML makes you want to wash yo mouth out with SOAP...) |
| 14:54 |
bshum |
ha ha |
| 14:56 |
Jaswinder |
One more thing, the results will be displayed in a separate tab along with other vendor tabs (if any). I have the db schema that contains library_id and keys that will be used for all API calls. |
| 14:56 |
Dyrcona |
Jaswinder: The vendor can usually supply you with records that you can load into the database of the items in your collection with them. |
| 14:59 |
jeffdavis |
Incorporating external search results would be pretty neat. We don't do that for Overdrive and OneClickdigital because integrating the results with EG search results runs into the difficulties JBoyer mentioned, but showing them separately from Evergreen search results ("bento box" style search) seems reasonable. |
| 15:01 |
jeffdavis |
And it has the advantage of not needing to import vendor records into Evergreen in advance. :) |
| 15:01 |
Jaswinder |
Jeff: Yes, I agree. |
| 15:02 |
Jaswinder |
This is approach that I taking |
| 15:02 |
jeffdavis |
I like it :) |
| 15:03 |
Jaswinder |
For the initial part, I need to first ensure that the user is authenticated by EG which I have done so. Then, I want to use the user email along with library id and key to make calls out. |
| 15:03 |
|
mmorgan1 joined #evergreen |
| 15:07 |
Jaswinder |
Do I have add an entry to this table: config.org_unit_setting_type to get library_id? |
| 15:07 |
Dyrcona |
You might, but you could always use the one for oneclick digital. |
| 15:08 |
Dyrcona |
That table provides the list of basic settings. |
| 15:08 |
* csharp |
tries to wrap his head around dereferencing perl arrays within an array :-/ |
| 15:08 |
Dyrcona |
You assign them values by making an entry in actor.org_unit_setting. |
| 15:08 |
Dyrcona |
csharp: Actuall arrays or arrays refs? |
| 15:08 |
jeff |
csharp: have a concrete example? |
| 15:09 |
Dyrcona |
csharp: If you installed perldoc: man perllol |
| 15:09 |
csharp |
https://pastebin.com/AG2qk2EA |
| 15:10 |
csharp |
this is towards my attempt to fix bug 1666987 |
| 15:10 |
pinesol_green |
Launchpad bug 1666987 in Evergreen "build/tools/relator_map no longer fetching relator codes" [Low,Confirmed] https://launchpad.net/bugs/1666987 - Assigned to Chris Sharp (chrissharp123) |
| 15:11 |
Dyrcona |
So, at columns is an array of arrays of 3 members? |
| 15:11 |
Jaswinder |
Is the library id uniquer per vendor or per user in EG? |
| 15:11 |
csharp |
Dyrcona: looks like 2 members - code and relator term |
| 15:11 |
jeffdavis |
Jaswinder: it depends on your vendor. |
| 15:11 |
Dyrcona |
Jaswinder: It is per vendor, but it could be unique per library & vendor combination if you have more than 1 library in your Evergreen. |
| 15:12 |
Jaswinder |
I see. so which tables I should be using to add library_id for a vendor? |
| 15:13 |
Dyrcona |
csharp: easy: $columns[0] is the first row $columns[0][0] is the first column of the first row, $columns[0][1] is the second column of the first row, and so one. |
| 15:13 |
csharp |
Dyrcona: ah - thanks for the pointer |
| 15:14 |
* csharp |
has the "alapaca" book open to the references section |
| 15:14 |
Dyrcona |
Jaswinder: you want to insert a row in actor.org_unit_setting. |
| 15:15 |
Dyrcona |
org_unit is the id of the library, for name use the name of the setting in config.org_unit_setting_type, and for value, it is a JSON value, so use double quotes if it is a string. |
| 15:17 |
Jaswinder |
Thanks! I am going to try this now |
| 15:17 |
Dyrcona |
The id will be generated automatically if you don't supply it or leave it null. |
| 15:18 |
jeff |
csharp: looking at the HTML::TableExtract docs, and depending on what you're trying to do, the rows() table method might be more useful for you: https://metacpan.org/pod/HTML::TableExtract |
| 15:19 |
csharp |
