| # | Time | Nick | Message |
|---|
| # | 00:23:36 | denials grabs 0535 for the authority indexes |
| # | 00:51:33 | denials likes 'git rebase -i' with 'fixup' |
| # | 02:24:15 | tater-laptop has quit IRC |
| # | 02:34:12 | tater-laptop has joined #evergreen |
| # | 02:55:01 | tater-laptop has quit IRC |
| # | 03:31:43 | bjwebb | gah fire drills |
| # | 04:10:52 | bjwebb has quit IRC |
| # | 07:29:34 | Sally has joined #evergreen |
| # | 08:08:07 | sylvar | at 3:3o? ouch |
| # | 08:17:40 | denials | sylvar: adjust for UK timezone |
| # | 08:18:05 | tsbere has known places to do fire drills in the middle of the night intentionally, though |
| # | 08:21:07 | Dyrcona has joined #evergreen |
| # | 08:26:34 | sylvar | Ah, that's not too bad then. |
| # | 08:28:37 | tsbere | sylvar: I have a friend who has horror stories of college dorm fire drills. At 2 am. |
| # | 08:28:55 | sylvar | tsbere: now you have two |
| # | 08:29:07 | sylvar | rah rah alma mater sis boom. |
| # | 08:29:09 | sylvar | bah. |
| # | 08:30:04 | tsbere | Apparently they felt that during the day when the dorm was empty wasn't a good drill? |
| # | 08:30:52 | Sally has quit IRC |
| # | 08:31:00 | sylvar | Frankly, the shenanigans that cause real fires are more likely at 2am. I knew some guys from a dorm floor that was diasporated (the ones who weren't evicted) after a dorm-furniture bonfire in the lounge. |
| # | 08:32:05 | sylvar | Late-night boredom. Some folks write a replacement Minix kernel, some people set fire to stuff. |
| # | 08:32:55 | gmcharlt | @quote add <sylvar> Late-night boredom. Some folks write a replacement Minix kernel, some people set fire to stuff. |
| # | 08:32:55 | pinesol_green | gmcharlt: The operation succeeded. Quote #9 added. |
| # | 08:33:35 | tsbere | Heh. That reminds me of a different horror story from my friend. 2:30 or so the night before the start of finals and someone's candle set their desk on fire. |
| # | 08:33:58 | ernieSimuro has joined #evergreen |
| # | 08:39:37 | sylvar | Here, before the caffeine kicks in, let me give you a topic: A hypothetical fault-tolerant (specifically, "Server? WHAT server?"-tolerant) EG staff client loses its connection. It switches to offline mode and alerts the staff member (different background, popup, something). When the server comes back, it uploads the offline file, attempts to auto-process it, and tells the staff member about exceptions. |
| # | 08:39:57 | sylvar | scuse |
| # | 08:39:58 | sylvar | It switches to offline mode and alerts the staff member (different background, popup, something). When the server comes back, it uploads the offline file, attempts to auto-process it, and tells the staff member about exceptions. |
| # | 08:41:13 | tsbere | sylvar: For various reasons, I feel that switching into and out of offline mode should be more manual. Also, if the issue is with just that computer they may want to stop using it. |
| # | 08:41:47 | tsbere | I would personally dislike a computer auto-accepting transactions for offline mode when, say, the network died because the memory is corrupting itself. |
| # | 08:43:15 | sylvar | Good idea. Could we usually test for that? (Server not responding to connections or pings. Traceroute dies one hop away. Offline! // Server not responding to connections or pings. Can't ping 127.0.0.1. Fail loud!) |
| # | 08:44:21 | sylvar | My inspiration for this: I was driving to work, listening to Pandora on my Chromebook over 3G. I get to the office, the Chromebook joins the wifi network, I manually turn off 3G data, and Pandora doesn't miss a beat. |
| # | 08:44:25 | sylvar | Literally. |
| # | 08:44:50 | tsbere | In theory, the staff client would act in a similar manner. |
| # | 08:44:50 | sylvar | SON, I AM IMPRESS |
| # | 08:45:17 | tsbere | Change networks and it should pick up right where it left off unless you were in the middle of something significant. |
| # | 08:45:33 | sylvar | I like that. |
| # | 08:45:34 | tsbere | and by singificant I mean "somehow uploading or downloading a few hundred megs" |
| # | 08:46:25 | asimon has joined #evergreen |
| # | 08:46:55 | asimon | Hello. I am attempting to load patrons with family group information. Is usr_grp_map the only table that pertains to family groups? |
| # | 08:47:34 | gmcharlt | yes |
| # | 08:47:43 | tsbere | Isn't that permission groups? |
| # | 08:47:54 | tsbere | Or are there multiple tables named that way in different schemas? |
| # | 08:48:49 | gmcharlt | asimon: pardon me - actor.usr.usrgroup |
| # | 08:49:49 | gmcharlt | patrons that have the same usrgroup value are in the same family group |
| # | 08:50:20 | gmcharlt | there's also a hack where a juvenile patron's ident_value2 can be interpreted as containing the name of a guardian, but I'd avoid that |
| # | 08:50:45 | sylvar | tsbere: I was gonna joke about a few hundred megs of config tables on startup, but I only measured about 630kB with wireshark |
| # | 08:50:54 | sylvar | those few seconds last forever |
| # | 08:52:31 | denials | sylvar: play swirling romantic music at staff client startup; make the moment last |
| # | 08:52:56 | sylvar | denials: hire Brian Eno for an Evergreen Sound |
| # | 08:53:24 | csharp | sylvar: the "auto-switch" to offline would be cool, but only for libraries that opt to use it. Many of our libraries will not touch standalone and trust the ol' "scan into notepad/wordpad/Word" procedure *shudder* |
| # | 08:53:34 | asimon | gmcharlt: So actor.usr.usrgroup contains a unique group value for all the members of a group, but that value is not reflected anywhere else in the schema? |
