emacs-orgmode@gnu.org archives
 help / color / mirror / code / Atom feed
From: David Dynerman <emperordali@block-party.net>
To: emacs-orgmode@gnu.org
Subject: Emacs hangs while loading org file with python blocks
Date: Tue, 06 Dec 2016 17:41:57 -0800	[thread overview]
Message-ID: <m2pol4cy6y.fsf@block-party.net> (raw)

[-- Attachment #1: Type: text/plain, Size: 3504 bytes --]

Dear list,

For several months I've been encountering a frustrating bug. My emacs hangs while initially loading an org file with ~3000 lines and around two dozen python blocks. If I press C-g during the lockup, emacs wakes up and the file is loaded. The hang does not re-occur after happening on initial load - the bug will reoccur only if I restart emacs and load the file again. The hang seems to be related to fontifying the python blocks: after I abort with C-g, some blocks will be unfontified. 

I previously emailed the list about this bug (Wed, 17 Aug 2016, subject 'Emacs hangs while loading .org file'), but didn't have an easy way to reproduce the problem.

I've now managed to extract a minimal org file that reproduces the hang:

#+BEGIN_SRC python
  """xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"""
#+END_SRC
#+BEGIN_SRC python
  class x:  
      def x(self):
          """
	  """
          pass
#+END_SRC

To reproduce:

1) Save above as bug.org
2) Open a fresh emacs process, and C-x f bug.org <RET>
3) Observe emacs is unresponsive. I've let it run as long as 30 mins without any change. 

I am able to reproduce with emacs -Q, only executing the following in *scratch*:

(add-to-list 'load-path "/usr/local/share/emacs/site-lisp/org") ;; org from git installed here
(require 'org)

This is with the latest org code from git (release_9.0-132-gd65aa3). The bug does NOT occur with org 8.2.10. I've been encountering this bug for several months, but unfortunately I don't remember exactly when I first saw it.

I'm running GNU Emacs 25.1.1 (x86_64-apple-darwin15.6.0, Carbon Version 157 AppKit 1404.47) of 2016-11-24.

I'm mystified as to what's going on here. It seems that removing any of the python code in the example above causes the bug to disappear, but I have no clue what the problem might be.

I've attached a profiler report captured during the hang. Here is the relevant part:

           - org-src-font-lock-fontify-block                    17419  46%
            - org-font-lock-ensure                              17240  45%
             - #<compiled 0x41086c73>                           17240  45%
              - font-lock-default-fontify-buffer                17240  45%
               - font-lock-fontify-region                       17240  45%
                - font-lock-default-fontify-region              17240  45%
                 - font-lock-fontify-syntactically-region              17240  45%
                  - python-font-lock-syntactic-face-function              17240  45%
                   - python-info-docstring-p                    17240  45%
                    - python-nav-backward-sexp                  17240  45%
                     - python-nav-forward-sexp                  17240  45%
                      - python-nav--forward-sexp                17240  45%
                       - python-info-end-of-block-p              17240  45%
                        - python-info-end-of-statement-p              17240  45%
                         - python-nav-end-of-statement              15444  41%
                            syntax-ppss                          8641  22%

It looks like emacs is getting stuck navigating the python blocks during fontification. The last org-mode call appears to be (org-font-lock-ensure).

If anyone has any suggestions on how to fix this, I'd be very interested to hearthem - I'm working with larger and larger org files containing python code, so I'm encountering this bug frequently.

Thank you very much,
David


[-- Attachment #2: fontify-bug-profile --]
[-- Type: application/octet-stream, Size: 300 bytes --]

+ ...                                                           19562  51%
+ redisplay_internal (C function)                               17420  46%
+ command-execute                                                 658   1%
+ timer-event-handler                                               9   0%

             reply	other threads:[~2016-12-07  1:42 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-07  1:41 David Dynerman [this message]
2016-12-07  3:41 ` Emacs hangs while loading org file with python blocks Clément Pit--Claudel
2016-12-07 13:41 ` Alan Schmitt
2016-12-07 18:33   ` David Dynerman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.orgmode.org/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=m2pol4cy6y.fsf@block-party.net \
    --to=emperordali@block-party.net \
    --cc=emacs-orgmode@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs/org-mode.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).