jeff: yep, that looks better |
| 15:22 |
jeff |
also, you probably want to test success using if ($response->is_success) instead of just if ($response) -- I didn't have LWP::Protocol::https installed, so if ($response) was true but the value of $content later on was "LWP will support https URLs if the LWP::Protocol::https module is installed." |
| 15:23 |
csharp |
jeff: thanks for that |
| 15:23 |
Dyrcona |
While on the subject of inserting into actor.org_unit_setting ... I made this handy, little skeleton for Emacs this morning: https://pastebin.com/Hzq1Q9J9 |
| 15:25 |
jeff |
csharp: and you can use something like this to output two columns for each row with a tab between them: map { print "@{$_}[0]\t@{$_}[1]\n" } $table->rows(); |
| 15:26 |
jeff |
csharp: that looks a bit dense, moreso because I put it on one line for irc. |
| 15:26 |
Dyrcona |
You can do a join, which is cleaner, IMNSHO. |
| 15:26 |
jeff |
sure, or a foreach, etc! |
| 15:26 |
csharp |
jeff: that's exactly what I'm after for the finished product - thank you |
| 15:27 |
jeff |
Dyrcona: oh, you meant a join within the map, right? :-) |
| 15:28 |
jeff |
that would look something like this: map { print join("\t", @{$_}), "\n" } $table->rows(); |
| 15:28 |
Dyrcona |
jeff: I haven't looked at the output of the function you mentioned but I was thinking print join('\t', @$_) |
| 15:28 |
Dyrcona |
yes, that, more or less. |
| 15:31 |
jeff |
in that same map block, you could grab the first value from the row and test for a leading hyphen, and then either remove it or skip output of the entire row, depending on what you want to do for "Discontinued" codes. |
| 15:31 |
jeff |
(there's apparently only one) |
| 15:31 |
csharp |
jeff: yes, I want to remove the leading hyphen to be in line with the script I'm replacing |
| 15:31 |
Dyrcona |
csharp: Would the current script not work if the useragent is set? |
| 15:32 |
Dyrcona |
I never got that far into fixing it because other things came up. |
| 15:32 |
csharp |
Dyrcona: dbwells was just asking me that privately - it appeared to break when I tested, but I'll try again |
| 15:33 |
Jaswinder |
What are these fields for: org_unit_setting_type: frm_class, view_perm, update_perm? |
| 15:33 |
jeff |
cloudflare-- |
| 15:33 |
Dyrcona |
csharp: I seem to recall that part of the problem was the pages had been reorganized, but I'm unclear. |
| 15:33 |
jeff |
loc_blindly_accepting_defaults-- |
| 15:34 |
Dyrcona |
Jaswinder: You view_perm and updates_perm can be set to say that the user must have those permissions to view or update the setting via the staff client. |
| 15:35 |
Dyrcona |
fm_class is a reference to a class in the fm_IDL.xml if one exists for that setting. |
| 15:37 |
Dyrcona |
typos-- # but it's IRC. |
| 15:38 |
* csharp |
confirms that changing LWP::Simple to LWP::UserAgent works without further work to the HTML parsing |
| 15:38 |
csharp |
so yay! |
| 15:38 |
Dyrcona |
csharp++ |
| 15:38 |
csharp |
I'll look into what jeff ran into with the HTTPS warning |
| 15:39 |
Dyrcona |
IIRC, I looked into it briefly around the time of the 2.9 release. |
| 15:40 |
csharp |
I'll test on a fresh server and see if we need to add new packager dependencies |
| 15:40 |
* csharp |
heads out for now |
| 15:41 |
dbwells |
csharp++ |
| 15:42 |
Dyrcona |
Guess I never made a Lp bug for it. |
| 15:42 |
jeff |
csharp: the relator_map script uses an http url, and LoC isn't redirecting to https, so you'll likely not hit that at all. Also, I was testing your script on a box that would not have have Evergreen pre-reqs installed, so the needed modules for https might already be there. |
| 15:44 |
jeff |
somewhat annoying that a single checkbox on LoC's cloudflare account configuration interface breaks us, but working around it rather than trying to get someone to untick the box is likely the better option for now. :-P |
