unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
@ 2021-02-22 22:22 Ryan Olson
  2021-02-22 22:36 ` Lars Ingebrigtsen
  0 siblings, 1 reply; 10+ messages in thread
From: Ryan Olson @ 2021-02-22 22:22 UTC (permalink / raw)
  To: 46713

In GNU Emacs 27.1 (build 1, x86_64-apple-darwin18.7.0, NS
appkit-1671.60 Version 10.14.6 (Build 18G95))
 of 2020-08-11 built on builder10-14.porkrind.org
Windowing system distributor 'Apple', version 10.3.1894
System Description:  Mac OS X 10.15.7

Recent messages:
Wrote /Users/CENSORED/CENSORED/src/CENSORED/CENSORED/CENSORED.js
Quit
2 (#o2, #x2, ?\C-b)
Saving file /Users/CENSORED/CENSORED/src/CENSORED/CENSORED/CENSORED.js...
Wrote /Users/CENSORED/CENSORED/src/CENSORED/CENSORED/CENSORED.js
Error during redisplay: (internal--syntax-propertize 2120) signaled
(error "Variable binding depth exceeds max-specpdl-size")
back-to-indentation: internal--syntax-propertize did not move
syntax-propertize--doneError during redisplay:
(internal--syntax-propertize 2645) signaled (error "Variable binding
depth exceeds max-specpdl-size")
Error during redisplay: (jit-lock-function 2121) signaled (error
"internal--syntax-propertize did not move syntax-propertize--done")
Error during redisplay: (jit-lock-function 3138) signaled (error
"Variable binding depth exceeds max-specpdl-size")
Error during redisplay: (jit-lock-function 3638) signaled (error
"Variable binding depth exceeds max-specpdl-size")

Configured using:
 'configure --with-ns '--enable-locallisppath=/Library/Application
 Support/Emacs/${version}/site-lisp:/Library/Application
 Support/Emacs/site-lisp' --with-modules'

Configured features:
NOTIFY KQUEUE ACL GNUTLS LIBXML2 ZLIB TOOLKIT_SCROLL_BARS NS MODULES
THREADS JSON PDUMPER

Important settings:
  value of $LANG: en_US.UTF-8
  locale-coding-system: utf-8-unix

Major mode: JavaScript[JSX]

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  line-number-mode: t
  transient-mark-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message rmc puny dired dired-loaddefs
format-spec rfc822 mml mml-sec password-cache epa derived epg epg-config
gnus-util rmail rmail-loaddefs text-property-search mm-decode mm-bodies
mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader sendmail
rfc2047 rfc2045 ietf-drums mm-util mail-prsvr mail-utils time-date
vc-git diff-mode easy-mmode js json subr-x map seq byte-opt gv bytecomp
byte-compile cconv imenu cc-mode cc-fonts easymenu cc-guess cc-menus
cc-cmds cc-styles cc-align cc-engine cc-vars cc-defs cl-loaddefs cl-lib
tooltip eldoc electric uniquify ediff-hook vc-hooks lisp-float-type
mwheel term/ns-win ns-win ucs-normalize mule-util term/common-win
tool-bar dnd fontset image regexp-opt fringe tabulated-list replace
newcomment text-mode elisp-mode lisp-mode prog-mode register page
tab-bar menu-bar rfn-eshadow isearch timer select scroll-bar mouse
jit-lock font-lock syntax facemenu font-core term/tty-colors frame
minibuffer cl-generic 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 charscript charprop case-table epa-hook jka-cmpr-hook help
simple abbrev obarray cl-preloaded nadvice loaddefs button faces
cus-face macroexp files text-properties overlay sha1 md5 base64 format
env code-pages mule custom widget hashtable-print-readable backquote
threads kqueue cocoa ns multi-tty make-network-process emacs)

Memory information:
((conses 16 77539 14591)
 (symbols 48 9435 1)
 (strings 32 25046 1242)
 (string-bytes 1 941698)
 (vectors 16 13840)
 (vector-slots 8 178575 12426)
 (floats 8 30 274)
 (intervals 56 717 4)
 (buffers 1000 13))





