| Time |
Nick |
Message |
| 00:42 |
|
sandbergja joined #evergreen |
| 03:04 |
|
mrisher joined #evergreen |
| 03:12 |
|
mrisher joined #evergreen |
| 04:18 |
|
mikerisher joined #evergreen |
| 04:46 |
pinesol |
News from qatests: Failed Installing OpenSRF pre-requisites <http://testing.evergreen-ils.org/~live//archive/2020-05/2020-05-06_04:00:03/test.7.html> |
| 07:15 |
|
JBoyer_ joined #evergreen |
| 07:17 |
|
rjackson_isl_hom joined #evergreen |
| 07:20 |
|
Dyrcona joined #evergreen |
| 07:22 |
|
rfrasur joined #evergreen |
| 07:23 |
|
JBoyer joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 08:45 |
Dyrcona |
mmorgan: A follow up from yesterday's discussion of quarantining items.... |
| 08:46 |
* mmorgan |
listens |
| 08:46 |
Dyrcona |
jamundson has figured out how to make only copies with the Available status show up on the printed pull list via a print template hack. Combining that with extending the time for the Reshelving status could work. |
| 08:47 |
Dyrcona |
Since it is a print template change, the libraries can do it themselves, and it is relatively easy to undo compared to a backend or staff client code hack. |
| 08:47 |
mmorgan |
Cool! |
| 08:48 |
Dyrcona |
Here's what he sent me in email: ng-show="hold_data.hold.pickup_lib.shortname==current_location.shortname && hold_data.copy.status==0" |
| 08:49 |
Dyrcona |
He added the condition for local pickup and only Available status. I'm not sure exactly where in the template that goes. |
| 08:54 |
|
dbwells_ joined #evergreen |
| 08:57 |
Dyrcona |
I'm probably going to bounce my access point, once I get signed into it, so I'm signing our for a little bit. |
| 09:06 |
|
mantis1 joined #evergreen |
| 09:07 |
|
mmorgan joined #evergreen |
| 09:08 |
* mmorgan |
's internet just took a coffee break |
| 09:08 |
mmorgan |
@coffee internet |
| 09:08 |
* pinesol |
brews and pours a cup of "Kona Sweet" 100% Kona, and sends it sliding down the bar to internet |
| 09:10 |
|
rjackson_isl_hom joined #evergreen |
| 09:10 |
Bmagic |
@coffee [someone] |
| 09:10 |
* pinesol |
brews and pours a cup of Fair Trade Organic Ethiopia Yirgacheffe, and sends it sliding down the bar to eby |
| 09:14 |
csharp |
postgresql/SQL question: (kind of looking at miker here) - is there an advantage of a materialized view vs. just creating a table? I'm looking to add the aged billing equavalent of money.materialized_billable_xact_summary (which I realize is not actually a materialized view) |
| 09:15 |
csharp |
creating a table seems kind of expensive, storage-wise, but I want our reports to actually finish before timing out :-/ |
| 09:16 |
csharp |
this is what I'm working on, for reference: https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/csharp/aged_billing_summary |
| 09:24 |
|
mmorgan joined #evergreen |
| 09:28 |
Bmagic |
csharp: "materialized" things are tables right? Kept in check with triggers |
| 09:28 |
miker |
csharp: a true materialized view is also a table, just one that the system updates via a specialized command |
| 09:29 |
Bmagic |
:) |
| 09:29 |
csharp |
ah - makes sense |
| 09:29 |
miker |
our "mat views" are, as Bmagic says, maintained by our own triggers rather than special system ones |
| 09:30 |
|
mmorgan1 joined #evergreen |
| 09:30 |
csharp |
ok - that makes more sense - I wasn't seeing a use case where you would want a mat view at all |
| 09:33 |
|
rfrasur joined #evergreen |
| 09:38 |
|
Dyrcona joined #evergreen |
| 09:44 |
|
mmorgan joined #evergreen |
| 09:57 |
|
sandbergja joined #evergreen |
| 10:01 |
Bmagic |
berick++ # ultimately, it's going to be easier to not use git I'm afraid |
| 10:02 |
Dyrcona |
Bmagic: Blasphemy! :P |
| 10:03 |
Bmagic |
