Package: emacs;
Reported by: epg <at> pretzelnet.org
Date: Thu, 5 Sep 2024 13:49:02 UTC
Severity: normal
Merged with 75906
Found in versions 30.0.90, 30.0.93
Done: Stefan Monnier <monnier <at> iro.umontreal.ca>
Bug is archived. No further changes may be made.
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 73041 in the body.
You can then email your comments to 73041 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
View this report as an mbox folder, status mbox, maintainer mbox
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 05 Sep 2024 13:49:02 GMT) Full text and rfc822 format available.epg <at> pretzelnet.org:bug-gnu-emacs <at> gnu.org.
(Thu, 05 Sep 2024 13:49:03 GMT) Full text and rfc822 format available.Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
From: epg <at> pretzelnet.org To: bug-gnu-emacs <at> gnu.org Subject: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 05 Sep 2024 08:28:53 -0500
I wish I could describe exactly what happened. I found
https://debbugs.gnu.org/cgi/bugreport.cgi?bug=70541 which sounds
very much like what I'm seeing, but I don't use any input method.
What I can say:
- I use rust-mode (not rust-ts-mode)
- I use eglot (config below)
- I use rust-analyzer 1.80.1 (3f5fd8d 2024-08-06)
- The problem seems to happen deep into a work day, after many
commits, reverts, etc.
My eglot configuration:
#+begin_src elisp
(setq
eglot-extend-to-xref t
eglot-ignored-server-capabilities
'(
:inlayHintProvider
:documentOnTypeFormattingProvider
)
)
#+end_src
Eventually, this pops into *Messages*:
#+begin_quote
cl--assertion-failed: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end))
#+end_quote
And this in *Warnings*:
#+begin_quote
⛔ Warning (emacs): Missing/incorrect calls to ‘before/after-change-functions’!!
Details logged to ‘track-changes--error-log’
#+end_quote
Over time, they just keep repeating, although I think killing the
buffer and reopening the file makes it stop.
If I have inspected ‘track-changes--error-log’ correctly, this is
what it contains (looks like some binary, hope it comes through
in some useful fashion or another):
#+begin_quote
(("lib.rs" (buffer-size 5254 5218) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer ... nil) (t funcall-interactively ... nil) (t call-interactively ... nil) (t command-execute ... nil)) [111 (nil . other-window) 103 (nil . recompile) 121 (nil . undefined) 24 96 (nil . next-error) 1 (nil . move-beginning-of-line) 11 ...]) ("lib.rs" (buffer-size 5256 5254) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer nil nil) (t #[257 "r\211q\210\300 )\207" [save-buffer] 2 ...] ... nil) (t map-y-or-n-p ... nil) (t save-some-buffers ... nil) (t recompile ... nil) ...) [19 (nil . save-buffer) 24 24 (nil . exchange-point-and-mark) 23 (nil . kill-region) 24 19 (nil . save-buffer) backspace (nil . delete-backward-char) ...]) ("lib.rs" (buffer-size 5278 5256) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer ... nil) (t funcall-interactively ... nil) (t call-interactively ... nil) (t command-execute ... nil)) [(nil . backward-word) 2 (nil . backward-char) 67108896 (nil . set-mark-command) 134217734 (nil . forward-sexp) 23 (nil . kill-region) 24 19 (nil . save-buffer) ...]) ("lib.rs" (buffer-size 5324 5278) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--track-changes-signal ... nil) (t #[771 "\211^BZ\211^HG^E]
#+end_quote
I wish I could reliably repeat it. Given that it occurs after
many hours of work, `emacs -Q` isn't an option.
Thanks!
In GNU Emacs 30.0.90 (build 9, x86_64-pc-linux-gnu, cairo version
1.18.0) of 2024-09-03 built on dogato
Repository revision: ae2463796f236b8ee2cef3b5e38bffa13abd2233
Repository branch: emacs-30
Windowing system distributor 'The X.Org Foundation', version 11.0.12101012
System Description: openSUSE Tumbleweed
Configured using:
'configure -C --prefix=/opt/emacs-30.0.90.5.ae2463796f2
--disable-silent-rules --with-x-toolkit=no --without-gsettings
--with-native-compilation=no'
Configured features:
ACL CAIRO DBUS FREETYPE GIF GLIB GMP GNUTLS HARFBUZZ JPEG LCMS2
LIBSELINUX LIBXML2 MODULES NOTIFY INOTIFY OLDXMENU PDUMPER PNG
RSVG SECCOMP SOUND SQLITE3 THREADS TIFF TREE_SITTER WEBP X11 XDBE
XIM XINPUT2 XPM ZLIB
Important settings:
value of $LC_CTYPE: en_US.UTF-8
value of $XMODIFIERS: @im=local
locale-coding-system: utf-8-unix
Major mode: ELisp/d
Minor modes in effect:
server-mode: t
editorconfig-mode: t
tooltip-mode: t
global-eldoc-mode: t
eldoc-mode: t
show-paren-mode: t
mouse-wheel-mode: t
file-name-shadow-mode: t
global-font-lock-mode: t
font-lock-mode: t
minibuffer-regexp-mode: t
line-number-mode: t
indent-tabs-mode: t
auto-composition-mode: t
auto-encryption-mode: t
auto-compression-mode: t
Load-path shadows:
None found.
Features:
(shadow sort mail-extr tabify cus-edit cus-start cus-load
cl-print help-fns radix-tree mh-utils emacsbug conf-mode
smerge-mode whitespace time-stamp perl-mode bug-reference
log-edit mule-util display-line-numbers dired-aux view pulse
color vc-annotate oc-basic org-element org-persist org-id
org-refile org-element-ast inline avl-tree generator ol-eww eww
xdg url-queue mm-url ol-rmail ol-mhe ol-irc ol-info ol-gnus
nnselect gnus-art mm-uu mml2015 gnus-sum shr pixel-fill kinsoku
url-file svg dom gnus-group gnus-undo gnus-start gnus-dbus dbus
xml gnus-cloud nnimap nnmail browse-url mail-source utf7 nnoo
parse-time gnus-spec gnus-int gnus-range message yank-media
rfc822 mailabbrev gmm-utils mailheader gnus-win gnus nnheader
range wid-edit ol-docview doc-view jka-compr image-mode exif
dired dired-loaddefs ol-bibtex bibtex iso8601 ol-bbdb ol-w3m
ol-doi org-link-doi org org-macro org-pcomplete org-list
org-footnote org-faces org-entities noutline outline org-version
ob-sql ob-emacs-lisp ob-C ob ob-tangle org-src sh-script smie
treesit executable ob-ref ob-lob ob-table ob-exp ob-comint
ob-core ob-eval cc-mode cc-fonts cc-guess cc-menus cc-cmds
cc-styles cc-align cc-engine cc-vars cc-defs org-cycle org-table
ol org-fold org-fold-core org-keys oc org-loaddefs cal-menu
calendar cal-loaddefs org-compat org-macs format-spec misearch
multi-isearch add-log cl-extra server shell pcomplete files-x
grep vc-cvs vc-rcs log-view pcvs-util vc-dir vc rust-utils
rust-mode rust-rustfmt rust-playpen rust-compile rust-cargo rx
eglot external-completion jsonrpc xref flymake thingatpt project
compat diff ert pp ewoc debug backtrace help-mode find-func
filenotify warnings compile comint ansi-osc ansi-color ring pcase
vc-git diff-mode track-changes easy-mmode vc-dispatcher url-cache
url url-proxy url-privacy url-expand url-methods url-history
url-cookie generate-lisp-file url-domsuf url-util url-parse
auth-source eieio eieio-core cl-macs icons json map byte-opt gv
bytecomp byte-compile url-vars mh-acros mh-folder which-func
imenu mh-comp sendmail mail-utils mh-scan mh-gnus mml mm-view
mml-smime mml-sec epa derived epg rfc6068 epg-config smime cl-seq
password-cache gnutls puny dig mailcap mm-decode mm-bodies
mm-encode mail-parse rfc2231 rfc2047 rfc2045 ietf-drums mm-util
mail-prsvr gnus-util text-property-search time-date subr-x mh-e
mh-buffers mh-loaddefs editorconfig editorconfig-core
editorconfig-core-handle editorconfig-fnmatch cl-loaddefs cl-lib
rmc iso-transl tooltip cconv eldoc paren electric uniquify
ediff-hook vc-hooks lisp-float-type elisp-mode mwheel term/x-win
x-win term/common-win x-dnd touch-screen tool-bar dnd fontset
image regexp-opt fringe tabulated-list replace newcomment
text-mode lisp-mode prog-mode register page tab-bar menu-bar
rfn-eshadow isearch easymenu timer select scroll-bar mouse
jit-lock font-lock syntax font-core term/tty-colors frame
minibuffer nadvice seq simple cl-generic indonesian philippine
cham georgian utf-8-lang misc-lang vietnamese tibetan thai
tai-viet lao korean japanese eucjp-ms cp51932 hebrew greek
romanian slovak czech european ethiopic indian cyrillic chinese
composite emoji-zwj charscript charprop case-table epa-hook
jka-cmpr-hook help abbrev obarray oclosure cl-preloaded button
loaddefs theme-loaddefs faces cus-face macroexp files window
text-properties overlay sha1 md5 base64 format env code-pages
mule custom widget keymap hashtable-print-readable backquote
threads dbusbind inotify lcms2 dynamic-setting
font-render-setting cairo xinput2 x multi-tty move-toolbar
make-network-process emacs)
Memory information:
((conses 16 407345 108770) (symbols 48 32008 1) (strings 32 102422 6760)
(string-bytes 1 2997195) (vectors 16 63439) (vector-slots 8 1388223 211290)
(floats 8 478 2399) (intervals 56 29367 817) (buffers 992 73))
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 05 Sep 2024 14:07:02 GMT) Full text and rfc822 format available.Message #8 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Eli Zaretskii <eliz <at> gnu.org> To: epg <at> pretzelnet.org, Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 05 Sep 2024 16:54:34 +0300
> From: epg <at> pretzelnet.org
> Date: Thu, 05 Sep 2024 08:28:53 -0500
>
> I wish I could describe exactly what happened. I found
> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=70541 which sounds
> very much like what I'm seeing, but I don't use any input method.