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-22 22:22 bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line Ryan Olson
@ 2021-02-22 22:36 ` Lars Ingebrigtsen
  2021-02-23  2:03   ` Ryan Olson
  0 siblings, 1 reply; 10+ messages in thread
From: Lars Ingebrigtsen @ 2021-02-22 22:36 UTC (permalink / raw)
  To: Ryan Olson; +Cc: 46713

Ryan Olson <ryan.olson.x@gmail.com> writes:

> (error "Variable binding depth exceeds max-specpdl-size")
> back-to-indentation: internal--syntax-propertize did not move
> syntax-propertize--doneError during redisplay:
> (internal--syntax-propertize 2645) signaled (error "Variable binding
> depth exceeds max-specpdl-size")

Do you have a recipe to reproduce this error, starting from "emacs -Q"?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-22 22:36 ` Lars Ingebrigtsen
@ 2021-02-23  2:03   ` Ryan Olson
  2021-02-23 15:36     ` Lars Ingebrigtsen
  0 siblings, 1 reply; 10+ messages in thread
From: Ryan Olson @ 2021-02-23  2:03 UTC (permalink / raw)
  To: Lars Ingebrigtsen; +Cc: 46713

Hi Lars,

I do. I'm not sure why it takes this, but for my specific js file
(with a few things replaced -- because it's company code) it has
issues.

Repro steps:
1. Open foo.js with emacs -Q --eval '(setq-default js-indent-level 2
indent-tabs-mode nil)' foo.js
2. Go to line 61
3. Delete brace at end of that line
4. Save
5. Add brace back
6. Press RET
7. Errors happen

foo.js:
import React, { useEffect } from 'react';
import { useDispatch, useSelector } from 'react-redux';
import { useHistory } from 'react-router';
import { Link } from 'react-router-dom';
import { Grid } from '@rmwc/grid';

import {
  getFoos,
  selectFoos,
  selectIsLoadingFoos,
  selectTotalTimesFoosFetched,
} from 'store/slices/foos.slice';
import { clearFoo } from 'store/slices/foo.slice';
import {
  filterDefinitions,
  getUniqueFooKey,
  searchProperties,
  getFeaturesForFoo,
} from './foosHelpers';
import { useFilters } from 'components/Filtering/useFilters';
import Fab from 'components/Fab';
import SortableDataTable from 'components/SortableDataTable';
import FilterBar from 'components/Filtering/FilterBar';
import useUpdateEffect from 'hooks/useUpdateEffect';

export default function FoosList() {
  const dispatch = useDispatch();
  const history = useHistory();
  const isLoading = useSelector(selectIsLoadingFoos);
  const foos = useSelector(selectFoos);
  const totalTimesFoosFetched = useSelector(
    selectTotalTimesFoosFetched
  );
  const filterBag = useFilters(
    'foos',
    foos,
    filterDefinitions,
    searchProperties
  );
  const { getFilterByKey, clearFilter } = filterBag;

  const canAddFoo = useSelector(state => state.permissions.canAddFoo);

  function requestFoos() {
    dispatch(getFoos({ dispatch }));
  }

  // Get foos from the server (if needed).
  // It attempts to skip reloading the foos, when it can.
  useEffect(() => {
    const isFirstLoad = totalTimesFoosFetched === 0;
    // Since we're only checking this on mount, I think
    // it's safe to say that if we don't have foos and it's
    // not our first load (while mounting), then this would be a
    // place where we've invalidated foos by clearing them out.
    // (to force a load)
    const foosInvalidated = !isFirstLoad && foos.length === 0;

    const shouldFetch = isFirstLoad || foosInvalidated;

    if (shouldFetch) {
      requestFoos();
    }
  }, []);

  // When show foos filter changes,
  // get foos from the server.
  useUpdateEffect(() => {
    requestFoos();
  }, []);

