From: rfengineer55 on
Hello,

I' working on another of a large collection of FCC Fortran programs
compiled in VAX VMS Fortran.

There are variables in this program named Class4 , Class3, Class2, and
Class1.They are declared as type Logical. There are other variables
such as class_out declared as type character. All of my variables with
class as part of their name, are generating compiler errors. Here is
one -

When the compiler hits these, I get an error saying that #5082 syntax
error, found IDENTIFIER '_out when expecting one of : (%:, ==>

The line in the code associated with this error is -

class_out = 'OA"


An error message I get associated with the Logical variables is $5082
Syntax error, found INTEGER_CONSTANT '4' when expecting one of :
(%:,,==>

The line of code associated with this error is -

class4 - .false.

I am not aware of anything in Fortran where Class has any kind of
reserved meaning, There is some mention of Class made in the Visual
Studio 2008 book, but I have not doug deeply into that. Before I do,
I thought to run this interesting compiler situation by the list
members.

What do you think?

Thanks beforehand, for the help.

Jeff

RF ENGINEER55

From: dpb on
rfengineer55 wrote:
> Hello,
>
> I' working on another of a large collection of FCC Fortran programs
> compiled in VAX VMS Fortran.
>
> There are variables in this program named Class4 , Class3, Class2, and
> Class1.They are declared as type Logical. There are other variables
> such as class_out declared as type character. All of my variables with
> class as part of their name, are generating compiler errors. Here is
> one -
>
> When the compiler hits these, I get an error saying that #5082 syntax
> error, found IDENTIFIER '_out when expecting one of : (%:, ==>
>
> The line in the code associated with this error is -
>
> class_out = 'OA"
>

There are mismatched character delimiters in the line as shown. If
that's really so, fix that first.

Otherwise, insufficient context given because in isolation the line
looks ok.

> An error message I get associated with the Logical variables is $5082
> Syntax error, found INTEGER_CONSTANT '4' when expecting one of :
> (%:,,==>
>
> The line of code associated with this error is -
>
> class4 - .false.

That's got a problem of an "-" instead of an "=" in what appears should
be an assignment statement.

From the above, I'm guessing this was optically scanned from a printed
listing and it's full of errors from that process.

--

From: Louis Krupp on
On 6/27/2010 9:28 AM, rfengineer55 wrote:
> Hello,
>
> I' working on another of a large collection of FCC Fortran programs
> compiled in VAX VMS Fortran.
>
> There are variables in this program named Class4 , Class3, Class2, and
> Class1.They are declared as type Logical. There are other variables
> such as class_out declared as type character. All of my variables with
> class as part of their name, are generating compiler errors. Here is
> one -
>
> When the compiler hits these, I get an error saying that #5082 syntax
> error, found IDENTIFIER '_out when expecting one of : (%:, ==>
>
> The line in the code associated with this error is -
>
> class_out = 'OA"
>
>
> An error message I get associated with the Logical variables is $5082
> Syntax error, found INTEGER_CONSTANT '4' when expecting one of :
> (%:,,==>
>
> The line of code associated with this error is -
>
> class4 - .false.
>
> I am not aware of anything in Fortran where Class has any kind of
> reserved meaning, There is some mention of Class made in the Visual
> Studio 2008 book, but I have not doug deeply into that. Before I do,
> I thought to run this interesting compiler situation by the list
> members.
>
> What do you think?
>
> Thanks beforehand, for the help.

A quick guess:

1. IVF recognizes a CLASS keyword.

2. Your compiler is treating blanks as insignificant, so it parses
"CLASS1" like "CLASS 1", and it's expecting something different after
"CLASS".

You'll have to find a way to make it treat blanks as significant (just
like in most other languages). This may be a matter of specifying
free-format source instead of fixed-format (or whatever they're called).
If your source files are really in fixed format, you may have to
change other things, like line continuations, but it will be worth it in
the end.

