|
|
 Up |
|
|
  |
Date: Aug 1, 2007 09:01
Hello - I have been given some 500 lines of Fortran code, that dates
back to 1992... I have not seen, much less, written any Fortran since
the mid-90's
I do not know what system or compiler it originated from, but I have
recompiled under windows with g77 compiler...
Now that I am trying to run it at the commandline, I get the error:
sfe: formatted io not allowed
apparent state: unit 11 named MARKt10
last format: list io
lately writing sequential formatted external IO
abnormal program termination
I debugged it and found that it is getting the error at:
Call IO_OPEN(TEMCOR,NUMTEM)
noting that the variable TEMCOR=37. and NUMTEM=31
Does anyone know where I can go to get some help on this?
thanks, mark
|
| |
|
| | 16 Comments |
|
|
Author: mecej4mecej4 Date: Aug 1, 2007 09:45
msp@duke wrote:
> Hello - I have been given some 500 lines of Fortran code, that dates
> back to 1992... I have not seen, much less, written any Fortran since
> the mid-90's
>
> I do not know what system or compiler it originated from, but I have
> recompiled under windows with g77 compiler...
>
> Now that I am trying to run it at the commandline, I get the error:
>
> sfe: formatted io not allowed
> apparent state: unit 11 named MARKt10
> last format: list io
> lately writing sequential formatted external IO
>
> abnormal program termination
>
> I debugged it and found that it is getting the error at:
> Call IO_OPEN(TEMCOR,NUMTEM)
> noting that the variable TEMCOR=37. and NUMTEM=31 ...
|
| Show full article (1.54Kb) |
|
| | no comments |
|
|
Date: Aug 1, 2007 12:57
On Aug 1, 12:45 pm, mecej4 operamail.com> wrote:> msp@duke wrote:
>> Hello - I have been given some 500 lines of Fortran code, that dates
>> back to 1992... I have not seen, much less, written any Fortran since
>> the mid-90's
>
>> I do not know what system or compiler it originated from, but I have
>> recompiled under windows with g77 compiler...
>
>> Now that I am trying to run it at the commandline, I get the error:
>
>> sfe: formatted io not allowed
>> apparent state: unit 11 named MARKt10
>> last format: list io
>> lately writing sequential formatted external IO
>
>> abnormal program termination
>
>> I debugged it and found that it is getting the error at:
>> Call IO_OPEN(TEMCOR,NUMTEM) ...
|
| Show full article (4.44Kb) |
| no comments |
|
|
Author: Dick HendricksonDick Hendrickson Date: Aug 1, 2007 13:17
msp@duke wrote:
> Hello - I have been given some 500 lines of Fortran code, that dates
> back to 1992... I have not seen, much less, written any Fortran since
> the mid-90's
>
> I do not know what system or compiler it originated from, but I have
> recompiled under windows with g77 compiler...
>
> Now that I am trying to run it at the commandline, I get the error:
>
> sfe: formatted io not allowed
> apparent state: unit 11 named MARKt10
> last format: list io
> lately writing sequential formatted external IO
>
> abnormal program termination
>
> I debugged it and found that it is getting the error at:
> Call IO_OPEN(TEMCOR,NUMTEM)
> noting that the variable TEMCOR=37. and NUMTEM=31 ...
|
| Show full article (2.30Kb) |
| no comments |
|
|
Author: George N. White IIIGeorge N. White III Date: Aug 1, 2007 13:39
On Wed, 1 Aug 2007, msp@duke wrote:
> [...]
>
> You raise some valid points but unfortunately, some are in an ideal
> setting :-/
>
> 1) As you said: '...harder for Usenet group readers to debug code that
> you don't show us.'
> - I concur but not being a Fortran programmer, per say, I am not sure
> if you really want to see 690 lines of code...
You need to construct a minimal example that demonstrates the problem.
Doing this may take some understanding of Fortran and also of the data.
You are faced with either learning a bit about Fortran or hoping that
someone will be willing to fix your 690 line program. In general, the
chances of the latter are small, particularly if the data are at all
messy.
|
| Show full article (1.09Kb) |
| no comments |
|
|
Author: glen herrmannsfeldtglen herrmannsfeldt Date: Aug 1, 2007 14:34
George N. White III wrote:
(snip)
> You need to construct a minimal example that demonstrates the problem.
