Closed soli closed 10 years ago
Thanks for the contributions :+1: They're merged via the command line - so don't worry when you see this pull request just got closed, the commits are in.
As for the 'emacs' copy mode update, I left it out because it seems it's not working. This line for vi copy-mode goes to the bottom of the scrollback buffer, not top.
About the emacs mode stuff, precisely, from the man page (I never use emacs mode) M-<
is the same as G
not M->
… but no problem, I'm still looking for a way to avoid that jumping completely (on my laptop it is cleary visible, i.e. the buffer is redrawn after G
and then again after jumping to the proper line…
Yep, the plugin is not perfect. I also get small "flicker" when jumping over the results. It's more noticable in emacs copy mode because it's slower.
Frankly, I think the only foolproof solution would be enabling regex search in tmux itself. If we had that, it would be easy to script everything else on top.
TBH, I currently run the plugin without the G
line, but with:
if [ "$TMUX_COPY_MODE" == "vi" ]; then
# vi copy mode, put cursor at bottom of the window
tmux send-keys L
fi
at the end of _copycat_go_to_line_with_jump
and I do notice less flicker. But I still have no idea how to do that for emacs :-/
That's cool if L
works for you.
I remember going back and forth about having G
, see removing it here,
then reverting the previous commit.
Unfortunately I didn't document the revert properly. I think the issue was that some jumps (n
and N
) were inconsistent and doing wrong highlights when there were no G
.
This goes back to tmux not handling jumps with :
in copy mode properly (read buggy as hell).
I had all sorts of bugs like this, mostly in tmux
, but in OS X versions of grep
and awk
tools too when working on the plugin.
Feel free to communicate if you have any code improvement in mind! I'd love to accept pull requests, but it would be a pity if you worked on something while not knowing about one of those bugs.
And kudos for getting into the code base. I thought no one ever will have the guts to dive into it.
Yeah, I indeed ran into all sorts of weird things when first removing G
completely. However L
seems stable for now, I'll keep you posted.
Thanks for writing that code base in the first place :wink:
Hi, I just want to notify you about the mistake I made with git - I had to rewrite the history. In case you'll hack on the project, this might mess your local git repo..
The explanation and a proposed solution is described here. I just wanted to personally let you know this since you showed interest in the project.
Hehe, no big deal, as they say "git happens" ;) I was a bit puzzled about the clone time too, so it's probably a good idea to have removed the video.
A few minor changes, mostly the docs part.
I changed the M-> to M-< for emacs copy mode to be coherent with vi. However, to avoid unnecessary visual glitches, I'd prefer to have
send-keys L
before adding padding instead of needing to scroll back and forth to the beginning. UnfortunatelyL
does not seem to have any emacs counterpart, nor any command name :(