@ntrel You seem to have misunderstood the point of my objection. The problem was the process, merging a change before others had the chance to assess it, not that there was a technical problem with the PR. Since as I noted I had not had time to see its contents I could not be commenting on the content.
Taken together though, my points justify the merge IMO.
This was not a crash problem or a sudden regression, it was an inconvenience that Geany users had been living with for a long time, there was no need to hurry a merge without giving others time to assess it. There is no release planned for tomorrow, so there is no rush to merge to enable distribution of the change to users. Therefore nothing justifies a quick merge.
There are no other cases. I thought we agreed that wrapping should not always be on, and this does not conflict with wrapping.
Again, I was not taking about the technical content of the PR, sorry if that was not clear. As I said above I was talking generally about the process, "cases" meant applying your sole judgement to the "obviousness" and complexity of the change to justify a quick merge for this and future PRs.
Having had a chance to look at the changes, why not put `reset_msgwin` in `ui_utils` and call it from both `build` and `search`, or is there some technical reason that the order of the calls needs to be different in build? If there is it probably should be commented so someone doesn't reorder them in the future.
Reverting is not the end of the world.
Agree totally, thats a very mature and sensible approach, reverting is not a criticism, it should be thought of in the same way as deleting a line of code if it needs to be changed, but some members of the Geany project seem to have a horror of it :-)