modules-interest — Discussion forum for modules usage.

Re: [Modules] Reorganization suggestion for modules/{version}/init directory

[<Xav...@CE...>]

RedHat has moved to Modules 4 for EL8, and if everything goes as
expected Modules 5 should be part of EL9.

openSUSE Leap is still relying on Modules 3.2, but openSUSE
Tumbleweed has version 4.8.

The repology website is really helpful to learn how Modules is
available across OSes:

https://repology.org/project/environment-modules/versions

Globally the new Modules versions are available on recent OS
releases.

Xavier


On Fri, 2021-10-15 at 10:28 -0500, Paul Markfort wrote:
> That works (the profile.d stuff should still be put together).
> 
> Maybe add a psuedo shell,  "list-all-shells"
> Output is the list of shells in a machine parse-able format (so the
> output could be used in shells).   Or that could be the output
> returned if there are no arguments.
> 
> 
> Now we have to figure out why Redhat and SUSE both seem to be stuck
> on modules 3.2.10
> 
> If it is a (perceived) performance issue, perhaps you need to bring
> back a binary version - in the form of the modulecmd.tcl command is
> compiled into modulecmd.bin (in the libexec folder) - or include
> instructions on how to do that.
> There are good reasons why most people would want to use the tcl
> script (cross platform compatibility) while others may want a
> binary (performance).
> 
> 
> 
> On 10/15/2021 12:11 AM, Xavier Delaruelle wrote:
> > Hi Paul,
> > 
> > Current organization of the init directory already gives a good
> > insight of the shells that are supported. Except for the profile
> > scripts, the first word of each file designates the shell.
> > Splitting
> > things with more directories will make it harder to get a full
> > overview. Also such change is a major one and v5.0 is still new
> > so
> > such file movement cannot happen before 6.0.
> > 
> > What could be done instead is to list the supported shells when
> > no
> > shell or an unsupported shell is set on $MODULES_CMD :
> > 
> > $ $MODULES_CMD
> > ERROR: Unknown shell type ''
> >    Supported shells are: sh, bash, ksh, zsh, csh, tcsh, fish,
> > cmd, tcl, perl,
> >     python, ruby, lisp, cmake and r
> > 
> > Regards,
> > Xavier
> > 
> > Le jeu. 14 oct. 2021 à 17:50, Paul Markfort <pau...@gm...>
> > a écrit :
> > > 
> > > 
> > > Xavier,
> > > 
> > > May I suggest a slight reorganization of the
> > > "modules/{version}/init" folder.
> > > 
> > > I note that the contents of this folder, is the easiest way to
> > > get a list of shells that a particular version of modules
> > > supports.
> > > 
> > > 
> > > First (and probably important), move profile.csh and profile.sh
> > > to a subfolder  profile.d (and rename them modules.*).
> > > This will make them easier to find and just use these by
> > > copying them to /etc/profile.d/
> > > (I also suggest you put a test in those scripts that they are
> > > effectively noop if the user is root).
> > > 
> > > Next, rename the *_* and *-* files and folders to *.* - adding
> > > a .d to the folders (tcsh.completion, zsh.functions.d ) - or
> > > create a subfolder (maybe auxiliary.d, misc.d, or extra.d) for
> > > all of them.  This part will make it easier to use the list of
> > > files to determine what shells modules supports (both by sight,
> > > and particularly in scripts).
> > > 
> > > 
> > > Thanks.
> > > 
> > > 
> > > --
> > > --------------------------------------------------------
> > > The views and opinions expressed above are strictly
> > > those of the author(s).  The content of this message has
> > > not been reviewed nor approved by any entity whatsoever.
> > > --------------------------------------------------------
> > > Paul FM         Info: http://paulfm.com/~paulfm/
> > > --------------------------------------------------------
> > > 
> > > 
> > > _______________________________________________
> > > Modules-interest mailing list
> > > Mod...@li...
> > > https://lists.sourceforge.net/lists/listinfo/modules-interest
> > 
> > 
> > _______________________________________________
> > Modules-interest mailing list
> > Mod...@li...
> > https://lists.sourceforge.net/lists/listinfo/modules-interest
> > 
> 
> 

Re: [Modules] Reorganization suggestion for modules/{version}/init directory

[Paul M. <pau...@gm...>]

That works (the profile.d stuff should still be put together).

