Re: rel-0-8-1; Default options are always used, even when package is loaded

David Kastrup <[email protected]> Mon, 21 Feb 2005 20:41:52 +0100
Newsgroups gmane.emacs.latex.preview.devel
Message-ID <x5sm3p8zun.fsf__580.83333210579$1109014894$gmane$org@lola.goethe.zz>
[email protected] (Frank K=FCster) writes:

> David Kastrup <[email protected]> wrote:
>
>> [email protected] (Frank K=FCster) writes:
>>
>>> Indeed the problem seems to be something different; it only showed
>>> up at a sectioning command. This is a minimal test file:
>>>
> [...]
>> There is no sectioning command here.
>
> Cool, right? Frank says that it doesn't need a sectioning command,
> only showed up by chance at a sectioning command, and in the minimal
> file there isn't one.

Well, _what_ problem?  Your original complaint was that preview-latex
was previewing sectioning commands although it shouldn't.  And you
illustrate this with an example without sectioning commands, saying
that the problem will be the same nevertheless.  How can the problem
be that sectioning commands get previewed?

>>> With this file, preview-latex runs _pdf_latex, and that seems to
>>> be the cause of the error.
>>
>> Of what error?
>
> The error in the output file I  attached.

I did not see any.

>> Why shouldn't preview-latex run pdflatex when you are working with
>> a PDFTeX mode document?
>
> I don't know.  There's not much about pdf in the manual, only some -Ppdf
> switches to dvips, and a remark in the Todo node.

Then I recommend that you replace the old manual that does not seem to
match your version of preview-latex (which is 0.8.1, according to the
log files you sent).

> But it seemed to me that it didn't work with pdf mode.  You say it
> should work in AUCTEX' pdf mode, and it should use pdftex, well
> fine. It does use it here, but it doesn't display a preview.

Then we should try to find out why it doesn't.

>> [Snipped the output, all of which looks perfectly fine for a PDFTeX
>>  run of preview-latex]
>
> There is an error in the output file you snipped. The help displayed is:
>
> ,----
> | ERROR: Preview: Snippet 1 started.
> |=20
> | --- TeX said ---
> | <-><->
> |=20=20=20=20=20=20=20
> | l.12 $
> |       a$
> | --- HELP ---
> | The auctex option to preview should not be applied manually.  If you
> | see this error message, either you did something too clever, or the
> | preview Emacs Lisp package something too stupid.
> `----

I'd say you did something too clever, namely asking for error help on
a preview run.

> and the relevant part of the output file is:
>
> ,----
> | No file _region_.aux.
> | Preview: Fontsize 10pt
> | Preview: PDFoutput 1
> |  !name(mini.tex) !offset(-3)=20
> | ! Preview: Snippet 1 started.
> | <-><->
> |=20=20=20=20=20=20=20
> | l.12 $
> |       a$
> | Preview: Tightpage -32891 -32891 32891 32891
> | ! Preview: Snippet 1 ended.(282168+0x346416).
> | <-><->
> |=20=20=20=20=20=20=20
> | l.12 $a$
> `----

Well, the FAQ has the following to offer:

    Preview causes all sort of strange error messages
    -------------------------------------------------

       When running preview-latex and taking a look at either log file or
    terminal output, lots of messages like

         ! Preview: Snippet 3 started.
         <-><->

         l.52 \item Sie lassen sich als Funktion $
                                                  y =3D f(x)$ darstellen.
         ! Preview: Snippet 3 ended.(491520+163840x2494310).
         <-><->

         l.52 \item Sie lassen sich als Funktion $y =3D f(x)$
                                                            darstellen.

       appear (previous versions generated messages looking even more like
    errors).  Those are not real errors (as will be noted in the log file).
    Or rather, while they *are* really TeX error messages, they are
    intentional.  This currently is the only reliable way to pass the
    information from the LaTeX run of preview-latex to its Emacs part about
    where the previews originated in the source text.  Since they are
    actual errors, you will also get AUCTeX to state
         Preview-LaTeX exited abnormally with code 1 at Wed Sep  4 17:03:30
       after the LaTeX run in the run buffer.  This merely indicates that
    errors were present, and errors will always be present when
    preview-latex is operating.  There might be also real errors, so in
    case of doubt, look for them explicitly in either run buffer or the
    resulting `.log' file.


> The result is that the roadwork sign continues to be displayed beside
> the math code, and there is no preview.

Ok, so here we are getting somewhat closer to the problem.  The last
thing you posted from the run buffer was


Preview-LaTeX exited abnormally with code 1 at Mon Feb 21 18:27:35
Running `Preview-PDF2DSC' with ``pdf2dsc _region_.pdf _region_.prv/tmp4728H=
oh/preview.dsc''

Preview-PDF2DSC finished at Mon Feb 21 18:27:35

****************************************

After we arrive there, GhostScript should be run.  I can see no
indication of that happening.  Which is strange as the PDF2DSC run
actually looks succesful, and the configuration data your original bug
report contained looked fine, with GhostScript available under the
configured executable name.

> You can look at it at
>
> http://www.kuesterei.ch/preview.gif

Sigh.  That is no roadworks sign.  That is an error indicator (Danger
ahead).  Ok, the mnemonic value of this icons suck.  Since it is
remarkably difficult to get a well-looking symmetric exclamation mark
and we are currently redesigning icons, anyway, what icon do you
propose instead for indicating an error?  A car falling off into
water?  A "STOP" sign?  Click it with middle mouse button.  The
resulting report should be interesting.

> Is the message in the output the usual thing with pdftex? Why does
> C-c ` go to that error message?

Because it is, after all, an error message.  Futzing around with
AUCTeX to steer it away from those messages is not trivial.  There was
some code to do this in preview-latex at one time, but it had some bad
interactions with other code, and preview runs don't tell you to use
C-c ` for errors, anyway (which will _always_ be present, see above).

--=20
David Kastrup, Kriemhildstr. 15, 44793 Bochum


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
preview-latex-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/preview-latex-bugs


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click