| # | Time | Nick | Message |
|---|
| # | 03:18:42 | atz has quit IRC |
| # | 04:39:47 | drdata_home has quit IRC |
| # | 04:45:23 | drdata_home has joined #evergreen |
| # | 07:50:39 | artunit_ has joined #evergreen |
| # | 07:51:46 | mrpeters-isl has joined #evergreen |
| # | 07:51:46 | artunit has quit IRC |
| # | 07:52:01 | artunit_ is now known as artunit |
| # | 08:25:51 | adbowling-isl has joined #evergreen |
| # | 08:31:59 | ernieSimuro has joined #evergreen |
| # | 08:35:14 | ernieSimuro has quit IRC |
| # | 08:35:46 | sfortin has quit IRC |
| # | 08:35:58 | sfortin has joined #evergreen |
| # | 08:36:32 | ernieSimuro has joined #evergreen |
| # | 08:38:51 | mrpeters-isl | ok - what am I doing wrong with git? fatal: Paths with -a does not make sense. |
| # | 08:38:57 | mrpeters-isl | git commit -as "LP#739444 Incrementing opac.dtd copyright date - Signed-off-by: Michael Peters <mrpeters@library.in.gov>" |
| # | 08:41:20 | Dyrcona has joined #evergreen |
| # | 08:42:28 | atheos has joined #evergreen |
| # | 08:43:31 | tsbere | mrpeters-isl: -a is "auto", -s is "signoff", you need a -m for "message" |
| # | 08:44:43 | mrpeters-isl | aha - wiki needs updating then |
| # | 08:44:52 | mrpeters-isl | git commit -as # and enter a useful comment |
| # | 08:48:39 | tsbere | If you don't put a comment on the command line it will bring up an editor |
| # | 08:50:25 | kmlussier has joined #evergreen |
| # | 08:55:56 | gdunbar has joined #evergreen |
| # | 09:04:37 | bradl_ is now known as bradl |
| # | 09:15:51 | yboston has joined #evergreen |
| # | 09:15:52 | Callender has quit IRC |
| # | 09:16:39 | jenny1 has joined #evergreen |
| # | 09:17:45 | dbs has joined #evergreen |
| # | 09:21:39 | dbs | mrpeters-isl: For some reason your sign-off was on the same line as your commit message; I'll fix that locally |
| # | 09:26:22 | dbs | applied |
| # | 09:28:27 | mrpeters-isl | dbs: sorry about that...its probably because i did my message from the command line |
| # | 09:28:38 | mrpeters-isl | ill let it pop up the external editor from now on |
| # | 09:29:21 | Callender has joined #evergreen |
| # | 09:31:18 | dbs | No sweat off my back |
| # | 09:31:45 | dbs | thanks for the patch! |
| # | 09:34:08 | mrpeters-isl | no problem |
| # | 09:34:11 | kmlussier has quit IRC |
| # | 09:35:30 | dbs | git question: once a branch has been merged into Evergreen or OpenSRF from working, should we / when should we then delete the remote branch? |
| # | 09:36:54 | mrpeters-isl | dbs: so, do you mean: create a "working_branch", do your patching, submit your patch, it gets accepted....then what? |
| # | 09:37:16 | Dyrcona | dbs: you can keep 'em or delete 'em, its up to you. |
| # | 09:37:20 | dbs | mrpeters-isl: right. |
| # | 09:37:29 | Dyrcona | I usually delete 'em. tsbere tends to keep 'em. |
| # | 09:37:29 | mrpeters-isl | what i was doing in that case is just reverting back to the beginning |
| # | 09:37:45 | mrpeters-isl | so i always have just 1 working directory that i make my patches in then "start over" |
| # | 09:37:46 | dbs | Dyrcona: well, technically yes, but I'm asking as a matter of project preference |
| # | 09:38:54 | dbs | Dyrcona: if everyone keeps all of their working branches forever, then it's going to get kind of hard to find the useful branches eventually, no? |
| # | 09:39:14 | Dyrcona | yes, pretty much. |
| # | 09:39:31 | tsbere | I have started leaning towards "If I don't think I will continue to work on that branch, delete it when it is merged" |
| # | 09:39:57 | Dyrcona | i was thinking more about local branches, but I see you're talking about on git.evergreen-ils.org. |
| # | 09:40:17 | Dyrcona | there, i'd say, delete them after they're merged. |
| # | 09:40:59 | collum has joined #evergreen |
| # | 09:41:44 | dbs | okay, sounds good to me |
| # | 09:41:55 | ernieSimuro | jeff: Are you any closer to checking in the operSRF-PHP version I know that when yoyu were on vacation you said that it only needed to be checked in, where do we stand? |
| # | 09:57:42 | phasefx | collab/phasefx/honor_unified_vol_copy_ui_setting, is that the sort of thing we want a 2nd sign-off on or is it trivial enough to just push to master/rel_2_1 without such? |
| # | 10:00:22 | phasefx | (given that I don't usually push untested code anyway) |
| # | 10:02:49 | pmplett has joined #evergreen |
| # | 10:13:35 | kmlussier has joined #evergreen |
| # | 10:26:39 | moodaepo has joined #evergreen |
| # | 10:28:00 | bshum | dbs: Hmm, broad searches within a specific copy location are apparently still failing for us. Like searching for the word "japan" in "juvenile nonfiction". Just spins forever. :( |
| # | 10:28:31 | bshum | On the plus side, specific searches in copy location = extremely fast. |
| # | 10:31:21 | mrpeters-isl | bshum: have you tried to find the actual SQL query that is performed and tried in in PSQL? |
| # | 10:31:38 | bshum | mrpeters-isl: That's a good idea, I will try to see if I can dig it out. |
| # | 10:31:50 | mrpeters-isl | maybe you could do a EXPLAIN ANALYZE on it |
