Why changes encoding of Symbol font ?

[Nguyen Nam <nn...@yahoo.fr>]

 Helllo ,

Could somone tell why does Fop v.0.23 change the encoding of the Symbol Font
?
Many Symbol characters becomes so invalid with this version wherase they are
valid in the version 0.21.

Is there any workaround ?

Thanks very much for any information !!

Nam




---------------------------------
Yahoo! Mail -- Une adresse @yahoo.fr gratuite et en français !

Re: Memory usage w/ FOP

["J.Pietschmann" <j3...@yahoo.de>]

Philip S. Constantinou wrote:
> This really helps...
> One of the things that I need to do is a mail merge creating thousands of
> one page letters. If the pages are simple with no table of
> contents/references/page numbers. Is it safe to assume that the FO processor
> will use roughly the same memory printing 10 letters as 1000?

It would be very hard to optimize momery consuption
for one page documents. Apart from this, it depends
whether you do it in parallel or sequentially. In
case of the latter, FOP should always consume the
same amount of memory if the documents are similar
enough, regardless of complexity. If you do it in
parallel, be prepared that 1000 parallel runs use
100 times the memory of 10 parallel runs.

J.Pietschmann

Re: Memory usage w/ FOP

["Philip S. Constantinou" <ph...@wayfinder.net>]

This really helps...
One of the things that I need to do is a mail merge creating thousands of
one page letters. If the pages are simple with no table of
contents/references/page numbers. Is it safe to assume that the FO processor
will use roughly the same memory printing 10 letters as 1000?

----- Original Message -----
From: "J.Pietschmann" <j3...@yahoo.de>
To: <fo...@xml.apache.org>
Sent: Monday, June 24, 2002 1:44 PM
Subject: Re: Memory usage w/ FOP


> Philip S. Constantinou wrote:
> > Hi -
> > If I'm using an XSL Transformer and an XML ByteArrayOutputStream to
create a
> > FO formatted document which is rendered as PDF or printed, what is a
good
> > assumption to make about the memory utilization? Is the entire merged FO
> > document stored in memory or can the FO processor stream out the PDF or
> > printed page?
>
> I'm not sure where the "XML ByteArrayOutputStream"
> comes into play.
> As for FOP itself: the processor has limited capabilities
> for streaming, it also depends on the layout.
> FOP starts reading the FO into an internal tree. After
> the end of a page sequence, it starts rendering pages,
> which can use a huge amount of memory. If you use forward
> references to page numbers (like "page X of N"), every
> page from the reference on is held in memory until the
> element the reference points to is rendered. Elements
> spanning multiple pages, in particular tables, are also
> held in memory at least until the page it ends is completed.
> Large images are also known as memory hogs.
>
> Some rules of thumb (roughly in this order)
> - Don't use forward references. Put TOCs at the end.
> - Use multiple page sequences.
> - Avoid large images, scale them before feeding them to
>    FOP (unless print quality matters). For some formats,
>    in particular JPEG, the file size matters, not the
>    height*width, therefore choosing a higher compression
>    can help.
> - Avoid large tables spanning multiple pages, use multiple
>    tables instead.
>
> HTH
> J.Pietschmann
>
>

Re: Memory usage w/ FOP

["J.Pietschmann" <j3...@yahoo.de>]

Philip S. Constantinou wrote:
> Hi -
> If I'm using an XSL Transformer and an XML ByteArrayOutputStream to create a
> FO formatted document which is rendered as PDF or printed, what is a good
> assumption to make about the memory utilization? Is the entire merged FO
> document stored in memory or can the FO processor stream out the PDF or
> printed page?

I'm not sure where the "XML ByteArrayOutputStream"
comes into play.
As for FOP itself: the processor has limited capabilities
for streaming, it also depends on the layout.
FOP starts reading the FO into an internal tree. After
the end of a page sequence, it starts rendering pages,
which can use a huge amount of memory. If you use forward
references to page numbers (like "page X of N"), every
page from the reference on is held in memory until the
element the reference points to is rendered. Elements
spanning multiple pages, in particular tables, are also
held in memory at least until the page it ends is completed.
Large images are also known as memory hogs.

Some rules of thumb (roughly in this order)
- Don't use forward references. Put TOCs at the end.
- Use multiple page sequences.
- Avoid large images, scale them before feeding them to
   FOP (unless print quality matters). For some formats,
   in particular JPEG, the file size matters, not the
   height*width, therefore choosing a higher compression
   can help.
- Avoid large tables spanning multiple pages, use multiple
   tables instead.

HTH
J.Pietschmann

Memory usage w/ FOP

["Philip S. Constantinou" <ph...@wayfinder.net>]

Hi -
If I'm using an XSL Transformer and an XML ByteArrayOutputStream to create a
FO formatted document which is rendered as PDF or printed, what is a good
assumption to make about the memory utilization? Is the entire merged FO
document stored in memory or can the FO processor stream out the PDF or
printed page?

Are there any good rules of thumb?

Thanks -
Phil

Re: Why changes encoding of Symbol font ?

["J.Pietschmann" <j3...@yahoo.de>]

Nguyen Nam wrote:
>  Helllo ,
> 
> Could somone tell why does Fop v.0.23 change the encoding of the Symbol Font
> ?

Where did you get the 0.23 version?

Seriously: Because it was not conforming to the standard.

> Many Symbol characters becomes so invalid with this version wherase they are
> valid in the version 0.21.
> 
> Is there any workaround ?
> 
FOP used native encoding. Now it uses Unicode encoding, as
it should. Look up the Unicode codepoints to use for example
at the Unicode consortium
  http://www.unicode.org/charts/charindex.html
or run docs/examples/fo/fonts.fo through FOP, choose the glyph
and look up the corresponding character reference in the FO
source.

J.Pietschmann