| # | 08:54:04 | gmcharlt | correct |
| # | 08:54:29 | gmcharlt | also, the lead member of a group should have actor.usr.master_account set to true |
| # | 08:56:09 | sylvar | csharp: As long as they don't have auto-spellcheck enabled... "Hey, why are our patrons all named ZIOIS something?" "Uh... I think that's 21015." |
| # | 08:56:28 | csharp | heh |
| # | 08:57:27 | csharp prefers his library card number font to be Comic Sans |
| # | 08:57:28 | sylvar | and now you can guess where I pick up my PINES holds :) |
| # | 09:00:54 | csharp did not guess, had to consult spreadsheet ;-) |
| # | 09:01:21 | gmcharlt | you don't have them all memorized? ;) |
| # | 09:02:13 | sylvar used to have a dozen or so memorized (not PINES, a multitype consortium that no longer hosts an ILS) |
| # | 09:02:45 | tsbere | So, any chance of getting someone to look at a couple of branches of code? I want to make sure they are acceptable (or close to it, and only for master, not 2.1) before we actually start using them in production. And we want to finalize our production codebase by the end of the week for migration headache reduction. |
| # | 09:04:29 | csharp | :-) |
| # | 09:15:39 | kmlussier has joined #evergreen |
| # | 09:28:48 | bjwebb has joined #evergreen |
| # | 09:28:49 | bjwebb has joined #evergreen |
| # | 09:29:27 | jenny has joined #evergreen |
| # | 10:18:34 | bshum | Anyone around with a 2.0 system? Got something odd I'd like someone else to check on |
| # | 10:18:56 | bshum | Catalogers have permission for update_volume set to depth of branch |
| # | 10:19:14 | bshum | But they seem to have the ability to edit volume labels regardless of where their actual working locations. |
| # | 10:19:58 | bshum | In holdings maintenance, if they pick another library within the consortium, highlight the row for volume and click "Edit Volume" from the options, it opens a little window that allows them to edit the label |
| # | 10:20:22 | bshum | If the permissions aren't set for allowing updates from a branch you don't work for, that seems like a bug in not respecting the permission. |
| # | 10:20:35 | jenny | bshum: I can check that for you in a moment |
| # | 10:20:44 | bshum | If someone else can confirm that behavior (I've tried it on our live and test environments, and a stock install of 2.0.6 as well) |
| # | 10:20:47 | bshum | Then I'll file a bug |
| # | 10:20:53 | bshum | jenny: Thanks, appreciate it :) |
| # | 10:23:22 | tsbere | So, eeevil, you around? |
| # | 10:23:25 | kmlussier | bshum: one of our catalogers noticed this recently, but I haven't had a chance to look into it yet. |
| # | 10:25:39 | bjwebb has quit IRC |
| # | 10:25:52 | bshum | kmlussier: Yeah it came up during our cataloging training, I guess our libraries using it so far haven't been poking at their fellow libraries. |
| # | 10:29:21 | eeevil | tsbere: I'm looking at lazy_circ2+master now |
| # | 10:29:32 | tsbere | eeevil++ |
| # | 10:34:05 | jamesrf has quit IRC |
| # | 10:35:41 | dbs has joined #evergreen |
| # | 10:35:41 | dbs has joined #evergreen |
| # | 10:38:57 | lisppaste | eeevil pasted "eyeballin' stuff" at http://paste.lisp.org/display/122061 |
| # | 10:39:08 | eeevil | tsbere: that's for you, sir |
| # | 10:41:02 | jenny | bshum: I tested the update_volume issue on our development server and it appears to be a bug. I was able to edit volumes at branches my catalog user was not registered to. This server is currently on 2.0.5. I don't have access to a 2.0.6 server at the moment. Permissions appear to work correctly on our 1.6.1 server (I was curious, so had to check) |
| # | 10:41:43 | bshum | jenny: Thanks for checking. I've started writing up a bug ticket for Launchpad on it. |
| # | 10:41:57 | bshum | And thanks for checking with 1.6.1 :) |
| # | 10:42:04 | bshum | I don't have any of those around anymore |
| # | 10:42:22 | jenny | we hope to get rid of ours this week :) |
| # | 10:42:55 | bshum | Awesome, heh |
| # | 10:45:01 | tsbere | eeevil: Yea. I apparently missed that, as my code says == but my installed version I was testing says != |
| # | 10:46:07 | tsbere | Correction: One copy of my code says == (as in the committed copy). I apparently fixed that locally *after* commit, and forgot to commit the extra change. tsbere-- |
| # | 10:47:04 | eeevil | tsbere: in a scan of the diff, that was the only thing that stuck right out at me ... there are varying levels of error handling going on, and I think they deserve some commentary as to why some are more robust than others (or robustification, if indeed the commentary suggest that) ... that said, I'm inclined to push this into trunk sooner rather than later, if you don't mind going through and commenting up some of the js (the call sites for get_barcode |
| # | 10:47:42 | tsbere | I can push that one fix and a pile of comments. Just have to write the comments. |
| # | 10:47:46 | eeevil | gah ... need to stop writing so much on one line |
| # | 10:48:30 | eeevil | not sure if that got truncated. last sentence was: there are only a couple call sites and IMO that would greatly improve maintenance efforts by others |
| # | 10:48:44 | eeevil | tsbere++ |
| # | 10:51:21 | sfortin has joined #evergreen |
| # | 10:56:48 | AaronZ-PLS has quit IRC |
| # | 10:57:09 | wlayton has joined #evergreen |
| # | 10:57:28 | Dyrcona has left #evergreen |
| # | 10:57:35 | eeevil | crap ... gmcharlt, in d9a4c8382cb65b36f201f12a7b06b32a3fc953c8 I'm seeing: |
| # | 10:57:36 | eeevil | copy from Open-ILS/src/sql/Pg/upgrade/0431.schema.hold_retarget.sql |