| # | 10:32:06 | mrpeters-isl | now, i dont know for sure what you'd be looking for |
| # | 10:32:11 | mrpeters-isl | but at least you could see its action plan |
| # | 10:33:29 | mrpeters-isl | how are you searching within a specific location? |
| # | 10:33:31 | mrpeters-isl | custom code? |
| # | 10:33:53 | bshum | Nah, it's an advanced search option |
| # | 10:34:05 | mrpeters-isl | really? i dont think we have that... |
| # | 10:34:08 | bshum | When you're scoped to a specific library |
| # | 10:34:18 | bshum | There's a new box that shows up below pub year |
| # | 10:34:22 | bshum | Called "Shelving Location" |
| # | 10:34:27 | bshum | That has all your copy locations in it. |
| # | 10:34:28 | mrpeters-isl | yeah, i am --- dont have that |
| # | 10:34:31 | mrpeters-isl | what version? |
| # | 10:34:37 | bshum | It's been there for awhile. |
| # | 10:34:40 | bshum | We had it in 2.0.1 |
| # | 10:34:47 | bshum | Still trying to use it in 2.0.6 now |
| # | 10:34:52 | mrpeters-isl | interesting....i dont have it |
| # | 10:34:57 | mrpeters-isl | wonder if its something you have to turn on |
| # | 10:35:33 | bshum | Hmm, I don't think we did anything too different. |
| # | 10:36:01 | bshum | Yeah it's in yours |
| # | 10:36:03 | bshum | I can see it.. |
| # | 10:36:10 | bshum | Switch to advanced search |
| # | 10:36:19 | mrpeters-isl | im there....and ive picked a library |
| # | 10:36:19 | bshum | Then go to the Search Library options |
| # | 10:36:28 | mrpeters-isl | oh i need to pick a branch |
| # | 10:36:29 | bshum | You have to select the specific branch, not the system |
| # | 10:36:38 | bshum | Right, cause the branch has locations, not the system :) |
| # | 10:37:10 | bshum | It's more evident in the staff client. The default search catalog brings you to advanced search set to the scope of a single library branch anyways. |
| # | 10:37:18 | mrpeters-isl | hmm taking a bit of time here too |
| # | 10:37:22 | bshum | Notably, that list of locations is unordered. |
| # | 10:37:31 | mrpeters-isl | yeah i noticed that |
| # | 10:37:41 | mrpeters-isl | not orderd by id maybe? |
| # | 10:38:00 | mrpeters-isl | looks like it spins here too |
| # | 10:38:30 | bshum | dbs put out some new function changes for that (and other things): http://git.evergreen-ils.org/?p=working/Evergreen.git;a=commit;h=095658b44ea6a64678fb68253eda1ef11a664863 |
| # | 10:39:03 | bshum | If you alter the search function to use that unnest() instead of search.explode_array() it does improve the searching. But apparently doesn't fix it entirely. |
| # | 10:51:20 | bshum | Ah, looks like it's choking on our search relevance ranking for keyword searches |
| # | 10:55:38 | jenny1 has quit IRC |
| # | 11:04:24 | dbs | mrpeters-isl / bshum: If you want to understand what's happening in search, you need to step through the logic in search.query_parser_fts() function manually - at least, that's the best approach that I've been able to come up with |
| # | 11:05:46 | dbs | bshum: like you, broad searches on shelving locations tend to time out for us as well |
| # | 11:06:23 | mrpeters-isl | i wondered if maybe it was just timing out, but would eventually finish at the db level |
| # | 11:06:24 | bshum | dbs: We recommended to our libraries to use one of the other search categories (not keyword). Using subject, etc. seemed to work slightly better. |
| # | 11:06:48 | dbs | mrpeters-isl: yes, due to the 60 second opensrf default time out. the query continues to run at the database level |
| # | 11:07:00 | mrpeters-isl | yeah cool thought it might |
| # | 11:07:18 | dbs | bshum: yep, that's the same shameful advice that we have to give |
| # | 11:07:41 | dbs | which is why I was diving into the search bits over the past week |
| # | 11:08:06 | mrpeters-isl | wonder if its "close" to 60 seconds to where we could increase it by 50% or so |
| # | 11:09:23 | dbs | mrpeters-isl: if your users are having to wait 60 seconds for search results, they won't be your users for long |
| # | 11:10:08 | mrpeters-isl | true, but wouldn't the other searches still run quicker? i'm thinking just as a temp workaround until its fixed |
| # | 11:13:31 | dbs | mrpeters-isl: the speed of the other searches should be unaffected, yes |
| # | 11:22:47 | jenny has joined #evergreen |
| # | 11:26:32 | phasefx | collab/phasefx/unified_vol_copy_ui_from_marc_editor_fast_add for those interested (requires a new local staff client) |
| # | 11:30:00 | dbs | phasefx: I'll take a peek at the code at least |
| # | 11:30:24 | phasefx | muchos gracias |
| # | 11:35:04 | dbs | phasefx++ # I can read the code! |
| # | 11:35:29 | phasefx | line breaks are our friend |
| # | 11:36:31 | dbs | One i18n bit ('Retrieving title...') to-do; it's in collab though so in theory I could do that |
| # | 11:37:03 | phasefx | that's a copy/paste job too, I think, so such i18n could be used elsewhere as well |
| # | 11:37:19 | james has joined #evergreen |
| # | 11:37:27 | james | Hi |
| # | 11:37:31 | james | anybody here? |
| # | 11:37:41 | phasefx | james: hiya |