  // Clear foo data from redux.
  useEffect(() => {
    // Clear foo itself.
    dispatch(clearFoo());
  }, []);

  // If we initially load foos and none come back that follow
  // our initial filter (only show centralized foos == true)
  // Remove that filter and fetch foos again.
  useEffect(() => {
    if (totalTimesFoosFetched === 1 && foos.length === 0) {
      // NOTE: this triggers a different effect to get foos that
      // match this filter.
      clearFilter();
    }
  }, [foos, totalTimesFoosFetched]);

  function getFooLink(foo) {
    if (hasValue(foo.fooCode)) {
      return `/foos/foo/fooCode/${foo.fooCode}`;
    }

    if (foo(foo)) {
      return `/foos/foo//${foo.id}`;
    }

    // Shouldn't get here. If all else fails, it shouldn't route anywhere.
    return '#';
  }

  return (
    <Grid>
      <FilterBar filterBag={filterBag}>
        < />
      </FilterBar>
      <SortableDataTable
        data={filterBag.filteredData}
        isLoading={isLoading}
        noRowsMessage="No foos to display."
        schema={{
          getRowKey: getUniqueFooKey,
          columns: [
            {
              headerName: 'Foo Name',
              getColumnValue: foo => (
                <Link
                  to={getFooLink(foo)}
                  className={sharedCss.colorPrimary}
                >
                  {foo.fooName}
                </Link>
              ),
              getSortValue: foo => foo.fooName,
            },
            {
              headerName: 'Foo Code',
              getColumnValue: foo => foo.fooCode,
              className: globalCss.width100,
            },
            {
              headerName: 'Site ID',
              // If there's no site id, it comes down as 0.
              getColumnValue: foo => foo.siteId || null,
              className: globalCss.width100,
            },
            {
              headerName: 'Master Foo ID',
              getColumnValue: foo => foo.id,
              className: globalCss.width100,
            },
          ].filter(Boolean),
        }}
      />
      {canAddFoo && (
        <Fab icon="add" onClick={() => history.push('/foos/new')} />
      )}
    </Grid>
  );
}

On Mon, Feb 22, 2021 at 3:36 PM Lars Ingebrigtsen <larsi@gnus.org> wrote:
>
> Ryan Olson <ryan.olson.x@gmail.com> writes:
>
> > (error "Variable binding depth exceeds max-specpdl-size")
> > back-to-indentation: internal--syntax-propertize did not move
> > syntax-propertize--doneError during redisplay:
> > (internal--syntax-propertize 2645) signaled (error "Variable binding
> > depth exceeds max-specpdl-size")
>
> Do you have a recipe to reproduce this error, starting from "emacs -Q"?
>
> --
> (domestic pets only, the antidote for overdose, milk.)
>    bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-23  2:03   ` Ryan Olson
@ 2021-02-23 15:36     ` Lars Ingebrigtsen
  2021-02-23 16:31       ` Ryan Olson
  2022-06-19 13:53       ` Lars Ingebrigtsen
  0 siblings, 2 replies; 10+ messages in thread
From: Lars Ingebrigtsen @ 2021-02-23 15:36 UTC (permalink / raw)
  To: Ryan Olson; +Cc: 46713

Ryan Olson <ryan.olson.x@gmail.com> writes:

> Repro steps:
> 1. Open foo.js with emacs -Q --eval '(setq-default js-indent-level 2
> indent-tabs-mode nil)' foo.js
> 2. Go to line 61
> 3. Delete brace at end of that line
> 4. Save
> 5. Add brace back
> 6. Press RET
> 7. Errors happen

Thanks for the clear recipe.  With it, I'm also seeing errors (in both
Emacs 27 and 28):

Debugger entered--Lisp error: (error "internal--syntax-propertize did not move syntax-pr...")
  backward-prefix-chars()
  back-to-indentation()
  current-left-margin()
  move-to-left-margin(nil t)
  #f(compiled-function () #<bytecode 0x155c01f42651>)()
  self-insert-command(1)
  newline(nil 1)
  funcall-interactively(newline nil 1)
  call-interactively(newline nil nil)

