Fortran jump command for fortran 95

  • Fortran
  • Thread starter buddhanath
  • Start date
  • #1
Even after writing the label , it is showing that label is not defined, please help!
my program reads...…...
……...
Fortran:
99 z = b + 2*c*(d - 0.0792079*(e + log((( sqrt(x) + 1.52569)**5.64616)/(((sqrt(x) - 1.26704)**5.93863)*((sqrt(x) - 0.258652)**50.2075)))) - 2 log(x))
zs = g + 2*i*(d - 0.0792079*(e + log((( sqrt(x) + 1.52569)**5.64616)/(((sqrt(x) - 1.26704)**5.93863)*((sqrt(x) - 0.258652)**50.2075)))) - 2 log(x)) + 2*j*(-0.0792079*( k-l +( (m-n-o)*p)/q + r))
h = -z/zs
tol= 0.0001
x= x+h
if (abs(h) .gt. tol) then
  goto 99
  else
  print*, "the root of the equaton is ", x
end if
end program
 
Last edited by a moderator:

Answers and Replies

  • #2
phinds
Science Advisor
Insights Author
Gold Member
17,123
8,329
PLEASE ... use the "code" tag so your post doesn't look so sloppy.
 
  • Like
Likes Wrichik Basu
  • #3
Wrichik Basu
Insights Author
Gold Member
2020 Award
1,780
1,613
I know nothing in Fortran, but it seems that the label 99 should be defined after it is mentioned in a goto statement.
 
  • Sad
Likes Vanadium 50
  • #4
FactChecker
Science Advisor
Gold Member
6,533
2,609
I know nothing in Fortran, but it seems that the label 99 should be defined after it is mentioned in a goto statement.
No. The compiler figures out the labels before it tries to run, so the label can be defined before it is referenced.
The original versions of FORTRAN were very column-dependent. Anything in column 1 was the beginning of a comment. There were a lot of column rules. I would check into that. The first 5 columns were for labels. Anything in column 6 meant that the following is a continuation of the prior row. Code was not allowed to go beyond column 72. I do not know about the modern versions.

It often helps to see the exact error message and what line and column it thinks the error is on.
It also often helps to know what version of FORTRAN you are using.
 
Last edited:
  • Like
  • Informative
Likes Vanadium 50, Wrichik Basu and berkeman
  • #5
Its FORTRAN 95. It only shows problem sometimes when I am using big equations
 
  • #6
FactChecker
Science Advisor
Gold Member
6,533
2,609
Clearly, this is just part of the program. It is not possible to know if everything was defined and compiled successfully. If long equations seem to cause a problem, see if breaking up the equation into a few steps solves the problem. If so, then you have narrowed down where to look. If not, something else went wrong.
 
  • #7
Vanadium 50
Staff Emeritus
Science Advisor
Education Advisor
27,395
11,505
A wise man once said:
"Check my work" is a really heavy ask.

You can make it easier with well-documented code, easy to understand variable names and logic structure, and the minimum example needed to demonstrate the problem.

  1. This is not complete code.
  2. "It only shows problem sometimes when I am using big equations" is annoying. Does this code show the problem or not?
  3. FORTRAN has no "jump" statement. Yes, we can figure out what you mean, but we shouldn't have to. Sloppy writing makes us work harder, just as sloppy thinking makes you work harder.
  4. If you are really using FORTRAN 95 you should not be messing with GOTO's. You should use some other structure, perhaps DO-WHILE. That is not some exotic function new to FIRTRAN 95 (itself a quarter-century old). It was in FORTRAN 77.
 
Last edited:
  • Like
Likes FactChecker
  • #8
FactChecker
Science Advisor
Gold Member
6,533
2,609
The GOTO is notorious because of extreme examples of "spaghetti code". But using it in simple ways should work with no problems. It certainly is frowned on by those who have been traumatized by the bad examples so it's a good habit to avoid it. In any case, you should always indent the interior of the loop so people (including you) can easily see it.
 
Last edited:
  • #9
35,230
7,048
If you are really using FORTRAN 95 you should not be messing with GOTO's.
This, especially (not that the other parts aren't also valid).

If you must use goto, I suspect that the problem is that the label 99 is not in the right place; i.e., somewhere in columns 1 through 5. Because it likely is not located correctly, the compiler doesn't know that you intend it to be a label.
 
  • Like
Likes Vanadium 50 and FactChecker
  • #10
35,230
7,048
It only shows problem sometimes when I am using big equations
These aren't equations -- they are assignment statements where the stuff on the right side is some expression you want to evaluate. As already mentioned, best practice is to break up these big expressions. Doing so makes debugging a lot easier.

You are using a debugger, right?
 
  • Like
Likes Vanadium 50 and FactChecker
  • #11
pbuk
Science Advisor
Gold Member
2,446
1,186
I agree with all the recommendations above; using an IDE with a debugger (or a syntax-checking code editor), breaking up long assignment statements, using do...while instead of goto, taking note of all compiler errors and warnings - these would all help you find the problem yourself.

But this once I'll find it for you - the first line you posted contains the code 2 log(x) which is not correct and prevents the line from compiling so the label 99 is not created.

This should presumably be 2 * log(x).
 
  • Like
Likes Wrichik Basu, Vanadium 50 and Ibix
  • #12
Vanadium 50
Staff Emeritus
Science Advisor
Education Advisor
27,395
11,505
Lessons from @pbuk's message:

0. I see you've been back but haven;t responded to the excellent advice you have received. That's not a good way to get more help.
1. He spotted a syntax error in Line 1. There's also one in Line 2. You never mentioned these to us - just the one on Line 7.
2. If you don't tell us all of the error messages you are getting, it will be very hard for us to figure out the problem.
 
  • #13
35,230
7,048
This should presumably be 2 * log(x).
Good catch! I didn't scroll to the right on these long code lines, so missed this one, and the similar error on the next line. This is another reason that argues against using overly long and complicated expressions in assignment statements.
 

Related Threads on Fortran jump command for fortran 95

  • Last Post
Replies
12
Views
7K
Replies
8
Views
3K
  • Last Post
Replies
2
Views
826
  • Last Post
Replies
1
Views
1K
  • Last Post
Replies
4
Views
4K
  • Last Post
Replies
7
Views
5K
  • Last Post
Replies
4
Views
2K
  • Last Post
Replies
1
Views
2K
Replies
3
Views
7K
  • Last Post
Replies
4
Views
2K
Top