| # | 11:37:45 | james is now known as Guest56672 |
| # | 11:38:00 | Guest56672 | so anybody here? |
| # | 11:38:45 | phasefx | just us and some grues |
| # | 11:39:53 | moodaepo | Guest56672: About 59 folks are here including bots and you. |
| # | 11:40:10 | Guest56672 | thanks for the reply |
| # | 11:40:48 | Guest56672 | i just posted an article about the Connecticut library migration at a forum thought it was a great example of the bazaar from the book of similar title |
| # | 11:41:59 | Guest56672 | to put things pithily, i'm a lifetime teacher in high schools in Missouri, full certifications in all sciences etc. bs ms not piled higher and deeper but also am a published author |
| # | 11:42:09 | Guest56672 | spent a lot of time writing instruction manuals |
| # | 11:42:15 | Meliss has joined #evergreen |
| # | 11:42:19 | Guest56672 | and thought i would volunteer to do so for you folks |
| # | 11:43:01 | dbs | phasefx: cool, so someone could cherry-pick 310db79e725ed539 from that branch if they want the i18n bit first |
| # | 11:43:38 | dbs | Guest56672: awesome! have you heard of the Evergreen Documentation Interest Group? |
| # | 11:43:57 | Guest56672 | no i looked at the site but don't remember that particularly |
| # | 11:44:04 | Guest56672 | if that is it then i could contact someone |
| # | 11:44:20 | dbs | there's a quick overview at http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:dig |
| # | 11:44:36 | phasefx | dbs++ # collab |
| # | 11:44:39 | dbs | the DIG people are very friendly |
| # | 11:45:20 | Guest56672 | aaahhh found it now, i was lookin under the wrong drop down. I'll noodle around there and if i need to will drop back, thanks very much for the quick reply |
| # | 11:46:24 | Guest56672 has quit IRC |
| # | 11:57:26 | phasefx | asset.call_number_class.normalizer populates asset.call_number.label_sortkey? |
| # | 11:59:03 | dbs | phasefx: it should |
| # | 11:59:09 | phasefx | gracias |
| # | 11:59:41 | dbs has seen cases where LC was applied to CODOC with result = NULL |
| # | 11:59:41 | phasefx | none of those normalizers probably handle labels like Volume 43 No. 11 (December 6, 2010) I bet |
| # | 12:00:05 | dbs | heh, well they "handle" them, just not the way people might expect |
| # | 12:00:18 | phasefx | stuff like that should get handled by serials functionality, I imagine |
| # | 12:00:25 | dbs | absolutely |
| # | 12:00:41 | phasefx | something I'm mostly ignorant of. :) thanks |
| # | 12:01:06 | dbs | Of course, someone could write a call number normalizer for that format if they were so inclined |
| # | 12:02:23 | dbs isn't very inclined |
| # | 12:02:38 | phasefx hears ya |
| # | 12:02:45 | berick | I know I'm late to the party, but where does ARRAY_ACUM come from? e.g. ARRAY_ACUM(evergreen.upgrade_list_applied_deprecates(my_db_patch)) |
| # | 12:02:54 | berick | or is it a typo? |
| # | 12:03:38 | dbs | That would be a typo |
| # | 12:04:05 | dbs | It should be spelled "array_agg()" in any case :) |
| # | 12:04:51 | berick does not know if that's a joke ;) |
| # | 12:05:00 | berick | i need to repair 002.schema.config.sql.. |
| # | 12:05:53 | dbs | berick: I have a branch to stamp out all use of ARRAY_ACCUM() in favour of the postgresql native ARRAY_AGG() function |
| # | 12:06:35 | dbs | and tsbere did me the favour of creating an unwrapped upgrade script for master: remotes/working/user/tsbere@mvlc.org/native-array-functions |
| # | 12:06:39 | berick | dbs: gotcha. OK if I fix that one typo for now so trunk will install? |
| # | 12:07:01 | dbs | I suggest fixing it as "array_agg()" |
| # | 12:07:10 | berick | that was my next question. thanks! |
| # | 12:07:28 | dbs | thank you, a working trunk is a good trunk |
| # | 12:19:06 | jamesrf has quit IRC |
| # | 12:20:52 | jamesrf has joined #evergreen |
| # | 12:25:05 | lisppaste | berick pasted "ARRAY_AGG error" at http://paste.lisp.org/display/122390 |
| # | 12:25:30 | berick | dbs: mean anything to you? ^-- |
| # | 12:25:53 | tsbere | berick: Means something to me! |
| # | 12:26:14 | dbs | It means pain? |
| # | 12:26:38 | berick | :) |
| # | 12:26:45 | tsbere | SELECT array_agg(var) FROM evergreen.upgrade_list_applied_deprecates(my_db_patch); with var being whatever it needs to be based on the function return |
| # | 12:29:49 | berick | hmm |
| # | 12:29:56 | dbs | looks painful in that context |
| # | 12:29:59 | berick | any suggestions on how that might look toward the bottom of http://git.evergreen-ils.org/?p=Evergreen.git;a=blob;f=Open-ILS/src/sql/Pg/002.schema.config.sql;h=822623b92761facfa01036414a5cc2988cbf99f5;hb=HEAD#l869 |
| # | 12:30:49 | dbs | can RAISE accept subselects? heh |
| # | 12:31:28 | berick | hm, maybe, or I could select into a variable |
| # | 12:33:26 | yboston has quit IRC |
| # | 12:33:35 | dbs | Or you could just use ARRAY_ACCUM() and make it work for now. |
| # | 12:33:48 | berick | dbs: tried that. same error |
| # | 12:33:54 | dbs | oh, hah |
| # | 12:34:04 | dbs | YOU CAN'T GET THERE FROM HERE |
| # | 12:34:06 | berick | probably this branch of code has never been run.. |
