New HTML exporter: Oddities with figure placement

New HTML exporter: Oddities with figure placement

[Richard Stanton]

It seems that figure placement has changed in HTML files generated by the
new (compared with the old) exporter.

For example, here's a simple org file:

---------

This is an org file

[[file:figure.png]]

* Abstract
This is an abstract

---------

When I export, all looks fine, except that the figure is centered (left to
right) rather than being left-justified, which I could have sworn it used
to be. Since I prefer left justification by default, I tried adding the
line

#+ATTR_HTML: align="left"

immediately before the figure link. Now when I export to HTML, the figure
is left justified as I want, but the following text appears to the right
of the figure, not below it, which isn't what I want. It seems odd that
this would happen even when I have a blank line separating the figure link
from the following text. Am I missing something?

Thanks.

Richard Stanton

Re: New HTML exporter: Oddities with figure placement

[Stefan Vollmar]

[-- Attachment #1: Type: text/plain, Size: 1657 bytes --]

Dear Richard,

On 25.02.2013, at 18:55, Richard Stanton wrote:

> It seems that figure placement has changed in HTML files generated by the
> new (compared with the old) exporter.
> 
> For example, here's a simple org file:
> 
> ---------
> 
> This is an org file
> 
> [[file:figure.png]]
> 
> * Abstract
> This is an abstract
> 
> ---------
> 
> When I export, all looks fine, except that the figure is centered (left to
> right) rather than being left-justified, which I could have sworn it used
> to be. Since I prefer left justification by default, I tried adding the
> line
[...]

You are right, the effect has to do with the new exporter but is easy to explain and change:

if you look at the HTML code generated by Org-mode you will find this line in the embedded CSS code (line 45):

  div.figure p { text-align: center; }

which causes the effect you have described.
This code used to be suppressed by

:style-include-default nil
:style-include-scripts nil

in the org-publish-project-alist - either this functionality is currently not available or the keyword has changed. Thanks for any help with this!

However, here is a simple fix for now: include this line in a CSS file (it will be parsed after the embedded CSS code and overwrites it):

  div.figure p { text-align: left; }

Warm regards,
 Stefan
-- 
Dr. Stefan Vollmar, Dipl.-Phys.
Head of IT group
Max-Planck-Institut für neurologische Forschung
Gleuelerstr. 50, 50931 Köln, Germany
Tel.: +49-221-4726-213  FAX +49-221-4726-298
Tel.: +49-221-478-5713  Mobile: 0160-93874279
Email: vollmar@nf.mpg.de   http://www.nf.mpg.de







[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 4409 bytes --]

Re: New HTML exporter: Oddities with figure placement

[Nicolas Goaziou]

Hello,

Stefan Vollmar <vollmar@nf.mpg.de> writes:

> if you look at the HTML code generated by Org-mode you will find this line in the embedded CSS code (line 45):
>
>   div.figure p { text-align: center; }
>
> which causes the effect you have described.
> This code used to be suppressed by
>
> :style-include-default nil
> :style-include-scripts nil
>
> in the org-publish-project-alist - either this functionality is
> currently not available or the keyword has changed. Thanks for any
> help with this!

It should be :html-style-include-default and :html-style-include-scripts
according to html back-end definition in ox-html.el (lines 122, 123).


Regards,

-- 
Nicolas Goaziou

Re: New HTML exporter: Oddities with figure placement

[Stefan Vollmar]

[-- Attachment #1: Type: text/plain, Size: 1174 bytes --]

Dear Nicolas,

On 26.02.2013, at 00:14, Nicolas Goaziou wrote:

>> [...] This code used to be suppressed by
>> 
>> :style-include-default nil
>> :style-include-scripts nil
>> 
>> in the org-publish-project-alist - either this functionality is
>> currently not available or the keyword has changed. Thanks for any
>> help with this!
> 
> It should be :html-style-include-default and :html-style-include-scripts
> according to html back-end definition in ox-html.el (lines 122, 123).


sorry, I could have looked that up myself - and, again, thank you (somehow these lightning fast responses and fixes of yours remind me of a thread several years ago in which there was speculation about the actual number of excellent programmers using the pseudonym "Carsten Dominik", and of course there also quite a few "Bastien's" out there...).