But I'm not seeing the "binding depth" one.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-23 15:36     ` Lars Ingebrigtsen
@ 2021-02-23 16:31       ` Ryan Olson
  2021-02-24 16:47         ` Lars Ingebrigtsen
  2022-06-19 13:53       ` Lars Ingebrigtsen
  1 sibling, 1 reply; 10+ messages in thread
From: Ryan Olson @ 2021-02-23 16:31 UTC (permalink / raw)
  To: Lars Ingebrigtsen; +Cc: 46713

[-- Attachment #1: Type: text/plain, Size: 1264 bytes --]

It was with something like this that I saw the error I reported. I’m not
sure how to reproduce that.

I guess we can instead fix this one (that _is_ reproducible)?

On Tue, Feb 23, 2021 at 8:36 AM Lars Ingebrigtsen <larsi@gnus.org> wrote:

> Ryan Olson <ryan.olson.x@gmail.com> writes:
>
> > Repro steps:
> > 1. Open foo.js with emacs -Q --eval '(setq-default js-indent-level 2
> > indent-tabs-mode nil)' foo.js
> > 2. Go to line 61
> > 3. Delete brace at end of that line
> > 4. Save
> > 5. Add brace back
> > 6. Press RET
> > 7. Errors happen
>
> Thanks for the clear recipe.  With it, I'm also seeing errors (in both
> Emacs 27 and 28):
>
> Debugger entered--Lisp error: (error "internal--syntax-propertize did not
> move syntax-pr...")
>   backward-prefix-chars()
>   back-to-indentation()
>   current-left-margin()
>   move-to-left-margin(nil t)
>   #f(compiled-function () #<bytecode 0x155c01f42651>)()
>   self-insert-command(1)
>   newline(nil 1)
>   funcall-interactively(newline nil 1)
>   call-interactively(newline nil nil)
>
> But I'm not seeing the "binding depth" one.
>
> --
> (domestic pets only, the antidote for overdose, milk.)
>    bloggy blog: http://lars.ingebrigtsen.no
>
-- 
Sent from my iPhone

[-- Attachment #2: Type: text/html, Size: 1932 bytes --]

^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-23 16:31       ` Ryan Olson
@ 2021-02-24 16:47         ` Lars Ingebrigtsen
  0 siblings, 0 replies; 10+ messages in thread
From: Lars Ingebrigtsen @ 2021-02-24 16:47 UTC (permalink / raw)
  To: Ryan Olson; +Cc: 46713

Ryan Olson <ryan.olson.x@gmail.com> writes:

> I guess we can instead fix this one (that _is_ reproducible)?

Yup.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2021-02-23 15:36     ` Lars Ingebrigtsen
  2021-02-23 16:31       ` Ryan Olson
@ 2022-06-19 13:53       ` Lars Ingebrigtsen
  2022-06-25 12:03         ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
  1 sibling, 1 reply; 10+ messages in thread
From: Lars Ingebrigtsen @ 2022-06-19 13:53 UTC (permalink / raw)
  To: Ryan Olson; +Cc: 46713, Stefan Monnier

Lars Ingebrigtsen <larsi@gnus.org> writes:

> Thanks for the clear recipe.  With it, I'm also seeing errors (in both
> Emacs 27 and 28):
>
> Debugger entered--Lisp error: (error "internal--syntax-propertize did
> not move syntax-pr...")
>   backward-prefix-chars()
>   back-to-indentation()
>   current-left-margin()
>   move-to-left-margin(nil t)
>   #f(compiled-function () #<bytecode 0x155c01f42651>)()
>   self-insert-command(1)
>   newline(nil 1)
>   funcall-interactively(newline nil 1)
>   call-interactively(newline nil nil)

Perhaps Stefan knows what's going on here; added to the CCs.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2022-06-19 13:53       ` Lars Ingebrigtsen
@ 2022-06-25 12:03         ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
  2022-06-25 12:43           ` Lars Ingebrigtsen
  0 siblings, 1 reply; 10+ messages in thread
From: Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors @ 2022-06-25 12:03 UTC (permalink / raw)
  To: Lars Ingebrigtsen; +Cc: Ryan Olson, 46713

> Perhaps Stefan knows what's going on here; added to the CCs.