Louis


From: rfengineer55 on
On Jun 27, 10:34 am, dpb <n...(a)non.net> wrote:
> rfengineer55 wrote:
> > Hello,
>
> > I' working on another of a large collection of FCC Fortran programs
> > compiled in VAX VMS Fortran.
>
> > There are variables in this program named Class4 , Class3, Class2, and
> > Class1.They are declared as type Logical. There are other variables
> > such as class_out declared as type character. All of my variables with
> > class as part of their name, are generating compiler errors. Here is
> > one -
>
> > When the compiler hits these, I get an error saying that #5082 syntax
> > error, found IDENTIFIER '_out when expecting one of :  (%:, ==>
>
> > The line in the code associated with this error is -
>
> >           class_out  = 'OA"
>
> There are mismatched character delimiters in the line as shown.  If
> that's really so, fix that first.
>
> Otherwise, insufficient context given because in isolation the line
> looks ok.
>
> > An error message I get associated with the Logical variables is $5082
> > Syntax error, found INTEGER_CONSTANT '4' when expecting one of :
> > (%:,,==>
>
> > The line of code associated with this error is -
>
> >          class4 - .false.
>
> That's got a problem of an "-" instead of an "=" in what appears should
> be an assignment statement.
>
>  From the above, I'm guessing this was optically scanned from a printed
> listing and it's full of errors from that process.
>
> --- Hide quoted text -
>
> - Show quoted text -


dpb,

Thanks. The errors you found are a result of my sloppy typing. In the
source code, the character delimiters are actually OK

class_out = 'AB'

Same for the other line. In the sorce, it actually reads -

class4 = .false.

I'm truly sorry for not catching this before I posted. I will be more
careful in interest of everyone's time.

Thanks,

Jeff
RF ENGINEER55
From: rfengineer55 on
On Jun 27, 11:42 am, Louis Krupp <lkrupp_nos...(a)indra.com.invalid>
wrote:
> On 6/27/2010 9:28 AM, rfengineer55 wrote:
>
>
>
>
>
> > Hello,
>
> > I' working on another of a large collection of FCC Fortran programs
> > compiled in VAX VMS Fortran.
>
> > There are variables in this program named Class4 , Class3, Class2, and
> > Class1.They are declared as type Logical. There are other variables
> > such as class_out declared as type character. All of my variables with
> > class as part of their name, are generating compiler errors. Here is
> > one -
>
> > When the compiler hits these, I get an error saying that #5082 syntax
> > error, found IDENTIFIER '_out when expecting one of :  (%:, ==>
>
> > The line in the code associated with this error is -
>
> >            class_out  = 'OA"
>
> > An error message I get associated with the Logical variables is $5082
> > Syntax error, found INTEGER_CONSTANT '4' when expecting one of :
> > (%:,,==>
>
> > The line of code associated with this error is -
>
> >           class4 - .false.
>
> > I am not aware of anything in Fortran where Class has any kind of
> > reserved meaning, There is some mention of Class made in the Visual
> > Studio 2008 book, but I have not doug deeply into that.  Before I do,
> > I thought to run this interesting compiler situation by the list
> > members.
>
> > What do you think?
>
> > Thanks beforehand, for the help.
>
> A quick guess:
>
> 1. IVF recognizes a CLASS keyword.
>
> 2. Your compiler is treating blanks as insignificant, so it parses
> "CLASS1" like "CLASS 1", and it's expecting something different after
> "CLASS".
>
> You'll have to find a way to make it treat blanks as significant (just
> like in most other languages).  This may be a matter of specifying
> free-format source instead of fixed-format (or whatever they're called).
>   If your source files are really in fixed format, you may have to
> change other things, like line continuations, but it will be worth it in
> the end.
>
> Louis- Hide quoted text -
>
> - Show quoted text -

Louis,

I think you're right on. Given your observations and suggestions, I'm
going to do some digging in the VS 2008 and find out more about the
Class references that I briefly noted. Hopefully there will be an easy
fix. I I can convince the compiler to treat class1 as written rather
than parsing it as class 1 like you're saying, I should be good to go.

This Fortran source I'm working with is large, so I hesitate to make
changes in the variable names because that could have far-reaching
consequences. I would normally just change the variable name and e
done with it, inthe case of a smaller program.

Louis, thank you for your thoughts on this.


Jeff

RF EGINEER55