| # | 12:34:21 | dbs | looks like it |
| # | 12:38:35 | dbs doesn't know how invoke that branch |
| # | 12:40:07 | berick | dbs: SELECT evergreen.upgrade_deps_block_check('XXXX', NULL); |
| # | 12:40:22 | dbs will give it a shot |
| # | 12:46:08 | dbs tries applying the updated 0526.schema.upgrade-dep-tracking.sql and gets ERROR: relation "db_patch_dependencies" already exists |
| # | 12:49:47 | mrpeters-isl | phasefx: 2 second xulrunner question |
| # | 12:50:09 | mrpeters-isl | "Restore Defaults" in the column picker should restore back to the default columns, right? or does it do something else? |
| # | 13:13:43 | dbs | berick: I think I've got something. feels nuts but it doesn't throw an error, so there's that |
| # | 13:16:25 | youdonotexist has joined #evergreen |
| # | 13:17:58 | jenny1 has joined #evergreen |
| # | 13:19:11 | phasefx has quit IRC |
| # | 13:19:53 | phasefx has joined #evergreen |
| # | 13:20:14 | jenny has quit IRC |
| # | 13:20:27 | bshum | Hmm, it looks like when an alert presents itself due to an entry in the messages area of the 2.0 patron interface, it takes away the coloring around the name of the patron. (or turns it into the same gray in the background). |
| # | 13:20:46 | bshum | I'm testing this behavior a bit more |
| # | 13:21:57 | bshum | Yeah, applying a standing penalty/message apparently takes away the color box around the patron name indicating their status :( |
| # | 13:22:13 | dbs | berick, check user/dbs/fix-dep-tracking |
| # | 13:22:13 | tsbere | mrpeters-isl: You are looking at the "they show an ID and not anything useful" launchpad bug? |
| # | 13:22:25 | mrpeters-isl | trying!!! |
| # | 13:22:39 | mrpeters-isl | im to the point where i want to test 'FM_PGT_RETRIEVE' : { 'app' : 'open-ils.actor', 'method' : 'open-ils.actor.groups.tree.retrieve', 'secure' : false }, |
| # | 13:22:43 | mrpeters-isl | with srfsh |
| # | 13:22:46 | mrpeters-isl | to see what i get back |
| # | 13:23:35 | mrpeters-isl | in util.js -- { return data.hash.pgt[ my.au.profile() ].name(); } |
| # | 13:23:58 | mrpeters-isl | so i'm thinking that takes au.profile and links it to a permission.grp_tree.name |
| # | 13:24:14 | tsbere | Some of the things that show up with that issue include addresses and such.....that exist as a set of fields further down the list. May want to double-check some of that. |
| # | 13:24:33 | mrpeters-isl | i noticed addresses came up as ID's but i wasnt sure if that was expected or not |
| # | 13:24:40 | mrpeters-isl | i kind of it recall always being that way |
| # | 13:25:29 | bshum | Oh, heh, maybe only horizontally. Have to retest with vertical summary display too. |
| # | 13:25:42 | mrpeters-isl | to be entirely honest, i'm not entirely sure where to go next |
| # | 13:26:47 | tsbere | bshum: Applying a penalty may indicate a new color to begin with? |
| # | 13:26:57 | mrpeters-isl | "Home Library" gives useful info, instead of an "id" but it doesnt look that different --- { return data.hash.aou[ my.au.home_ou() ].shortname(); } |
| # | 13:27:13 | mrpeters-isl | so, actor.usr.home_ou = actor.org_unit.shortname i guess |
| # | 13:27:38 | bshum | tsbere: I'm checking that out next. The "box" surrounding the name of the patron seems to go away and the name becomes flush to the upper left of the patron display instead of being centered inside the box. Which leads me to assume that the box is somehow gone. |
| # | 13:28:10 | tsbere | mrpeters-isl: The data.hash.aou part is "a pre-cached list of org units by id populated at login" |
| # | 13:28:46 | mrpeters-isl | ok, so it "works" in the search results becuase its populated when you pull down the org tree for WS registration? |
| # | 13:29:13 | tsbere | It "works" there because it pulls the org tree down at login no matter what, because having the org tree locally saves a lot of potential lookups. |
| # | 13:29:42 | mrpeters-isl | ok, so we probably need to pull down permisson.grp_tree and config.ident_type data too at login/ |
| # | 13:29:58 | mrpeters-isl | config.identification_type, rather |
| # | 13:30:08 | tsbere | They might exist somewhere alerady. Or you can add them to the flesh list. |
| # | 13:31:26 | mrpeters-isl | hmm where would i find the "flesh" list |
| # | 13:32:03 | tsbere | I am unsure |
| # | 13:32:12 | tsbere isn't looking at that |
| # | 13:32:16 | mrpeters-isl | 10-4 |
| # | 13:32:34 | tsbere has too many post-migration headaches, and is avoiding digging through code |
| # | 13:32:41 | mrpeters-isl | ill put this into the bug in hopes maybe someone more experienced can pick up from here later on |
| # | 13:33:40 | youdonotexist has quit IRC |
| # | 13:34:09 | tsbere | mrpeters-isl: A hint: See how it is loading the address and barcode. You want to duplicate *them* |
| # | 13:34:54 | mrpeters-isl | aha - s = patron.util.retrieve_fleshed_au_via_id(ses(),s,["card"]); |
| # | 13:35:00 | mrpeters-isl | i assume that's something in the "flesh" list |
| # | 13:36:47 | tsbere | if you replace ["card"] with, say, ["card","ident_type"] you should find the type is no longer a value, but an object. With a name() attached. |