[ The recipe works without the js-indent-level and
  indent-tabs-mode settings.  ]

I think the crux of the matter is the following:

                 (run-hook-wrapped
                  'syntax-propertize-extend-region-functions
                  (lambda (f)
                    (let ((new (funcall f start end))
                          ;; Avoid recursion!
                          (syntax-propertize--done most-positive-fixnum))
                      (if (or (null new)
                              (and (>= (car new) start) (<= (cdr new) end)))
                          nil

Where the binding of `syntax-propertize--done` should be around the call
to `f` (apparently I introduced this bug in commit
3928ef2dd5b8febf3b1d9c1bfb22af3698d16bea).

But the bug also shows that it's difficult for
syntax-propertize-functions to know when they use `syntax-ppss` since it
can happen as a side-effect of forward-sexp and hence I'm beginning to
think it's not practical to require them to call
`syntax-ppss-flush-cache` manually.

I'm thinking of installing the patch below on `master`.  We could
install a simpler patch on `emacs-28` which just swaps the two vars in
the above `let` (and changes it into a `let*`).

WDYT?


        Stefan


    * lisp/emacs-lisp/syntax.el: Rework the handling of nested calls.

    Nested calls to `syntax-ppss` and `syntax-propertize` can easily
    happen unexpectedly via ondemand propertizing or `forward-sexp`.
    Refine the handling of nested calls so we detect them more reliably
    (e.g. also within `syntax-propertize-extend-region-functions`)
    and so that the `syntax-ppss` cache is automatically flushed in case
    it might have been filled with data that's become obsolete since.

    (syntax-propertize--inhibit-flush): Delete var.
    (syntax-propertize--in-process-p): New function to replace it.
    (syntax-ppss-flush-cache): Use it.
    (syntax-ppss--updated-cache): New var.
    (syntax-propertize): Make `syntax-propertize--done` binding apply to
    `syntax-propertize-extend-region-functions` as well, as intended (fixes
    bug#46713).  Use `syntax-ppss--updated-cache` to flush
    syntax-ppss cache at the end when needed.
    Don't bind `syntax-propertize--inhibit-flush` any more.
    (syntax-ppss): Set `syntax-ppss--updated-cache` when applicable.


diff --git a/lisp/emacs-lisp/syntax.el b/lisp/emacs-lisp/syntax.el
index a4d7beade13..36b0c56e953 100644
--- a/lisp/emacs-lisp/syntax.el
+++ b/lisp/emacs-lisp/syntax.el
@@ -345,10 +345,16 @@ syntax-propertize-via-font-lock
 (defvar-local syntax-ppss-table nil
   "Syntax-table to use during `syntax-ppss', if any.")
 
-(defvar-local syntax-propertize--inhibit-flush nil
-  "If non-nil, `syntax-ppss-flush-cache' only flushes the ppss cache.
-Otherwise it flushes both the ppss cache and the properties
-set by `syntax-propertize'")
+(defun syntax-propertize--in-process-p ()
+  "Non-nil if we're inside `syntax-propertize'.
+This is used to avoid infinite recursion as well as to handle cases where
+`syntax-ppss' is called when the final `syntax-table' properties have not
+yet been setup, in which case we may end up putting invalid info into the cache.
+It's also used so that `syntax-ppss-flush-cache' can be used from within
+`syntax-propertize' without ruining the `syntax-table' already set."
+  (eq syntax-propertize--done most-positive-fixnum))
+
+(defvar-local syntax-ppss--updated-cache nil)
 