>
> What I can say:
>
> - I use rust-mode (not rust-ts-mode)
> - I use eglot (config below)
> - I use rust-analyzer 1.80.1 (3f5fd8d 2024-08-06)
> - The problem seems to happen deep into a work day, after many
> commits, reverts, etc.
>
> My eglot configuration:
>
> #+begin_src elisp
> (setq
> eglot-extend-to-xref t
> eglot-ignored-server-capabilities
> '(
> :inlayHintProvider
> :documentOnTypeFormattingProvider
> )
> )
> #+end_src
>
> Eventually, this pops into *Messages*:
>
> #+begin_quote
> cl--assertion-failed: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end))
> #+end_quote
>
> And this in *Warnings*:
>
> #+begin_quote
> ⛔ Warning (emacs): Missing/incorrect calls to ‘before/after-change-functions’!!
> Details logged to ‘track-changes--error-log’
> #+end_quote
>
> Over time, they just keep repeating, although I think killing the
> buffer and reopening the file makes it stop.
>
> If I have inspected ‘track-changes--error-log’ correctly, this is
> what it contains (looks like some binary, hope it comes through
> in some useful fashion or another):
>
> #+begin_quote
> (("lib.rs" (buffer-size 5254 5218) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer ... nil) (t funcall-interactively ... nil) (t call-interactively ... nil) (t command-execute ... nil)) [111 (nil . other-window) 103 (nil . recompile) 121 (nil . undefined) 24 96 (nil . next-error) 1 (nil . move-beginning-of-line) 11 ...]) ("lib.rs" (buffer-size 5256 5254) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer nil nil) (t #[257 "r\211q\210\300 )\207" [save-buffer] 2 ...] ... nil) (t map-y-or-n-p ... nil) (t save-some-buffers ... nil) (t recompile ... nil) ...) [19 (nil . save-buffer) 24 24 (nil . exchange-point-and-mark) 23 (nil . kill-region) 24 19 (nil . save-buffer) backspace (nil . delete-backward-char) ...]) ("lib.rs" (buffer-size 5278 5256) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--signal-textDocument/didChange nil nil) (t eglot--signal-textDocument/didSave nil nil) (t run-hooks ... nil) (t basic-save-buffer ... nil) (t save-buffer ... nil) (t funcall-interactively ... nil) (t call-interactively ... nil) (t command-execute ... nil)) [(nil . backward-word) 2 (nil . backward-char) 67108896 (nil . set-mark-command) 134217734 (nil . forward-sexp) 23 (nil . kill-region) 24 19 (nil . save-buffer) ...]) ("lib.rs" (buffer-size 5324 5278) ((t track-changes--recover-from-error ... nil) (t track-changes-fetch ... nil) (t eglot--track-changes-fetch ... nil) (t eglot--track-changes-signal ... nil) (t #[771 "\211^BZ\211^HG^E]
> #+end_quote
>
> I wish I could reliably repeat it. Given that it occurs after
> many hours of work, `emacs -Q` isn't an option.
Adding Stefan.
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Sun, 08 Sep 2024 23:07:02 GMT) Full text and rfc822 format available.Message #11 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Eli Zaretskii <eliz <at> gnu.org> Cc: 73041 <at> debbugs.gnu.org, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Sun, 08 Sep 2024 19:06:09 -0400
[Message part 1 (text/plain, inline)]
>> I wish I could describe exactly what happened. I found
>> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=70541 which sounds
>> very much like what I'm seeing, but I don't use any input method.
Thanks for your description of the problem. A reproducible recipe is of
course better, but sadly, we can't always have what we want.
Let's hope we find that eventually we'll get what we need.
>> Eventually, this pops into *Messages*:
>>
>> #+begin_quote
>> cl--assertion-failed: Assertion failed: (or track-changes--before-no (<=
>> track-changes--before-beg (track-changes--state-beg track-changes--state)
>> beg end (track-changes--state-end track-changes--state)
>> track-changes--before-end))
>> #+end_quote
Hmm... this clearly points to a bug in track-changes.
>> #+begin_quote
>> ⛔ Warning (emacs): Missing/incorrect calls to ‘before/after-change-functions’!!
>> Details logged to ‘track-changes--error-log’
>> #+end_quote
Barring bugs in track-changes, this points to a bug in some other part
of Emacs, which prevents track-changes from getting the info it needs.
Track-changes (and Eglot) are supposedly equipped to deal with such
problems, because we know we still have many of them.
>> Over time, they just keep repeating,
Hmm... `track-changes--recover-from-error` should (as the name suggests)
"recover" so they shouldn't just keep repeating, unless the source of
the problem keeps repeating as well. Or unless there's a bug in
track-changes.
>> although I think killing the
>> buffer and reopening the file makes it stop.
>>
>> If I have inspected ‘track-changes--error-log’ correctly, this is
>> what it contains (looks like some binary, hope it comes through
>> in some useful fashion or another):
It's a list of debug info, one per "recovery". Each debug info is made
of the buffer name, the inconsistency encountered, then a backtrace, and
an extract of (raw) `view-lossage`. Here's your data reformatted:
(("lib.rs" (buffer-size 5254 5218)
((t track-changes--recover-from-error ... nil)
(t track-changes-fetch ... nil)
(t eglot--track-changes-fetch ... nil)
(t eglot--signal-textDocument/didChange nil nil)
(t eglot--signal-textDocument/didSave nil nil)
(t run-hooks ... nil)
(t basic-save-buffer ... nil)
(t save-buffer ... nil)
(t funcall-interactively ... nil)
(t call-interactively ... nil)
(t command-execute ... nil))
[111 (nil . other-window)
103 (nil . recompile)
121 (nil . undefined)
24 96 (nil . next-error)
1 (nil . move-beginning-of-line)
11 ...])