| # | 10:57:37 | eeevil | copy to Open-ILS/src/sql/Pg/upgrade/0534.schema.fix_hold_permit_test.sql |
| # | 10:57:50 | eeevil | gmcharlt: that's ... ungood, I think, no? |
| # | 10:58:16 | tsbere | eeevil: Ironically, in re-reading, both versions of my end were wrong. the == was correct. barcode_object was missing a .ilsevent though. |
| # | 10:58:19 | MarkEIN has joined #evergreen |
| # | 10:59:29 | eeevil | tsbere: ahh, ok ... that formulation was used elsewhere |
| # | 10:59:54 | tsbere | Which is why I noticed. Got to that for adding comments, went "wait, that doesn't look right based on what I just commented" |
| # | 10:59:59 | gmcharlt | eeevil: see where, precisely? |
| # | 11:00:26 | eeevil | gmcharlt: well, I noticed when pulling master into my local lazy_circ2 a "removing blah" message |
| # | 11:01:08 | eeevil | and then, in the diff for that commit, it looks like it's renaming the 0431 upgrade script and changing it a bit |
| # | 11:01:44 | eeevil | but ... 0431 is still in the working tree here, so I guess it's just more scary git messages (I hope, anyway) |
| # | 11:04:35 | eeevil | gmcharlt: ah HA ... nevermind, I'm straighe |
| # | 11:04:38 | eeevil | straight |
| # | 11:05:16 | gmcharlt | :) |
| # | 11:06:37 | gmcharlt | just see an artifact of Git's similiarity detection, I imagine |
| # | 11:06:40 | tsbere prepares for a quick test to make sure his commenting didn't break anything |
| # | 11:07:25 | tsbere | eeevil: To hopefully correct for that bit, I can git pull to ensure lazy_circ2 is more up to date against master as well. But that might make cherry-pick harder (but is fine if you just do merge) |
| # | 11:07:40 | eeevil | gmcharlt: yes, that threw me off ... easing into this, though |
| # | 11:08:20 | eeevil | tsbere: I'm just pulling your branch ... I've pulled master in locally too, all's well so far |
| # | 11:10:41 | Dyrcona_ has joined #evergreen |
| # | 11:11:22 | Dyrcona_ | I am trying to get the OpenSRF/OpenILS modules working for clients on my workstation, and I get the following error when I try to run a test program: |
| # | 11:11:23 | tsbere | eeevil: Did a git pull anyway. And pushed. |
| # | 11:11:25 | Dyrcona_ | Could not create file parser context for file "/openils/conf/fm_IDL.xml": No such file or directory at /usr/local/share/perl/5.10.1/OpenILS/Utils/Fieldmapper.pm line 199 |
| # | 11:12:54 | Dyrcona_ | ah, never mind: sudo ln -s /home/jason/openils /openils |
| # | 11:12:56 | dbs | Dyrcona_: /openils/conf/fm_IDL.xml exists and is readable by the user? |
| # | 11:12:59 | Dyrcona_ | fixed it. |
| # | 11:12:59 | dbs | aha |
| # | 11:13:29 | yboston has joined #evergreen |
| # | 11:19:26 | collsk12 has joined #evergreen |
| # | 11:20:47 | eeevil | tsbere: awesome, thanks |
| # | 11:21:06 | tsbere couldn't resist having fun with one comment, though |
| # | 11:23:04 | eeevil | tsbere: as a git noob, I'll "let it sit for a bit" ... if you want to push a rebased branch with condensed commits, feel free. if not, I'll push this ... soonish |
| # | 11:23:33 | tsbere | I can rebase it down to one commit if you want. Makes the core log easier to deal with. |
| # | 11:25:35 | eeevil | I think the security commit is worth keeping separate, as documentation for other ... "that stuff is important!" ... the others I'm less concerned with as they're mainly cleanup from the first, no? |
| # | 11:26:42 | tsbere | If my security commit's log message was more than "Security checks, error checks, etc" maybe. I can add notes about the security stuff in the single-commits log message, though. |
| # | 11:27:52 | eeevil | since it's not a ginormous pile of code I'm happy to handle it how you like, really. your pref |
| # | 11:33:42 | tsbere | eeevil: Well, the rebased version is in branch lazy_circ2_rebased. Don't care which you use. |
| # | 11:35:33 | tsbere figured it was "safer" to push the rebase to a new branch pub-side |
| # | 11:36:03 | eeevil | indeed, thanks for that :) |
| # | 11:38:14 | tsbere | since using git locally I do all my dev work in piles of branches that don't know about each other's changes (usually) anyway. Thus, I also don't usually have issues with merge conflicts with local version compared to pushed version. But I know others differ there. |
| # | 11:38:28 | wlayton has left #evergreen |
| # | 11:38:43 | youdonotexist_ has joined #evergreen |
| # | 11:40:28 | collsk12 | Good morning, all. Could anyone help me figure out why my book cover images stopped loading this morning? They were fine yesterday. |
| # | 11:40:49 | dbs | collsk12: what added content provider? |
| # | 11:40:51 | tsbere | added content provider account expire? ;) |
| # | 11:41:22 | dbs | openlibrary - the default for Evergreen as of 2.0 - experienced an outage recently |
| # | 11:41:33 | collsk12 | That's the one! |
| # | 11:41:41 | dbs | http://blog.openlibrary.org/2011/05/16/openlibrary-org-is-currently-offline/ |
| # | 11:41:52 | dbs | sounds like covers are currently affected |
| # | 11:42:24 | dbs | note that with the new code in trunk, covers wouldn't be affected (because we're not using covers.openlibrary.org by ISBN anymore) |
| # | 11:42:37 | ernieSimuro | jeff_: did you get to upload the PHP code for openSRF? |
| # | 11:42:43 | collsk12 | dbs: You rock! Thank you! |
| # | 11:43:42 | dbs | collsk12: aww, you're sweet :) if you want to try http://git.evergreen-ils.org/?p=Evergreen.git;a=blob;f=Open-ILS/src/perlmods/lib/OpenILS/WWW/AddedContent/OpenLibrary.pm;h=2be38a0c6ebe9516ac32e4639b2d679bfc193758;hb=HEAD it should be a drop-in replacement for the current OpenLibrary.pm module |