 (defun syntax-propertize (pos)
   "Ensure that syntax-table properties are set until POS (a buffer point)."
@@ -370,21 +376,24 @@ syntax-propertize
         (with-silent-modifications
           (with-syntax-table (or syntax-ppss-table (syntax-table))
             (make-local-variable 'syntax-propertize--done) ;Just in case!
+            ;; Make sure we let-bind it only buffer-locally.
+            (make-local-variable 'syntax-ppss--updated-cache)
             (let* ((start (max (min syntax-propertize--done (point-max))
                                (point-min)))
                    (end (max pos
                              (min (point-max)
                                   (+ start syntax-propertize-chunk-size))))
                    (first t)
-                   (repeat t))
+                   (repeat t)
+                   (syntax-ppss--updated-cache nil))
               (while repeat
                 (setq repeat nil)
                 (run-hook-wrapped
                  'syntax-propertize-extend-region-functions
                  (lambda (f)
-                   (let ((new (funcall f start end))
-                         ;; Avoid recursion!
-                         (syntax-propertize--done most-positive-fixnum))
+                   ;; Bind `syntax-propertize--done' to avoid recursion!
+                   (let* ((syntax-propertize--done most-positive-fixnum)
+                          (new (funcall f start end)))
                      (if (or (null new)
                              (and (>= (car new) start) (<= (cdr new) end)))
                          nil
@@ -399,20 +408,26 @@ syntax-propertize
               ;; Flush ppss cache between the original value of `start' and that
               ;; set above by syntax-propertize-extend-region-functions.
               (syntax-ppss-flush-cache start)
-              ;; Move the limit before calling the function, so the function
-              ;; can use syntax-ppss.
+              ;; Move the limit before calling the function, so it's
+              ;; done in case of errors.
               (setq syntax-propertize--done end)
               ;; (message "syntax-propertizing from %s to %s" start end)
               (remove-text-properties start end
                                       '(syntax-table nil syntax-multiline nil))
-              ;; Make sure we only let-bind it buffer-locally.
-              (make-local-variable 'syntax-propertize--inhibit-flush)
-              ;; Let-bind `syntax-propertize--done' to avoid infinite recursion!
-              (let ((syntax-propertize--done most-positive-fixnum)
-                    ;; Let `syntax-propertize-function' call
-                    ;; `syntax-ppss-flush-cache' without worries.
-                    (syntax-propertize--inhibit-flush t))
-                (funcall syntax-propertize-function start end)))))))))
+              ;; Bind `syntax-propertize--done' to avoid recursion!
+              (let ((syntax-propertize--done most-positive-fixnum))
+                (funcall syntax-propertize-function start end)
+                (when syntax-ppss--updated-cache
+                  ;; `syntax-ppss' was called and updated the cache while we
+                  ;; were propertizing so we need to flush the part of the
+                  ;; cache that may have been rendered out-of-date by the new
+                  ;; properties.
+                  ;; We used to require syntax-propertize-functions to do that
+                  ;; manually when applicable, but nowadays the `syntax-ppss'
+                  ;; cache can be updated by too many functions, so the author
+                  ;; of the syntax-propertize-function may not be aware it
+                  ;; can happen.
+                  (syntax-ppss-flush-cache start))))))))))
 
 ;;; Link syntax-propertize with syntax.c.
 
@@ -487,10 +502,10 @@ syntax-ppss-narrow-start
 
 (define-obsolete-function-alias 'syntax-ppss-after-change-function
   #'syntax-ppss-flush-cache "27.1")
-(defun syntax-ppss-flush-cache (beg &rest ignored)
+(defun syntax-ppss-flush-cache (beg &rest _ignored)
   "Flush the cache of `syntax-ppss' starting at position BEG."
   ;; Set syntax-propertize to refontify anything past beg.
-  (unless syntax-propertize--inhibit-flush
+  (unless (syntax-propertize--in-process-p)
     (setq syntax-propertize--done (min beg syntax-propertize--done)))
   ;; Flush invalid cache entries.
   (dolist (cell (list syntax-ppss-wide syntax-ppss-narrow))
@@ -517,10 +532,16 @@ syntax-ppss-flush-cache
        (setcdr cell cache)))
     ))
 
