🚀 go-pugleaf

In article <j5t5iv$lb7$1@reader1.panix.com>, rodd@panix.com (Rod Dorman) wrote: 

> In article <j5scc5$pak$1@dont-email.me>,
> NoHtmlMailsPlease <UsePlainText@dog.edu> wrote:
> >"Rod Dorman" <rodd@panix.com> wrote in message 
> >news:j5ildf$cgu$1@reader1.panix.com...
> >>  ...
> >> There isn't any concept of "next line" in PostScript.
> >
> >My brief look at the docus suggests that <newLine> could be programmed as:
> >/Nline <decr y-coordinate per font size>
> >  <move abs x-coordinate to line-start> def
> 
> Define what you mean by "per font size". Is that something your
> PostScript emitting program is tracking or are you expecting the
> PostScript RIP to somehow figure it out for you?
> 
This thread has become a good example of what happens when
the OP is truncated for economy.
The aim is simply convert plain-text to pdf, for adobe's text-to-speech;
but to control the position of the <page breaks>.
Therefore the <font> is irrelevant.
So AFAICS the algorith is SIMPLY:---
SetConvenientFont
SetConvenientXYstartOfPage
<WriteLineOfAcii>
<NextLine>perConvenientFont
...

> Consider a "line" that starts off with a 16 point high ornamental
> character EPS followed by a mix of 12 point Helvetica and 10 point
> Palatino-Bold characters.
> 
> You'll need to "decr y-coordinate" by at least 16 points (probably a
> little more to avoid decenders) before emitting the PostScript code
> for that "line".
> 
> In PostScript you first position the baseline before drawing the text
> which means you have to figure out how tall that line is going to be
> before you can do the move.
> -- 
With the Postscript-priests pretending that their craft is magic,
I was bamboozled into over looking the obvious solution of using
an editor which gives pdf output, and just manually syncronising
page-breaks with para-breaks.

Then I hit the interesting problem of 
Linux: ascii to pdf
could be rendered on Win7 abode, but not 'spoken'.
And linux: pdf to ascii 
confirmed that pfd2txt(txt2pdf(originalAscii))
[or an alternative notation:
 txt2 -> pdf -> txt2]
showed:
"........N"
instead of the original line text.
And the clever adobe-lady spoke:
"dot dot dot dot EN" .

So now the problem has evolve to
'what format does <the linux editor> translate the
original text to pdf, so that it is viewable OK but NOT 
translatable back to the original ascii?

Which asks "what pdf formats are there that would
give a ascii to pdf, which didn't translate back to the 
original ascii" ?

And even better: has anyone done direct 
<text to speech> on Win7?

== TIA