this is the command that would work: git apply -p2 --directory=docs-antora/modules/admin/pages/ ~/patch.patch - but only when my patch file contains changes to files that are two folders deep. In this example, I am targeting the two files that were changed in the admin tree, but if the patch contains changes to anything in a folder depth less than 2, it will fail to apply completely |
| 10:03 |
Bmagic |
Therefore, it seems, I would need to split my patch file into chunks for each folder depth, and rerun the command with a --directory prefix for each target subfolder |
| 10:05 |
Bmagic |
Looking at a manual diff of the docs tree between the collab branch and master - I can clearly see the few lines in each file that need to be carried over. It will take me all of 5-10 minutes to copy/paste those lines |
| 10:05 |
Bmagic |
plus, my patch, for some reason, didn't get the carousels.adoc at all (probably didn't go back far enough in master when running git diff) |
| 10:06 |
Dyrcona |
Not having seen your patch, I can't really comment, but I do often apply patches with patch instead of git apply. |
| 10:06 |
berick |
https://imgs.xkcd.com/comics/automation.png |
| 10:07 |
Bmagic |
yeah, pretty much :) |
| 10:07 |
Bmagic |
I'll paste it |
| 10:07 |
pastebot |
"Bmagic" at 168.25.130.30 pasted "docs patch for antora branch" (721 lines) at http://paste.evergreen-ils.org/10142 |
| 10:08 |
Dyrcona |
berick: My answer to that is, "Your'e doing it wrong." :) |
| 10:08 |
Bmagic |
I'm happy to learn - check the patch file out and see if that helps understand things |
| 10:10 |
Dyrcona |
You did a git diff between two branches and you want to apply that a different file hierarchy, right? |
| 10:10 |
Bmagic |
the RELEASE_NOTES changes are only one folder deep, so passing -p2 will bork the command - so I removed all of those patches from the file |
| 10:10 |
Bmagic |
I used this: git diff 3b067004fc95c8b92b757a8...d338486c851339b --relative docs |
| 10:10 |
pinesol |
Bmagic: [evergreen|Bill Erickson] LP1824391 Hatch File Writer release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3b06700> |
| 10:10 |
pinesol |
Bmagic: [evergreen|Chris Sharp] LP#1873048 - Stamp upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d338486> |
| 10:11 |
Dyrcona |
And, you want to apply that to a different hierarchy. Does that hierarchy match the docs/ hierarchy? |
| 10:11 |
Bmagic |
Dyrcona: yes in essense |
| 10:11 |
Bmagic |
not at all |
| 10:12 |
Bmagic |
for example: this file "docs/admin/lsa-work_log.adoc" - needs applied to" docs-antora/modules/admin/pages/lsa-work_log.adoc" |
| 10:14 |
Dyrcona |
You'll be better off with individual patch files, I think. 'Cause I'm not sure you can slice the hierarchy in the middle like that without intervention on the patch lines. |
| 10:14 |
Bmagic |
that's what I was thinking, which brought me to: "faster to do it manually" |
| 10:15 |
Dyrcona |
If the hierarchy changes are regular, then sed patch the patch for you. |
| 10:15 |
Bmagic |
Using Beyond Compare to point out the differences between master:docs/* and our collab:docs/* - I can see exactly what I need to do |
| 10:16 |
Dyrcona |
I dropped some words on my sed comment. Basically, sed could fix the patch files so that they apply if the new file hierarchy is internally consistent relative to the old one. |
| 10:17 |
Bmagic |
Drycona: alright, let me work out a clever sed command and see where that takes me |
| 10:19 |
Dyrcona |
s docs/\([^/]+\)/\(.+\)@docs-antora/modules/\1/pages\2@ <- Ought to work for your example |
| 10:19 |
Dyrcona |
Oops. missing a / after pages... |
| 10:20 |
Bmagic |
ah, I was going a different direction, I see what you mean |
| 10:22 |
Bmagic |
hmm, that didn't make any changes |
| 10:22 |
Dyrcona |
