; Silence byte compiler in erc-netsplit-JOIN

* etc/ERC-NEWS: Partially revert edit from c9f1ad2a87 "Revive option
erc-query-on-unjoined-chan-privmsg".
* lisp/erc/erc-netsplit.el (erc-netsplit-JOIN): Silence byte compiler
warning re ignored return value from `delete' when removing nicks.
Could probably suppress rather than reconstitute since the CAR of an
`erc-netsplit-list' entry originates from the trailing "reason" param
of the instigating "QUIT" command and should look something like
"irc.example.org chat.example.org", which cannot be confused for a
nickname.
* test/lisp/erc/resources/erc-scenarios-common.el: Fix wording in
Commentary.
This commit is contained in:
F. Jason Park 2023-05-18 23:47:27 -07:00
parent 71622d70e8
commit 8120f5156c
3 changed files with 6 additions and 4 deletions

View file

@ -366,8 +366,8 @@ In an effort to help further tame ERC's complexity, the variable
'erc-default-recipients' is now expected to hold but a single target.
As a consequence, functions like 'erc-add-default-channel' that
imagine an alternate, aspirational model of buffer-target relations
have been deprecated. Grep for their names in ChangeLog.4 for
details.
have been deprecated. For specifics, see entries in Emacs'
ChangeLog.4 from around June 30, 2022.
A number of less consequential deprecations also debut in this
release. For example, the function 'erc-auto-query' was deemed too

View file

@ -117,7 +117,9 @@ join from that split has been detected or not.")
parsed 'notice (process-buffer proc)
'netjoin-done ?s (car elt))
(setq erc-netsplit-list (delq elt erc-netsplit-list)))
(delete nick elt))
;; Avoid `ignored-return-value' warning for `delete'.
(let ((tail (nthcdr 2 elt))) ; (t n1 ... nN)
(setcdr tail (delete nick (cdr tail)))))
(setq no-next-hook t))))
no-next-hook))

View file

@ -51,7 +51,7 @@
;; argument, a `let*'-style VAR-LIST. Relying on such a macro is
;; unfortunate because in many ways it actually hampers readability by
;; favoring magic over verbosity. But without it (or something
;; similar), any failing test would cause all subsequent tests in this
;; similar), any failing test would cause all subsequent tests in a
;; file to fail like dominoes (making all but the first backtrace
;; useless).
;;