The following comment is from issue #3224 that I opened because I didn't find this issue when I did a superficial search.
The history --delete command is broken. If your history contains multiline commands each line is listed as an individual entry you can select for deletion. That's obviously bogus. It might be possible to fix that. However, I'm wondering if it wouldn't make more sense to simply change the UI to eliminate selecting individual entries for deletion. Just give the user the choice of deleting all or none of the matching entries.
Also, including the recently introduced -t or --with-time flag produces output that makes selecting an individual entry for deletion impossible. Yet another reason to remove that capability.
This fixes several problems with how the builtin `history` command handles
arguments. It now complains and refuses to do anything if the user specifies
incompatible actions (e.g., `--search` and `--clear`). It also fixes a
regression introduced by previous changes with regard to invocations that
don't explicitly specify `--search` or a search term.
Enhances the history man page to clarify the behavior of various options.
This change is already far larger than I like so unit tests will be added
in a separate commit.
Note: This fixes only a couple problems with the interactive `history
--delete` command in the `history` function. The main problem will be
dealt with via issue #31.
It seems like the only sensible solution is adding another flag to request that the history search output have newlines converted to \n and backslashes escaped (i.e., \ -> \\). Similar to, if not identical to, running each history item through string escape --no-quoted before writing it to stdout. The caller can then easily and safely unescape the search output back to its original form for display and other processing. Even if --with-time or -t was also used.
It is unlikely this will be implemented in time for the 2.4.0 release but I'm going to focus on fixing this any way. Fixing this will make a nice finale to the other history improvements that have been made in the past couple of months.
Yes. I was worried about potentially having an embedded NUL in the history but that's impossible today and we should prohibit it in the future if we ever switch from "char *" style strings to C++ string objects in all the history code paths.