| # | 11:44:34 | collsk12 | dbs: Are you a mind reader or something? I appreciate it. |
| # | 11:45:16 | dbs awaits jeff's code for added content by record ID so that we can do more fun stuff :) |
| # | 11:45:29 | collsk12 | Uh oh... |
| # | 11:45:33 | collsk12 | :) |
| # | 11:52:21 | moodaepo | ernieSimuro: I believe jeff is traveling today so maybe tomorrow. |
| # | 11:53:09 | collsk12 | Since you have been so helpful today, could you possibly explain to me why the little icons located next to the search refinement options, in the left column, disappeared too? This was a couple of days ago and I'm not quite sure how to get them back. |
| # | 11:53:33 | moodaepo also awaits jeff release code to do pulls from evergreen which he has locked up in one of his VMs |
| # | 11:54:10 | ernieSimuro | moodaepos: Thanks I thought he was due back today will check tomorrow |
| # | 11:54:26 | moodaepo | collsk12: Mind sharing a link to the opac? |
| # | 11:55:17 | tsbere | eeevil: Any thoughts of looking at my sip_statcats branch today? Or recommendations for poking someone else about it? |
| # | 11:55:17 | collsk12 | Sorry, I don't have it in the DMZ yet. I just built the system last week and will probably go live this summer. |
| # | 11:56:45 | collsk12 | I have objects like the Author names, number of matches, etc. I just lost the icons. |
| # | 11:59:41 | dbs | collsk12: if this is Evergreen 2.0, I don't think there are any icons on the left |
| # | 11:59:53 | dbs | Evergreen 1.6 had little icons on the left |
| # | 12:00:33 | collsk12 | Oh...that could explain it. No biggie. Thanks again, dbs. |
| # | 12:01:34 | moodaepo | dbs: Would you know what time the governance meeting is today? |
| # | 12:01:56 | dbs | moodaepo: evergreen-ils.org/calendar would |
| # | 12:02:04 | moodaepo | Oh right |
| # | 12:02:16 | dbs | oh, that didn't make the jump |
| # | 12:02:37 | moodaepo | There seems to be some major misunderstanding of the student funding I've been pushing for with the IMLS folks. |
| # | 12:02:50 | dbs | http://evergreen-ils.org/dokuwiki/doku.php?id=calendar:start works |
| # | 12:03:01 | dbs | moodaepo: yes, you're reading the governance mailing list? |
| # | 12:03:36 | gmcharlt | yep, I've been pointing moodaepo that way, given the misunderstanding that's going on |
| # | 12:03:37 | moodaepo | Hmm I used the link in the menu |
| # | 12:03:46 | moodaepo | But yea gmcharlt forwarded the mail to me |
| # | 12:08:20 | bshum | Hmm |
| # | 12:08:26 | bshum | Those emails back and forth about OpenSRF |
| # | 12:08:33 | dbs | moodaepo: we used to have a redirect for the easy-to-remember evergreen-ils.org/calendar to the wiki page, that's all |
| # | 12:08:35 | bshum | Makes me feel that maybe step 9 should be before step 8 for the instructions. |
| # | 12:08:59 | bshum | cp opensrf examples before you register the ejabberd accounts? |
| # | 12:09:48 | moodaepo | dbs: Setting it up |
| # | 12:10:04 | dbs | bshum: either way, you have a chicken and the egg scenario |
| # | 12:10:23 | bshum | dbs: True. I'm thinking if there's any way to clean up the language there now. |
| # | 12:10:24 | dbs | edit opensrf_core.xml to have accounts that don't exist yet? |
| # | 12:11:37 | dbs | bshum: rather than refining language, writing a script to handle all of the configuration would probably be more efficient |
| # | 12:11:54 | bshum | dbs: Well, sure :) |
| # | 12:13:47 | youdonotexist_ is now known as youdonotexist |
| # | 12:16:09 | dbs | bshum: the one concession I would make is to add 'opensrf_core.xml file that you will edit in the next step' - for people who don't skim over all of the instructions before beginning at step 1, or are unable to use CTRL-F in their browser to find another reference to something mysterious, or who don't scan ahead to the next step |
| # | 12:16:49 | csharp | yeah - I think the instructions are pretty clear about that as-is (though a language barrier may be at play here too ;-) ) |
| # | 12:17:02 | csharp | bshum answered just as I was typing my email |
| # | 12:17:16 | csharp | bshum is a faster typist, as I now know ;-) |
| # | 12:17:46 | moodaepo | dbs: evergreen-ils.org/calendar is now setup |
| # | 12:17:54 | jamesrf has joined #evergreen |
| # | 12:17:58 | dbs | moodaepo++ |
| # | 12:19:46 | bshum | csharp: Eh, faster isn't always better :( |
| # | 12:23:15 | dbs | I closed off milestone 2.0.6 and added 2.0.7 for anybody interested in changing 13 bugs to "Fix released" :) |
| # | 12:24:27 | bshum | That's like... the fun part! |
| # | 12:24:36 | asimon has quit IRC |
| # | 12:54:37 | natschil has joined #evergreen |
| # | 12:58:06 | tsbere finds it interesting that GitHub thinks Evergreen is Javascript |
| # | 12:58:41 | Dyrcona_ | :) |
| # | 12:58:52 | Dyrcona_ | more lines of javascript than perl maybe? |
| # | 13:02:38 | AaronZ-PLS has joined #evergreen |
| # | 13:03:06 | moodaepo_ has joined #evergreen |
| # | 13:16:34 | jenny has quit IRC |
| # | 13:20:58 | eeevil | tsbere: re sip_statcats, I won't have time today, and I'd like berick or another sip'r (dbs, senator, djfiander?) to look at it (berick, in particular, is on vacation ATM) |
| # | 13:23:22 | eeevil | tsbere: I'm probably going to use your rebased branch, beat it around a bit for the upgrade script and such, and rebase again adding some more detail to the commit message. that is, assuming, that rebase -i will (can) retain your sign-off ... not sure, do you know? |
| # | 13:23:32 | eeevil | (thats re lazy circ, obv) |
| # | 13:24:04 | tsbere | eeevil: It will retain sign-off. Dunno about author, though. |
