pre-funny being, well, funny...

pre-funny being, well, funny...

[flip phillips]

Now that the mpost mystery has been solved, I am getting a wee crash 
when trying to use
pre-funny:

\usemodule[pre-funny]
\starttext
\TitlePage
   {A Few Nice Quotes\\
    A Simple Style Demo\\
    Hans Hagen, August 2000}

\Topic {Douglas R. Hofstadter} \input douglas \page
\stoptext

gives a rousing:

layout : calc bgrounds
Randomizer initialized... [MP to PDF] (./mpgraph.8000
!Arithmetic overflow.

(sorry for any typos, there seems to be a bug in the TeXShop console 
window that is keeping me from selecting the error for copy / paste... )

--
flip phillips
http://www.skidmore.edu/~flip/

Re: pre-funny being, well, funny...

[Johannes Hüsing]

On Tue, Apr 09, 2002 at 12:51:41PM -0400, flip phillips wrote:
[...]
> 
> layout : calc bgrounds
> Randomizer initialized... [MP to PDF] (./mpgraph.8000
> !Arithmetic overflow.
> 

The problem seems to be in Metapost not accepting high integers.
Context with embedded Metapost code generates distinct mp files
whose file names all start with mpgraph. 

For some reason the new version of context starts with a high 
integer as suffix and counts down.

Another thing I get in the second run of pdftex: 

! Use of \@@liTopicinteractie doesn't match its definition.
\doifvalue ...f \!!stringa {\csname #1\endcsname }
                                                  \edef \!!stringb {#2}\ifx ...

\dosomelijstelement ...1\c!interactie }{\v!tekst }
                                                   {\setlistparameter {#1}\c...

\dodolijstelement ...ment {#1}{#2}{#3}{#4}{#5}{#6}
                                                  \global \utilitydonetrue 
l.12 ...linischer Studien}{2::0:2:0:0:0:0:0::6}{7}

What is happening here?

Greetings

Johannes
-- 
Johannes Hüsing   There is something fascinating about science. One gets
hannes@ruhrau.de  such wholesale returns of conjecture from such a 
                  trifling investment of fact.                Mark Twain

Re: pre-funny being, well, funny...

[Hans Hagen]

At 09:18 PM 4/9/2002 +0200, Johannes Hüsing wrote:
>On Tue, Apr 09, 2002 at 12:51:41PM -0400, flip phillips wrote:
>[...]
> >
> > layout : calc bgrounds
> > Randomizer initialized... [MP to PDF] (./mpgraph.8000
> > !Arithmetic overflow.
> >
>
>The problem seems to be in Metapost not accepting high integers.
>Context with embedded Metapost code generates distinct mp files
>whose file names all start with mpgraph.

strange, because my version accepts this; is this a linux problem?

anyway, in cont-sys.tex you can say:

\maxnofMPgraphics = 4000

if this problem is persistent, i can default to 4000 (and set it to 8000 here)

The count up-down has a reason: graphics generated during typesetting the 
pagebody don't influence the text flow; by using a different range (count 
down) a change mid document of a graphic that is part of the background 
(say, that you add one), will now have no
influence on the page breaks (due to a possible change in order, and 
thereby swapping of graphics); this can save many runs as well as 
oscillating [i hope you understand the problem, it's hard to explain]

imagine (--automp):

page one : text graphic + background graphic
page two : text graphic + background graphic

first  run: no graphics
second run: text graphic 1, page grows, graphic 2 moved to page 3
third run: extra background graphic, old background graphic takes place of 
text graphic

etc etc; pretty tricky but it works quite well now (you cannot imagine on 
what weird things i spent my time)

>For some reason the new version of context starts with a high
>integer as suffix and counts down.
>
>Another thing I get in the second run of pdftex:
>
>! Use of \@@liTopicinteractie doesn't match its definition.
>\doifvalue ...f \!!stringa {\csname #1\endcsname }
>                                                   \edef \!!stringb 
> {#2}\ifx ...
>
>\dosomelijstelement ...1\c!interactie }{\v!tekst }
>                                                    {\setlistparameter 
> {#1}\c...
>
>\dodolijstelement ...ment {#1}{#2}{#3}{#4}{#5}{#6}
>                                                   \global \utilitydonetrue
>l.12 ...linischer Studien}{2::0:2:0:0:0:0:0::6}{7}
>
>What is happening here?

