| Time |
Nick |
Message |
| 00:08 |
|
Christineb joined #evergreen |
| 05:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:20 |
|
agoben joined #evergreen |
| 08:48 |
|
mmorgan joined #evergreen |
| 08:54 |
|
Dyrcona joined #evergreen |
| 09:12 |
|
Bmagic joined #evergreen |
| 09:29 |
|
jvwoolf joined #evergreen |
| 09:38 |
jvwoolf |
Hi everybody! I'm trying to add a new page to our OPAC. It's working in test, but I have to sign in to view it. I've scoured the docs, but I can't figure out how to make it public. Any tips? |
| 09:40 |
|
collum joined #evergreen |
| 09:46 |
Dyrcona |
jvwoolf: I think that's largely a function of where you put the file. |
| 09:46 |
tsbere |
jvwoolf: You will need to edit the TPAC perl code to add your page before the auth check |
| 09:51 |
tsbere |
jvwoolf: More specifically, Open-ILS/src/perlmods/lib/OpenILS/WWW/EGCatLoader.pm, load function, probably add a return $self->load_simple("page") if $path =~ m|opac/page|; line before the "Everything below here requires SSL" block |
| 09:52 |
tsbere |
(unless you want it to require SSL, then jam it before the "Everything below here requires authentication" block instead) |
| 09:56 |
Dyrcona |
tsbere++ |
| 09:58 |
* tsbere |
wonders if that should, in fact, be documented >_> |
| 10:00 |
mmorgan |
Back to what Dyrcona said, if the file isn't put under the myopac directory, will it be visible without a sign in? |
| 10:01 |
|
barbara joined #evergreen |
| 10:02 |
tsbere |
mmorgan: The fallthrough ends up triggering SSL and then auth for any unknowns in the entire /opac tree. Outside of the /opac tree will, of course, vary. |
| 10:03 |
tsbere |
The /kpac tree, for example, has a different list of pages and is in EGKPacLoader.pm instead |
| 10:05 |
Dyrcona |
mmorgan jvwoolf: Yes, you can just put the file in /openils/var/web, but it will be outside of the /opac area. |
| 10:06 |
Dyrcona |
robots.txt works from there, for instance. |
| 10:10 |
jvwoolf |
tsbere++ |
| 10:10 |
mmorgan |
So just to clarify, if a new file, somefile.tt2 is put in /openils/var/templates/opac/parts, will an edit to EGCatLoader.pm as tsbere describes be required to access somefile.tt2 without authentication? |
| 10:10 |
jvwoolf |
Dyrcona++ |
| 10:11 |
tsbere |
mmorgan: That is indeed the situation. Other features may require further edits as well. |
| 10:11 |
tsbere |
mmorgan: Note that "including" somefile.tt2 in a pre-existing file won't apply, though. Only the initially accessed template needs editing done. |
| 10:13 |
mmorgan |
Interesting. Then I would agree with tsbere that it should be documented. |
| 10:13 |
mmorgan |
Dyrcona++ |
| 10:13 |
mmorgan |
tsbere++ |
| 10:19 |
Dyrcona |
Oh, yeah. I assumed we're talking about plain html. |
| 10:20 |
Dyrcona |
If you're doing template, it has to go in with the templates and things edited as tsbere said, just to be perfectly clear. |
| 10:24 |
|
bmills joined #evergreen |
| 10:43 |
jvwoolf |
OK, I edited the perl script. Do I need to restart the whole shebang? |
| 10:44 |
Dyrcona |
jvwoolf: You shouldn't have to restart anything. |
| 10:44 |
Dyrcona |
Err... Wait... That's for templates. You may have to reload apache. |
| 10:45 |
tsbere |
jvwoolf: To clarify further: You edited the perl script, did you install the new version? |
| 10:45 |
* Dyrcona |
assumed she edited it in place, which are both terrible ideas: (asusming and editing in place). |
| 10:49 |
jvwoolf |
tsbere: The new version of perl? |
| 10:49 |
jvwoolf |
Dyrcona: It's a tt2 file |
| 10:49 |
tsbere |
jvwoolf: The EGCatLoader.pm file |
