Default values the way the docs suggest we do#1
Conversation
Codecov Report
@@ Coverage Diff @@
## master #1 +/- ##
============================================
+ Coverage 68.94% 69.32% +0.38%
- Complexity 58 60 +2
============================================
Files 1 1
Lines 161 163 +2
============================================
+ Hits 111 113 +2
Misses 50 50 |
|
@tvdijen since you've made a number of changes, I want to make sure this bug fix isn't going backwards on anything you've done? Also at what point do we create a new release? |
|
Any changes I've made were related upgrading dependencies or taking advantage of typehinting... I see I've also split some code to reduce complexity.. No real changes. |
|
I've changed the dependency on SSP from |
|
Yeah, I realised when I did some testing that we couldn't tag as-is (hence the draft). I was trying to work out which of your changes were compatible with the current 1.x branch and merge those into a release-1.0 branch. Started on that, but hadn't finished :) |
|
You want your change to be in a 1.19 compatible version? In this case I would check out v1.1.2, push it to a release-1.0 branch and cherry-pick this PR on top.. Then tag it as v1.1.3. |
|
Yup, that's what I'm doing :) |
|
The release-1.0 branch now includes this patch, and tests fine against SSP 1.19.1. The draft is tagged against that now. |
2 participants
The documentation says we default process name and facility, but this doesn't happen if logdest=remote and a logconfig is set to point somewhere other than localhost (as is the case when sending to eduGAIN!). This change makes the behaviour match the documentation.