-;;; FIXME: Explain this variable.  Currently only its last (5th) slot is used.
-;;; Perhaps the other slots should be removed?
+;; FIXME: Explain this variable.  Currently only its last (5th) slot is used.
+;; Perhaps the other slots should be removed?
+;; This variable is only used when `syntax-begin-function' is used and
+;; will hence be removed together with `syntax-begin-function'.
 (defvar syntax-ppss-stats
-  [(0 . 0) (0 . 0) (0 . 0) (0 . 0) (0 . 0) (2 . 2500)])
+  [(0 . 0) (0 . 0) (0 . 0) (0 . 0) (0 . 0) (2 . 2500)]
+  "Statistics about which case is more/less frequent in `syntax-ppss'.
+The 5th slot drives the heuristic to use `syntax-begin-function'.
+The rest is only useful if you're interested in tweaking the algorithm.")
+
 (defun syntax-ppss-stats ()
   (mapcar (lambda (x)
 	    (condition-case nil
@@ -658,6 +679,7 @@ syntax-ppss
 	       ;; populate the cache so we won't need to do it again soon.
 	       (t
 		(syntax-ppss--update-stats 3 pt-min pos)
+                (setq syntax-ppss--updated-cache t)
 
 		;; If `pt-min' is too far, add a few intermediate entries.
 		(while (> (- pos pt-min) (* 2 syntax-ppss-max-span))
@@ -692,6 +714,7 @@ syntax-ppss
 			(push pair ppss-cache)
 		      (setcar ppss-cache pair)))))))))
 
+          (setq syntax-ppss--updated-cache t)
 	  (setq ppss-last (cons pos ppss))
           (setcar cell ppss-last)
           (setcdr cell ppss-cache)






^ permalink raw reply related	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2022-06-25 12:03         ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
@ 2022-06-25 12:43           ` Lars Ingebrigtsen
  2022-06-26 13:00             ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
  0 siblings, 1 reply; 10+ messages in thread
From: Lars Ingebrigtsen @ 2022-06-25 12:43 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: Ryan Olson, 46713

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> I'm thinking of installing the patch below on `master`.  We could
> install a simpler patch on `emacs-28` which just swaps the two vars in
> the above `let` (and changes it into a `let*`).
>
> WDYT?

The fix on master sounds good, but we're just gearing up for 28.2 this
weekend, so while the change looks innocuous, I'd rather hold off until
after 28.2.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no





^ permalink raw reply	[flat|nested] 10+ messages in thread

* bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line
  2022-06-25 12:43           ` Lars Ingebrigtsen
@ 2022-06-26 13:00             ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
  0 siblings, 0 replies; 10+ messages in thread
From: Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors @ 2022-06-26 13:00 UTC (permalink / raw)
  To: Lars Ingebrigtsen; +Cc: Ryan Olson, 46713

> The fix on master sounds good, but we're just gearing up for 28.2 this
> weekend, so while the change looks innocuous, I'd rather hold off until
> after 28.2.

OK, pushed to `master`,


        Stefan






^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2022-06-26 13:00 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-22 22:22 bug#46713: 27.1; Variable binding depth exceeds max-specpdl-size in js-mode with js-indent-level 2 and indent-tabs-mode nil on new line Ryan Olson
2021-02-22 22:36 ` Lars Ingebrigtsen
2021-02-23  2:03   ` Ryan Olson
2021-02-23 15:36     ` Lars Ingebrigtsen
2021-02-23 16:31       ` Ryan Olson
2021-02-24 16:47         ` Lars Ingebrigtsen
2022-06-19 13:53       ` Lars Ingebrigtsen
2022-06-25 12:03         ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-06-25 12:43           ` Lars Ingebrigtsen
2022-06-26 13:00             ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).