Maybe add a psuedo shell,  "list-all-shells"
Output is the list of shells in a machine parse-able format (so the output could be used in shells).   Or that could be the output returned if there are no arguments.


Now we have to figure out why Redhat and SUSE both seem to be stuck on modules 3.2.10

If it is a (perceived) performance issue, perhaps you need to bring back a binary version - in the form of the modulecmd.tcl command is compiled into modulecmd.bin (in the libexec folder) - or include instructions on how to do that.
There are good reasons why most people would want to use the tcl script (cross platform compatibility) while others may want a binary (performance).



On 10/15/2021 12:11 AM, Xavier Delaruelle wrote:
> Hi Paul,
> 
> Current organization of the init directory already gives a good
> insight of the shells that are supported. Except for the profile
> scripts, the first word of each file designates the shell. Splitting
> things with more directories will make it harder to get a full
> overview. Also such change is a major one and v5.0 is still new so
> such file movement cannot happen before 6.0.
> 
> What could be done instead is to list the supported shells when no
> shell or an unsupported shell is set on $MODULES_CMD :
> 
> $ $MODULES_CMD
> ERROR: Unknown shell type ''
>    Supported shells are: sh, bash, ksh, zsh, csh, tcsh, fish, cmd, tcl, perl,
>     python, ruby, lisp, cmake and r
> 
> Regards,
> Xavier
> 
> Le jeu. 14 oct. 2021 à 17:50, Paul Markfort <pau...@gm...> a écrit :
>>
>>
>> Xavier,
>>
>> May I suggest a slight reorganization of the "modules/{version}/init" folder.
>>
>> I note that the contents of this folder, is the easiest way to get a list of shells that a particular version of modules supports.
>>
>>
>> First (and probably important), move profile.csh and profile.sh to a subfolder  profile.d (and rename them modules.*).
>> This will make them easier to find and just use these by copying them to /etc/profile.d/
>> (I also suggest you put a test in those scripts that they are effectively noop if the user is root).
>>
>> Next, rename the *_* and *-* files and folders to *.* - adding a .d to the folders (tcsh.completion, zsh.functions.d ) - or create a subfolder (maybe auxiliary.d, misc.d, or extra.d) for all of them.  This part will make it easier to use the list of files to determine what shells modules supports (both by sight, and particularly in scripts).
>>
>>
>> Thanks.
>>
>>
>> --
>> --------------------------------------------------------
>> The views and opinions expressed above are strictly
>> those of the author(s).  The content of this message has
>> not been reviewed nor approved by any entity whatsoever.
>> --------------------------------------------------------
>> Paul FM         Info: http://paulfm.com/~paulfm/
>> --------------------------------------------------------
>>
>>
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
> 
> 
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] Reorganization suggestion for modules/{version}/init directory

[Xavier D. <xav...@gm...>]

Hi Paul,

Current organization of the init directory already gives a good
insight of the shells that are supported. Except for the profile
scripts, the first word of each file designates the shell. Splitting
things with more directories will make it harder to get a full
overview. Also such change is a major one and v5.0 is still new so
such file movement cannot happen before 6.0.

What could be done instead is to list the supported shells when no
shell or an unsupported shell is set on $MODULES_CMD :

$ $MODULES_CMD
ERROR: Unknown shell type ''
  Supported shells are: sh, bash, ksh, zsh, csh, tcsh, fish, cmd, tcl, perl,
   python, ruby, lisp, cmake and r

Regards,
Xavier

Le jeu. 14 oct. 2021 à 17:50, Paul Markfort <pau...@gm...> a écrit :
>
>
> Xavier,
>
> May I suggest a slight reorganization of the "modules/{version}/init" folder.
>
> I note that the contents of this folder, is the easiest way to get a list of shells that a particular version of modules supports.
>
>
> First (and probably important), move profile.csh and profile.sh to a subfolder  profile.d (and rename them modules.*).
> This will make them easier to find and just use these by copying them to /etc/profile.d/
> (I also suggest you put a test in those scripts that they are effectively noop if the user is root).
>
> Next, rename the *_* and *-* files and folders to *.* - adding a .d to the folders (tcsh.completion, zsh.functions.d ) - or create a subfolder (maybe auxiliary.d, misc.d, or extra.d) for all of them.  This part will make it easier to use the list of files to determine what shells modules supports (both by sight, and particularly in scripts).
>
>
> Thanks.
>
>
> --
> --------------------------------------------------------
> The views and opinions expressed above are strictly
> those of the author(s).  The content of this message has
> not been reviewed nor approved by any entity whatsoever.
> --------------------------------------------------------
> Paul FM         Info: http://paulfm.com/~paulfm/
> --------------------------------------------------------
>
>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest

Re: [Modules] Parsing module result

[Paul M. <pau...@gm...>]

I realized the bash aliases are better written as:

	alias test-module-anyshell='tclsh "${MODULES_CMD}" $* 2>&1'
	alias test-module-bash='tclsh "${MODULES_CMD}" bash $* 2>&1'

(the tclsh is in case either the wrong tclsh is at the top of "${MODULES_CMD}" tclsh script, or "${MODULES_CMD}" is not executable).

You can alsp (usually) run the test by hand (no alias) with:
(example)
	$MODULES_CMD bash load junk





On 10/14/2021 10:15 AM, Paul Markfort wrote:
> You could also get complicated and implement this as a script, that tests that the first argument is a valid shell name (that list is a bit long).
> And then either echos a warning, or assumes a default if it is not a shell (you should have it echo out some help if there are no arguments).
> 
> 
> 
> On 10/14/2021 9:45 AM, Paul Markfort wrote:
>>
>> I use a (BASH) alias for that (the second one makes it easy to specifically test in bash).
>>
>> alias test-module-anyshell='tclsh $MODULESHOME/default/libexec/modulecmd.tcl  $* 2>&1'
>> alias test-module-bash='tclsh $MODULESHOME/default/libexec/modulecmd.tcl bash $* 2>&1'
>>
>> Basically, it echos out what will be sent to the "eval" command of the shell.
>>
>> So if you want to know what a module will actually do, you just use test-module-bash {command} {module_name}.
>>
>> example:
>>
>>      test-module-bash load junk
>>
>> Or:
>>      test-module-anyshell bash load junk
>>      test-module-anyshell csh load junk
>>
>>
>>
>> The output can be redirected to a file (it will include all the stdout and stderr output).
>>
>>
>> On 10/14/2021 7:33 AM, Sternberg, Michael G. via Modules-interest wrote:
>>> Xavier,
>>>
>>> Indeed, thank you for integrating this!
>>>
>>> Now that "module sh-to-mod" is implemented internal to modulecmd.tcl, an interesting question and opportunity comes up:  Would it be useful – or even necessary for consistency – to have sh-to-mod *filter out* (i.e., not issue) modifications of module-internal variables, like I did pragmatically with grep?  Such internals would not be of interest for later deployment of the generated module, and make for an "Inception" challenge to modulecmd on eventual load. Having module invocations already occur in the input is certainly an edge case, but one that could well come up in porting situations.
>>>
>>> Michael Strelnikov: Could you illustrate the motivation of your question?
>>>
>>>
>>> With best regards,
>>> Michael
>>>
>>>
>>>> On Oct 13, 2021, at 23:50, Xavier Delaruelle <xav...@gm...> wrote:
>>>> […] Please note that the createmodule.py script has been replaced by the
>>>> "sh-to-mod" sub-command of module:
>>>> https://modules.readthedocs.io/en/latest/MIGRATING.html#sh-to-mod-sub-command
>>>> Regards,
>>>> Xavier
>>>>
>>>> Le jeu. 14 oct. 2021 à 02:10, Sternberg, Michael G. via
>>>> Modules-interest <mod...@li...> a écrit :
>>>>>
>>>>> […] you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:
>>>>>     echo module load foo > /tmp/foo.sh
>>>>>     createmodule.py /tmp/foo.sh \
>>>>>         | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'
>>>
>>>
>>> _______________________________________________
>>> Modules-interest mailing list
>>> Mod...@li...
>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>
>>
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

[Modules] Reorganization suggestion for modules/{version}/init directory

[Paul M. <pau...@gm...>]

Xavier,

May I suggest a slight reorganization of the "modules/{version}/init" folder.

I note that the contents of this folder, is the easiest way to get a list of shells that a particular version of modules supports.


First (and probably important), move profile.csh and profile.sh to a subfolder  profile.d (and rename them modules.*).
This will make them easier to find and just use these by copying them to /etc/profile.d/
(I also suggest you put a test in those scripts that they are effectively noop if the user is root).