("lib.rs" (buffer-size 5256 5254)
((t track-changes--recover-from-error ... nil)
(t track-changes-fetch ... nil)
(t eglot--track-changes-fetch ... nil)
(t eglot--signal-textDocument/didChange nil nil)
(t eglot--signal-textDocument/didSave nil nil)
(t run-hooks ... nil)
(t basic-save-buffer ... nil)
(t save-buffer nil nil)
(t #[257 "r\211q\210\300 )\207" [save-buffer] 2 ...]
(t map-y-or-n-p ... nil)
(t save-some-buffers ... nil)
(t recompile ... nil)
...)
[19 (nil . save-buffer)
24 24 (nil . exchange-point-and-mark)
23 (nil . kill-region)
24 19 (nil . save-buffer)
backspace (nil . delete-backward-char) ...])
("lib.rs" (buffer-size 5278 5256)
((t track-changes--recover-from-error ... nil)
(t track-changes-fetch ... nil)
(t eglot--track-changes-fetch ... nil)
(t eglot--signal-textDocument/didChange nil nil)
(t eglot--signal-textDocument/didSave nil nil)
(t run-hooks ... nil)
(t basic-save-buffer ... nil)
(t save-buffer ... nil)
(t funcall-interactively ... nil)
(t call-interactively ... nil)
(t command-execute ... nil))
[(nil . backward-word)
2 (nil . backward-char)
67108896 (nil . set-mark-command)
134217734 (nil . forward-sexp)
23 (nil . kill-region)
24 19 (nil . save-buffer)
...])
("lib.rs" (buffer-size 5324 5278)
((t track-changes--recover-from-error ... nil)
(t track-changes-fetch ... nil)
(t eglot--track-changes-fetch ... nil)
(t eglot--track-changes-signal ... nil)
(t #[771 "\211^BZ\211^HG^E] ...)
...)
...)
...)
This suggests the error is always detected "late" when we find that the
buffer-size is not the one we expected, so some buffer change presumably
happened without informing track-changes.
That doesn't explain the assertion failure.
Also, I don't see anything "unusual" in the view-lossage.
>> I wish I could reliably repeat it. Given that it occurs after
>> many hours of work, `emacs -Q` isn't an option.
Thanks.
I stared at the code for a bit trying to see how that assertion failure
can occur, but nothing came to mind yet. I think we'll need more data.
Can you try to install the patch below, then run with
`debug-on-error` enabled?
Hopefully when the error occurs, this will give us more info (ideally
including a backtrace).
Stefan
[track-changes.patch (text/x-diff, inline)]
diff --git a/lisp/emacs-lisp/track-changes.el b/lisp/emacs-lisp/track-changes.el
index 92d14959763..325a92317ca 100644
--- a/lisp/emacs-lisp/track-changes.el
+++ b/lisp/emacs-lisp/track-changes.el
@@ -229,6 +229,7 @@ track-changes-register
(push tracker track-changes--clean-trackers)
(when disjoint
(push tracker track-changes--disjoint-trackers))
+ (cl-assert (track-changes--sane-state-p))
tracker))
(defun track-changes-unregister (id)
@@ -253,7 +254,8 @@ track-changes-unregister
track-changes--buffer-size
track-changes--before-clean
track-changes--state))
- (remove-hook 'after-change-functions #'track-changes--after t)))
+ (remove-hook 'after-change-functions #'track-changes--after t))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes-fetch (id func)
"Fetch the pending changes for tracker ID pass them to FUNC.
@@ -372,7 +374,8 @@ track-changes-fetch
(funcall func beg end (or before lenbefore)))
;; Re-enable the tracker's signal only after running `func', so
;; as to avoid nested invocations.
- (cl-pushnew id track-changes--clean-trackers))))
+ (cl-pushnew id track-changes--clean-trackers)
+ (cl-assert (track-changes--sane-state-p)))))
(defun track-changes-inconsistent-state-p ()
"Return whether the current buffer is in an inconsistent state.
@@ -387,6 +390,19 @@ track-changes-inconsistent-state-p
;;;; Auxiliary functions.
+(defun track-changes--sane-state-p ()
+ (and (equal track-changes--buffer-size (buffer-size))
+ (or track-changes--before-no
+ (pcase track-changes--before-clean
+ ('unset t)
+ ('set (<= track-changes--before-beg track-changes--before-end))
+ ('nil
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ (_ nil)))))
+
(defun track-changes--clean-state ()
(cond
((null track-changes--state)
@@ -551,7 +567,8 @@ track-changes--before
(length track-changes--before-string))))))
(setf track-changes--before-end new-bend)
(cl-callf concat track-changes--before-string
- (buffer-substring-no-properties old-bend new-bend)))))))))
+ (buffer-substring-no-properties old-bend new-bend))))))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--after (beg end len)
(cl-assert track-changes--state)
@@ -576,10 +593,26 @@ track-changes--after
;; BEG..END is not covered by previous `before-change-functions'!!
(track-changes--recover-from-error `(unexpected-after ,beg ,end ,len))
;; Note the new changes.
+ (let ((orig-beg (track-changes--state-beg track-changes--state))
+ (orig-end (track-changes--state-end track-changes--state)))
(when (< beg (track-changes--state-beg track-changes--state))
(setf (track-changes--state-beg track-changes--state) beg))
(cl-callf (lambda (old-end) (max end (+ old-end offset)))
(track-changes--state-end track-changes--state))
+ (cl-assert (or track-changes--before-no
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)))
+ nil "<=? %S %S (was %S)"
+ track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ orig-beg)
+ (cl-assert (or track-changes--before-no
+ (<= (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ nil "<=? %S %S (was %S)"
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end
+ orig-end))
(cl-assert (or track-changes--before-no
(<= track-changes--before-beg
(track-changes--state-beg track-changes--state)
@@ -591,7 +624,8 @@ track-changes--after
(if (track-changes--tracker-immediate tracker)
(funcall (track-changes--tracker-signal tracker) tracker)
(run-with-timer 0 nil #'track-changes--call-signal
- (current-buffer) tracker)))))
+ (current-buffer) tracker))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--call-signal (buf tracker)
(when (buffer-live-p buf)
diff --git a/lisp/progmodes/eglot.el b/lisp/progmodes/eglot.el
index 82e99a2c920..a2c9f73fc73 100644
--- a/lisp/progmodes/eglot.el
+++ b/lisp/progmodes/eglot.el
@@ -2813,6 +2813,8 @@ eglot--signal-textDocument/didChange
(defun eglot--signal-textDocument/didOpen ()
"Send textDocument/didOpen to server."
+ ;; Flush any potential pending change.
+ (eglot--track-changes-fetch eglot--track-changes)
(setq eglot--recent-changes nil
eglot--versioned-identifier 0
eglot--TextDocumentIdentifier-cache nil)
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Sat, 21 Sep 2024 09:12:01 GMT) Full text and rfc822 format available.Message #14 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Eli Zaretskii <eliz <at> gnu.org> To: epg <at> pretzelnet.org, Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Sat, 21 Sep 2024 12:11:21 +0300
Ping! epg, could you please try Stefan's patches and report back?
> From: Stefan Monnier <monnier <at> iro.umontreal.ca>
> Cc: epg <at> pretzelnet.org, 73041 <at> debbugs.gnu.org
> Date: Sun, 08 Sep 2024 19:06:09 -0400
>
>
> >> I wish I could describe exactly what happened. I found
> >> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=70541 which sounds
> >> very much like what I'm seeing, but I don't use any input method.
>
> Thanks for your description of the problem. A reproducible recipe is of
> course better, but sadly, we can't always have what we want.
> Let's hope we find that eventually we'll get what we need.
>
> >> Eventually, this pops into *Messages*:
> >>
> >> #+begin_quote
> >> cl--assertion-failed: Assertion failed: (or track-changes--before-no (<=
> >> track-changes--before-beg (track-changes--state-beg track-changes--state)
> >> beg end (track-changes--state-end track-changes--state)
> >> track-changes--before-end))
> >> #+end_quote
>
> Hmm... this clearly points to a bug in track-changes.
