Disable codecov comments #1949
Disable codecov comments #1949
Conversation
Signed-off-by: Morris Jobke <[email protected]>
|
@MorrisJobke, thanks for your PR! By analyzing the history of the files in this pull request, we identified @LukasReschke to be a potential reviewer. |
|
👍 |
|
👍 🙏 thanks |
|
Why? I do dislike the removal hugely 👎 |
|
Its an easy way to see which new files have received tests and which don't etc… And most people here do simply ignore the results in the bar below. I'm totally against removing this. |
| @@ -6,6 +6,4 @@ coverage: | |||
| round: down | |||
| range: "70...100" | |||
|
|
|||
| comment: | |||
| layout: "header, diff, changes, sunburst, uncovered, tree" | |||
There was a problem hiding this comment.
At least header should really stay… Removing this is just another easy way to ignore coverage. Awesome! 🚀
There was a problem hiding this comment.
(don't care about the other values)
And "Jan is annoyed by being notified once on each of his pull request" is NOT a valid reason before that comes up. 😉 |
|
@LukasReschke I'm open to reconsidered, as I think there is a conflict of interest between developers focus on testing and others that are annoyed of the comments:grin:. But if you want to see changes, it should be the tree or diff view, not the header. |
|
I'm with @LukasReschke here. It is an easy overview. |
|
@LukasReschke @rullzer the reason for removing this is because it makes the discussion of a pull request hugely technical, and intimidating for any new contributors.
If you ignore the results in the bar below, then pushing them up everyone’s faces is not going to change that. It will just result in being annoyed. |
|
It’s like having a bot posting to an IRC channel – misunderstanding the medium. A Github thread is a discussion between humans, not a space for bots to dump their results. |
|
Well if people changed code they should add/change tests. Having a clear overview of what was added etc is a lot easier then telling. Click on the status of codecov. Then see where you need to add coverage. A github PR is the perfect place for that info. It is where we discuss if a new piece of code should go into the project. I'm sorry but a pull request is technical. |
|
This is just leading to having coverage ignored… Just had another PR today especially because the status simply got ignored. Jesus. Folks. This is proper software development and not a YOLO here. Just because some of you are like "OMG I GET NOTIFIED ONCE" we can't just throw away important tools ❗️❗️❗️ |
You gotta be kidding me, right? A PULL REQUEST IS TECHNICAL. THIS IS ABOUT CODE. Sorry to say it that clear. But this is bullshit. We do software development and not a "let's keep away any tools that tell you your code is bad." |
|
Revert in #1982 |
The feedback in the result bar is DIFFERENT from what is shown in the list. The feedback in the bar will also be green even if coverage effectively decreased (because it could relatively increase). See #1347 for the green bar which effectively is NOT good if you look at the actual comment by the bot. That people ignore it: Well. I don't care. At least other people that don't can actually do the right thing and tell people to fix their stuff before merging. I have a very low amount of tolerance for removing tools that assure code quality. |
Technical as in not (instantly) human readable. I think there is nothing wrong with the information itself, but with the way how a (obviously) machine-generated output mixes with a conversation of (hopefully) real people. It doesn't feel right, because it is embedded in the conversation context (and not distinguishable from people's comments), but not really part of that conversation. One or two sentences, describing the current status in human readable language (and those could of course be machine-generated) and a maybe link for more details would be just enough. But I guess that not a configurable option, so for now I would support the revert as well. |
Was requested in #1903 (comment)