Next, rename the *_* and *-* files and folders to *.* - adding a .d to the folders (tcsh.completion, zsh.functions.d ) - or create a subfolder (maybe auxiliary.d, misc.d, or extra.d) for all of them.  This part will make it easier to use the list of files to determine what shells modules supports (both by sight, and particularly in scripts).


Thanks.


-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] Parsing module result

[Paul M. <pau...@gm...>]

You could also get complicated and implement this as a script, that tests that the first argument is a valid shell name (that list is a bit long).
And then either echos a warning, or assumes a default if it is not a shell (you should have it echo out some help if there are no arguments).



On 10/14/2021 9:45 AM, Paul Markfort wrote:
> 
> I use a (BASH) alias for that (the second one makes it easy to specifically test in bash).
> 
> alias test-module-anyshell='tclsh $MODULESHOME/default/libexec/modulecmd.tcl  $* 2>&1'
> alias test-module-bash='tclsh $MODULESHOME/default/libexec/modulecmd.tcl bash $* 2>&1'
> 
> Basically, it echos out what will be sent to the "eval" command of the shell.
> 
> So if you want to know what a module will actually do, you just use test-module-bash {command} {module_name}.
> 
> example:
> 
>      test-module-bash load junk
> 
> Or:
>      test-module-anyshell bash load junk
>      test-module-anyshell csh load junk
> 
> 
> 
> The output can be redirected to a file (it will include all the stdout and stderr output).
> 
> 
> On 10/14/2021 7:33 AM, Sternberg, Michael G. via Modules-interest wrote:
>> Xavier,
>>
>> Indeed, thank you for integrating this!
>>
>> Now that "module sh-to-mod" is implemented internal to modulecmd.tcl, an interesting question and opportunity comes up:  Would it be useful – or even necessary for consistency – to have sh-to-mod *filter out* (i.e., not issue) modifications of module-internal variables, like I did pragmatically with grep?  Such internals would not be of interest for later deployment of the generated module, and make for an "Inception" challenge to modulecmd on eventual load. Having module invocations already occur in the input is certainly an edge case, but one that could well come up in porting situations.
>>
>> Michael Strelnikov: Could you illustrate the motivation of your question?
>>
>>
>> With best regards,
>> Michael
>>
>>
>>> On Oct 13, 2021, at 23:50, Xavier Delaruelle <xav...@gm...> wrote:
>>> […] Please note that the createmodule.py script has been replaced by the
>>> "sh-to-mod" sub-command of module:
>>> https://modules.readthedocs.io/en/latest/MIGRATING.html#sh-to-mod-sub-command
>>> Regards,
>>> Xavier
>>>
>>> Le jeu. 14 oct. 2021 à 02:10, Sternberg, Michael G. via
>>> Modules-interest <mod...@li...> a écrit :
>>>>
>>>> […] you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:
>>>>     echo module load foo > /tmp/foo.sh
>>>>     createmodule.py /tmp/foo.sh \
>>>>         | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'
>>
>>
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] Parsing module result

[Paul M. <pau...@gm...>]

I use a (BASH) alias for that (the second one makes it easy to specifically test in bash).

alias test-module-anyshell='tclsh $MODULESHOME/default/libexec/modulecmd.tcl  $* 2>&1'
alias test-module-bash='tclsh $MODULESHOME/default/libexec/modulecmd.tcl bash $* 2>&1'

Basically, it echos out what will be sent to the "eval" command of the shell.

So if you want to know what a module will actually do, you just use test-module-bash {command} {module_name}.

example:

	test-module-bash load junk
	
Or:
	test-module-anyshell bash load junk
	test-module-anyshell csh load junk



The output can be redirected to a file (it will include all the stdout and stderr output).


On 10/14/2021 7:33 AM, Sternberg, Michael G. via Modules-interest wrote:
> Xavier,
> 
> Indeed, thank you for integrating this!
> 
> Now that "module sh-to-mod" is implemented internal to modulecmd.tcl, an interesting question and opportunity comes up:  Would it be useful – or even necessary for consistency – to have sh-to-mod *filter out* (i.e., not issue) modifications of module-internal variables, like I did pragmatically with grep?  Such internals would not be of interest for later deployment of the generated module, and make for an "Inception" challenge to modulecmd on eventual load. Having module invocations already occur in the input is certainly an edge case, but one that could well come up in porting situations.
> 
> Michael Strelnikov: Could you illustrate the motivation of your question?
> 
> 
> With best regards,
> Michael
> 
> 
>> On Oct 13, 2021, at 23:50, Xavier Delaruelle <xav...@gm...> wrote:
>> […] Please note that the createmodule.py script has been replaced by the
>> "sh-to-mod" sub-command of module:
>> https://modules.readthedocs.io/en/latest/MIGRATING.html#sh-to-mod-sub-command
>> Regards,
>> Xavier
>>
>> Le jeu. 14 oct. 2021 à 02:10, Sternberg, Michael G. via
>> Modules-interest <mod...@li...> a écrit :
>>>
>>> […] you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:
>>>     echo module load foo > /tmp/foo.sh
>>>     createmodule.py /tmp/foo.sh \
>>>         | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'
> 
> 
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] Parsing module result

[Sternberg, M. G. <ste...@an...>]

Xavier,

Indeed, thank you for integrating this!

Now that "module sh-to-mod" is implemented internal to modulecmd.tcl, an interesting question and opportunity comes up:  Would it be useful – or even necessary for consistency – to have sh-to-mod *filter out* (i.e., not issue) modifications of module-internal variables, like I did pragmatically with grep?  Such internals would not be of interest for later deployment of the generated module, and make for an "Inception" challenge to modulecmd on eventual load. Having module invocations already occur in the input is certainly an edge case, but one that could well come up in porting situations.

Michael Strelnikov: Could you illustrate the motivation of your question?


With best regards,
Michael


> On Oct 13, 2021, at 23:50, Xavier Delaruelle <xav...@gm...> wrote:
> […] Please note that the createmodule.py script has been replaced by the
> "sh-to-mod" sub-command of module:
> https://modules.readthedocs.io/en/latest/MIGRATING.html#sh-to-mod-sub-command
> Regards,
> Xavier
> 
> Le jeu. 14 oct. 2021 à 02:10, Sternberg, Michael G. via
> Modules-interest <mod...@li...> a écrit :
>> 
>> […] you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:
>>    echo module load foo > /tmp/foo.sh
>>    createmodule.py /tmp/foo.sh \
>>        | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'

Re: [Modules] Parsing module result

[Xavier D. <xav...@gm...>]

Hi,

Please note that the createmodule.py script has been replaced by the
"sh-to-mod" sub-command of module:

https://modules.readthedocs.io/en/latest/MIGRATING.html#sh-to-mod-sub-command

Regards,
Xavier

Le jeu. 14 oct. 2021 à 02:10, Sternberg, Michael G. via
Modules-interest <mod...@li...> a écrit :
>
> Try the module command for the purpose:
>
>         module show foo
>
> The output is meant for human inspection and it does not recurse for sub-loads. To get nitty-gritty details after all, you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:
>
>     echo module load foo > /tmp/foo.sh
>
>     createmodule.py /tmp/foo.sh \
>         | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'
>
>
> createmodule.py is a helper script in $PATH, available in all recent Modules versions. It needs an explicit file to work on, unfortunately, so a pipe won't work.
> Also available and in $PATH is "createmodule.sh", in previous Modules versions included under contrib/. It is notably slower than the Python variant.
>
>
> With best regards,
> --
> Michael Sternberg, Ph.D.
> Principal Scientific Computing Administrator
> Center for Nanoscale Materials
> Argonne National Laboratory
>
>
>
> > On Oct 13, 2021, at 16:11 , Michael Strelnikov <mic...@gm...> wrote:
> >    I have a module that loads internal modules and sets variables.
> >    Is there a way to "dry load" module and get an indication what other modules were loaded and variables set?
>
>
>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest

Re: [Modules] Parsing module result

[Sternberg, M. G. <ste...@an...>]

Try the module command for the purpose:

	module show foo

The output is meant for human inspection and it does not recurse for sub-loads. To get nitty-gritty details after all, you could let Modules itself do the work of gathering and interpreting the accumulated changes of environment vars:

    echo module load foo > /tmp/foo.sh

    createmodule.py /tmp/foo.sh \
        | grep -Ev '(_modshare|MODULES_LM|_LMFILES_|LOADEDMODULES)'


createmodule.py is a helper script in $PATH, available in all recent Modules versions. It needs an explicit file to work on, unfortunately, so a pipe won't work.
Also available and in $PATH is "createmodule.sh", in previous Modules versions included under contrib/. It is notably slower than the Python variant.