>
> >> #+begin_quote
> >> ⛔ Warning (emacs): Missing/incorrect calls to ‘before/after-change-functions’!!
> >> Details logged to ‘track-changes--error-log’
> >> #+end_quote
>
> Barring bugs in track-changes, this points to a bug in some other part
> of Emacs, which prevents track-changes from getting the info it needs.
> Track-changes (and Eglot) are supposedly equipped to deal with such
> problems, because we know we still have many of them.
>
> >> Over time, they just keep repeating,
>
> Hmm... `track-changes--recover-from-error` should (as the name suggests)
> "recover" so they shouldn't just keep repeating, unless the source of
> the problem keeps repeating as well. Or unless there's a bug in
> track-changes.
>
> >> although I think killing the
> >> buffer and reopening the file makes it stop.
> >>
> >> If I have inspected ‘track-changes--error-log’ correctly, this is
> >> what it contains (looks like some binary, hope it comes through
> >> in some useful fashion or another):
>
> It's a list of debug info, one per "recovery". Each debug info is made
> of the buffer name, the inconsistency encountered, then a backtrace, and
> an extract of (raw) `view-lossage`. Here's your data reformatted:
>
> (("lib.rs" (buffer-size 5254 5218)
> ((t track-changes--recover-from-error ... nil)
> (t track-changes-fetch ... nil)
> (t eglot--track-changes-fetch ... nil)
> (t eglot--signal-textDocument/didChange nil nil)
> (t eglot--signal-textDocument/didSave nil nil)
> (t run-hooks ... nil)
> (t basic-save-buffer ... nil)
> (t save-buffer ... nil)
> (t funcall-interactively ... nil)
> (t call-interactively ... nil)
> (t command-execute ... nil))
> [111 (nil . other-window)
> 103 (nil . recompile)
> 121 (nil . undefined)
> 24 96 (nil . next-error)
> 1 (nil . move-beginning-of-line)
> 11 ...])
> ("lib.rs" (buffer-size 5256 5254)
> ((t track-changes--recover-from-error ... nil)
> (t track-changes-fetch ... nil)
> (t eglot--track-changes-fetch ... nil)
> (t eglot--signal-textDocument/didChange nil nil)
> (t eglot--signal-textDocument/didSave nil nil)
> (t run-hooks ... nil)
> (t basic-save-buffer ... nil)
> (t save-buffer nil nil)
> (t #[257 "r\211q\210\300 )\207" [save-buffer] 2 ...]
> (t map-y-or-n-p ... nil)
> (t save-some-buffers ... nil)
> (t recompile ... nil)
> ...)
> [19 (nil . save-buffer)
> 24 24 (nil . exchange-point-and-mark)
> 23 (nil . kill-region)
> 24 19 (nil . save-buffer)
> backspace (nil . delete-backward-char) ...])
> ("lib.rs" (buffer-size 5278 5256)
> ((t track-changes--recover-from-error ... nil)
> (t track-changes-fetch ... nil)
> (t eglot--track-changes-fetch ... nil)
> (t eglot--signal-textDocument/didChange nil nil)
> (t eglot--signal-textDocument/didSave nil nil)
> (t run-hooks ... nil)
> (t basic-save-buffer ... nil)
> (t save-buffer ... nil)
> (t funcall-interactively ... nil)
> (t call-interactively ... nil)
> (t command-execute ... nil))
> [(nil . backward-word)
> 2 (nil . backward-char)
> 67108896 (nil . set-mark-command)
> 134217734 (nil . forward-sexp)
> 23 (nil . kill-region)
> 24 19 (nil . save-buffer)
> ...])
> ("lib.rs" (buffer-size 5324 5278)
> ((t track-changes--recover-from-error ... nil)
> (t track-changes-fetch ... nil)
> (t eglot--track-changes-fetch ... nil)
> (t eglot--track-changes-signal ... nil)
> (t #[771 "\211^BZ\211^HG^E] ...)
> ...)
> ...)
> ...)
>
> This suggests the error is always detected "late" when we find that the
> buffer-size is not the one we expected, so some buffer change presumably
> happened without informing track-changes.
> That doesn't explain the assertion failure.
>
> Also, I don't see anything "unusual" in the view-lossage.
>
> >> I wish I could reliably repeat it. Given that it occurs after
> >> many hours of work, `emacs -Q` isn't an option.
>
> Thanks.
> I stared at the code for a bit trying to see how that assertion failure
> can occur, but nothing came to mind yet. I think we'll need more data.
>
> Can you try to install the patch below, then run with
> `debug-on-error` enabled?
> Hopefully when the error occurs, this will give us more info (ideally
> including a backtrace).
>
>
> Stefan
>
> diff --git a/lisp/emacs-lisp/track-changes.el b/lisp/emacs-lisp/track-changes.el
> index 92d14959763..325a92317ca 100644
> --- a/lisp/emacs-lisp/track-changes.el
> +++ b/lisp/emacs-lisp/track-changes.el
> @@ -229,6 +229,7 @@ track-changes-register
> (push tracker track-changes--clean-trackers)
> (when disjoint
> (push tracker track-changes--disjoint-trackers))
> + (cl-assert (track-changes--sane-state-p))
> tracker))
>
> (defun track-changes-unregister (id)
> @@ -253,7 +254,8 @@ track-changes-unregister
> track-changes--buffer-size
> track-changes--before-clean
> track-changes--state))
> - (remove-hook 'after-change-functions #'track-changes--after t)))
> + (remove-hook 'after-change-functions #'track-changes--after t))
> + (cl-assert (track-changes--sane-state-p)))
>
> (defun track-changes-fetch (id func)
> "Fetch the pending changes for tracker ID pass them to FUNC.
> @@ -372,7 +374,8 @@ track-changes-fetch
> (funcall func beg end (or before lenbefore)))
> ;; Re-enable the tracker's signal only after running `func', so
> ;; as to avoid nested invocations.
> - (cl-pushnew id track-changes--clean-trackers))))
> + (cl-pushnew id track-changes--clean-trackers)
> + (cl-assert (track-changes--sane-state-p)))))
>
> (defun track-changes-inconsistent-state-p ()
> "Return whether the current buffer is in an inconsistent state.
> @@ -387,6 +390,19 @@ track-changes-inconsistent-state-p
>
> ;;;; Auxiliary functions.
>
> +(defun track-changes--sane-state-p ()
> + (and (equal track-changes--buffer-size (buffer-size))
> + (or track-changes--before-no
> + (pcase track-changes--before-clean
> + ('unset t)
> + ('set (<= track-changes--before-beg track-changes--before-end))
> + ('nil
> + (<= track-changes--before-beg
> + (track-changes--state-beg track-changes--state)
> + (track-changes--state-end track-changes--state)
> + track-changes--before-end))
> + (_ nil)))))
> +
> (defun track-changes--clean-state ()
> (cond
> ((null track-changes--state)
> @@ -551,7 +567,8 @@ track-changes--before
> (length track-changes--before-string))))))
> (setf track-changes--before-end new-bend)
> (cl-callf concat track-changes--before-string
> - (buffer-substring-no-properties old-bend new-bend)))))))))
> + (buffer-substring-no-properties old-bend new-bend))))))))
> + (cl-assert (track-changes--sane-state-p)))
>
> (defun track-changes--after (beg end len)
> (cl-assert track-changes--state)
> @@ -576,10 +593,26 @@ track-changes--after
> ;; BEG..END is not covered by previous `before-change-functions'!!
> (track-changes--recover-from-error `(unexpected-after ,beg ,end ,len))
> ;; Note the new changes.