| # | 13:37:18 | phasefx | mrpeters-isl: also, the whole data.hash thing uses fieldmapper hint/class names by convention. There should be a hash.pgt and a hash.cit |
| # | 13:37:49 | tsbere | phasefx: Assuming either of those is cached. Are they? |
| # | 13:38:03 | phasefx | yeah, pretty sure |
| # | 13:38:39 | phasefx | yeah, pgt, cit, citm, ccs, aou, aout, crahp |
| # | 13:38:46 | phasefx | I love that last one |
| # | 13:38:51 | mrpeters-isl | lol |
| # | 13:38:52 | tsbere | mrpeters-isl: If they are cached, don't flesh them. |
| # | 13:39:18 | mrpeters-isl | ok - i'm not sure i fillow the difference |
| # | 13:39:24 | mrpeters-isl | cached means they're pulled down at login? |
| # | 13:39:48 | tsbere | On the other hand, you need to use the proper things for them. hash.cit[ my.au.item_type() ].name() for example? |
| # | 13:39:56 | tsbere | ident_type, not item_type |
| # | 13:39:58 | tsbere | <_< |
| # | 13:40:23 | mrpeters-isl | 'primary' : false, 'hidden' : true, 'editable' : false, 'render' : function(my) { return data.hash.cit[ my.au.ident_type() ].name(); } |
| # | 13:40:30 | mrpeters-isl | looks like it is using the proper thing |
| # | 13:41:15 | mrpeters-isl | im trying to compare this to barcode....but honestly it looks the same to me |
| # | 13:41:45 | phasefx | mrpeters-isl: it's more defensive to do something like return typeof my.au.ident_type() == 'object' ? my.au.ident_type().name() : data.hash.cit[ my.au.ident_type() ].name(); |
| # | 13:41:49 | jenny has joined #evergreen |
| # | 13:41:56 | phasefx | in case the data is already fleshed |
| # | 13:42:06 | Rural has joined #evergreen |
| # | 13:43:38 | phasefx | patron search, incidentally, no longer uses the column definitions from a util.js file, but uses the fm_columns method from list.js, which uses fm_IDL.xml |
| # | 13:43:59 | youdonotexist has joined #evergreen |
| # | 13:44:17 | jenny1 has quit IRC |
| # | 13:44:53 | mrpeters-isl | ok. this may be getting a bit beyond my skillset. |
| # | 13:45:46 | phasefx | though such definitions can be overridden on a per column basis |
| # | 13:48:07 | mrpeters-isl | i see a fm_columns for aua, ac, aua again, and au |
| # | 13:48:19 | mrpeters-isl | do we need similar for pgt and cid? |
| # | 13:49:40 | berick | dbs++ # fix-dep-tracking works on my dev db |
| # | 13:49:53 | phasefx | mrpeters-isl: that's one way to do (and then in 'retrieve_row' you would have to add row.my.pgt and row.my.cit as data being fed to the list), but it could also be overkill since by default you would have a column for every field from those objects |
| # | 13:50:07 | dbs | AFAICT, Cookies.js is no longer used by anything. I'm going to delete it and remove all calls to Cookie.js |
| # | 13:50:16 | dbs | berick: yay, please sign-off :) |
| # | 13:50:22 | berick | dbs: will do |
| # | 13:51:00 | dbs | hmm - not sure if we need to explicitly DROP TYPE evergreen_patch; outside the transaction to support repeated --create-schemas actually |
| # | 13:52:55 | berick | dbs: perhaps creaete the type in the evergreen schema? |
| # | 13:53:01 | berick | so it will get dropped automatically |
| # | 13:53:58 | dbs | berick: yes, good plan |
| # | 13:54:02 | mrpeters-isl tries |
| # | 13:57:18 | phasefx | mrpeters-isl: what would be cool, is to extend fm_columns (in list.js) around where it determines that my_field.datatype == 'link', and have it look for a data.hash to use if the object isn't fleshed. That'd probably take some thought |
| # | 13:59:36 | phasefx tries it real quick |
| # | 14:03:13 | phasefx | doesn't fieldmapper have a way of indicating a default or preferred "display" field for a given class? |
| # | 14:04:37 | tsbere | The reporter has a "lookup via this instead of the primary key in display interfaces" setting. I think dojo interfaces know how to use it. |
| # | 14:07:11 | mrpeters-isl | ok, i think i'm totally whiffing on this |
| # | 14:07:11 | mrpeters-isl | http://pastie.org/1999377 |
| # | 14:07:42 | mrpeters-isl | and then i added row.my.pgt = row.my.au.profile(); to "retrieve_row" |
| # | 14:07:52 | mrpeters-isl | but i still get an id back for Main (Permission) Profile Group |
| # | 14:08:34 | phasefx | mrpeters-isl: the dataobj field, you can remove it. That would tell fm_columns to look for my.au_profile instead of my.pgt |
| # | 14:09:04 | phasefx | useful for when have multiple fields of the same fieldmapper class, like mailing address and billing address |
| # | 14:09:47 | mrpeters-isl | and, am i correct that the "hidden" is for display in staff client, by default? |
| # | 14:10:17 | phasefx | mrpeters-isl: also, for row.my.pgt, you want to pass a fleshed object. So something like row.my.pgt = typeof row.my.au.profile() == 'object' ? row.my.au.profile() : obj.data.hash.pgt[ row.my.au.profile() ]; |
| # | 14:11:20 | phasefx | mrpeters-isl: right, but in this case, you're unhiding every field for the pgt class. It's best to hide them all, and then override a specific one to be unhidden |