| 10:49 |
Dyrcona |
jvwoolf: You edited EGCatLoader.pm, right? That's what tsbere and I are asking about. |
| 10:50 |
Dyrcona |
If you change a tt2 file, you don't have to do anything. The changes should just show up the next time you load the page. |
| 10:50 |
Dyrcona |
If you change a Perl module, you'll need to reload Apache. |
| 10:51 |
tsbere |
And if you changed it in your source directory you need to install the new version |
| 10:51 |
tsbere |
(that applies to tt2 and perl, technically) |
| 10:54 |
jvwoolf |
Yes, I edited EGCatLoader.pm. This is a new tt2 file that didn't exist before. |
| 10:54 |
jvwoolf |
Can you clarify why I need the new version of perl? Just curious. |
| 10:54 |
tsbere |
jvwoolf: Not perl itself, but EGCatLoader.pm needs to be updated in the place it gets installed to |
| 10:55 |
Dyrcona |
jvwoolf: You need to tell EGCatLoader.pm how to load the tt2 file so that it knows where it shows up in the path. |
| 10:55 |
Dyrcona |
If you don't edit EGCatLoader.pm, it doesn't know the tt2 file exists. |
| 10:56 |
Dyrcona |
Pretty much everything under /eg/opac goes through EGCatLoader.pm and family. |
| 10:56 |
tsbere |
jvwoolf: For example of what I am saying, you may have a copy under /usr/local/share/perl/<version>/OpenILS/WWW/EGCatLoader.pm (with <version> being the relevant perl version) - That is the one that will be used, not the one in your source directory. |
| 10:57 |
tsbere |
The make install step updates that stuff for you, or you can see about manually updating it. |
| 10:57 |
jvwoolf |
OK, that's where I edited it. |
| 10:58 |
Dyrcona |
And, that's what I thought....You'll want to make a copy of that file and make a diff of the changes or you'll lose them at the next upgrade. |
| 10:58 |
Dyrcona |
If you run bricks, you'll need to update it on every brick, too. |
| 10:59 |
jvwoolf |
Ahh. Gotcha. |
| 10:59 |
jvwoolf |
I'm only doing this in test at the moment. |
| 10:59 |
Dyrcona |
OK, but for production, I'd recommend doing as tsbere suggests and making the change in the source and then copying the final version into place. |
| 11:00 |
Dyrcona |
Git really helps with managing little changes like this, too. |
| 11:02 |
jvwoolf |
I've got it now. Thank you. |
| 11:02 |
jvwoolf |
Dyrcona++ |
| 11:02 |
jvwoolf |
tsbere++ |
| 11:31 |
|
jihpringle joined #evergreen |
| 11:48 |
|
dbwells__ joined #evergreen |
| 13:04 |
|
jvwoolf joined #evergreen |
| 13:08 |
|
NawJo joined #evergreen |
| 13:11 |
_bott_ |
Badges seem to be giving us some issues. Seemingly random searches fail and throw an error relating to a copy_location_group |
| 13:11 |
_bott_ |
http://pastebin.com/raw/y0XmXTYU |
| 13:12 |
_bott_ |
Search yields no results. Moments later, the same search will succeed. |
| 13:13 |
tsbere |
_bott_: I have seen that kind of thing in my own scripts when things weren't loading in the correct order |
| 13:15 |
tsbere |
_bott_: Though I do wonder about the copy_location_group bit, as I don't see anything specific jumping out in the search that would require looking one up in the first place? |
| 13:15 |
_bott_ |
tsbere: we do have a couple location groups, but nothing relating to the search, no |
| 13:20 |
|
Christineb joined #evergreen |
| 13:26 |
Dyrcona |
So, why would an action_trigger.event be set to complete and have a null complete_time? I seem to have a lot more of these that I do complete events with a complete_time actually set. |
| 13:27 |
tsbere |
I dunno. I considered digging into that at one point but decided it wasn't worth figuring out. |