> + (let ((orig-beg (track-changes--state-beg track-changes--state))
> + (orig-end (track-changes--state-end track-changes--state)))
> (when (< beg (track-changes--state-beg track-changes--state))
> (setf (track-changes--state-beg track-changes--state) beg))
> (cl-callf (lambda (old-end) (max end (+ old-end offset)))
> (track-changes--state-end track-changes--state))
> + (cl-assert (or track-changes--before-no
> + (<= track-changes--before-beg
> + (track-changes--state-beg track-changes--state)))
> + nil "<=? %S %S (was %S)"
> + track-changes--before-beg
> + (track-changes--state-beg track-changes--state)
> + orig-beg)
> + (cl-assert (or track-changes--before-no
> + (<= (track-changes--state-end track-changes--state)
> + track-changes--before-end))
> + nil "<=? %S %S (was %S)"
> + (track-changes--state-end track-changes--state)
> + track-changes--before-end
> + orig-end))
> (cl-assert (or track-changes--before-no
> (<= track-changes--before-beg
> (track-changes--state-beg track-changes--state)
> @@ -591,7 +624,8 @@ track-changes--after
> (if (track-changes--tracker-immediate tracker)
> (funcall (track-changes--tracker-signal tracker) tracker)
> (run-with-timer 0 nil #'track-changes--call-signal
> - (current-buffer) tracker)))))
> + (current-buffer) tracker))))
> + (cl-assert (track-changes--sane-state-p)))
>
> (defun track-changes--call-signal (buf tracker)
> (when (buffer-live-p buf)
> diff --git a/lisp/progmodes/eglot.el b/lisp/progmodes/eglot.el
> index 82e99a2c920..a2c9f73fc73 100644
> --- a/lisp/progmodes/eglot.el
> +++ b/lisp/progmodes/eglot.el
> @@ -2813,6 +2813,8 @@ eglot--signal-textDocument/didChange
>
> (defun eglot--signal-textDocument/didOpen ()
> "Send textDocument/didOpen to server."
> + ;; Flush any potential pending change.
> + (eglot--track-changes-fetch eglot--track-changes)
> (setq eglot--recent-changes nil
> eglot--versioned-identifier 0
> eglot--TextDocumentIdentifier-cache nil)
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Wed, 25 Sep 2024 01:56:01 GMT) Full text and rfc822 format available.Message #17 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Eric Gillespie <epg <at> pretzelnet.org> To: Eli Zaretskii <eliz <at> gnu.org> Cc: 73041 <at> debbugs.gnu.org, Stefan Monnier <monnier <at> iro.umontreal.ca> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Tue, 24 Sep 2024 20:36:59 -0500
Eli Zaretskii <eliz <at> gnu.org> writes: > Ping! epg, could you please try Stefan's patches and report back? Sorry for the delay. I applied that patch the next time I built, and it definitely had an impact, but not what I expected. I now have a trivial -Q test case for a bug, though I fear it may not be the bug I originally reported here. 1. Run 'cargo new foo' to create a new cargo project 2. emacs -Q foo/src/main.rs 3. M-x rust-ts-mode RET 4. M-x eglot RET 5. C-x k RET Expected: Buffer main.rs is killed. Actual: (cl-assert (track-changes--sane-state-p)) assertion triggered. Who knows, maybe these two bugs share a root cause. > From: Stefan Monnier <monnier <at> iro.umontreal.ca> >> Hmm... this clearly points to a bug in track-changes. In fact, I was able to trace this kill-buffer behavior all the way back to the birth of track-changes in commit d7a83e23d47ca9e3e6ca70078e956e31301e5e6d (the patch needed a tiny bit of conflict resolution to apply that far back, but it was no problem). Thanks! -- Eric Gillespie <*> epg <at> pretzelnet.org
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Fri, 27 Sep 2024 14:57:01 GMT) Full text and rfc822 format available.Message #20 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Eric Gillespie <epg <at> pretzelnet.org> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Fri, 27 Sep 2024 10:55:58 -0400
[Message part 1 (text/plain, inline)]
> 1. Run 'cargo new foo' to create a new cargo project
> 2. emacs -Q foo/src/main.rs
> 3. M-x rust-ts-mode RET
> 4. M-x eglot RET
> 5. C-x k RET
>
> Expected:
> Buffer main.rs is killed.
>
> Actual:
> (cl-assert (track-changes--sane-state-p)) assertion triggered.
Duh, sorry. Try this patch instead.
Stefan
[track-changes.patch (text/x-diff, inline)]
diff --git a/lisp/emacs-lisp/track-changes.el b/lisp/emacs-lisp/track-changes.el
index 92d14959763..24df0aac6bb 100644
--- a/lisp/emacs-lisp/track-changes.el
+++ b/lisp/emacs-lisp/track-changes.el
@@ -213,6 +213,7 @@ track-changes-register
that may block, do as little work as possible, ...
When IMMEDIATE is non-nil, the SIGNAL should probably not always call
`track-changes-fetch', since that would defeat the purpose of this library."
+ (cl-assert (track-changes--sane-state-p))
(when (and nobefore disjoint)
;; FIXME: Without `before-change-functions', we can discover
;; a disjoint change only after the fact, which is not good enough.
@@ -229,6 +230,7 @@ track-changes-register
(push tracker track-changes--clean-trackers)
(when disjoint
(push tracker track-changes--disjoint-trackers))
+ (cl-assert (track-changes--sane-state-p))
tracker))
(defun track-changes-unregister (id)
@@ -236,6 +238,7 @@ track-changes-unregister
Trackers can consume resources (especially if `track-changes-fetch' is
not called), so it is good practice to unregister them when you don't
need them any more."
+ (cl-assert (track-changes--sane-state-p))
(unless (memq id track-changes--trackers)
(error "Unregistering a non-registered tracker: %S" id))
(setq track-changes--trackers (delq id track-changes--trackers))
@@ -253,7 +256,8 @@ track-changes-unregister
track-changes--buffer-size
track-changes--before-clean
track-changes--state))
- (remove-hook 'after-change-functions #'track-changes--after t)))
+ (remove-hook 'after-change-functions #'track-changes--after t))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes-fetch (id func)
"Fetch the pending changes for tracker ID pass them to FUNC.
@@ -372,7 +376,8 @@ track-changes-fetch
(funcall func beg end (or before lenbefore)))
;; Re-enable the tracker's signal only after running `func', so
;; as to avoid nested invocations.
- (cl-pushnew id track-changes--clean-trackers))))
+ (cl-pushnew id track-changes--clean-trackers)
+ (cl-assert (track-changes--sane-state-p)))))
(defun track-changes-inconsistent-state-p ()
"Return whether the current buffer is in an inconsistent state.
@@ -387,6 +392,26 @@ track-changes-inconsistent-state-p
;;;; Auxiliary functions.
+(defun track-changes--sane-state-p ()
+ (if (null track-changes--trackers)
+ (and (zerop track-changes--before-beg)
+ (zerop track-changes--before-end)
+ (equal track-changes--before-string "")
+ (null track-changes--buffer-size)
+ (eq track-changes--before-clean 'unset)
+ (null track-changes--state))
+ (and (equal track-changes--buffer-size (buffer-size))
+ (or track-changes--before-no
+ (pcase track-changes--before-clean
+ ('unset t)
+ ('set (<= track-changes--before-beg track-changes--before-end))
+ ('nil
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ (_ nil))))))
+
(defun track-changes--clean-state ()
(cond
((null track-changes--state)
@@ -551,7 +576,8 @@ track-changes--before
(length track-changes--before-string))))))
(setf track-changes--before-end new-bend)
(cl-callf concat track-changes--before-string
- (buffer-substring-no-properties old-bend new-bend)))))))))
+ (buffer-substring-no-properties old-bend new-bend))))))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--after (beg end len)
(cl-assert track-changes--state)
@@ -576,10 +602,26 @@ track-changes--after
;; BEG..END is not covered by previous `before-change-functions'!!
(track-changes--recover-from-error `(unexpected-after ,beg ,end ,len))
;; Note the new changes.