| # | 14:11:31 | mrpeters-isl | right |
| # | 14:11:39 | mrpeters-isl | ok, so i end up with an empty "Group Name" column |
| # | 14:11:57 | mrpeters-isl | pretty sure that's the "correct" one to use now, instead of main (profile) permission group |
| # | 14:12:08 | phasefx | e.g. '*' : { 'hidden' : true }, 'pgt_name' : { 'hidden' : false } |
| # | 14:12:29 | mrpeters-isl | ah cool |
| # | 14:13:07 | phasefx | the fm_columns for 'ac' being a good template |
| # | 14:14:01 | phasefx | mrpeters-isl: and it looks like that file is older than the obj.data convention, it's obj.OpenILS.data |
| # | 14:14:24 | phasefx | for your row.my.pgt line |
| # | 14:16:21 | moodaepo | dbs: Regarding the TT hackfest @ Windsor any chance it could be attended on IRC with an audio/video feed online? |
| # | 14:18:44 | mrpeters-isl | -= THIS MESSAGE NOT LOGGED =- |
| # | 14:18:52 | mrpeters-isl | pasting search_result.js diff now |
| # | 14:20:31 | dbs | moodaepo: IRC, for sure; dunno how much audio/video will help though |
| # | 14:21:41 | phasefx | mrpeters-isl: if you have spidermonkey-bin installed, could run js search_result.js on the command-line and find any syntax errors |
| # | 14:21:58 | mrpeters-isl | ok |
| # | 14:26:32 | mrpeters-isl | it works! |
| # | 14:26:47 | phasefx | rock |
| # | 14:26:48 | mrpeters-isl | i assume i can rinse and repeat for the ident type |
| # | 14:27:07 | phasefx | mrpeters-isl: not quite |
| # | 14:27:41 | phasefx | there are two ident types, so you'll need to mimic what billing address and mailing address are doing with their column definitions, using a specific dataobj |
| # | 14:28:26 | phasefx | there's ident_type and ident_type2 |
| # | 14:28:35 | mrpeters-isl | ok |
| # | 14:30:21 | mrpeters-isl | so, 'dataobj' : 'ident_au' and 'dataobj' : 'ident2_au' perhaps? |
| # | 14:31:03 | mrpeters-isl | assuming the dataobj can be called anything? |
| # | 14:34:02 | mrpeters-isl | another important thing will be to remove the "Main (Permission) Profile" column from the picker, somehow |
| # | 14:34:06 | phasefx | mrpeters-isl: yeah, they can be called anything. I'd go with ident_type and ident_type2, and use row.my.ident_type and row.my_ident_type2 |
| # | 14:35:05 | phasefx | mrpeters-isl: use 'remove_me' : true in the definition |
| # | 14:35:15 | phasefx | in the place where you would normally put 'hidden' : true |
| # | 14:36:21 | mrpeters-isl | row.my.cit = typeof row.my.au.ident_type() == 'object' ? row.my.au.ident_type() : obj.OpenILS.data.hash.pgt[ row.my.au.ident_type() ]; correcT? |
| # | 14:36:32 | mrpeters-isl | err row.my.cit = typeof row.my.au.ident_type() == 'object' ? row.my.au.ident_type() : obj.OpenILS.data.hash.cit[ row.my.au.ident_type() ]; |
| # | 14:37:50 | phasefx | instead of "row.my.cit", you need to use "row.my.foo", where foo is whatever you specified as the dataobj |
| # | 14:38:00 | mrpeters-isl | aha |
| # | 14:38:10 | phasefx | assuming you want to flesh both ident_type and ident_type2 |
| # | 14:40:10 | mrpeters-isl | i think i need to prefix ident_type2 differently too, right? it loosk like its using the same heading as ident_type |
| # | 14:40:36 | mrpeters-isl | 'label_prefix' : $('patronStrings').getString('staff.patron.search_result.ident_type2_column_label_prefix'), perhaps? |
| # | 14:44:04 | phasefx | try it :) |
| # | 14:44:50 | collum has quit IRC |
| # | 14:45:48 | Rural | Well... I've munged the OpenSRF install on our test Evergreen install. It passed the tests at the end of the OpenSRF part of the installation document on the wiki. |
| # | 14:45:53 | mrpeters-isl | ok, i think i have this knocked out except hiding these columns |
| # | 14:46:44 | Rural | Now, using 'request opensrf.math add 2 2' gets me "Received no data from the server." |
| # | 14:46:51 | mrpeters-isl | any hings on where i'm looking to do "hidden : true" ? |
| # | 14:49:15 | Rural | My srfsh.log is complaining that there is no one to send the message to. |
| # | 14:49:34 | dbs | Rural: did you by any chance shut down the server without shutting down opensrf first? |
| # | 14:49:57 | Rural | Yes. There is a very good chance of that. |
| # | 14:50:03 | phasefx | mrpeters-isl: in the object for the '*' key |
| # | 14:50:26 | dbs | Rural: if you're running OpenSRF 2.0.0, it will recover nicely from that |
| # | 14:50:49 | dbs | if not, 'rm /openils/var/run/*.pid' and restart opensrf services will probably do the trick |
| # | 14:51:13 | Rural | dbs: Ok. Here goes. |
| # | 14:52:06 | mrpeters-isl | phasefx: still in search_result.js? |
| # | 14:52:21 | phasefx | paste your diff |
| # | 14:52:28 | mrpeters-isl | ok |
| # | 14:53:54 | Rural | dbs: Woo-hoo! Yup, that's all it was. Makes sense too. |
| # | 14:54:01 | dbs | yay |
| # | 14:55:48 | phasefx | weird, my patron search can't find the stock admin user, but can find newly registered users |
| # | 14:56:17 | Rural | Ok. Now onto the staff client. |
| # | 14:56:26 | mrpeters-isl | phasefx: http://pastie.org/1999594 |
| # | 14:56:37 | mrpeters-isl | hoping i did that right....first "big" patch with git |