| 13:27 |
tsbere |
I suspect it has to do with grouped events, though |
| 13:28 |
Dyrcona |
Yes, I think the grouped events might have something to do with it. |
| 13:29 |
Dyrcona |
One class of these seem to be print overdue notices that are generated after the overdue notice event. |
| 13:29 |
* tsbere |
has not dug any deeper than that, and that is from memory |
| 13:29 |
Dyrcona |
OK. I'll not worry about it. |
| 13:30 |
tsbere |
Grouping may be a false positive on figuring that out too, for that matter. It wasn't important enough to me to figure out ;) |
| 13:30 |
Dyrcona |
I asked because I was considering changing a script that uses run_time to use complete_time, but none of these event have a complete_time. |
| 13:30 |
Dyrcona |
:) |
| 13:31 |
tsbere |
What about start_time? I don't think I have seen that one skipped over before. |
| 13:31 |
Dyrcona |
Right. I was just starting to type that I may switch to start_time instead. |
| 13:31 |
Dyrcona |
Maybe throw in a state = 'complete' clause as well. |
| 13:33 |
Dyrcona |
Might recover better from failures like we had this weekend using start_time instead of run_time. I dunno for sure, though. |
| 13:37 |
_bott_ |
Odd. It seems my search issues may begin when someone does include a location_group in their search. Subsequent searches fail. |
| 13:41 |
JBoyer |
_bott_: maybe a url parameter not being cleared out from subsequent searches? That's been an issue in the past for some things. |
| 13:48 |
_bott_ |
My quick kludge is to null out any location_group in the badge section of metabib.pm. Seems to be effective in eliminating my problem. |
| 14:30 |
miker |
_bott_: do you have a DB error log entry? |
| 14:33 |
berick |
miker: he pasted something up around 13:11 |
| 14:34 |
_bott_ |
miker: based on the query duration, I do have a full query as well |
| 14:34 |
_bott_ |
http://pastebin.com/raw/Jn7ew6YA |
| 14:34 |
miker |
so he did... OK, that looks like a cdbi-and-friends issue |
| 14:35 |
miker |
I'll see if I can find a minute to look a that later today |
| 14:48 |
miker |
_bott_: I think http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/miker/aclg-cdbi-for-badges will help you... |
| 14:52 |
_bott_ |
ah, makes sense. I'll give it a whirl. |
| 16:00 |
mmorgan |
I'm looking at the 2.11 SIP feature that treats location as workstation. I'm not getting it to work on our test system. |
| 16:01 |
mmorgan |
I can make connections and perform transactions, but the location isn't being recognized as the workstation. |
| 16:01 |
mmorgan |
Any ideas? |
| 16:03 |
mmorgan |
Also tried adding the location_code to the config file for the staff user and that's not working either. |
| 16:05 |
tsbere |
mmorgan: Just to be sure, did you make sure that both Evergreen and SIPServer are up to date? |
| 16:09 |
mmorgan |
Yes, I checked Sip.pm and SIPServer.pm, both are up to date. |
| 16:11 |
tsbere |
and you either defined client_location_code as "true" for the institution block (passing in a valid workstation name in the login step) or location_code to a valid workstation name? |
| 16:14 |
mmorgan |
I have a valid workstation, but maybe it's the client_location_code = true that's missing. Checking that. |
| 16:40 |
mmorgan |
Hmm. Still no luck. Will need to revisit this tomorrow |
| 16:41 |
mmorgan |
tsbere++ |
| 16:41 |
mmorgan |
Thanks for the suggestions. |
| 17:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 17:02 |
|
mmorgan left #evergreen |
| 17:16 |
|
jvwoolf left #evergreen |
| 17:33 |
|
bmills joined #evergreen |