Warm regards, 
 Stefan
-- 
Dr. Stefan Vollmar, Dipl.-Phys.
Head of IT group
Max-Planck-Institut für neurologische Forschung
Gleuelerstr. 50, 50931 Köln, Germany
Tel.: +49-221-4726-213  FAX +49-221-4726-298
Tel.: +49-221-478-5713  Mobile: 0160-93874279
Email: vollmar@nf.mpg.de   http://www.nf.mpg.de







[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 4409 bytes --]

HTML export: no date in postamble unless set manually

[Richard Stanton]

I have  org-export-html-postamble set to t. When I export a simple org
file that has no explicit #+date line (using the new exporter), I get a
line that says "Date:" at the bottom of the resulting HTML file, but
there's no date listed next to it. If I add an explicit #+date line, it is
correctly transferred to the bottom of the HTML file, but wouldn't it make
sense to have today's date listed if no explicit alternative is provided?
At any rate, adding a line that says "Date:" with nothing after the colon
doesn't seem like the optimal default setting.

Thanks.

Richard Stanton

Re: HTML export: no date in postamble unless set manually

[Nick Dokos]

Richard Stanton <stanton@haas.berkeley.edu> wrote:

> I have  org-export-html-postamble set to t. When I export a simple org
> file that has no explicit #+date line (using the new exporter), I get a
> line that says "Date:" at the bottom of the resulting HTML file, but
> there's no date listed next to it. If I add an explicit #+date line, it is
> correctly transferred to the bottom of the HTML file, but wouldn't it make
> sense to have today's date listed if no explicit alternative is provided?
> At any rate, adding a line that says "Date:" with nothing after the colon
> doesn't seem like the optimal default setting.
> 

The backend-specific variables have been renamed in the new exporter: try
org-html-postamble. Check the FAQ:

http://orgmode.org/worg/org-faq.html#new-exporter-switch

for more details.

Nick

Re: HTML export: no date in postamble unless set manually

[Nick Dokos]

Nick Dokos <nicholas.dokos@hp.com> wrote:

> Richard Stanton <stanton@haas.berkeley.edu> wrote:
> 
> > I have  org-export-html-postamble set to t. When I export a simple org
> > file that has no explicit #+date line (using the new exporter), I get a
> > line that says "Date:" at the bottom of the resulting HTML file, but
> > there's no date listed next to it. If I add an explicit #+date line, it is
> > correctly transferred to the bottom of the HTML file, but wouldn't it make
> > sense to have today's date listed if no explicit alternative is provided?
> > At any rate, adding a line that says "Date:" with nothing after the colon
> > doesn't seem like the optimal default setting.
> > 
> 
> The backend-specific variables have been renamed in the new exporter: try
> org-html-postamble. Check the FAQ:
> 
> http://orgmode.org/worg/org-faq.html#new-exporter-switch
> 
> for more details.
> 

OK, now that I've tried it a bit, I can see that the variable name is
not the reason: I can reproduce the empty date after setting
org-html-postamble to t.  The code looks like this
(ox-html.el:org-html--build-postamble):

,----
| 	       (let ((date (if (not (plist-get info :with-date)) ""
| 			     (org-export-data (plist-get info :date) info)))
`----

:with-date (i.e. org-export-with-date) is set to t but there is no
default for an absent :date.

This is probably another case of too many cooks (or variables in this
case: org-export-with-date, org-html-postamble-format, and the :date
entry in the plist have to cooperate in order to produce the desired
result), so the behavior is difficult to explain and it's not clear that
all the different combinations make sense - e.g. if org-export-with-date
is nil, but the format includes a %d, we probably should not have an
empty date field, whereas if org-export-with-date is t but there is no
date supplied, it should probably default to today.

Apologies for rushing to judgment.

Nick