| # | 14:57:29 | mrpeters-isl | ew, something nasty happened in there |
| # | 14:57:34 | mrpeters-isl | hang on |
| # | 14:57:48 | kmlussier has quit IRC |
| # | 14:58:54 | mrpeters-isl is really sucking at git |
| # | 15:00:20 | phasefx | mrpeters-isl: I don't see anything obviously wrong/bad. What do you want to do now again? |
| # | 15:00:51 | mrpeters-isl | line 45 thru 52 are bad |
| # | 15:01:09 | mrpeters-isl | i accidentally got 2 commits in there |
| # | 15:01:17 | mrpeters-isl | 1 was a bad old one....i forgot to go back to a clean branch |
| # | 15:01:31 | mrpeters-isl | what i need to do is hide the columns that only display id values |
| # | 15:01:56 | mrpeters-isl | Main (Permission) Profile, Primary Identification, Secondary Identification |
| # | 15:02:27 | phasefx | in the fm_columns for 'au', add 'au_profile' : { 'hidden' : true }, 'au_ident_type' : { 'hidden' : true }, etc. |
| # | 15:02:40 | phasefx | if you want to remove them from the column picker altogether, use remove_me instead of hidden |
| # | 15:03:01 | mrpeters-isl | yikes...ok i dont know why i was making it so hard |
| # | 15:03:02 | mrpeters-isl | thanks |
| # | 15:04:08 | dbs | mrpeters-isl: you also need to set your actual name in git config |
| # | 15:04:17 | mrpeters-isl | dbs: yep saw that |
| # | 15:13:51 | phasefx | one bad thing about typeof foo == 'object'.. it catches nulls. I keep forgetting about that |
| # | 15:15:11 | mrpeters-isl | patch posted |
| # | 15:15:14 | mrpeters-isl | hope it fixes things |
| # | 15:16:38 | dbs | heh, maybe good we didn't go with fossil: http://sheddingbikes.com/posts/1306005291.html (although I love his take on git) |
| # | 15:17:13 | drdata_home_ has joined #evergreen |
| # | 15:18:02 | drdata_home has quit IRC |
| # | 15:18:09 | drdata_home_ is now known as drdata_home |
| # | 15:18:53 | Rural | xulrunner is barfing on 'xulrunner ~/Evergreen-ILS-2.0.6/Open-ILS/xul/staff_client/application.ini'. Pops up a window which complains about an XML parsing error. Any ideas? |
| # | 15:20:06 | Rural | I am using an Ubuntu 11.04 machine for the staff client. Maybe that is contributing to the problem. |
| # | 15:24:27 | phasefx | Rural: wrong application.ini |
| # | 15:24:34 | phasefx | use build/application.ini |
| # | 15:26:02 | Rural | phasefx: Bah! ;) |
| # | 15:26:51 | phasefx | used to work back in the day, but we kept making things more complicated :) |
| # | 15:28:03 | phasefx | fun thing, if we do move to all local xul, we can get rid of that stamp id business |
| # | 15:42:44 | phasefx | there's a collab/phasefx/fm_columns that integrates OpenILS.data.hash with util.list.fm_columns |
| # | 15:54:26 | Rural | So I'm guessing that trying to do your first login from another mausing the staff client is best done on the same machine as the server is running on. I'm having no luck from my desktop to the server running on a VM. |
| # | 15:54:38 | Rural | mausing -> machine |
| # | 15:59:38 | Meliss has quit IRC |
| # | 16:01:35 | phasefx | Rural: are you using NAT for the vm's networking? that would stop you |
| # | 16:01:57 | sfortin has quit IRC |
| # | 16:03:09 | Rural | phasefx: Ya. The VM is on a virtual bridge. I guess I could create another VM on the same network. |
| # | 16:08:22 | berick | can anyone suggest a reasonable value for the "cat.marc_control_number_identifier" org setting? |
| # | 16:10:03 | berick was pleasantly surprised to see the staff client telling him to set it |
| # | 16:10:34 | dbs | berick: ESILIBRARY would be fine |
| # | 16:11:06 | berick | dbs: gotcha, thanks! |
| # | 16:11:56 | dbs | Rural: can you see http://hostname/xul in a web browser? |
| # | 16:12:30 | Rural | dbs: I can. |
| # | 16:13:02 | dbs | Rural: two possible problems: 1) you need to add the SSL certificate (near login button) |
| # | 16:13:15 | dbs | 2) you need to restart Apache, it needs to be started after OpenSRF |
| # | 16:17:50 | Rural | I added the SSL exception and restarted Apache. I get a "Login failed." |
| # | 16:18:45 | dbs | that's progress |
| # | 16:18:47 | Rural | Actually, using srfsh, the same credentials fail. Hmmm... |
| # | 16:18:53 | dbs | what credentials? |
| # | 16:19:11 | dbs | were they the --admin-user --admin-pass values you set with eg_db_config.pl? |
| # | 16:19:59 | Rural | Yes. |
| # | 16:20:22 | tsbere should probably be avoiding anything work related, such as IRC, given the morning he had |
| # | 16:20:23 | Rural | One second. There is a possible discrepency in the password. |
| # | 16:20:33 | tsbere pokes berick about sip_statcats2 anyway |
| # | 16:21:44 | Rural | Ok. I've confirmed the admin password using srfsh. The same credential fails on the staff client. |
| # | 16:26:15 | Rural | It bothers me that the staff client has a message about the workstation not being configured. |
| # | 16:27:22 | mrpeters-isl | that is normal |
| # | 16:27:29 | mrpeters-isl | you have to register the workstation to one of the branches |
| # | 16:27:50 | mrpeters-isl | http://www.open-ils.org/dokuwiki/doku.php?id=evergreen-user:registering_staff_client_to_different_working_locations |