Sometimes, these things are more easily done manually when the hierarchy changes. I used Emacs' emerge-mode a lot on the cwmars custom template hierarchy. |
| 10:23 |
Dyrcona |
Well, depends on the version of sed and the flags used. Try removing the \ in front of the parenthesis. |
| 10:23 |
Dyrcona |
I probably typed what works with FreeBSD sed with no options, which probably doesn't work with GNU sed the same way. |
| 10:23 |
Bmagic |
broke it |
| 10:24 |
Bmagic |
it likes the escaped ( |
| 10:24 |
Dyrcona |
Add -E with no backslashes. |
| 10:24 |
Dyrcona |
sed -E -e .... |
| 10:25 |
Bmagic |
though it should be a capture group which makes me think it shouldn't be escaped |
| 10:25 |
Dyrcona |
Anyway, it may be better to do manually or with tools designed to merge different files. There is a way in git to merge two file hierarchies. |
| 10:25 |
|
mmorgan joined #evergreen |
| 10:26 |
Dyrcona |
Bmagic: Depends on the options you use and the "brand" of sed, unfortunately. Depends on regular vs. extended regex options, and it's some old UNIX baggage. |
| 10:26 |
Bmagic |
sed -E sdocs/([^/]+)/(.+)@docs-antora/modules/\1/pages/\2@ < ~/patch.patch > sed.patch |
| 10:27 |
Bmagic |
gets mad at the lack of escaped ( |
| 10:27 |
Bmagic |
putting the escaped ( back in and it works but the result is: "diff --git a/docs/RELEASE_NOTES_3_3.adoc b/docs/RELEASE_NOTES_3_3.adoc" becomes "diff --git a/docs-antora/modules/1/pages/2" |
| 10:28 |
Dyrcona |
Quote the command string. It's the shell complaining. |
| 10:31 |
Bmagic |
now it doesn't like the \2 |
| 10:31 |
Bmagic |
tried double and single quotes |
| 10:31 |
Dyrcona |
Is this in your antora branch? I'm going to check it out and play with git. I remember doing something like this entirely in git, but that was comparing Open-ILS/src/templates with Open-ILS/src/cwopac-templates where the hierarchy was the same. |
| 10:32 |
Dyrcona |
Double quotes would require \\, I don't think single quotes should. |
| 10:32 |
Bmagic |
the patch file isn't - but everything else you need should be there |
| 10:32 |
Bmagic |
https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/collab/blake/LP1848524_antora_ize_docs |
| 10:35 |
Dyrcona |
yeah. got it. thanks! |
| 10:37 |
Bmagic |
I think I got it working |
| 10:37 |
Bmagic |
the pattern, though, needs to include the second half of the diff clause |
| 10:42 |
Bmagic |
this isn't working but something like this? sed -E 'sa/docs/([^/]+)/(.+)\sb/docs/([^/]+)/(.*)@a/docs-antora/modules/\1/pages/\2 b/docs-antora/modules/\3/pages/\4@' |
| 10:44 |
csharp |
re: the automation thing, I went into a rabbit hole looking at the launchpadlib python library to see if it would be easy to write scripts to manage some of the bug wrangling stuff |
| 10:44 |
csharp |
iirc, Dyrcona investigated that a while back |
| 10:44 |
csharp |
didn't get very far with it |
| 10:45 |
Dyrcona |
I had script that worked, but stopped working sometime around 2016. |
| 10:45 |
Dyrcona |
gmcharlt was sent copies. |
| 10:46 |
* gmcharlt |
reads up - do you need that dug out? |
| 10:46 |
csharp |
some of the tools referred to in the docs appear to rely on Python 2.X |
| 10:46 |
csharp |
gmcharlt: if it's not too difficult, I'm just curious :-) |
| 10:49 |
Dyrcona |
csharp: I can send them to you later. They were Python 2.x sciprts, IIRC. |
| 10:49 |
csharp |
Dyrcona: thanks! |
| 10:50 |
csharp |
again - not a proirity :-) |
| 10:50 |
gmcharlt |
yeah, I don't appear to have my copy immediately at hand |
| 10:51 |
Dyrcona |
Bmagic: I have some command line magic that may help: https://pastebin.com/Y36xiGDH |
| 10:52 |
Dyrcona |
It gets a bunch of the files, but not all. |
| 10:52 |
Bmagic |
I'll give it a whirl |
| 10:54 |
Bmagic |