I don't know, can you make a minimal example showing the problem? Maybe 
solved already. (time to post a new beta)

hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-ade.nl
-------------------------------------------------------------------------

Re: pre-funny being, well, funny...

[Daniel Pittman]

On Tue, 09 Apr 2002, Hans Hagen wrote:
> At 09:18 PM 4/9/2002 +0200, Johannes Hüsing wrote:
>>On Tue, Apr 09, 2002 at 12:51:41PM -0400, flip phillips wrote:
>>[...]
>> >
>> > layout : calc bgrounds
>> > Randomizer initialized... [MP to PDF] (./mpgraph.8000
>> > !Arithmetic overflow.
>> >
>>
>>The problem seems to be in Metapost not accepting high integers.
>>Context with embedded Metapost code generates distinct mp files
>>whose file names all start with mpgraph.
> 
> strange, because my version accepts this; is this a linux problem?

It works fine for me on:

Linux inanna 2.5.6 #3 Mon Mar 11 23:34:33 EST 2002 i686 unknown

with:

This is pdfeTeX, Version 3.14159-1.00a-pretest-20011114-ojmw-2.1 (Web2C 7.3.7)
ConTeXt  ver: 2002.3.27  fmt: 2002.4.10  int: english  mes: english

        Daniel

-- 
Solum certum nihil esse certi.
the only certainty is that nothing is certain
        -- Pliny the Elder, _Historia Naturalis_

Re[2]: pre-funny being, well, funny...

[Giuseppe Bilotta]

Wednesday, April 10, 2002 Daniel Pittman wrote:

DP> It works fine for me on:

DP> Linux inanna 2.5.6 #3 Mon Mar 11 23:34:33 EST 2002 i686 unknown

DP> with:

DP> This is pdfeTeX, Version 3.14159-1.00a-pretest-20011114-ojmw-2.1 (Web2C 7.3.7)
DP> ConTeXt  ver: 2002.3.27  fmt: 2002.4.10  int: english  mes: english

What about MetaPost and the MetaFun format?

Re: pre-funny being, well, funny...

[Daniel Pittman]

On Wed, 10 Apr 2002, Giuseppe Bilotta wrote:
> Wednesday, April 10, 2002 Daniel Pittman wrote:
> 
> DP> It works fine for me on:
> 
> DP> Linux inanna 2.5.6 #3 Mon Mar 11 23:34:33 EST 2002 i686 unknown
> 
> DP> with:
> 
> DP> This is pdfeTeX, Version 3.14159-1.00a-pretest-20011114-ojmw-2.1
> DP> (Web2C 7.3.7) ConTeXt ver: 2002.3.27 fmt: 2002.4.10 int: english
> DP> mes: english
> 
> What about MetaPost and the MetaFun format?

I couldn't find anything but: This is MetaPost, Version 0.641 (Web2C 7.3.7)
This is the beta downloaded on the 28th of March.

...and whadda ya know. It broke for me with an integer overflow, just as
the OP reported. Error follows, but it's the same as he mentioned in the
thread.

This *definitely* worked for me earlier tonight, though, with exactly
the same setup and all. That's ... odd.

Ah. Found it. It only crashes that way if I am running off PDF output
with a '% output=pdftex' in the file -- running off to DVI works fine.

So, there you go. That makes a substantial difference. Sorry if I
confused the issue at all. It seems that running the output with PDF
output hits:

systems        : begin file test at line 3
layout         : calculating backgrounds
Randomizer initialized to 1638256672. [MP to PDF] (./mpgraph.8000
! Arithmetic overflow.
<to be read again> 
                   \relax 
\doMPconcat ...expanddimen 18 \doMPexpanddimen 18 
                                                  \divide \dimen 12 \dimen 1...