+ (let ((orig-beg (track-changes--state-beg track-changes--state))
+ (orig-end (track-changes--state-end track-changes--state)))
(when (< beg (track-changes--state-beg track-changes--state))
(setf (track-changes--state-beg track-changes--state) beg))
(cl-callf (lambda (old-end) (max end (+ old-end offset)))
(track-changes--state-end track-changes--state))
+ (cl-assert (or track-changes--before-no
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)))
+ nil "<=? %S %S (was %S)"
+ track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ orig-beg)
+ (cl-assert (or track-changes--before-no
+ (<= (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ nil "<=? %S %S (was %S)"
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end
+ orig-end))
(cl-assert (or track-changes--before-no
(<= track-changes--before-beg
(track-changes--state-beg track-changes--state)
@@ -591,7 +633,8 @@ track-changes--after
(if (track-changes--tracker-immediate tracker)
(funcall (track-changes--tracker-signal tracker) tracker)
(run-with-timer 0 nil #'track-changes--call-signal
- (current-buffer) tracker)))))
+ (current-buffer) tracker))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--call-signal (buf tracker)
(when (buffer-live-p buf)
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 03 Oct 2024 19:22:01 GMT) Full text and rfc822 format available.Message #23 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Eric Gillespie <epg <at> pretzelnet.org> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 03 Oct 2024 15:21:24 -0400
[Message part 1 (text/plain, inline)]
Hi Eric,
> Duh, sorry. Try this patch instead.
I pushed to Emacs `master` a patch to `track-changes.el` which doesn't
fix any bug but allows it to record more information to help find the
root cause of errors.
If you're able to try Emacs `master` (which would be helpful), then the
patch I sent doesn't apply to it (any more), so please use the patch
below instead and add `(setq track-changes-record-errors 'trace)` to your
init file so as to activate the additional recording.
Stefan
[track-changes.patch (text/x-diff, inline)]
diff --git a/lisp/emacs-lisp/track-changes.el b/lisp/emacs-lisp/track-changes.el
index 1b0f64f544d..245b7b89ae5 100644
--- a/lisp/emacs-lisp/track-changes.el
+++ b/lisp/emacs-lisp/track-changes.el
@@ -219,6 +219,7 @@ track-changes-register
When IMMEDIATE is non-nil, the SIGNAL should probably not always call
`track-changes-fetch', since that would defeat the purpose of this library."
(track-changes--trace)
+ (cl-assert (track-changes--sane-state-p))
(when (and nobefore disjoint)
;; FIXME: Without `before-change-functions', we can discover
;; a disjoint change only after the fact, which is not good enough.
@@ -235,6 +236,7 @@ track-changes-register
(push tracker track-changes--clean-trackers)
(when disjoint
(push tracker track-changes--disjoint-trackers))
+ (cl-assert (track-changes--sane-state-p))
tracker))
(defun track-changes-unregister (id)
@@ -243,6 +245,7 @@ track-changes-unregister
not called), so it is good practice to unregister them when you don't
need them any more."
(track-changes--trace)
+ (cl-assert (track-changes--sane-state-p))
(unless (memq id track-changes--trackers)
(error "Unregistering a non-registered tracker: %S" id))
(setq track-changes--trackers (delq id track-changes--trackers))
@@ -260,7 +263,8 @@ track-changes-unregister
track-changes--buffer-size
track-changes--before-clean
track-changes--state))
- (remove-hook 'after-change-functions #'track-changes--after t)))
+ (remove-hook 'after-change-functions #'track-changes--after t))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes-fetch (id func)
"Fetch the pending changes for tracker ID pass them to FUNC.
@@ -380,7 +384,8 @@ track-changes-fetch
(funcall func beg end (or before lenbefore)))
;; Re-enable the tracker's signal only after running `func', so
;; as to avoid nested invocations.
- (cl-pushnew id track-changes--clean-trackers))))
+ (cl-pushnew id track-changes--clean-trackers)
+ (cl-assert (track-changes--sane-state-p)))))
(defun track-changes-inconsistent-state-p ()
"Return whether the current buffer is in an inconsistent state.
@@ -418,6 +423,26 @@ track-changes--trace
(track-change--backtrace
10 #'track-changes--trace)))))
+(defun track-changes--sane-state-p ()
+ (if (null track-changes--trackers)
+ (and (zerop track-changes--before-beg)
+ (zerop track-changes--before-end)
+ (equal track-changes--before-string "")
+ (null track-changes--buffer-size)
+ (eq track-changes--before-clean 'unset)
+ (null track-changes--state))
+ (and (equal track-changes--buffer-size (buffer-size))
+ (or track-changes--before-no
+ (pcase track-changes--before-clean
+ ('unset t)
+ ('set (<= track-changes--before-beg track-changes--before-end))
+ ('nil
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ (_ nil))))))
+
(defun track-changes--clean-state ()
(cond
((null track-changes--state)
@@ -586,7 +611,8 @@ track-changes--before
(length track-changes--before-string))))))
(setf track-changes--before-end new-bend)
(cl-callf concat track-changes--before-string
- (buffer-substring-no-properties old-bend new-bend)))))))))
+ (buffer-substring-no-properties old-bend new-bend))))))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--after (beg end len)
(track-changes--trace)
@@ -612,10 +638,26 @@ track-changes--after
;; BEG..END is not covered by previous `before-change-functions'!!
(track-changes--recover-from-error `(unexpected-after ,beg ,end ,len))
;; Note the new changes.
+ (let ((orig-beg (track-changes--state-beg track-changes--state))
+ (orig-end (track-changes--state-end track-changes--state)))
(when (< beg (track-changes--state-beg track-changes--state))
(setf (track-changes--state-beg track-changes--state) beg))
(cl-callf (lambda (old-end) (max end (+ old-end offset)))
(track-changes--state-end track-changes--state))
+ (cl-assert (or track-changes--before-no
+ (<= track-changes--before-beg
+ (track-changes--state-beg track-changes--state)))
+ nil "<=? %S %S (was %S)"
+ track-changes--before-beg
+ (track-changes--state-beg track-changes--state)
+ orig-beg)
+ (cl-assert (or track-changes--before-no
+ (<= (track-changes--state-end track-changes--state)
+ track-changes--before-end))
+ nil "<=? %S %S (was %S)"
+ (track-changes--state-end track-changes--state)
+ track-changes--before-end
+ orig-end))
(cl-assert (or track-changes--before-no
(<= track-changes--before-beg
(track-changes--state-beg track-changes--state)
@@ -627,7 +669,8 @@ track-changes--after
(if (track-changes--tracker-immediate tracker)
(funcall (track-changes--tracker-signal tracker) tracker)
(run-with-timer 0 nil #'track-changes--call-signal
- (current-buffer) tracker)))))
+ (current-buffer) tracker))))
+ (cl-assert (track-changes--sane-state-p)))
(defun track-changes--call-signal (buf tracker)
(when (buffer-live-p buf)
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 10 Oct 2024 16:22:02 GMT) Full text and rfc822 format available.Message #26 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Eric Gillespie <epg <at> pretzelnet.org> To: Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 10 Oct 2024 11:21:21 -0500
Hi Stefan!
Stefan Monnier <monnier <at> iro.umontreal.ca> writes:
> If you're able to try Emacs `master` (which would be helpful), then the
> patch I sent doesn't apply to it (any more), so please use the patch
> below instead and add `(setq track-changes-record-errors 'trace)` to your
> init file so as to activate the additional recording.
I applied that patch to master and have been using it every day.
This morning I FINALLY hit it... on a NEW FILE, where I'm
switching in and out of a branch with a new file. I'm not sure
all the previous occurrences were on new files, but I could
believe they were.
Anyway, with that clue in hand, I was able to narrow it down to a
simple reproduction recipe.
I'm running emacs built from this commit to master:
commit ef587bf6b46b2ea3ef91b260ac2542666081260d
Author: Stefan Monnier <monnier <at> iro.umontreal.ca>
Date: Thu Oct 3 13:32:09 2024
With your cl-assert patch applied on top of that.
GNU Emacs 31.0.50 (build 1, x86_64-pc-linux-gnu, cairo version 1.18.2) of 2024-10-04
(We won't use git in the recipe, but eglot requires project-root
which I can't get to work without git (which also seems like a
bug, albeit unrelated...))