| # | 13:24:47 | gmcharlt | it will retain author as well |
| # | 13:25:58 | eeevil | gmcharlt: even on squash? |
| # | 13:26:00 | tsbere rarely rebases someone else's commits |
| # | 13:26:54 | eeevil | hrm... but ... gmcharlt, you warned against rebasing after a post-patch cleanup commit? |
| # | 13:28:07 | tsbere | eeevil: He did that for the "stop merge conflicts from happening on the author's machine" I think. |
| # | 13:28:15 | gmcharlt | yes |
| # | 13:28:20 | tsbere | And I won't be working on those branches again, and thus don' |
| # | 13:28:26 | tsbere | *don't care about merge conflicts ;) |
| # | 13:28:29 | eeevil | ahh... righto |
| # | 13:28:31 | gmcharlt | there's not necessarily a good reeason to squash, though |
| # | 13:28:51 | gmcharlt | if all you're doing is tacking on upgrade scripts and the like |
| # | 13:28:57 | eeevil | no, there's not ... I'll just squash my own (if needed) |
| # | 13:29:28 | tsbere | gmcharlt: Well, he likely wants to wipe out the XXXX upgrade script as well. I could see that as a reason to squash. |
| # | 13:29:58 | gmcharlt | maybe, or just do a git mv and commit the rename |
| # | 13:30:02 | eeevil | grabbing 0536 |
| # | 13:30:11 | eeevil | right |
| # | 13:30:22 | eeevil | and other minor adjustments |
| # | 13:30:38 | tsbere | Like replacing XXXX with 0536 inside the file? ;) |
| # | 13:31:06 | eeevil | and in 002.schema.config.sql |
| # | 13:31:25 | tsbere knows that the XXXX in the upgrade script was missed once, he thinks by phasefx |
| # | 13:31:58 | tsbere | Hmmm. |
| # | 13:32:51 | tsbere | Would having a different commit for the upgrade script help for future patches? |
| # | 13:33:19 | gmcharlt | it would |
| # | 13:34:05 | tsbere will try and remember to do that in the future |
| # | 13:34:31 | eeevil | tsbere: there's a script for generating the boilerplate stuff .../Pg/make-db-patch.pl |
| # | 13:35:07 | tsbere | eeevil: I rigged this up myself at one point: http://paste.lisp.org/display/121956 |
| # | 13:37:12 | tsbere | Isn't perfect, granted, but it is great for grabbing the exact defs for new stuff |
| # | 13:37:22 | moodaepo_ | dbs++ # software bounty |
| # | 13:37:54 | Dyrcona_ | where would i find the code that determines search relevance of results? |
| # | 13:38:25 | Dyrcona_ not finding what he's looking for. |
| # | 13:39:13 | eeevil | tsbere: make-db-patch.pl does some other stuff that trunkish upgrades need to do, but yours could be easily modified or used within that |
| # | 13:39:55 | tsbere | eeevil: Mine came around in part (less friendly) before make-db-patch.pl. Was considering submitting a patch to make-db-patch.pl to have it do some of what mine does. |
| # | 13:40:07 | moodaepo_ has quit IRC |
| # | 13:40:18 | gmcharlt | tsbere: go for it |
| # | 13:42:27 | Dyrcona_ thinks he found what he's looking for, unlike a certain Irish singer. |
| # | 13:48:36 | yboston has quit IRC |
| # | 13:53:42 | eeevil | gmcharlt: when editing a cherry-picked commit message, do I need to uncomment the Author: line to retain that info? |
| # | 13:53:50 | eeevil | (I assume no, but...) |
| # | 13:55:36 | eeevil | we'll find out.. :) |
| # | 13:55:53 | eeevil | nope ... I assumed correctly |
| # | 14:01:12 | jenny has joined #evergreen |
| # | 14:01:52 | tsbere | gmcharlt / eeevil : Any thoughts? http://git.mvlcstaff.org/?p=tsbere/ILS.git;a=shortlog;h=refs/heads/mdp-improve |
| # | 14:03:14 | gmcharlt | tsbere: looks good; I'll test it now and push if it passes |
| # | 14:03:25 | tsbere has quit IRC |
| # | 14:03:41 | tsbere has joined #evergreen |
| # | 14:03:51 | tsbere dislikes random disconnects |
| # | 14:04:13 | gmcharlt | did you catch what I just said to your, tsbere? |
| # | 14:04:33 | eeevil | gmcharlt: so, pull caused conflict (same file edited) ... is there a way to restart the pull after solving the conflict and using git-add? |
| # | 14:04:43 | eeevil | something like git merge --continue? |
| # | 14:04:59 | tsbere | gmcharlt: Last I saw was my own "Any thoughts?" |
| # | 14:05:11 | gmcharlt | <gmcharlt> tsbere: looks good; I'll test it now and push if it passes |
| # | 14:06:46 | eeevil | gmcharlt: or is the answer 'git commit -am "dummy message"' ? |
| # | 14:06:49 | senator | eeevil: if you've solved a conflict and git-add'd your changes, git commit next |
| # | 14:06:54 | gmcharlt | eeevil: one just fixes the merge conflicts, then does a commit -c |
| # | 14:07:34 | moodaepo | Will importing a pg_dump from 8.3 to a 9.0 server be an issue? |
| # | 14:07:39 | eeevil | git commit next == git commit -c ? |
| # | 14:07:59 | senator | -c is if you care about editing the commit message |
| # | 14:08:14 | eeevil | moodaepo: maybe not :) ... (use the 9.0 version of pg_dump if you can) |
| # | 14:08:16 | eeevil | thanks |
| # | 14:08:37 | eeevil | hrm.. fatal: cannot do a partial commit during a merge. |
| # | 14:08:45 | moodaepo | eeevil: Will do.. |
| # | 14:08:56 | senator | eeevil: you don't have everything added then, i think |
| # | 14:09:01 | gmcharlt | eeevil: have you resolved all of the merge conflicts? |
| # | 14:09:05 | eeevil | git status says I do |
| # | 14:09:19 | senator | eeevil: using colors? if so, everything's green no red? |
| # | 14:09:36 | eeevil | senator: no colors, but green, yes. wanna see? :) |
| # | 14:09:49 | eeevil | (please? ;) ) |
| # | 14:09:54 | senator | sure |
| # | 14:10:04 | senator moseys |
| # | 14:11:10 | dbs | moodaepo: I've been doing 8.3 -> 9.0, seems to be okay |
| # | 14:11:34 | dbs | eeevil: cmon, fire up a public VNC session so we can all see! |
| # | 14:11:45 | moodaepo | dbs: I'm considering a pg_dump not a pg_upgrade or pg_upgrade_cluster |