\flushconcatMPsegment ... {\gMPs 1}\a {\gMPs 2}\b 
                                                  \doMPconcat {\gMPs 3}\c {\...

\body ->\flushconcatMPsegment 
                              \advance \nofMPsegments 1 \ifnum \nofMPsegment...

\iterate ->\body 
                 \expandafter \iterate \else \fi 
\handleMPsequence ...\fi \fi \fi \fi \flushMPpath 
                                                  \closeMPpath \finishMPpath...
l.37 ...0.7071 -0.35356 0.7071 0 0] concat stroke 
                                                  grestore
?

        Daniel

-- 
We have nothing to lose because we don't have anything.
        ... Anything we want anyway ...
        -- KMFDF, _Dogma_

Re[2]: pre-funny being, well, funny...

[Giuseppe Bilotta]

Wednesday, April 10, 2002 Daniel Pittman wrote:

DP> So, there you go. That makes a substantial difference. Sorry if I
DP> confused the issue at all. It seems that running the output with PDF
DP> output hits:

DP> systems        : begin file test at line 3
DP> layout         : calculating backgrounds
DP> Randomizer initialized to 1638256672. [MP to PDF] (./mpgraph.8000
DP> ! Arithmetic overflow.
DP> <to be read again> 
DP>                    \relax 
DP> \doMPconcat ...expanddimen 18 \doMPexpanddimen 18 
DP>                                                   \divide \dimen 12 \dimen 1...

DP> \flushconcatMPsegment ... {\gMPs 1}\a {\gMPs 2}\b 
DP>                                                   \doMPconcat {\gMPs 3}\c {\...

DP> \body ->\flushconcatMPsegment 
DP>                               \advance \nofMPsegments 1 \ifnum \nofMPsegment...

DP> \iterate ->\body 
DP>                  \expandafter \iterate \else \fi 
DP> \handleMPsequence ...\fi \fi \fi \fi \flushMPpath 
DP>                                                   \closeMPpath \finishMPpath...
DP> l.37 ...0.7071 -0.35356 0.7071 0 0] concat stroke 
DP>                                                   grestore
DP> ?

Smells like an MPtoPDF bug to me!

--
Giuseppe "Oblomov" Bilotta

Re[2]: pre-funny being, well, funny...

[Hans Hagen]

At 08:04 PM 4/10/2002 +0200, Giuseppe Bilotta wrote:

>Wednesday, April 10, 2002 Daniel Pittman wrote:
>
>DP> So, there you go. That makes a substantial difference. Sorry if I
>DP> confused the issue at all. It seems that running the output with PDF
>DP> output hits:
>
>DP> systems        : begin file test at line 3
>DP> layout         : calculating backgrounds
>DP> Randomizer initialized to 1638256672. [MP to PDF] (./mpgraph.8000
>DP> ! Arithmetic overflow.
>DP> <to be read again>
>DP>                    \relax
>DP> \doMPconcat ...expanddimen 18 \doMPexpanddimen 18
>DP>                                                   \divide \dimen 12 
>\dimen 1...
>
>DP> \flushconcatMPsegment ... {\gMPs 1}\a {\gMPs 2}\b
>DP>                                                   \doMPconcat {\gMPs 
>3}\c {\...
>
>DP> \body ->\flushconcatMPsegment
>DP>                               \advance \nofMPsegments 1 \ifnum 
>\nofMPsegment...
>
>DP> \iterate ->\body
>DP>                  \expandafter \iterate \else \fi
>DP> \handleMPsequence ...\fi \fi \fi \fi \flushMPpath
>DP>                                                   \closeMPpath 
>\finishMPpath...
>DP> l.37 ...0.7071 -0.35356 0.7071 0 0] concat stroke
>DP>                                                   grestore
>DP> ?
>
>Smells like an MPtoPDF bug to me!

Is that the latest version? The latest release follows a different method 
for high values (keep in mind that tex has no real calc); i need to 
distinguish, otherwise we loose to much precission (spent daaayyyys omn this)

Hans

-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-ade.nl
-------------------------------------------------------------------------

Re: pre-funny being, well, funny...

