Re: [bug] void-variable org-special-blocks-line when exporting to HTML

["Sebastien Vauban" <wxhgmqzgwmuf-geNee64TY+gS+FvcfC7Uqw@public.gmane.org>]

Hi Bastien,

Bastien wrote:
> "Sebastien Vauban" <wxhgmqzgwmuf-geNee64TY+gS+FvcfC7Uqw@public.gmane.org> writes:
>
>> I'm not out of the blue yet... I still can't understand why:
>>
>> - you were unable to reproduce it?
>>   What's the difference between you and me? ;-)
>>   I mean why does Org behave differently between us?
>>
>> - nobody else seemed hit by this?
>>   I'm surprised by the absence of reactions while HTML export was completely
>>   failing for a couple of days. Once again, why only me?
>>
>> - this wasn't trapped by the ERT test suite?
>>   There are a lot of HTML exports done in the tests. Why did they succeed?
>>
>> Have you hints on this?
>
> I guess that's because not everybody uses org-special-blocks.el as a
> module.

By "using org-special-blocks.el as a module", if you mean "having a line such
as":

    (add-to-list 'org-modules 'org-special-blocks)

I did not have one either.

I just had:

    (require 'org-special-blocks)

But, IIUC, that's more or less equivalent:

- being in `org-modules' means that the package will be required[1] many times
  before trying to do different operations (such as inserting a link, etc.),

  and that there is a protection mechanism in case the package can't be loaded

- being explicitly required in my .emacs does only try to load it once.

Apart from that, I don't see differences.

> The ERT suite doesn't make any test against this.

Well against exporting HTML. But I did not realize that I was more or less the
only one in the world ;-) to load `org-special-blocks'[2] while exporting to HTML.

This must explain that. Case is closed.

Best regards,
  Seb

Footnotes:

[1] Though, really only loaded once.

[2] And I must admit I maybe have used it once or twice, not more... Though, I
still find it an interesting feature.

-- 
Sebastien Vauban