| Time |
Nick |
Message |
| 07:31 |
|
collum joined #evergreen |
| 07:51 |
|
RFrasur joined #evergreen |
| 08:14 |
|
mantis1 joined #evergreen |
| 08:44 |
|
Dyrcona joined #evergreen |
| 09:43 |
Dyrcona |
Hmm. Anyone else noticed that %z isn't working in TT2's Date plugin? |
| 09:43 |
Dyrcona |
I wonder if it is the underlying implementation, i.e. the POSIX Perl lib, or even a bug in the O/S C library? |
| 09:44 |
Dyrcona |
S'pose I could write programs to check those two conditins... |
| 09:54 |
Dyrcona |
strftime works from C. |
| 10:00 |
Dyrcona |
%Z works in the Date plugin, but I want %z. |
| 10:08 |
Dyrcona |
%z works in POSIX::stftime... |
| 10:08 |
Dyrcona |
stftime...stupid fingers. |
| 10:08 |
Dyrcona |
Ha! /me gives up. |
| 10:11 |
Dyrcona |
I suppose that I can hard code this for the given program, but this is bothersome. |
| 10:12 |
Dyrcona |
Looks like it is a problem on Ubuntu 18.04, 20.04, and 22.04. |
| 10:24 |
JBoyer |
And a bunch of Debian releases. I wanted to use it in the Date header in email A/T templates but couldn't, so we ended up with a hard-coded -0000 instead, since that means (if memory serves) that the tz offset is meaningless (as opposed to +0000 which is UTC) |
| 10:25 |
JBoyer |
(meaningless as defined by whatever email RFC mentions the Date header) |
| 10:27 |
Dyrcona |
JBoyer++ |
| 10:27 |
Dyrcona |
Well, since %Z works, I was thinking of doing a check for EDT or EST and setting it to -0400 or -0500 depending... |
| 10:28 |
Dyrcona |
Since I'm going to run the thing that I'm working on in a week or two, I'll hard code -0400 for now. |
| 10:29 |
Dyrcona |
I may try the latest TT2 from github to see if that makes a difference, but not right now. |
| 10:31 |
Dyrcona |
Or maybe CPAN. |
| 10:33 |
Dyrcona |
I also intended to use this in our email notices later. |
| 10:40 |
|
RFrasur joined #evergreen |
| 11:27 |
|
jihpringle joined #evergreen |
| 11:35 |
Dyrcona |
Hm... Joins on IN don't seem to be working.... |
| 11:38 |
Dyrcona |
Well, no, they might be working. My data dump is kind of big, and it gets bigger with the IN.... It seems our org hierarchy is a bit off. |
| 11:42 |
Dyrcona |
Lol, what.... diff tells me two lines are different, but I don't see it. |
| 11:46 |
Dyrcona |
Ok. The different join criteria was causing the rows to have a slightly different order. Fixed that. |
| 11:51 |
Dyrcona |
It also looks like the IN doesn't do what I wanted. The orgs that didn't have settings before still don't have settings which means we never set the settings for those orgs. |
| 15:36 |
|
jihpringle joined #evergreen |
| 15:44 |
Dyrcona |
Two many nots ties your query optimizer up in knots. |
| 15:51 |
Dyrcona |
Tiu? Too much typing for one day. |
| 16:01 |
miker |
Dyrcona: ugh, that's annoying, because: "The Date plugin provides an easy way to generate formatted time and date strings by delegating to the POSIX strftime() routine." ... allegedly? |
| 16:02 |
Dyrcona |
Yeahp. And, whatever I just did to my query makes it run a bit longer.... |
| 16:03 |
Dyrcona |
Too many "in"s and outs. |
| 16:03 |
miker |
Dyrcona: oh! "To maintain backwards compatibility, using the %z placeholder in the format string (to output the UTC offset) currently requires the use_offset parameter to be set to a true value. This can also be passed as the fifth parameter to format (but the former will probably be clearer)." ... so, maybe `date.format( ..., use_offset = 1)` in the template? |
| 16:04 |
Dyrcona |
Oh. I missed that. i'll give it a try. |
| 16:04 |
miker |
and the "that" I'm referring to in both is T::P::Date |
| 16:05 |
Dyrcona |
Yeah. Also, my query is a lot faster without the extra in... :) |
| 16:05 |
miker |
well, that's good, too! :) |
| 16:09 |
Dyrcona |
[% date.format(format = '%a, %d %b %Y %H:%M:%S %z', use_offset = 1) %] doesn't seem to be working for me. |
| 16:15 |
jeff |
metacpan should support T::P::Date style search queries (it doesn't) |
| 16:15 |
jeff |
(my brain wasn't quite jumping to the right place to get to Template::Plugin::Date) |
| 16:18 |
jeff |
changelog indicates that %z was introduced in 3.005 in 2020 |
| 16:20 |
Dyrcona |
libtemplate-perl is 2.27 on Ubuntu 20.04 |
| 16:21 |
jeff |
stable libraries are stable. |
| 16:21 |
Dyrcona |
Same on ubuntu 22.04. :( |
| 16:21 |
Dyrcona |
Guess I could install it from CPAN. |
| 16:21 |
Dyrcona |
jeff++ miker++ |
| 16:21 |
jeff |
but occasionally I'm tempted to run everything from a non-system perl using perlbrew or similar. :-) |
| 16:21 |
Dyrcona |
But, that's for tomorrow. |
| 16:22 |
Dyrcona |
I install all but the base perl from CPAN on my manjaro lapt. i keep it all in ~/perl5 |
| 16:22 |
Dyrcona |
OOf... fingers won't cooperate. I think a mutiny is at hand. |
| 16:23 |
miker |
but /which/ hand?! |
| 16:23 |
Dyrcona |
:) |
| 16:27 |
* Dyrcona |
calls it a day. got a long drive home. |
| 16:29 |
JBoyer |
jeff++ |
| 16:30 |
JBoyer |
I was just about to post that, but too slow. :) |
| 17:30 |
Bmagic |
B&T created an alternate server that is working in PASV mode, still in test mode ATM. But, FYI RFrasur Dyrcona (niether are here, but hey, there's IRC logs) |
| 17:30 |
Bmagic |
/niether/neither |
| 17:43 |
jmurray-isl |
Bmagic++ |
| 19:57 |
|
Dyrcona joined #evergreen |
| 19:59 |
Dyrcona |
miker: I got the timezone to work with Template::Toolkit 3.100 installed from CPAN. I had to remove use_offset=1: [% date.format(date.now, format = '%a, %d %b %Y %H:%M:%S %z') %] |
| 20:06 |
Dyrcona |
For the logs, I had to install Template::Toolkit and Template::Plugin::POSIX. |
| 20:11 |
Dyrcona |
Anyway, that's all I've got for now. More tomorrow. |