1. cargo new --vcs git foo
2. echo 'mod foo;' >> foo/src/main.rs
3. emacs -Q foo/src/foo.rs
(foo.rs DOES NOT EXIST YET -- that is part of the bug)
4. M-x rust-ts-mode RET
5. (progn (setq track-changes-record-errors 'trace) (setq debug-on-error t))
6. C-x C-f main.rs RET
7. M-x eglot RET
8. echo 'struct foo;' > foo/src/foo.rs
9. C-x C-f foo.rs RET
10. yes RET
BOOM TODAY
Note that #8 must be an edit behind emacs's back. In my daily
life, what's happening there is I had the buffer 'foo.rs' open
earlier while working on the branch where I added it, I'd
switched to a branch without it, and then I switch back, and
everything fell apart when I reverted it.
Debugger entered--Lisp error: (cl-assertion-failed ((track-changes--sane-state-p) nil))
cl--assertion-failed((track-changes--sane-state-p))
track-changes--before(1 13)
track-changes--after(1 13 0)
insert-file-contents("/home/epg/tmp/emacsbug73041/foo/src/foo.rs" t nil nil if-regular)
revert-buffer-insert-file-contents--default-function("/home/epg/tmp/emacsbug73041/foo/src/foo.rs" nil)
revert-buffer--default(t t)
revert-buffer(t t)
find-file-noselect("~/tmp/emacsbug73041/foo/src/foo.rs" nil nil t)
find-file("~/tmp/emacsbug73041/foo/src/foo.rs" t)
funcall-interactively(find-file "~/tmp/emacsbug73041/foo/src/foo.rs" t)
call-interactively(find-file nil nil)
command-execute(find-file)
I hope this is enough to go on now.
Thanks!
--
Eric Gillespie <*> epg <at> pretzelnet.org
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 17 Oct 2024 15:47:01 GMT) Full text and rfc822 format available.Message #29 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Eric Gillespie <epg <at> pretzelnet.org> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 17 Oct 2024 11:45:24 -0400
[Message part 1 (text/plain, inline)]
> 1. cargo new --vcs git foo
> 2. echo 'mod foo;' >> foo/src/main.rs
> 3. emacs -Q foo/src/foo.rs
> (foo.rs DOES NOT EXIST YET -- that is part of the bug)
> 4. M-x rust-ts-mode RET
> 5. (progn (setq track-changes-record-errors 'trace) (setq debug-on-error t))
> 6. C-x C-f main.rs RET
> 7. M-x eglot RET
> 8. echo 'struct foo;' > foo/src/foo.rs
> 9. C-x C-f foo.rs RET
> 10. yes RET
Thanks, I managed to reproduce it locally with:
rm foo.c
emacs -Q foo.c -f eglot
echo 'struct foo;' > foo.c
M-x revert-buffer RET
and that pointed to a bug in `track-changes.el` which could explain the
problem you reported originally.
So, I pushed the patch below to `emacs-30`.
It seems part of the problem also comes from `insert-file-contents`
which apparently fails to run the `before-change-functions` in the above
recipe, but I haven't investigated this any further yet, and
track-changes *should* tolerate this (potentially emitting a warning
along the way), and with the patch, the above recipe now works for me
without error nor warning.
Stefan
[track-changes.patch (text/x-diff, inline)]
diff --git a/lisp/emacs-lisp/track-changes.el b/lisp/emacs-lisp/track-changes.el
index 92d14959763..1d5c5e9917a 100644
--- a/lisp/emacs-lisp/track-changes.el
+++ b/lisp/emacs-lisp/track-changes.el
@@ -555,16 +555,16 @@ track-changes--before
(defun track-changes--after (beg end len)
(cl-assert track-changes--state)
- (and (eq track-changes--before-clean 'unset)
- (not track-changes--before-no)
- ;; This can be a sign that a `before-change-functions' went missing,
- ;; or that we called `track-changes--clean-state' between
- ;; a `before-change-functions' and `after-change-functions'.
- (track-changes--before beg end))
- (setq track-changes--before-clean nil)
(let ((offset (- (- end beg) len)))
- (cl-incf track-changes--before-end offset)
(cl-incf track-changes--buffer-size offset)
+ (if (and (eq track-changes--before-clean 'unset)
+ (not track-changes--before-no))
+ ;; This can be a sign that a `before-change-functions' went missing,
+ ;; or that we called `track-changes--clean-state' between
+ ;; a `before-change-functions' and `after-change-functions'.
+ (track-changes--before beg end)
+ (cl-incf track-changes--before-end offset))
+ (setq track-changes--before-clean nil)
(if (not (or track-changes--before-no
(save-restriction
(widen)
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Fri, 10 Jan 2025 18:26:01 GMT) Full text and rfc822 format available.Message #32 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Morgan Willcock <morgan <at> ice9.digital> To: Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>, Eric Gillespie <epg <at> pretzelnet.org> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Fri, 10 Jan 2025 18:25:21 +0000
I am still seeing similar issues with Eglot and Phpactor with Emacs 30.0.93, with this warning occurring fairly regularly: track-changes--after: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end)) I haven't figured out a way to reproduce it yet, but I've set track-changes-record-errors to 'trace to try and get more information the next time that it occurs. -- Morgan Willcock
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Sat, 25 Jan 2025 08:42:02 GMT) Full text and rfc822 format available.Message #35 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Eli Zaretskii <eliz <at> gnu.org> To: monnier <at> iro.umontreal.ca, Morgan Willcock <morgan <at> ice9.digital> Cc: 73041 <at> debbugs.gnu.org, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Sat, 25 Jan 2025 10:41:13 +0200
> From: Morgan Willcock <morgan <at> ice9.digital> > Cc: Eric Gillespie <epg <at> pretzelnet.org>, 73041 <at> debbugs.gnu.org, Eli > Zaretskii <eliz <at> gnu.org> > Date: Fri, 10 Jan 2025 18:25:21 +0000 > > I am still seeing similar issues with Eglot and Phpactor with Emacs > 30.0.93, with this warning occurring fairly regularly: > > track-changes--after: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end)) > > I haven't figured out a way to reproduce it yet, but I've set > track-changes-record-errors to 'trace to try and get more information > the next time that it occurs. Stefan, any suggestions?
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Sun, 26 Jan 2025 00:17:02 GMT) Full text and rfc822 format available.Message #38 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Eli Zaretskii <eliz <at> gnu.org> Cc: 73041 <at> debbugs.gnu.org, epg <at> pretzelnet.org, Morgan Willcock <morgan <at> ice9.digital> Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Sat, 25 Jan 2025 19:16:18 -0500
> I am still seeing similar issues with Eglot and Phpactor with Emacs
> 30.0.93, with this warning occurring fairly regularly:
>
> track-changes--after: Assertion failed: (or track-changes--before-no (<=
> track-changes--before-beg (track-changes--state-beg track-changes--state)
> beg end (track-changes--state-end track-changes--state)
> track-changes--before-end))
>
> I haven't figured out a way to reproduce it yet, but I've set
> track-changes-record-errors to 'trace to try and get more information
> the next time that it occurs.
Thanks. Send us the contents of `track-changes--error-log` (but note
that it can contain sensitive information) next time it occurs.
Stefan
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Wed, 29 Jan 2025 18:32:02 GMT) Full text and rfc822 format available.Message #41 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Morgan Willcock <morgan <at> ice9.digital> To: Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Wed, 29 Jan 2025 18:31:25 +0000
Stefan Monnier <monnier <at> iro.umontreal.ca> writes:
>> I haven't figured out a way to reproduce it yet, but I've set
>> track-changes-record-errors to 'trace to try and get more information
>> the next time that it occurs.
>
> Thanks. Send us the contents of `track-changes--error-log` (but note
> that it can contain sensitive information) next time it occurs.
I encountered the issue again while doing a query-replace in a buffer
narrowed to a single function, after moving to the beginning of the
buffer.
