You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
A clear and concise description of what the bug is.
To Reproduce
Steps to reproduce the behavior:
Add 2+ messages with file and line to Dangerfile.swift
Run danger-swift ci on gitlab instance
Expected behavior
The first inline comment is created, then overriden by the second as a general comment
Screenshots
Your Environment
danger-swift 3.17.1
software
version
installation method
homebrew
Operating System
macOS
Additional context
As I understand there is a connected issue fixed for Github. Is it possible to adopt the same fix for Gitlab or is it completely different?
The text was updated successfully, but these errors were encountered:
As I understand there is a connected issue for Github. Is it possible to adopt the same fix for Gitlab or is it completely different?
I would certainly examine the linked PR to see if it teaches you anything, but I wouldn’t expect it to necessarily be exactly related. The fundamentals of cross-provider DangerJS behavior has been pretty stable for a long time. GitLab is a relatively newer provider so I’d expect the issue to be in one of these places:
I think more likely, the problem is a misunderstanding about GitLab’s automatic behaviors, for example maybe GitLab has only one “slot” for a line-level comment, which means the last one always wins. In that case, you might be able to preprocess all the line-level comments, and merge all the ones that target the same/file-line into a single comment. (I even believe this is doable in user-space)
Describe the bug
A clear and concise description of what the bug is.
To Reproduce
Steps to reproduce the behavior:
Expected behavior
The first inline comment is created, then overriden by the second as a general comment
Screenshots
Your Environment
danger-swift 3.17.1
Additional context
As I understand there is a connected issue fixed for Github. Is it possible to adopt the same fix for Gitlab or is it completely different?
The text was updated successfully, but these errors were encountered: