info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Is there good IDE tool for fortran?
372 views
Skip to first unread message
fortran user
Jan 21, 2023, 3:17:58 PM1/21/23
to
Dear fortran users,
I am also a user. I work on some quite old fortran programs. I need IDE tools to catch the structure of that program. Which do you suggest to use?
I use Pycharm because I use python too. I found there is a plugin for fortran, but when it loads the fortran files, it is very slow and failed to recgnise some fortran lines, like #define.
Thanks for your help.
I am also a user. I work on some quite old fortran programs. I need IDE tools to catch the structure of that program. Which do you suggest to use?
I use Pycharm because I use python too. I found there is a plugin for fortran, but when it loads the fortran files, it is very slow and failed to recgnise some fortran lines, like #define.
Thanks for your help.
Thomas Koenig
Jan 21, 2023, 5:50:33 PM1/21/23
to
fortran user <mj198...@gmail.com> schrieb:
a spin. Apart from having many checks, it also generates
a calltree of files.
It is not an IDE, though, it is text-based.
> I use Pycharm because I use python too. I found there is a plugin
> for fortran, but when it loads the fortran files, it is very slow
> and failed to recgnise some fortran lines, like #define.
#define is not a Fortran line. The (traditional) C preprocessor is
often used because Fortran lacks a standardized preprocessor, but
I don't think that ftnchek supports that.
Caveat: I have used ftncheck a few decades ago, but not since
I got access to a working Fortran 95 compiler with modules.
> Dear fortran users,
>
> I am also a user. I work on some quite old fortran programs. I
> need IDE tools to catch the structure of that program. Which do
> you suggest to use?
If the program is old (pre-Fortran 90) you might give ftnchek
>
> I am also a user. I work on some quite old fortran programs. I
> need IDE tools to catch the structure of that program. Which do
> you suggest to use?
a spin. Apart from having many checks, it also generates
a calltree of files.
It is not an IDE, though, it is text-based.
> I use Pycharm because I use python too. I found there is a plugin
> for fortran, but when it loads the fortran files, it is very slow
> and failed to recgnise some fortran lines, like #define.
often used because Fortran lacks a standardized preprocessor, but
I don't think that ftnchek supports that.
Caveat: I have used ftncheck a few decades ago, but not since
I got access to a working Fortran 95 compiler with modules.
Lynn McGuire
Jan 21, 2023, 8:11:57 PM1/21/23
to
to the GCC and GFortran compilers.
Your other choice is Intel Fortran using the Community version of Visual
Studio 2019. The iFort integration is not very well done.
If you want to use an old F77 compiler, Open Watcom has a very good
visual debugger for Windows 32 bit. But the debugger has problems with
F77 code on Windows 10 x64. And the OW compilers do not generate 64 bit
code.
http://openwatcom.org/
Good luck, you are going to need it. My experience was not good so I am
moving my 800,000 lines of F77 code to C++. It is going slowly, I am
only at 73,000 lines converted so far.
Lynn
Eberhard Franz
Jan 22, 2023, 10:34:17 AM1/22/23
to
Greetings
muppets
Sergey Budaev
Jan 28, 2023, 9:44:52 AM1/28/23
to
I also vote for Code:Blocks fortran, I am using it for many years on
Windows and Linux and it works better than alternatives. It is also
rather lightweight.
What structure of the program do you need? Code:Blocks shows a panel
with modules, variables, derived types, subroutines, functions in a
thee-like fashion. It also has a nicely snmart autocompletion for
derived types and objects.
Geany editor is also nice and very light. It has a similar panel with
the program components. But code completion is much more primitive.
Best regards,
Sergey
--
Sergey V. Budaev
Universitetet i Bergen, Institutt for biovitenskap, Teoretisk økologi,
Windows and Linux and it works better than alternatives. It is also
rather lightweight.
What structure of the program do you need? Code:Blocks shows a panel
with modules, variables, derived types, subroutines, functions in a
thee-like fashion. It also has a nicely snmart autocompletion for
derived types and objects.
Geany editor is also nice and very light. It has a similar panel with
the program components. But code completion is much more primitive.
Best regards,
Sergey
Sergey V. Budaev
Universitetet i Bergen, Institutt for biovitenskap, Teoretisk økologi,
Lynn McGuire
Jan 28, 2023, 3:29:29 PM1/28/23
to
Thanks,
Lynn
Sergey Budaev
Jan 28, 2023, 4:01:00 PM1/28/23
to
>Does Code::Blocks work with Intel Fortran and can you debug with it ?
>
>Thanks,
>Lynn
Yes on Linux, but I recall I had problems on Windows. But the newer
>
>Thanks,
>Lynn
versions of both Code:Blocks and ifort's debugger.
Sergey
Lynn McGuire
Jan 28, 2023, 8:24:33 PM1/28/23
to
On 1/28/2023 3:00 PM, Sergey Budaev wrote:
>> Does Code::Blocks work with Intel Fortran and can you debug with it ?
>>
>> Thanks,
>> Lynn
>
> Yes on Linux, but I recall I had problems on Windows. But the newer
> versions of both Code:Blocks and ifort's debugger.
>
> Sergey
Do you have mixed Fortran and C/C++ in your app ?
>> Does Code::Blocks work with Intel Fortran and can you debug with it ?
>>
>> Thanks,
>> Lynn
>
> Yes on Linux, but I recall I had problems on Windows. But the newer
> versions of both Code:Blocks and ifort's debugger.
>
> Sergey
Thanks,
Lynn
Sergey Budaev
Jan 29, 2023, 6:58:26 AM1/29/23
to
>Do you have mixed Fortran and C/C++ in your app ?
>
>Thanks,
>Lynn
Not C++, but C. On Linux it invokes gdb which works with C,C++ and
>
>Thanks,
>Lynn
Fortran and should not be any problem. On Windows this may be a
problem for ifort, I guess.
Best,
FortranFan
Jan 30, 2023, 12:30:23 AM1/30/23
to
On Sunday, January 29, 2023 at 6:58:26 AM UTC-5, Sergey Budaev wrote:
> .. On Windows this may be a
Note the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code which is highly nonstandard per any of Fortran (and earlier FORTRAN version) standards and not portable and likely has a host of other questionable aspects and issues. Like most nonquality craftspeople, @Lynn McGuire then proceeds to blame the tools ..
Otherwise, as a company like WinSim Inc. that strives to sell software and services to chemical processing and related industries whose market capitalization is a trillion plus US$, with just a bit of discipline and humility and purpose and enterprise, any running code in earlier dialects of FORTRAN, even if billions of lines let alone the piddly millions LOC mentioned often by @Lynn McGuire, can readily be modernized to work with Microsoft Visual Studio 2022 and Intel Fortran 2023 in conjunction with Microsoft C/C++ compiler 2022 and also benefit from *graphical debugging* as well as making use of Fortran 2018 standard and C++20 standard along with some C++23 features. The competitors of WinSim Inc. are already using Visual Studio IDE successfully on Windows OS with Intel Fortran, Microsoft C/C++ compiler, etc.
And the bulk of the computations toward chemical process modeling and simulations, especially the most important aspects of thermophysical property and phase equilibrium computations using equation-of-state and activity coefficient models and other thermodynamic number-crunching can all remain in modern Fortran where they are best suited. Whereas many of the graphical user interface (UI), data processing, and other aspects can make use of a host of languages and platforms including C++, C#/VB, Python, etc. and also Fortran.
A teeny tiny flavor of what is possible on Windows OS with Intel Fortran 2023 compiler version supporting all of Fortran 2018 and Microsoft C/C++ compiler supporting C++20 (and C++23) features can be seen in this example I posted at the Fortran Discourse site on how to consume "a string of arrays" in Fortran in C++:
https://fortran-lang.discourse.group/t/return-an-array-of-strings-from-fortran-to-c/5100/7?u=fortranfan
Surely there are some problems with Intel Fortran integration with Microsoft Visual Studio in that there can be many initial missteps during installation and that the support for Fortran is *without all the creature comfort** that the users of Microsoft's own C# (and Visual Basic) languages receive. Plus there are some outstanding bugs Intel has not fixed now for close to 10 years. But these issues can be worked around, the benefits to be gained with a good build system for Windows and also graphical debugging are huuuge.
Again, the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code ..
> .. On Windows this may be a
> problem for ifort, I guess.
> ..
Note the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code which is highly nonstandard per any of Fortran (and earlier FORTRAN version) standards and not portable and likely has a host of other questionable aspects and issues. Like most nonquality craftspeople, @Lynn McGuire then proceeds to blame the tools ..
Otherwise, as a company like WinSim Inc. that strives to sell software and services to chemical processing and related industries whose market capitalization is a trillion plus US$, with just a bit of discipline and humility and purpose and enterprise, any running code in earlier dialects of FORTRAN, even if billions of lines let alone the piddly millions LOC mentioned often by @Lynn McGuire, can readily be modernized to work with Microsoft Visual Studio 2022 and Intel Fortran 2023 in conjunction with Microsoft C/C++ compiler 2022 and also benefit from *graphical debugging* as well as making use of Fortran 2018 standard and C++20 standard along with some C++23 features. The competitors of WinSim Inc. are already using Visual Studio IDE successfully on Windows OS with Intel Fortran, Microsoft C/C++ compiler, etc.
And the bulk of the computations toward chemical process modeling and simulations, especially the most important aspects of thermophysical property and phase equilibrium computations using equation-of-state and activity coefficient models and other thermodynamic number-crunching can all remain in modern Fortran where they are best suited. Whereas many of the graphical user interface (UI), data processing, and other aspects can make use of a host of languages and platforms including C++, C#/VB, Python, etc. and also Fortran.
A teeny tiny flavor of what is possible on Windows OS with Intel Fortran 2023 compiler version supporting all of Fortran 2018 and Microsoft C/C++ compiler supporting C++20 (and C++23) features can be seen in this example I posted at the Fortran Discourse site on how to consume "a string of arrays" in Fortran in C++:
https://fortran-lang.discourse.group/t/return-an-array-of-strings-from-fortran-to-c/5100/7?u=fortranfan
Surely there are some problems with Intel Fortran integration with Microsoft Visual Studio in that there can be many initial missteps during installation and that the support for Fortran is *without all the creature comfort** that the users of Microsoft's own C# (and Visual Basic) languages receive. Plus there are some outstanding bugs Intel has not fixed now for close to 10 years. But these issues can be worked around, the benefits to be gained with a good build system for Windows and also graphical debugging are huuuge.
Again, the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code ..
Lynn McGuire
Jan 30, 2023, 2:36:26 PM1/30/23
to
On 1/29/2023 11:30 PM, FortranFan wrote:
> On Sunday, January 29, 2023 at 6:58:26 AM UTC-5, Sergey Budaev wrote:
>
>> .. On Windows this may be a
>> problem for ifort, I guess.
>> ..
>
> Note the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code which is highly nonstandard per any of Fortran (and earlier FORTRAN version) standards and not portable and likely has a host of other questionable aspects and issues. Like most nonquality craftspeople, @Lynn McGuire then proceeds to blame the tools ..
Hello Vipul Parekh,
> On Sunday, January 29, 2023 at 6:58:26 AM UTC-5, Sergey Budaev wrote:
>
>> .. On Windows this may be a
>> problem for ifort, I guess.
>> ..
>
> Note the problem for @Lynn McGuire is almost entirely @Lynn McGuire's code which is highly nonstandard per any of Fortran (and earlier FORTRAN version) standards and not portable and likely has a host of other questionable aspects and issues. Like most nonquality craftspeople, @Lynn McGuire then proceeds to blame the tools ..
Please seek help for your obsession with me and my business. You direly
need mental help.
Sincerely,
Lynn McGuire
Lynn McGuire
Jan 30, 2023, 2:39:04 PM1/30/23
to
On 1/29/2023 5:58 AM, Sergey Budaev wrote:
>> Do you have mixed Fortran and C/C++ in your app ?
>>
>> Thanks,
>> Lynn
>
> Not C++, but C. On Linux it invokes gdb which works with C,C++ and
> Fortran and should not be any problem. On Windows this may be a
> problem for ifort, I guess.
>
> Best,
>
> Sergey
Yes, I have F77, C, and C++ all mixed up in my application. It is quite
>> Do you have mixed Fortran and C/C++ in your app ?
>>
>> Thanks,
>> Lynn
>
> Not C++, but C. On Linux it invokes gdb which works with C,C++ and
> Fortran and should not be any problem. On Windows this may be a
> problem for ifort, I guess.
>
> Best,
>
> Sergey
the mess and a serious challenge for the IDEs.
Thanks,
Lynn
FortranFan
Jan 30, 2023, 4:47:56 PM1/30/23
to
On Monday, January 30, 2023 at 2:39:04 PM UTC-5, Lynn McGuire wrote:
> ..
@Lynn McGuire,
Why don't you first stop spreading misinformation online?
Mixed language solutions on Windows OS involving Fortran, C, and C++ projects are *NOT* "a serious challenge for the IDEs", that is again misinformation you are trying to spread.
If you would just procure the right services or show a bit of discipline and humility and purpose and enterprise yourself, things would get working for WinSim.
Instead you have been at this over 10 years and again and again, you post erroneous information online from your experiences where you ostensibly tried to do things in unfathomable manner and failed.
You then repeatedly post online that in essence involves blaming the tools instead of fixing up your questionable code and workflow. Your posts are again and again simply spreading the wrong information about the capabilities that are available for mixed language applications on Windows OS with C++, C, and Fortran.
> ..
> Yes, I have F77, C, and C++ all mixed up in my application. It is quite
> the mess and a serious challenge for the IDEs. ..
@Lynn McGuire,
Why don't you first stop spreading misinformation online?
Mixed language solutions on Windows OS involving Fortran, C, and C++ projects are *NOT* "a serious challenge for the IDEs", that is again misinformation you are trying to spread.
If you would just procure the right services or show a bit of discipline and humility and purpose and enterprise yourself, things would get working for WinSim.
Instead you have been at this over 10 years and again and again, you post erroneous information online from your experiences where you ostensibly tried to do things in unfathomable manner and failed.
You then repeatedly post online that in essence involves blaming the tools instead of fixing up your questionable code and workflow. Your posts are again and again simply spreading the wrong information about the capabilities that are available for mixed language applications on Windows OS with C++, C, and Fortran.
Lynn McGuire
Jan 30, 2023, 5:22:49 PM1/30/23
to
FortranFan
Jan 30, 2023, 9:58:27 PM1/30/23
to
On Monday, January 30, 2023 at 2:39:04 PM UTC-5, Lynn McGuire wrote:
> ..
> ..
> Yes, I have F77, C, and C++ all mixed up in my application. It is quite
> the mess and a serious challenge for the IDEs. ..
@Lynn McGuire, get into the 1990s ..
https://snoweye.github.io/R_note/inc_menu/reference/mixed_lang_progr.htm
0 new messages