Here is the relevant part of my messages buffer:
Beginning of buffer
track-changes--after: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end))
Mark set
track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers))
Undo
Saving file /home/mwillcock/path/to/file/UserChooser.php...
Use C-x 4 s to update shadows.
Wrote /home/mwillcock/path/to/file/UserChooser.php
track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers))
The undo was because I had tried to answer "y" to the next replacement,
but I imagine that the track-changes error had exited the query-replace
process because the "y" was inserted into the buffer.
Here is the value of track-changes--error-log:
(("UserChooser.php" #1=(buffer-size 3999 4001)
((t track-changes--recover-from-error (#1#) nil)
(t track-changes-fetch
(#2=
#s(track-changes--tracker :signal eglot--track-changes-signal
:state
#s(track-changes--state :beg 1156
:end 1858
:before error
:next nil)
:nobefore nil :immediate nil)
#<subr F616e6f6e796d6f75732d6c616d626461_anonymous_lambda_135>)
nil)
(t eglot--track-changes-fetch (#2#) nil)
(t eglot--track-changes-signal (#2# 112) nil)
(t track-changes--before (1292 1298) nil)
(t replace-match-maybe-edit
(#3="is_null" nil t nil (1292 1298 #<buffer UserChooser.php>)
nil)
nil)
(t perform-replace
(#4="!isset" #3# t nil nil nil nil nil nil nil nil) nil)
(t query-replace (#4# #3# nil nil nil nil nil) nil)
(t funcall-interactively
(query-replace #4# #3# nil nil nil nil nil) nil)
(t command-execute (query-replace) nil))
[110 (nil . self-insert-command) 117 (nil . self-insert-command) 108
(nil . self-insert-command) 108 (nil . self-insert-command) 108
(nil . self-insert-command) backspace
(nil . c-electric-backspace) 59 (nil . c-electric-semi&comma)
134217765 (nil . query-replace) return (nil . exit-minibuffer)
121 121]))
If it is relevant, I have also seen this being logged by Eglot in a
previous Emacs session:
[eglot] Server reports (type=2): Diagnostic provider "php" errored with "Col number cannot be less than 1 (got "0")", removing from pool
The Eglot version is 1.18.
The LSP server is Phpactor version "2024.11.28.1".
The major-mode is php-ts-mode, as supplied with Emacs 30.0.93.
--
Morgan Willcock
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 30 Jan 2025 18:34:01 GMT) Full text and rfc822 format available.Message #44 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Morgan Willcock <morgan <at> ice9.digital> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 30 Jan 2025 13:33:05 -0500
>>> I haven't figured out a way to reproduce it yet, but I've set
>>> track-changes-record-errors to 'trace to try and get more information
>>> the next time that it occurs.
Hmm... looks like you're using the `emacs-30` code, where
`track-changes-record-errors` doesn't yet support `trace`. 🙁
> I encountered the issue again while doing a query-replace in a buffer
> narrowed to a single function, after moving to the beginning of the
> buffer.
>
> Here is the relevant part of my messages buffer:
>
> Beginning of buffer
> track-changes--after: Assertion failed: (or track-changes--before-no (<= track-changes--before-beg (track-changes--state-beg track-changes--state) beg end (track-changes--state-end track-changes--state) track-changes--before-end))
> Mark set
> track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers))
> Undo
> Saving file /home/mwillcock/path/to/file/UserChooser.php...
> Use C-x 4 s to update shadows.
> Wrote /home/mwillcock/path/to/file/UserChooser.php
> track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers))
>
> The undo was because I had tried to answer "y" to the next replacement,
> but I imagine that the track-changes error had exited the query-replace
> process because the "y" was inserted into the buffer.
Thanks for the details, it's very helpful.
These seem to point not just to errors in other parts of Emacs, but in
errors in `track-changes.el` itself. I'll have to look more closely to
try and figure out where it might come from.
Just to rule some obvious candidate, could you clarify exactly which
version of Emacs you were running? Is it built from the Emacs-30.0.93
tarball or from a particular commit in Git, or does it come from elsewhere?
Stefan
bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Thu, 30 Jan 2025 19:55:01 GMT) Full text and rfc822 format available.Message #47 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Morgan Willcock <morgan <at> ice9.digital> To: Stefan Monnier <monnier <at> iro.umontreal.ca> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Thu, 30 Jan 2025 19:54:24 +0000
Stefan Monnier <monnier <at> iro.umontreal.ca> writes: >>>> I haven't figured out a way to reproduce it yet, but I've set >>>> track-changes-record-errors to 'trace to try and get more information >>>> the next time that it occurs. > > Hmm... looks like you're using the `emacs-30` code, where > `track-changes-record-errors` doesn't yet support `trace`. 🙁 > >> I encountered the issue again while doing a query-replace in a buffer >> narrowed to a single function, after moving to the beginning of the >> buffer. >> >> Here is the relevant part of my messages buffer: >> >> Beginning of buffer >> track-changes--after: Assertion failed: (or >> track-changes--before-no (<= track-changes--before-beg >> (track-changes--state-beg track-changes--state) beg end >> (track-changes--state-end track-changes--state) >> track-changes--before-end)) >> Mark set >> track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers)) >> Undo >> Saving file /home/mwillcock/path/to/file/UserChooser.php... >> Use C-x 4 s to update shadows. >> Wrote /home/mwillcock/path/to/file/UserChooser.php >> track-changes-fetch: Assertion failed: (not (memq id track-changes--clean-trackers)) >> >> The undo was because I had tried to answer "y" to the next replacement, >> but I imagine that the track-changes error had exited the query-replace >> process because the "y" was inserted into the buffer. > > Thanks for the details, it's very helpful. > > These seem to point not just to errors in other parts of Emacs, but in > errors in `track-changes.el` itself. I'll have to look more closely to > try and figure out where it might come from. > > Just to rule some obvious candidate, could you clarify exactly which > version of Emacs you were running? Is it built from the Emacs-30.0.93 > tarball or from a particular commit in Git, or does it come from elsewhere? The Emacs I am using is from the Emacs-30.0.93 tarball. -- Morgan Willcock
Stefan Monnier <monnier <at> iro.umontreal.ca>
to control <at> debbugs.gnu.org.
(Sun, 02 Feb 2025 23:15:02 GMT) Full text and rfc822 format available.bug-gnu-emacs <at> gnu.org:bug#73041; Package emacs.
(Mon, 03 Feb 2025 02:34:02 GMT) Full text and rfc822 format available.Message #52 received at 73041 <at> debbugs.gnu.org (full text, mbox):
From: Stefan Monnier <monnier <at> iro.umontreal.ca> To: Morgan Willcock <morgan <at> ice9.digital> Cc: 73041 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>, epg <at> pretzelnet.org Subject: Re: bug#73041: 30.0.90; track-changes-mode logs warnings (#70541 regression? not actually fixed?) Date: Sun, 02 Feb 2025 21:33:06 -0500
> Here is the value of track-changes--error-log:
>
> (("UserChooser.php" #1=(buffer-size 3999 4001)
> ((t track-changes--recover-from-error (#1#) nil)
> (t track-changes-fetch
> (#2=
> #s(track-changes--tracker :signal eglot--track-changes-signal
> :state
> #s(track-changes--state :beg 1156
> :end 1858
> :before error
> :next nil)
> :nobefore nil :immediate nil)
> #<subr F616e6f6e796d6f75732d6c616d626461_anonymous_lambda_135>)
> nil)
BTW, this error seems to come not from `track-changes.el` but from some
other part of Emacs. The fix I suggested in the other bug should
hopefully fix the consequence of this problem but of course not
its origin.
To find its origin, I'd need more info (like the one we could get with
`(setq track-changes-record-errors 'trace)` on `master`, or some way to
reproduce the bug).
Stefan
Debbugs Internal Request <help-debbugs <at> gnu.org>
to internal_control <at> debbugs.gnu.org.
(Tue, 04 Mar 2025 12:24:06 GMT) Full text and rfc822 format available.
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.