| # | 14:12:06 | eeevil | screen -S githatesme |
| # | 14:12:07 | dbs | moodaepo: yes, that's what I did. pg_dump from the 9.0 machine |
| # | 14:12:26 | senator | it was my bad for being unclear: 'next' wasn't supposed to be part of the actual command |
| # | 14:12:27 | moodaepo | Good to know! |
| # | 14:14:03 | moodaepo | dbs: Did you use the 9.0 pg_dump as eeevil suggested or 8.3 one? I don't have 9.0 installed on the 8.3 server (might have to do that to be safe) |
| # | 14:14:08 | eeevil | dbs: I'm tracking your social branch, btw ... keeping that at state at http://git.esilibrary.com/?p=evergreen-equinox.git;a=shortlog;h=refs/heads/social so I can poke at it a bit :) |
| # | 14:15:03 | dbs | moodaepo: yes. pg_dump works over the network, doesn't need to be locally installed |
| # | 14:15:26 | moodaepo | Ah I didn't think of that...graci |
| # | 14:16:10 | dbs | eeevil: cool! I was planning on pushing that to git.evergreen-ils.org proper; I think it's the only branch that I care about on gitorious (and it's been pretty much untouched for ages, sigh) |
| # | 14:17:16 | eeevil | er ... s/at state at/at master state at/ ... not much better, but at least contains all the words |
| # | 14:17:41 | dbs | hmm. where should the tiny little governance directory from ILS-Contrib live? |
| # | 14:19:49 | gmcharlt | minigov.git |
| # | 14:20:56 | eeevil | Dyrcona_: so does the (now main?) SIPServer repo at git.evergreen-ils.org now have fees/fines support? |
| # | 14:21:46 | Dyrcona_ | eeevil: yes in that all that's required on the SIPServer side is the fix to the configuration of the 37/38 message. |
| # | 14:21:58 | Dyrcona_ | the rest of it is on the OpenILS |
| # | 14:22:13 | eeevil | rad |
| # | 14:24:37 | gmcharlt | tsbere: pushed your mdp change, along with a tweak |
| # | 14:25:41 | tsbere | gmcharlt++ |
| # | 14:25:50 | Dyrcona_ | eeevil: I still have to work on your suggestions for the fee payment code. |
| # | 14:26:23 | tsbere sees an svn commit go by |
| # | 14:27:46 | gmcharlt | dbs: to answer your question seriously, just eg-governance.git ? |
| # | 14:28:06 | dbs | tsbere: using it while it's still open |
| # | 14:28:32 | dbs | gmcharlt: almost wondering whether it should be a subset of Evergreen.git or Evergreen-Documentation.git |
| # | 14:28:38 | tsbere | dbs: Was more of a "wait, wha?" moment for me when reading the subject line |
| # | 14:29:02 | tsbere | dbs: Evergreen.git I think no. Maybe the docs, but probably still no. |
| # | 14:30:48 | dbs | I don't really care, as long as we don't end up having dozens of top-level git repos |
| # | 14:31:40 | gmcharlt | we could stick it under contrib |
| # | 14:31:43 | gmcharlt | contrib/governance.git |
| # | 14:31:45 | tsbere | dbs: Awww, you don't want a list like this? http://git.kernel.org/ |
| # | 14:32:01 | dbs | tsbere: heh, if we had that many contributors I would be cool with that |
| # | 14:32:05 | gmcharlt would the level of activity that such a list would imply |
| # | 14:32:46 | gmcharlt | s/would/would love/ |
| # | 14:33:15 | dbs | governance contribs to evergreen - that seems about right |
| # | 14:33:31 | gmcharlt will make it so |
| # | 14:34:27 | natschil has quit IRC |
| # | 14:36:50 | natschil has joined #evergreen |
| # | 14:37:27 | moodaepo_ has joined #evergreen |
| # | 14:46:35 | gmcharlt | dbs: http://git.evergreen-ils.org/?p=contrib/governance.git;a=summary |
| # | 14:47:15 | dbs | gmcharlt++ # thanks! |
| # | 14:48:08 | tsbere | How many more things need to be pulled out of ILS-Contrib? |
| # | 14:48:49 | dbs | constrictor needs a home |
| # | 14:49:00 | gmcharlt | yep, once berick is back from vacation |
| # | 14:49:02 | tsbere notes some things haven't changed for years |
| # | 14:49:12 | gmcharlt | and some, like upei, that are empty |
| # | 14:49:29 | dbs | servres (Syrup) will need a home, given that there's an RFP for it now :) |
| # | 14:49:54 | dbs | most of ILS-Contrib is a ghost town though |
| # | 14:50:17 | Dyrcona_ watches tumbleweeds roll by. |
| # | 14:51:07 | Dyrcona_ | i suggest syrup getting its own repo. |
| # | 14:53:38 | jenny has quit IRC |
| # | 14:53:40 | dbs | Well, crap. Time to move off of launchpad. "Currently, only the HEAD branch of a given Git repository may be imported." lame. |
| # | 14:54:35 | eeevil | gah |
| # | 14:55:07 | eeevil | dbs: blithly disregarding tsbere's concern, we could have a repo-per-branch and link to each ;) |
| # | 14:55:26 | tsbere | Wait, my concern? |
| # | 14:55:36 | tsbere | Wasn't it dbs's concern? :P |
| # | 14:56:03 | eeevil | tsbere: it was, nm |
| # | 14:56:26 | dbs | Also, going to have to find out if we can either bulk-change bugs currently pointing at trunk to point at master, or get the Launchpadians to do it for us |
| # | 14:56:43 | dbs | having repos-per-branch sounds rather horrible to me |
| # | 14:56:52 | eeevil | dbs: it does, I was totally joking |
| # | 14:57:28 | dbs rips the hook out of his mouth |
| # | 14:57:48 | dbs | can't believe the foot that's usually in there didn't protect me |
| # | 14:58:13 | tsbere just had a crazy idea to fool git-daemon into thinking we had a pile of repos with different heads for each. It should probably not even be attempted. |
| # | 15:09:03 | AaronZ-PLS has quit IRC |
| # | 15:09:12 | AaronZ-PLS has joined #evergreen |
| # | 15:18:43 | kmlussier has quit IRC |
| # | 15:31:02 | kmlussier has joined #evergreen |
| # | 15:37:06 | Dyrcona_ has quit IRC |
| # | 15:44:18 | collsk12 | Heading home for the night. Just wanted to say thanks again and have a great day. |