| # | 16:33:31 | Rural | I guess I could just install a desktop environment onto the server and go from there. |
| # | 16:36:35 | dbs | Rural: I doubt that you'll get anywhere, if you can connect but not login there's some other problem |
| # | 16:36:51 | dbs | Maybe HTTPS connectivity isn't allowed |
| # | 16:43:22 | Rural | I don't think it's that. I can point my web-browser at the Evergreen server and it is serving up HTTPS content. |
| # | 16:44:05 | Dyrcona | Rural: try logging into the opac with the account you tried from the staff client. |
| # | 16:44:46 | Rural | opac? |
| # | 16:47:02 | Dyrcona | that's what you get when you go to https://your-server-name |
| # | 17:00:13 | Rural | Dyrcona: Ah! Ya, clicking on the "Log in" link gets me...nothing at all. I'm playing with my Javascript settings now. |
| # | 17:00:48 | dbs | Rural: JavaScript Console would actually be helpful. |
| # | 17:01:04 | dbs | No "playing with JavaScript settings" should be required |
| # | 17:01:14 | dbs | oh. Did you run "autogen.sh -u"? |
| # | 17:02:42 | Rural | dbs: Maybe not. I'll check. |
| # | 17:05:34 | Rural | dbs: Yes. I did. Step 3 under "Starting Evergreen", right? |
| # | 17:06:09 | Rural | dbs: On http://open-ils.org/dokuwiki/doku.php?id=server:2.0:install |
| # | 17:06:29 | dbs | Sounds like it |
| # | 17:06:55 | dbs adds "pullrequest" as an official tag to bugs.launchpad.net/evergreen |
| # | 17:09:53 | phasefx | dbs++ |
| # | 17:16:39 | dbs | https://bugs.launchpad.net/evergreen/+bugs?field.tag=pullrequest is starting to fill up, now if only phasefx and I could find somebody to signoff / commit these bits... :) |
| # | 17:17:29 | bjwebb has joined #evergreen |
| # | 17:17:29 | bjwebb has joined #evergreen |
| # | 17:19:19 | dbs | hey bjwebb - how are those exams going? |
| # | 17:19:43 | bjwebb | dbs: slowly |
| # | 17:20:14 | bjwebb | dbs: they've been fairly okay so far, but i have 3 on consecutive days coming up |
| # | 17:20:45 | dbs | bjwebb: well, nice to know that you're still thinking of us |
| # | 17:21:14 | dbs | btw - you pointed me at a branch with Makefile changes - is that ready for merging, in your opinion? or still a work in progress? |
| # | 17:23:42 | bjwebb | i'm not sure, i've not had chance to test it a lot, but it should be okay |
| # | 17:26:53 | dbs | Okay. I'll try to give it a slight workout |
| # | 17:28:29 | Dyrcona has quit IRC |
| # | 17:29:08 | jenny has left #evergreen |
| # | 17:33:42 | dbs | berick: I pushed another commit into user/dbs/fix-dep-tracking to move the TYPE to evergreen.patch and made the upgrade script work a bit more aggressively. |
| # | 17:50:17 | dbs has quit IRC |
| # | 17:50:38 | dbwells | grabbing 0542 |
| # | 17:51:11 | phasefx | dbwells: want to sign off on a collab branch which uses 0542? :-) |
| # | 17:51:18 | Rural | dbs was correct. I get the same issues when running a browser on the server. |
| # | 17:52:26 | dbwells | phasefx: I'll certainly take a look at least. Where is it? |
| # | 17:52:32 | phasefx | dbwells: collab/phasefx/missing_cat_perms |
| # | 17:52:55 | Rural | Where can I find the passwords in the database? (It's been a while since I did any serious work with Postgres.) |
| # | 17:53:35 | phasefx | dbwells: I can't tell if I doing upgrade scripts correctly anymore. I certainly don't have the knack of applying them and having them work |
| # | 18:04:29 | phasefx | dbwells: if need be, I can just bump the version later (or just go ahead and push it and let this dual-sign-off thing happen more slowly). don't mean to derail you |
| # | 18:06:37 | tsbere advocates pushing *unwrapped* upgrade scripts, with the committer to master or whatever wrapping them with a version # |
| # | 18:07:32 | phasefx | that's what I do within Equinox, but I was just going to push this, and doing collab instead was an after-thought |
| # | 18:08:29 | phasefx | or, rather, instead of unwrapped, I do something like upgrade/xxx.foo.sql, and 'test' as a version. I haven't used the make db patch thing yet |
| # | 18:08:33 | Dyrcona has joined #evergreen |
| # | 18:09:10 | shadowspar has quit IRC |
| # | 18:09:17 | shadowspar has joined #evergreen |
| # | 18:15:14 | phasefx is bowing out for a while |
| # | 18:15:20 | dbwells | phasefx: I will hold off on my commit and try to figure out signing-off on yours early tomorrow. I have no clue what the workflow is supposed to be. |
| # | 18:16:10 | phasefx | if someone offers a good one, I'll wiki it |
| # | 18:47:50 | Rural has quit IRC |
| # | 18:48:54 | mrpeters-isl has quit IRC |
| # | 19:22:22 | youdonotexist has quit IRC |
| # | 19:27:57 | b_bonner has left #evergreen |
| # | 20:21:28 | finnx has joined #evergreen |
| # | 21:05:33 | Dyrcona has quit IRC |
| # | 21:59:00 | dbs has joined #evergreen |
| # | 21:59:01 | dbs has joined #evergreen |
| # | 22:38:01 | finnx has quit IRC |
| # | 22:55:20 | adbowling-isl has quit IRC |
| # | 23:26:35 | jamesrf has quit IRC |
| # | 23:33:47 | jamesrf has joined #evergreen |
April 10-13