[Daniel Pittman]

On Thu, 11 Apr 2002, Hans Hagen wrote:
> At 08:04 PM 4/10/2002 +0200, Giuseppe Bilotta wrote:
>>Wednesday, April 10, 2002 Daniel Pittman wrote:
>>
>>DP> So, there you go. That makes a substantial difference. Sorry if I
>>DP> confused the issue at all. It seems that running the output with
>>DP> PDF output hits:

[...]

>>Smells like an MPtoPDF bug to me!
> 
> Is that the latest version? 

As far as I can tell, yes. No more recent release than the 27th of March?

> The latest release follows a different method for high values (keep in
> mind that tex has no real calc); i need to distinguish, otherwise we
> loose to much precission (spent daaayyyys omn this)

Ack. Poor chap. :)
        Daniel

-- 
Across clinical benches with nothing to talk
Breathing tea and biscuits and the Serenity Prayer
While the bones of our child crumble like chalk
O where do we go now but nowhere
        -- Nick Cave, _Where Do We Go Now But Nowhere?_

Re: Re[2]: pre-funny being, well, funny...

[flip phillips]

So, I'm still having this sort of problem, even with the latest version 
of teTeX under MacOS X (with the countdown starting at 4000 now).
The MPtoPDF in my release is 1.2 apparently.

Any progress? Ideas?

On Thursday, April 11, 2002, at 04:27  AM, Hans Hagen wrote:

> At 08:04 PM 4/10/2002 +0200, Giuseppe Bilotta wrote:
>
>> Wednesday, April 10, 2002 Daniel Pittman wrote:
>>
>> DP> So, there you go. That makes a substantial difference. Sorry if I
>> DP> confused the issue at all. It seems that running the output with 
>> PDF
>> DP> output hits:
>>
>> DP> systems        : begin file test at line 3
>> DP> layout         : calculating backgrounds
>> DP> Randomizer initialized to 1638256672. [MP to PDF] (./mpgraph.8000
>> DP> ! Arithmetic overflow.
>> DP> <to be read again>
>> DP>                    \relax
>> DP> \doMPconcat ...expanddimen 18 \doMPexpanddimen 18
>> DP>                                                   \divide \dimen 
>> 12 \dimen 1...
>>
>> DP> \flushconcatMPsegment ... {\gMPs 1}\a {\gMPs 2}\b
>> DP>                                                   \doMPconcat 
>> {\gMPs 3}\c {\...
>>
>> DP> \body ->\flushconcatMPsegment
>> DP>                               \advance \nofMPsegments 1 \ifnum 
>> \nofMPsegment...
>>
>> DP> \iterate ->\body
>> DP>                  \expandafter \iterate \else \fi
>> DP> \handleMPsequence ...\fi \fi \fi \fi \flushMPpath
>> DP>                                                   \closeMPpath 
>> \finishMPpath...
>> DP> l.37 ...0.7071 -0.35356 0.7071 0 0] concat stroke
>> DP>                                                   grestore
>> DP> ?
>>
>> Smells like an MPtoPDF bug to me!
>
> Is that the latest version? The latest release follows a different 
> method for high values (keep in mind that tex has no real calc); i need 
> to distinguish, otherwise we loose to much precission (spent daaayyyys 
> omn this)
>
> Hans
>
>
> -------------------------------------------------------------------------
>                                   Hans Hagen | PRAGMA ADE | 
> pragma@wxs.nl
>                       Ridderstraat 27 | 8061 GH Hasselt | The 
> Netherlands
>  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-
> ade.com
> -------------------------------------------------------------------------
>                                   fall-back web server: www.pragma-
> ade.nl
> -------------------------------------------------------------------------
>
>
--
flip phillips
http://www.skidmore.edu/~flip/

Re: Re[2]: pre-funny being, well, funny...

