Agreed, thank you! I'm just trying things out at the moment on my spare time to see what's feasible. I'll add incremental regex once the basic stuff works and is accepted. C-r will have the existing emacs behaviour, of course.
On Sun, Nov 24, 2019 at 10:45 AM Nicholas Marriott < [email protected]> wrote: > If you want to add it, go for it, but the default C-r needs to stay > non-regexp like emacs. > > I suggest doing these changes in pieces rather than in one go. > > Thanks > > > On Sun, Nov 24, 2019 at 10:43:00AM -0800, Anindya Mukherjee wrote: > > Hi, in vi the incremental search is regex aware by default (unless we > use > > the nomagic option), while in emacs we can do C-r for non-regex and > > C-Meta-r for regex incremental search. The latter is actually quite > > useful, and I wonder if we can add this mode to tmux as well. Apart > from a > > slight refactoring it should not be too much additional trouble to > > implement. > > Best, > > Anindya > > On Thu, Nov 21, 2019 at 7:44 AM Anindya Mukherjee < > [email protected]> > > wrote: > > > > Ah great, sorry I misunderstood. That makes sense, thanks! > > On Thu, Nov 21, 2019 at 12:09 AM Nicholas Marriott > > <[email protected]> wrote: > > > > No that is not incremental search, that is search-again, it > should use > > regexps. Incremental search is search-forward-incremental where it > > updates the search every time you change the search term not only > when > > you press Enter. Try C-r with mode-keys set to Emacs. > > > > On Thu, 21 Nov 2019, 07:07 Anindya Mukherjee, < > [email protected]> > > wrote: > > > > Thanks! So say I start a / or Ctrl-S search with the regular > > expression "a..d", to pick a concrete example. If the text has > > multiple matches such as "axxd", "ayzd", they will all be > > highlighted. Then if the user presses n, incremental search runs > > with the first exact match, in this case "axxd". So in this > case all > > "axxd" will be highlighted and "ayzd" will not. Basically > > incremental will run with the first match encountered using > that as > > the search string (not regexp). Does that make sense? > > On Wed, Nov 20, 2019 at 1:01 PM Nicholas Marriott > > <[email protected]> wrote: > > > > I think search-forward/backward should change to use regexps > but > > incremental should not. > > On Wed, 20 Nov 2019, 20:59 Anindya Mukherjee, > > <[email protected]> wrote: > > > > Following up, is it desirable to add a new command to do > regexp > > search and leave the old commands as they are? This would > mean > > window_copy_regexp_search_... variants of some of the > existing > > functions and associated UI hookup. > > I was able to fix the line wrapped search issue mentioned > above, > > still doing character wise comparison, of course. Wondering > if > > it deserves a separate commit to the existing search > > infrastructure? > > Thanks! > > On Tue, Nov 12, 2019 at 9:29 AM Anindya Mukherjee > > <[email protected]> wrote: > > > > Yes my thoughts exactly :D > > On Tue, Nov 12, 2019 at 8:50 AM Nicholas Marriott > > <[email protected]> wrote: > > > > If you are adding regex search it would be a perfect > > opportunity to > > fix that also :-). > > > > On Tue, 12 Nov 2019 at 16:27, Anindya Mukherjee > > <[email protected]> wrote: > > > > > > Thanks for clarifying! > -- You received this message because you are subscribed to the Google Groups "tmux-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/tmux-users/CAN%2Bi5iNQXk_H_-e6cLPbxKPhk-U1XU_C6vVRCAzGOKG1PqfkAw%40mail.gmail.com.