| 15:46 |
Jaswinder |
Dyrcona: Where do I add the library id then? Which field? |
| 15:46 |
Jaswinder |
Sorry, I am bit confused here |
| 15:47 |
|
derekz joined #evergreen |
| 15:48 |
* jeff |
tries asking LoC anyway |
| 15:53 |
|
stephengwills joined #evergreen |
| 16:02 |
jeffdavis |
INSERT INTO config.org_unit_setting_type (name, label, grp, datatype) VALUES ('ebook_api.ebooks.com.library_id','Ebooks.com Library ID', 'ebook_api', 'string'); |
| 16:02 |
jeffdavis |
INSERT INTO actor.org_unit_setting (org_unit, name, value) VALUES (1, 'ebook_api.ebooks.com.library_id', '"123"'); |
| 16:02 |
jeffdavis |
Jaswinder: I would do something like the above. |
| 16:02 |
jeffdavis |
First you create a new library setting type for Ebooks.com library_id in the config.org_unit_setting_type table, then you add the actual library_id value for your library in actor.org_unit_setting. |
| 16:03 |
Jaswinder |
Thanks Jeff |
| 16:07 |
dbwells |
Jaswinder: Just to double-check and turn your question back around, whatever you are planning to store in "library_id" is per library, not per user? |
| 16:09 |
dbwells |
Nevermind, I think see where you mentioned combining it with the user's email. I think I get what is going on now. |
| 16:10 |
Jaswinder |
yea, got it dbwells. |
| 16:11 |
|
stephengwills joined #evergreen |
| 16:12 |
Jaswinder |
Jeff, the insert statement to org_unit_settings table has org_unit as 1. Why the org_unit is 1 here? |
| 16:13 |
dbwells |
Jaswinder: that is just the default "root" org_unit |
| 16:13 |
dbwells |
It comes from the ids in the actor.org_unit table. |
| 16:13 |
Jaswinder |
So it is not tied to any vendor? |
| 16:15 |
dbwells |
That row just says "for org_unit 1, for the setting of "ebook_api.ebooks.com.library_id", the value is "123". So the second column is what ties it to a vendor, so to speak. |
| 16:15 |
Jaswinder |
Got it, this is org unit for the libraries that user selects |
| 16:16 |
dbwells |
Yes, however that happens in your case (manual selection, home ou, etc.) |
| 16:20 |
|
khuckins joined #evergreen |
| 16:20 |
Dyrcona |
Ah, no. I was messing more with the marc_tooltips script and not relator codes now that I think about it, but neither one seemed to work. |
| 16:23 |
jeff |
Dyrcona: loc.gov is behind cloudflare now, and they have the (default, I believe) "Browser Integrity Check" enabled, so certain user-agent strings (or lack of user-agent string) are rejected: https://support.cloudflare.com/hc/en-us/articles/200170086-What-does-the-Browser-Integrity-Check-do- |
| 16:24 |
Dyrcona |
This was in 2015 and they were blocking based on UserAgent. |
| 16:24 |
Dyrcona |
Dunno if they were at cloudflare, then, but the useragent checks were in place. |
| 16:24 |
Dyrcona |
Kind of dumb, really. I write bots all the time that claim to be Firefox. |
| 16:25 |
Dyrcona |
And, I'm usually just interested in the text, so actually less stress on the site than a normal user who grabs images, etc. |
| 16:26 |
Dyrcona |
Anyway, I'm looking at the scripts, now, and thinking about commenting on bug 1746121. |
| 16:26 |
pinesol_green |
Launchpad bug 1746121 in Evergreen "Wishlist: Add MARC field help to the MARC editor" [Wishlist,New] https://launchpad.net/bugs/1746121 |
| 16:27 |
Dyrcona |
IIRC, the one that grabs the French translations from LAC-BAC worked the last time I tried it. |
| 16:28 |
Dyrcona |
Maybe I'll comment some other time. |
| 16:28 |
jeff |
Dyrcona: I'm going to pester webmasterloc to see if they'll add a page rule, but that's no reason not to fix it on our end. :-) |
| 16:30 |
Dyrcona |
I have no problem lying about a useragent string. :) |
| 16:38 |
* Dyrcona |
