Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.postscript > #1555

Re: Checking the stack is empty at the end of the program

From Cecil Westerhof <Cecil@decebal.nl>
Newsgroups comp.lang.postscript
Subject Re: Checking the stack is empty at the end of the program
Organization Decebal Computing
References (4 earlier) <0034540c-4d60-42b6-9d32-f3357f719636@googlegroups.com> <87pptx8zip.fsf@Aspire.decebal.nl> <d7307707-edf3-4dde-805b-e789439e340b@googlegroups.com> <a9d5917e-62ba-4a6f-bf29-4dc5e1166ee2@googlegroups.com> <b2a30f33-795a-4a93-ba63-2e3338b92c53@googlegroups.com>
Date 2013-08-01 15:45 +0200
Message-ID <87d2px8qzi.fsf@Aspire.decebal.nl> (permalink)

Show all headers | View raw


Op donderdag 1 aug 2013 13:23 CEST schreef luser:

> On Thursday, August 1, 2013 5:56:09 AM UTC-5, luser- -droog wrote:
>> On Thursday, August 1, 2013 5:53:20 AM UTC-5, luser- -droog wrote:
>>
>>> On Thursday, August 1, 2013 5:40:46 AM UTC-5, Cecil Westerhof wrote:
>>
>>
>>
>>>> Works, thanks. Only one thing: I already showed the stack. Is it
>>
>>>> possible to circumvent the stack dumps?
>>
>>>
>>
>>>
>>
>>> Well, yeah. But aren't you going waaay overboard on this?
>>
>>>
>>
>>>
>>
>>>
>>
>>> Be that as it may, you can install your own error handler
>>
>>> in errordict.
>>
>>>
>>
>>> errordict/unregistered{
>>
>>> (Error: unregistered in ) print =
>>
>>
>>
>> interrupt %= stop
>>
>>
>>
>>> } put
>>
>>>
>>
>>>
>>
>>>
>>
>>> This is the procedure that signalerror looks up and calls.
>>
>>
>>
>> [remember to call `interrupt` or `stop` in the error handler
>>
>> or else it won't bail-out of the execution.]
>
> Actually, after looking at the error code again,
>
> http://code.google.com/p/xpost/source/browse/err.ps
>
> (borrowed from the book Inside Postscript),
> it's errordict/handleerror that you'd want to override.
> That's the procedure that prints out the error report.
>
> Although the above will probably work (maybe not, since
> we're explicitly calling signalerror), it bypasses 
> several steps.

Now it works:
    % to be called at the end of a program
    % if stack empty (as it should be) say this
    % otherwise signal this and show the stack
    % ends the program to circumvent/signal it being called to soon
    /endProgram {
        count 0 eq {
            (Stack empty at end of program) ==
            quit
        } {
            count dup
            1 eq {
                (ERROR: object on stack at the end of the program)   ==
            } {
                (ERROR: objects on stack at the end of the program)  ==
            } ifelse
            {
                (Unexpected object on stack: ) print
                ==
            } repeat
            errordict/handleerror {
                interrupted
            } put
            /STACK-NOT-EMPTY cvx /unregistered signalerror
        } ifelse
        (This message should never be shown)       ==
        (If it is, the quit operater is overruled) ==
    } def

And now I get:
    $ psConvert.sh brainDump.ps 
    15:35:01: converting postscript
    15:35:06: generating PDF
    (ERROR: object on stack at the end of the program)
    Unexpected object on stack: (test endProgram)
    $

And when I cleanup the stack I get (default the PDF is converted to
PNG('s)):
    $ psConvert.sh brainDump.ps 
    15:39:44: converting postscript
    15:39:48: generating PDF
    (Stack empty at end of program)
    15:39:49: generating graphics
    15:39:50: brainDump.ps converted
    $

I am happy now: thanks.

By the way: what is the difference between interrupted and stop?

-- 
Cecil Westerhof
Senior Software Engineer
LinkedIn: http://www.linkedin.com/in/cecilwesterhof

Back to comp.lang.postscript | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread


Thread

Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 00:16 +0200
  Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-07-31 15:41 -0700
    Re: Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 02:36 +0200
  Re: Checking the stack is empty at the end of the program Dan Coby <adcoby@earthlink.net> - 2013-07-31 18:36 -0700
    Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-07-31 23:53 -0700
      Re: Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 10:41 +0200
        Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 02:37 -0700
          Re: Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 12:40 +0200
            Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 03:53 -0700
              Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 03:56 -0700
                Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 04:23 -0700
                Re: Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 15:45 +0200
              Re: Checking the stack is empty at the end of the program Cecil Westerhof <Cecil@decebal.nl> - 2013-08-01 13:18 +0200
                Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 09:33 -0700
                Re: Checking the stack is empty at the end of the program luser- -droog <mijoryx@yahoo.com> - 2013-08-01 18:25 -0700

csiph-web