Re: fot2pdf release 1.02

Subject: Re: fot2pdf release 1.02
From: "Christof Drescher" <drescher@xxxxxxxxxxxx>
Date: Thu, 2 Nov 2000 10:31:20 +0100
Hi Thomas, hi DSSSL-users,

Thank you for your reply - and don't mind posting it to the DSSSL-list; bugs
and wishes should be no secret...

I try to comment on all of your remarks in detail:

>I have one more thing that I would like to
>see your Java backend generate, and that is the "Bookmarks" in the
>left column in Acroreader.  We get bookmarks in our jadetex solution.

So would I, but unfortunately, this is not possible at the moment. This
feature would depend on a change in the Jade code which would have to export
the "user-defined" attribute "heading-level" to the fot-backend, which it
does not at the moment. So, other for people generating the input directely
or changing it afterwards, an implementation on my side would have no
effect, which is sad but true.
I brought this to the attention of the devel-list people and even got a
reply at first, but actually no one seemed to be able or interested in that
code change, and I cannot do it myself. If anyone reads this and thinks he
(or she) could do it, I would be VERY HAPPY - you are not the only one who
requested it, Thomas.

>I get:
>
>WARNING! Unsupported FlowObject 'text' skipped!
>WARNING! Unsupported FlowObject 'text' skipped!
>WARNING! Unsupported FlowObject 'text' skipped!
>
>I think you have commented on this before, but I don't quite remember.
>What is the problem?

The problem is that some text is not properly nested into a sequence of

<simple-page-sequence>
  <paragraph>
    <text>blabla</text>
  </paragraph>
</simple-page-sequence>

but is somehow "on its own", which should not be. Other backends simply drop
them (which is what fot2pdf does as well, but issues a warning) or display
them with some "default" method - but it is not clear if this is ok, at
least to me. Anyway, it should be seen as a "hint" to have the stylesheet
handle such things more properly, right?

>ERROR! Could not read &lt;OSFILE
SOIBASE='../graphics_pdf.ent'&gt;psm_arch.pdf
>ERROR! Could not read &lt;OSFILE
SOIBASE='../graphics_pdf.ent'&gt;modstruc.pdf
>
><external-graphic entity-system-id="&lt;OSFILE
SOIBASE='../graphics_pdf.ent'&gt;psm_arch.pdf" notation-system-id="PDF"
scale-x="0.75" scale-y="0.75" display="true" display-alignment="center"/>

>We use entities to refer to graphics files (to resolve the format
>dependencies to the different backends).  It would be nice if your
>solution could handle this.

Hm, I might be good at programming a PDF formatter, but I'm no wizard with
entities or SGML in general. So this is the first time I see a construct
like this, sorry... :-)
If you (or anyone else) could tell me what this magic chant means, and how I
can get a "normal" filename from it, it should be no problem implementing
it.

>But the most important thing that does not workis that <screen>
>elements that is line specific (that should adhere to the newlines in
>the source!) is reformatted!  This is a sample fot output:
>
><paragraph space-before="11pt" space-after="11pt" font-size="7.201pt"
line-spacing="9.362pt" start-indent="48pt" font-family-name="Courier New"
font-weight="medium" font->posture="upright" first-line-start-indent="0pt"
lines="asis" input-whitespace-treatment="preserve">
><a name="338"/>
><text>
></text>
><text>                  PSM Resources Installation
></text>
><text>                  ==========================
></text>
><text>
></text>
><text>This script will configure and install PSM Resources Software.
></text>
><text>
>
>The output does not adhere to the newlines (the text elemts).

OK, the "lines"-characteristic is not implemented as the moment, but I see
it is necessary for something like it. I just wonder if this is really the
proper way of handling it - if you want a new line, why not have a
"paragraph-break" or a completely new "paragraph" for it? Is this the way it
is defined in DSSSL? As I conclude from the standard, I thought a value of
"asis" should only prevent the formatter from "wrapping" (e.g. simply
truncate a line if it does not fit), but I wonder if it is correct to assume
a new <text> elemment means a newline should be inserted?!

I deliberately posted this to the list as well, since the original message
went there too and there are some questions from my side which could be
answered by others as well. Especially, I would welcome a big "ME!" on the
question of adding the support for the "heading-level" in Jade as mentioned
above... :-)))

Greetings,
Christof
--------------------------------------------
Pro Image GbR, Marburg
Tel. 06421-66800, Fax 06421-66820
email: drescher@xxxxxxxxxxxx


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist


Current Thread