> Doing this may take some understanding of Fortran and also of the data.
> You are faced with either learning a bit about Fortran or hoping that
> someone will be willing to fix your 690 line program. In general, the
> chances of the latter are small, particularly if the data are at all
> messy.
It seems the OP doesn't have all the source, which will make
recompiling harder. I suspect, though, that since it has not been
changed for many years that it is trying to give an error message,
and so using code that wasn't properly debugged. It might be
possible to figure out the possible error condition without
completely understanding the code.
> In my experience, any time you are faced with processing an unfamiliar
> data format, it is helpful to write a program that just reads a bit of
> data and writes a short summary. This might be a good exercise for you
> to brush up on your Fortran skills, and may just give you the sort of
> short example where usenet can be effective.
|
| Show full article (1.11Kb) |
| no comments |
|
|
Author: TerenceTerence Date: Aug 1, 2007 15:33
On Aug 2, 8:29 am, glen herrmannsfeldt ugcs.caltech.edu> wrote:> George N. White III wrote:
>
> (snip)
>
>> You need to construct a minimal example that demonstrates the problem.
>> Doing this may take some understanding of Fortran and also of the data.
>> You are faced with either learning a bit about Fortran or hoping that
>> someone will be willing to fix your 690 line program. In general, the
>> chances of the latter are small, particularly if the data are at all
>> messy.
>
> It seems the OP doesn't have all the source, which will make
> recompiling harder. I suspect, though, that since it has not been
> changed for many years that it is trying to give an error message,
> and so using code that wasn't properly debugged. It might be
> possible to figure out the possible error condition without
> completely understanding the code.
>
>> In my experience, any time you are faced with processing an unfamiliar ...
|
| Show full article (2.14Kb) |
| no comments |
|
|
Date: Aug 2, 2007 06:07
On Aug 1, 6:33 pm, Terence cantv.net> wrote:> On Aug 2, 8:29 am, glen herrmannsfeldt ugcs.caltech.edu> wrote:
>
>
>
>
>>> George N. White III wrote:
>
>> (snip)
>
>>> You need to construct a minimal example that demonstrates the problem.
>>> Doing this may take some understanding of Fortran and also of the data.
>>> You are faced with either learning a bit about Fortran or hoping that
>>> someone will be willing to fix your 690 line program. In general, the
>>> chances of the latter are small, particularly if the data are at all
>>> messy.
>
>> It seems the OP doesn't have all the source, which will make
>> recompiling harder. I suspect, though, that since it has not been ...
|
| Show full article (4.68Kb) |
| no comments |
|
|
Author: Dick HendricksonDick Hendrickson Date: Aug 2, 2007 11:28
msp@duke wrote:
> On Aug 1, 6:33 pm, Terence cantv.net> wrote:>> On Aug 2, 8:29 am, glen herrmannsfeldt ugcs.caltech.edu> wrote:
>>
>>
>>
>>
>>>>> George N. White III wrote:
>>> (snip)
>>>> You need to construct a minimal example that demonstrates the problem.
>>>> Doing this may take some understanding of Fortran and also of the data.
>>>> You are faced with either learning a bit about Fortran or hoping that
>>>> someone will be willing to fix your 690 line program. In general, the
>>>> chances of the latter are small, particularly if the data are at all
>>>> messy.
>>> It seems the OP doesn't have all the source, which will make
>>> recompiling harder. I suspect, though, that since it has not been
>>> changed for many years that it is trying to give an error message,
>>> and so using code that wasn't properly debugged. It might be ...
|
| Show full article (6.32Kb) |
| no comments |
|
|
|
|
|
Author: dpbdpb Date: Aug 2, 2007 11:57
msp@duke wrote:
...
> Read (5,'(q,a64)',end=999) length,filename
>
> Again, not knowing Fortran, I am thinking that:
> "5" = the unit that the file in read in
> "'(q,a64)'" = I HAVE NO IDEA!
> "end=999" = when eof is reached, go to line
> numbered 999
> "length" = the 1st field read in?
> "filename" = the 2nd field read in?
...
From CVF help files...
/snip
Character Count Editing (Q)
The character count edit descriptor returns the remaining number of
characters in the current input record.
The corresponding I/O list item must be of type integer or logical. For
example, suppose the following statements are specified:
|
| Show full article (1.59Kb) |
| no comments |
|
|
|
|
