Inappropriate django indentation

[Oleg Lomaka]

There is indent after {% load commnts %} token. The indentation looks like that:

{% extends "site.html" %}
{% block title %}{{ object.title }}{% endblock %}
{% block content %}
  {% load comments %}
    <div class="content">
    </div>
  {% endblock %}

[lborgman (lennart-borgman)]

Thanks Oleg, could you please try the new beta 2.09 and see if that is a bit better?

[Oleg Lomaka (olomix)]

Thanks for help. I've installed latest nightly build of emacs with rev 744, 2.09 beta (as beta doesn't work with emacs 23.1).

Looks like indent problem is fixed. If I use reserved words as part of my custom tags names, the indentation work as expected. But highlighting still incorrect. I'm not sure it concerns exactly this issue. Just in case, I've attached example screenshot.

Also I've experienced another bug with using templates variables and mumamo in beta and submitted another but #692290.

[lborgman (lennart-borgman)]

Thanks, Oleg.

Could you please tell me what problem you have using nXhtml with Emacs 23.1? (I can test myself here, but it would be good to know what problems you have seen.)

I have tried to fix the highlighting. Please try the new beta 2.09.

[Oleg Lomaka (olomix)]

Looks like highlighting issue is resolved. Thanks.

As to latest stable emacs 23.2, mumamo uses a new function syntax-propertize-function, that was introduced in emacs cvs brunch and thus don't exists on 23.2 version. Actually this is another issue, I can file another ticket, if you want. I was thinking 2.09 beta doesn't suppose to work with non-cvs emacs.

If it would be helpful, here is the backtrace from emacs 23.2 opening a html file with 2.09 beta nxhtml:

MU:mumamo-fontify-region: Symbol's value as variable is void: syntax-propertize-function
** In buffer 1.html
  backtrace()
  (let ((standard-output standard-output)) (backtrace))
  (progn (let (...) (backtrace)) (with-current-buffer standard-output (buffer-string)))
  (unwind-protect (progn (let ... ...) (with-current-buffer standard-output ...)) (kill-buffer standard-output))
  (let ((standard-output ...)) (unwind-protect (progn ... ...) (kill-buffer standard-output)))
  (with-output-to-string (backtrace))
  (let ((format-string2 ...) (bt ...)) (mumamo-message-with-face (concat ... "
" ... bt) (quote highlight)) (apply (quote message) format-string2 lwarn-type args) (run-with-idle-timer 1 nil (quote mumamo-show-report-message)) (when mumamo-display-error-stop (setq font-lock-mode nil) (when ... ...) (when ... ...) (apply ... format-string2 lwarn-type args)))
  mumamo-display-error(mumamo-fontify-region "%s" "Symbol's value as variable is void: syntax-propertize-function")
  (condition-case err (mumamo-fontify-region-1 start end verbose) (error (mumamo-display-error ... "%s" ...)))
  (if (not mumamo-use-condition-case) (let* (... ...) (mumamo-fontify-region-1 start end verbose)) (condition-case err (mumamo-fontify-region-1 start end verbose) (error ...)))
  (mumamo-condition-case err (mumamo-fontify-region-1 start end verbose) (error (mumamo-display-error ... "%s" ...)))
  (if (and mumamo-just-changed-major nil (= 0 ...)) (mumamo-display-error (quote mumamo-fontify-region) "Just changed major, should not happen") (mumamo-condition-case err (mumamo-fontify-region-1 start end verbose) (error ...)))
  mumamo-fontify-region(1 105 nil)
  font-lock-fontify-region(1 105)
  run-hook-with-args(font-lock-fontify-region 1 105)
  byte-code("\302\303 #\207" [start next run-hook-with-args jit-lock-functions] 4)
  jit-lock-fontify-now(1 501)
  jit-lock-function(1)
  (let ((ret ...)) (mumamo-msgfntfy "mumamo-jit-lock-function EXIT %s, ff=%s, just-changed=%s" start (when start ...) mumamo-just-changed-major) ret)
  mumamo-jit-lock-function(1)

[lborgman (lennart-borgman)]

Thanks. The intention is to keep backward compatibility as long as possible if it it not too much trouble.

I have tried to fix this problem with backward compatibility now.

[Oleg Lomaka (olomix)]

I see you have committed the patch for versions less then 23.1. But I have the current stable 23.2. And it still doesn't work here, as syntax-propertize-function was introduced in cvs (23.3 or 24.0).