On Mon, Sep 16, 2019 at 10:54:10PM +0200, SZEDER Gábor wrote: > > Yes on that final bit. We could always fall back to (4) if the terminal > > information is not available, but given that the benefit is mostly in > > simplifying the code, I don't know if it's worth carrying around _two_ > > solutions. > > Ok, so here is a patch to revert 5b12e3123b (progress: use > term_clear_line(), 2019-06-24) with proper explanation. > > As a bonus there is a new test script exercising the progress display > as well, in particular how it covers up the previous progress line, so > we may have a bit more confidence in it. Thanks for doing this. It's especially nice to get test coverage for the progress meters, which have traditionally been neglected. I'm a little late on my review, as it looks like this has already hit next, but it looks pretty good to me. My only complaint is that I think putting the new "private" bits of the progress API into the header (with a comment) is a lesser evil than re-declaring them in test-progress.c (if only because the compiler could tell us if the two get out of sync). But I can live with it either way. -Peff