| # | 15:44:30 | collsk12 has quit IRC |
| # | 15:50:05 | eeevil | dbs: re your new indexes ... it doesn't look like the index on deleted was necessary ... could you confirm that that's true? (your pasted explain analyze didn't use that, just the restrictive value index, IIRC) |
| # | 15:56:29 | dbs | eeevil: it depends on how many matches you get for the restrictive value index |
| # | 15:57:04 | eeevil | dbs: ok ... you wouldn't want it on (id)where not deleted? (since that's the join condition |
| # | 15:57:31 | dbs | eeevil: if you want to refine it further, go ahead |
| # | 15:57:47 | eeevil | dbs: well, I don't have a big dataset loaded to test |
| # | 15:58:13 | eeevil | just looking for ways to make the index more generally useful ... may not be worth looking |
| # | 16:00:39 | dbs | eeevil: sure. sorry, sluggish context-switching from copyright concerns over wiki -> docs. If you have something concrete you want me to test with multiple conditions, I'll be happy to do so |
| # | 16:05:45 | Dyrcona has joined #evergreen |
| # | 16:18:40 | bgoble has joined #evergreen |
| # | 16:23:49 | csharp | :q |
| # | 16:23:53 | csharp | oops |
| # | 16:30:26 | natschil has quit IRC |
| # | 16:31:53 | Dyrcona | :q! |
| # | 16:31:59 | Dyrcona | heh. wrong window. :) |
| # | 16:32:06 | Dyrcona | j/k |
| # | 16:32:26 | gmcharlt | ZZ |
| # | 16:32:45 | natschil has joined #evergreen |
| # | 16:37:11 | Callender has quit IRC |
| # | 16:38:24 | natschil has quit IRC |
| # | 16:43:30 | natschil has joined #evergreen |
| # | 16:43:46 | bjwebb has joined #evergreen |
| # | 16:43:46 | bjwebb has joined #evergreen |
| # | 16:50:44 | dbs has quit IRC |
| # | 16:54:15 | sfortin has quit IRC |
| # | 16:58:32 | moodaepo_ has quit IRC |
| # | 16:59:05 | natschil has quit IRC |
| # | 17:01:40 | jenny has joined #evergreen |
| # | 17:05:16 | AaronZ-PLS has quit IRC |
| # | 17:12:15 | kmlussier has quit IRC |
| # | 17:32:15 | Dyrcona has quit IRC |
| # | 18:03:05 | jenny has left #evergreen |
| # | 18:36:23 | bjwebb has quit IRC |
| # | 18:52:32 | bshum | Looks like we may be missing a permission called "MARK_ITEM_MISSING_PIECES" |
| # | 18:53:01 | bshum | Trying to see if this has been addressed in a more recent version of 2.0. We're a bit behind. |
| # | 19:01:11 | tsbere | bshum: Did you want to learn how to push your git work to the Evergreen working repo (instead of uploading patch files)? |
| # | 19:01:25 | bshum | tsbere: That sounds pretty intriguing. |
| # | 19:01:34 | bshum | tsbere: I have a few moments, sure! |
| # | 19:01:44 | tsbere | ok |
| # | 19:02:03 | tsbere | step 1: git remote add working git@git.evergreen-ils.org:working/Evergreen |
| # | 19:02:18 | tsbere | That adds the working repo as an easily targeted remote |
| # | 19:02:41 | bshum | Doing that within the same directory I have Evergreen cloned to, I presume? |
| # | 19:02:46 | tsbere | Yea. |
| # | 19:03:00 | bshum | Okay |
| # | 19:03:27 | tsbere | Then, to be sure it works, git fetch remote |
| # | 19:04:14 | tsbere | sorry |
| # | 19:04:18 | tsbere | git fetch working |
| # | 19:04:22 | tsbere | or git fetch --all |
| # | 19:04:36 | tsbere had to answer the phone while typing |
| # | 19:04:47 | tsbere | And even worse, it was a wrong number |
| # | 19:06:25 | bshum | Heh |
| # | 19:06:31 | bshum | Okay, fetch appears good |
| # | 19:06:46 | tsbere | Ahh, good |
| # | 19:07:43 | tsbere | Next up, push! specifically, "git push working <localname>:user/USER/<remotename>" where <localname> gets replaced with your local branch name, <remotename> with the remote branch name, and USER with your git username. |
| # | 19:09:49 | bshum | What would a local branch name be? Oh like "working" as defined by default in the short git instruction guide we have on the wiki? |
| # | 19:09:58 | bshum | Or should I be making it more constructive |
| # | 19:10:31 | tsbere | The local branch name is "the name of the branch you did the work in" |
| # | 19:11:16 | bshum | Gotcha |
| # | 19:11:30 | tsbere | The remote branch name is the name you want the public to see |
| # | 19:12:30 | bshum | Hmm, are there recommendations on how to name a remote branch? :) |
| # | 19:12:37 | bshum | For public viewing |
| # | 19:12:53 | tsbere | I usually try and make it related to the work that happened in it. |
| # | 19:13:07 | ernieSimuro has quit IRC |
| # | 19:13:16 | bshum | Aha, I see. |
| # | 19:13:16 | tsbere | Beyond that, whatever you are comfortable with |
| # | 19:15:05 | bshum | Oh spooky |
| # | 19:15:09 | bshum | I think it did something |
| # | 19:15:31 | tsbere | http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/bshum@biblio.org/google-link-fix |
| # | 19:15:45 | bshum | Neat. |
| # | 19:17:08 | bshum | tsbere: Thanks for introducing me to this. |
| # | 19:17:10 | tsbere | Now you can push to the working repo and share the branch names with people |
| # | 19:19:18 | tsbere | The bigger the change the better help I think it is, but this way if they pull from there you get flagged in git as the author ;) |
| # | 19:27:05 | tsbere | bshum: In regards to that permission, even the master branch doesn't have it in the seed scripts. |
| # | 19:27:18 | tsbere | There are a number missing, actually. |
| # | 19:27:29 | bshum | tsbere: That's what I was afraid of. |
| # | 19:28:08 | bshum | tsbere: At some point I think we patched our system locally to fix that feature but never noticed the missing permission since all of our HQ staff have the -1 everything perm |
| # | 19:30:35 | bshum | Those are all initially created in the 950.data.seed-values.sql file, right? |