has written some rather sophisticated link checking bots, including one that parses a JSON object out of the JavaScript on the page that comes back and checks one of it's fields to determine if the link is still valid. Yes, someone has gone that far in making their site "dynamic." |
| 16:38 |
Dyrcona |
They can't just return a 404 like normal people. |
| 16:41 |
jeff |
we have at least one vendor where their removed/deleted titles cannot be determined until you make an attempt to view/circ the title. they are of course a CPC vendor, so the only way to do a definitive check for validity costs a non-trivial amount of money. :-) |
| 16:42 |
Dyrcona |
That's a bummer. |
| 17:00 |
|
mmorgan joined #evergreen |
| 17:06 |
|
mmorgan left #evergreen |
| 17:10 |
Jaswinder |
Guys, is there a way I can write test case to run srfsh commands. I don't like to regenerate sessions every hour |
| 17:19 |
dbwells |
Jaswinder: srfsh is just a handy way to run OpenSRF commands. Our live tests are written in Perl rather than using srfsh. You can find them in Evergreen/Open-ILS/src/perlmods/live_t (for inspiration) |
| 17:23 |
Jaswinder |
I am looking at this and this is good |
| 17:23 |
Jaswinder |
How do I retrieve email id of the user from table? Also, how DB calls are made to retrieve information? |
| 17:32 |
dbwells |
Jaswinder: I need to head out shortly, but there are no good quick answers to those questions. Here is a quick answer which might be okay: |
| 17:33 |
dbwells |
There are separate APIs for fetching many things. These are not uniform, but particular to certain needs. |
| 17:33 |
dbwells |
To get the email for a usr, you need to get at the usr object. |
| 17:34 |
dbwells |
One way to do this, as mentioned earlier, is via 'open-ils.auth.session.retrieve'. |
| 17:34 |
Jaswinder |
okay, So for newly created tables, I can still use existing API to get data? |
| 17:36 |
dbwells |
Actually, it looks like the whole usr object might already be on the 'requestor' for the CStore editor, so $editor->requestor->email might be all you need here. |
| 17:37 |
dbwells |
In addition to "custom" APIs, there are uniform APIs which are autogenerated from something called the IDL. |
| 17:37 |
dbwells |
An entirely new table would also need an entry in the IDL to get these methods autogenerated. |
| 17:38 |
Jaswinder |
Good to know that information. I will research on this as how I can leverage IDL to get the data |
| 17:40 |
dbwells |
Jaswinder: if you go back to here and look under "Dealing with Data", you might be able to pick it up. http://library.calvin.edu/devdocs_project/doku.php?id=start Use the live tests for examples as well. |
| 17:41 |
dbwells |
Lack of tuits have prevented that page from maturing, but having you asking these questions is the right kind of inspriation. |
| 17:41 |
dbwells |
Basically, for your needs, "cstore" and "CstoreEditor" are the things to focus on. PCRUD is similar, but more useful for external/client-side access to things. |
| 17:42 |
Jaswinder |
thanks! I am going to review these docs now |
| 17:46 |
dbwells |
I know the docs are still jumbled and incomplete. You seem savvy enough, so that information combined with trolling through the code for "cstore" and "$editor->" calls might get you there. Good luck! Folks are generally most in the channel during 9-5 Eastern, though a few are further west. |
| 17:46 |
* dbwells |
heads out |
| 18:31 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:33 |
Dyrcona |
"And, there was much rejoicing." |
| 18:46 |
ngf42 |
http://i.imgur.com/oLYSHuu.gif |
| 18:47 |
Dyrcona |
Indeed! |
| 18:47 |
Dyrcona |
ngf42++ |
| 18:48 |
ngf42 |
EveryoneWhoDidActualWorkOnMakingThatHappen++ Dyrcona++ |
| 18:52 |
|
abowling joined #evergreen |
| 18:54 |
Dyrcona |
It's weird some of the stuff you find if you just latest pates on Pastebin. |
| 20:07 |
|
Dyrcona joined #evergreen |
| 20:21 |
|
stephengwills joined #evergreen |