[Johannes Hüsing]

On Mon, Apr 15, 2002 at 10:14:41PM -0400, flip phillips wrote:
> So, I'm still having this sort of problem, even with the latest version 
> of teTeX under MacOS X (with the countdown starting at 4000 now).
                                                         ~~~~
> The MPtoPDF in my release is 1.2 apparently.
> 
> Any progress? Ideas?
> 
> On Thursday, April 11, 2002, at 04:27  AM, Hans Hagen wrote:
> 
> >At 08:04 PM 4/10/2002 +0200, Giuseppe Bilotta wrote:
> >
> >>Wednesday, April 10, 2002 Daniel Pittman wrote:
> >>
> >>DP> So, there you go. That makes a substantial difference. Sorry if I
> >>DP> confused the issue at all. It seems that running the output with 
> >>PDF
> >>DP> output hits:
> >>
> >>DP> systems        : begin file test at line 3
> >>DP> layout         : calculating backgrounds
> >>DP> Randomizer initialized to 1638256672. [MP to PDF] (./mpgraph.8000
                                                                     ~~~~

Hrm.

Have you deleted all mpgraph* stuff before retrying?

Greetings

Johannes
-- 
Johannes Hüsing   There is something fascinating about science. One gets
hannes@ruhrau.de  such wholesale returns of conjecture from such a 
                  trifling investment of fact.                Mark Twain

Re: pre-funny being, well, funny...

[Hans Hagen]

At 12:51 PM 4/9/2002 -0400, flip phillips wrote:
>Now that the mpost mystery has been solved, I am getting a wee crash when 
>trying to use
>pre-funny:
>
>\usemodule[pre-funny]
>\starttext
>\TitlePage
>   {A Few Nice Quotes\\
>    A Simple Style Demo\\
>    Hans Hagen, August 2000}
>
>\Topic {Douglas R. Hofstadter} \input douglas \page
>\stoptext
>
>
>gives a rousing:
>
>layout : calc bgrounds
>Randomizer initialized... [MP to PDF] (./mpgraph.8000
>!Arithmetic overflow.

Can you try:

\maxnofMPgraphics = 4000
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-ade.nl
-------------------------------------------------------------------------

Re: pre-funny being, well, funny...

[flip phillips]

On Tuesday, April 9, 2002, at 05:45  PM, Hans Hagen wrote:

> At 12:51 PM 4/9/2002 -0400, flip phillips wrote:
>> Now that the mpost mystery has been solved, I am getting a wee crash 
>> when trying to use
>> pre-funny:
>>
>> \usemodule[pre-funny]
>> \starttext
>> \TitlePage
>>   {A Few Nice Quotes\\
>>    A Simple Style Demo\\
>>    Hans Hagen, August 2000}
>>
>> \Topic {Douglas R. Hofstadter} \input douglas \page
>> \stoptext
>>
>>
>> gives a rousing:
>>
>> layout : calc bgrounds
>> Randomizer initialized... [MP to PDF] (./mpgraph.8000
>> !Arithmetic overflow.
>
> Can you try:
>
> \maxnofMPgraphics = 4000
>

Did so (in meta-ini.tex actually, couldn't find it referenced anywhere 
but there), rebuilt formats, count-down numbering now starts at 4000, as 
predicted. Same error unfortunately.

> --
flip phillips
http://www.skidmore.edu/~flip/

Re: pre-funny being, well, funny...

[Johannes Hüsing]

On Wed, Apr 10, 2002 at 12:37:12PM -0400, flip phillips wrote:

> On Tuesday, April 9, 2002, at 05:45  PM, Hans Hagen wrote:

> >Can you try:
> >
> >\maxnofMPgraphics = 4000
> >
> 
> Did so (in meta-ini.tex actually, couldn't find it referenced anywhere 
> but there), rebuilt formats, count-down numbering now starts at 4000, as 
> predicted. Same error unfortunately.
> 

For me, it worked :-) Thank you.

As for the othr problem, I seemed to have reduced the minimal example
beyond the max, I'll see what must be done to reproduce the error.

Groetjes

Johannes
-- 
Johannes Hüsing   There is something fascinating about science. One gets
hannes@ruhrau.de  such wholesale returns of conjecture from such a 
                  trifling investment of fact.                Mark Twain