Dyrcona: very nice - only issue is that many many of these changes were intentional on the antora side |
| 10:54 |
Bmagic |
namely the asciidoc heading changes to moustache style |
| 10:54 |
Dyrcona |
Well, then, you're on your own. |
| 10:55 |
Bmagic |
the changes that we are interested in are those that are different between master:/docs/* and collab/HEAD~1/docs/* (before rebase) |
| 10:56 |
Bmagic |
the sed command was getting real close |
| 10:56 |
Dyrcona |
Well, keep working at it, then. |
| 10:56 |
Bmagic |
Dyrcona++ |
| 10:59 |
Bmagic |
we might have a winner: sed -E 's@^diff\s+.*?git\s+a/docs/([^/]+)/(.+)\s+b/docs/([^/]+)/(.*)$@diff --git a/docs-antora/modules/\1/pages/\2 b/docs-antora/modules/\3/pages/\4@' < ~/patch.patch > sed.patch |
| 10:59 |
Dyrcona |
csharp: I have them in a zip file. I'll share it on Google Drive. |
| 10:59 |
Dyrcona |
Bmagic++ # Looks tricky enough to be correct. :) |
| 11:00 |
Bmagic |
haha |
| 11:06 |
csharp |
Dyrcona++ # thanks! |
| 11:07 |
Dyrcona |
YW! |
| 11:07 |
|
sandbergja joined #evergreen |
| 11:12 |
|
sandbergja joined #evergreen |
| 11:18 |
|
sandbergja joined #evergreen |
| 12:08 |
|
mrisher joined #evergreen |
| 12:13 |
|
jihpringle joined #evergreen |
| 12:24 |
|
rjackson_isl_hom joined #evergreen |
| 12:29 |
|
khuckins joined #evergreen |
| 14:17 |
|
sandbergja joined #evergreen |
| 14:31 |
Dyrcona |
gmcharlt: Are these still the operative instructions for making an OpenSRF release, modulo the version numbers? https://wiki.evergreen-ils.org/doku.php?id=dev:release_process:opensrf:2.0 |
| 14:31 |
gmcharlt |
Dyrcona: yes |
| 14:31 |
Dyrcona |
Thanks! |
| 15:24 |
|
Dyrcona joined #evergreen |
| 15:34 |
|
mantis1 left #evergreen |
| 15:51 |
|
jihpringle joined #evergreen |
| 15:59 |
|
sandbergja joined #evergreen |
| 16:19 |
|
rfrasur joined #evergreen |
| 16:45 |
Bmagic |
What methods are folks using to prevent holds from causing items to transit? In other words: we want patrons to be able to place holds - just not fill the hold with anything other than an item at the pickujp location |
| 16:49 |
mmorgan |
Bmagic: We plan on using closed dates and library settings to only target local holds for the pull list. |
| 16:49 |
mmorgan |
Won't stop checked in items from transiting, just keep Available items from being pulled to transit. |
| 16:52 |
berick |
hard hold boundaries? |
| 16:56 |
mmorgan |
berick: That would require updating all the holds, right? |
| 16:57 |
berick |
mmorgan: yes, I believe it would |
| 17:02 |
Bmagic |
This library setting seems to be exactly the right thing: "Suppress hold transits group" - but we've never seen it in action |
| 17:04 |
mmorgan |
Bmagic: If I'm remembering right, that will put items on the holdshelf at the checkin library, even if the pickup point was elsewhere. |
| 17:04 |
mmorgan |
Won't keep the hold from capturing. |
| 17:05 |
Bmagic |
I see - so it's like a "single holds shelf for the group of specified libraries" ? |
| 17:05 |
Bmagic |
it's still on hold for the remote patron - just physically sitting on the local shelf |
| 17:06 |
mmorgan |
Bmagic: If I'm remembering correctly, yes. |
| 17:06 |
Bmagic |
makes sense |
| 17:08 |
mmorgan |
We considered that functionality when it was first introduced, thinking to have Adult and Children's in the same library be different org units. |
| 17:09 |
mmorgan |
You could suppress transits and hold transits between those two org units because they're really in the same building. |
| 17:13 |
|
jvwoolf left #evergreen |
| 17:22 |
|
mmorgan left #evergreen |
| 17:54 |
|
rjackson_isl_hom joined #evergreen |
| 18:00 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 21:16 |
|
sandbergja joined #evergreen |
| 22:11 |
|
sandbergja joined #evergreen |
| 22:26 |
|
sandbergja joined #evergreen |