| # | 19:30:52 | bshum | And fixes added via upgrade scripts between point releases |
| # | 19:30:57 | bshum | ? |
| # | 19:31:13 | tsbere | yea. |
| # | 19:31:19 | bshum | Fun, fun |
| # | 19:31:50 | tsbere | I just found a few that are missing |
| # | 19:32:27 | tsbere | CREATE_PURCHASE_REQUEST MARK_ITEM_MISSING_PIECES UPDATE_HOLD_REQUEST_TIME UPDATE_PICKLIST UPDATE_WORKSTATION |
| # | 19:33:44 | tsbere | Those are just ones I found via scanning for any permissions the perl layer asks about in a basic fashion and then asking for those that don't have an entry in 950.data.seed-values.sql, that doesn't include ones set elsewhere. |
| # | 19:34:09 | youdonotexist has quit IRC |
| # | 19:34:13 | tsbere should look into a quick automated script to indicate permissions checked for in various ways |
| # | 19:38:47 | tater-laptop has joined #evergreen |
| # | 19:40:54 | bshum | Oh well. Have to tackle that tomorrow morning I guess. |
| # | 19:41:04 | bshum | Thanks again for showing me more git goodness tsbere! |
| # | 19:41:12 | bshum | Have a good night. |
| # | 19:41:15 | tsbere | night |
| # | 20:04:37 | moodaepo | Any preferences folks? > http://bethesignal.org/projects/irclog2html/style_tt.html || http://bethesignal.org/projects/irclog2html/style_simpletable.html |
| # | 20:07:15 | moodaepo | bshum and I have been looking at prettyfication and he found this script which prettyfies. It is not a bot so supy/pinsol_green will still be creating the raw logs. We think we can just set it up so supy creates the log files and this script copies it over prettyfied to the public location say every minute/5 minutes/whatever. |
| # | 20:07:23 | moodaepo goes to walk the dog |
| # | 20:09:31 | eeeevil has joined #evergreen |
| # | 20:31:04 | tsbere throws a branch out for review: http://git.mvlcstaff.org/?p=tsbere/ILS.git;a=shortlog;h=refs/heads/perms_check |
| # | 20:31:23 | tsbere may bring it up again tomorrow |
| # | 20:33:42 | eeeevil | have you seen the xlst that attempts to do that? (may not work anymore, haven't tested it recently) |
| # | 20:35:13 | tsbere | Does it scan the perl for allowed('permission' and allowed(['permission','permission'] entries? |
| # | 20:35:37 | tsbere | Mine isn't perfect (it doesn't understand perl, for example) but I tried to get the easy pickings |
| # | 20:36:38 | tsbere | Also, this is intended to find *missing* permissions. Not just list them. |
| # | 20:36:51 | tsbere | Based on bshum bringing up one that he found was missing |
| # | 20:41:39 | eeeevil | no, it was only intended to do the ILD-supplied permissions that permacrud adds |
| # | 20:42:08 | bgoble has quit IRC |
| # | 20:43:06 | eeeevil | since it understands xml, it may be useful to fold that in to your tool ... being native. but the IDL perms are much simpler to parse, too |
| # | 20:43:18 | eeeevil runs away to watch LOST |
| # | 20:48:33 | jeffdavis has quit IRC |
| # | 20:58:40 | eeeevil has quit IRC |
| # | 21:35:42 | pmplett has quit IRC |
| # | 21:36:28 | pmplett has joined #evergreen |
| # | 21:45:02 | phasefx | evergreen=# \i 0536.schema.lazy_circ-barcode_lookup.sql |
| # | 21:45:02 | phasefx | BEGIN |
| # | 21:45:02 | phasefx | psql:0536.schema.lazy_circ-barcode_lookup.sql:8: ERROR: syntax error at or near ":" |
| # | 21:45:05 | phasefx | LINE 1: SELECT evergreen.update_deps_block_check('0536', :eg_version... |
| # | 22:10:13 | tsbere | phasefx: My understanding (limited as it is) of that is "there is a script that is supposed to call those for you now" |
| # | 22:10:21 | phasefx | ah |
| # | 22:10:28 | tsbere didn't even build the script with that stuff in it, actually |
| # | 22:10:39 | phasefx | suspected that, but didn't really pay attention when folks were talking about it |
| # | 22:12:20 | tsbere | I think you can "psql -v eg_version=NULL blah" |
| # | 22:13:36 | phasefx | opensrf@dev141:~/git/Evergreen/Open-ILS/src/sql/Pg/upgrade (masslnc)$ psql -d evergreen -U evergreen -v eg_version=NULL -f 0536.schema.lazy_circ-barcode_lookup.sql |
| # | 22:13:39 | phasefx | BEGIN |
| # | 22:13:42 | phasefx | psql:0536.schema.lazy_circ-barcode_lookup.sql:8: ERROR: function evergreen.update_deps_block_check(unknown, unknown) does not exist |
| # | 22:13:45 | phasefx | LINE 1: SELECT evergreen.update_deps_block_check('0536', NULL); ^ |
| # | 22:13:47 | phasefx | HINT: No function matches the given name and argument types. You might need to add explicit type casts. |
| # | 22:15:03 | phasefx is using another db built from seed data, so this isn't a roadblock for him |
| # | 22:24:23 | phasefx | eww, scary. I was using a bare git push from master, and it complained about rel_2_1 (because that was behind some commits). I didn't expect a bare git push to try to push all branches for the remote |
| # | 22:25:36 | tsbere | phasefx: Yea....I dislike that. Default is to push anything you have local that the remote also has. |
| # | 22:28:39 | tsbere | @later tell eeevil In regards to the new trunk/master/whatever way of doing DB upgrade scripts....would it be wise to make that method work before we start committing scripts using it? phasefx had issues, and I can't find any scripts that look to apply updates properly? |
| # | 22:28:39 | pinesol_green | tsbere: The operation succeeded. |
| # | 22:53:17 | bshum | @later tell moodaepo Hmm, that logs2html part is tricky. Apparently it dislikes the date scheme and will only accept YYYY-MM-DD. |
| # | 22:53:17 | pinesol_green | bshum: The operation succeeded. |
| # | 23:28:30 | pmplett is now known as pmpafk |
| # | 23:38:50 | tater-laptop has quit IRC |
April 10-13