On 10/28/2015 9:34 PM, Orchid Win7 v1 wrote:
> On 28/10/2015 06:51 PM, Stephen wrote:

>
> Interesting. Now how about the Cray 2, the VAX, UNIAC and all the other
> legendary systems?
>

What did your last slave die of?

> To be clear, I'm sure all the data is out there. It's just rather a lot
> of effort to collate it all into a short summary.
>

Dinkums! Too important to do a bit of work, are we? :P


>>> PostScript. And it wasn't *once*. ;-)
>
> Did I tell you about the time I wrote a text to PostScript converter as
> an MS-DOS batch script?
>

No but I am sure you will. ;)

> Actually, the script just concatenates the text file with some
> hand-written PostScript code; the text file itself becomes a giant text
> string in a PostScript program. The program itself is a semi-complete
> formatting engine, which adds a filename, page number and date of
> printing to each page, and scans the characters looking for line breaks.
> It even line-wraps the text and adds little arrows when it does so. (I
> think there *might* be a bug if it line-wraps on the last line on a page
> though...)
>
> Damn, my talents are completely unappreciated! :-(
>

Not completely, I'm sure.
Your mother loves you and your boss hasn't sacked you. :)

>>> Actually, I did start trying to write a PostScript *interpreter*, to
>>> make debugging this stuff easier! It turns out the language has a lot
>>> more edge-cases than you'd think though...
>>
>> One edge-case is more than I ken.
>> I would ask what an edge-case is but then you would say that I can't
>> Google. :-)
>
> It's simple. An edge-case is where a function is supposed to do XYZ,
> *except* in this one really obscure case where if A, B and C happen
> simultaneously, it does something totally different.


Are you sure about that?
Do all the cases (A,B & C) have to happen simultaneously?

My reading is when one of them is on the edge of the design parameters.

Like when you
> configure how many pages you want to print, but if you put in -1 it
> prints *all* the pages. Not -1 pages.
>
> PostScript does some slightly strange things with token parsing. You can
> take a text string and ask the PostScript interpreter to parse out the
> first token, giving you back the token and the remainder of the string.
> But the specification gives some really inconsistent rules for whether
> it should or shouldn't skip spaces...

Hit it with a shifting spanner then.
I can't stand uppity software. ;)

-- 

Regards
     Stephen

Post a reply to this message