With best regards,
-- 
Michael Sternberg, Ph.D.
Principal Scientific Computing Administrator
Center for Nanoscale Materials
Argonne National Laboratory



> On Oct 13, 2021, at 16:11 , Michael Strelnikov <mic...@gm...> wrote:
>    I have a module that loads internal modules and sets variables.
>    Is there a way to "dry load" module and get an indication what other modules were loaded and variables set?

[Modules] Parsing module result

[Michael S. <mic...@gm...>]

Hi,

   I have a module that loads internal modules and sets variables.
   Is there a way to "dry load" module and get an indication what other
modules were loaded and variables set?

Thanks.

Re: [Modules] nested .modulerc

[Xavier D. <xav...@gm...>]

Hi Robbin,

In your example, you define "local" variable with the "set" Tcl
command. Such variable is cleaned at the end of modulefile or modulerc
evaluation.

I would suggest you create modulefile to represent projects then once
a project modulefile is seen loaded, global modulerc of the tools
modulepath adapts the default version to use :

$ tree -a --charset=ascii ./mp
./mp
|-- .modulerc
|-- prj
|   |-- A
|   `-- B
|-- X
|   |-- 1
|   |-- 2
|   `-- 3
`-- Y
    |-- 1
    |-- 2
    `-- 3

$ cat mp/.modulerc
#%Module

switch -- [module-info loaded prj] {
    prj/A {
        module-version X/2 default
        module-version Y/2 default
    }
    prj/B {
        module-version X/3 default
        module-version Y/3 default
    }
    default {
        module-version X/1 default
        module-version Y/1 default
    }
}


$ ml av -o sym -t --color=never
prj/A
prj/B
X/1(default)
X/2
X/3
Y/1(default)
Y/2
Y/3
$ ml prj/A
$ ml av -o sym -t --color=never
prj/A
prj/B
X/1
X/2(default)
X/3
Y/1
Y/2(default)
Y/3
$ ml switch prj/B
$ ml av -o sym -t --color=never
prj/A
prj/B
X/1
X/2
X/3(default)
Y/1
Y/2
Y/3(default)

You can of course split this setup in multiple modulepaths to have the
prj modulefiles in one path and the other modulefiles in other paths.

Regards,
Xavier

Le ven. 8 oct. 2021 à 02:44, Robbin Bonthond <ro...@bo...> a écrit :
>
> PS:
>
> $ module config implicit_default 0
> $ module avail -t --color=never
> /tmp/tool:
> X/v1(default)
> X/v2
> X/v3
> Y/v1(default)
> Y/v2
> $ module load X
> WARNING: Symbolic version 'X/default' already defined
> ERROR: No default version defined for 'X'
>
>
> On Thu, Oct 7, 2021 at 5:52 PM Robbin Bonthond <ro...@bo...> wrote:
>>
>> so we have various tool versions installed in a central place, and a .modulerc in the root there contains all defaults.
>>
>> something like
>>
>> $ tree -a --charset=ascii /tmp/tool
>> /tmp/tool
>> |-- .modulerc
>> |-- X
>> |   |-- .common
>> |   |-- v1 -> .common
>> |   |-- v2 -> .common
>> |   `-- v3 -> .common
>> `-- Y
>>     |-- .common
>>     |-- v1 -> .common
>>     `-- v2 -> .common
>>
>> That /tmp/tool/.modulerc contains something like this:
>>
>> $ cat /tmp/tool/.modulerc
>> #%Module
>> if { ![info exists X_DEFAULT ] } {
>>   set X_DEFAULT X/v2
>>   module-version $X_DEFAULT default
>> }
>> if { ![info exists Y_DEFAULT ] } {
>>   set Y_DEFAULT Y/v1
>>   module-version $Y_DEFAULT default
>> }
>>
>> Then we have various projects that can take a while to complete, and each can have it's own set of defaults.
>>
>> We thought about doing that by having each project have it's own .modulerc with a set of defaults, and adding that to the MODULEPATH
>> Now ideally we would just define a module-version command in a .modulerc, and once that's set, all subsequent module-versions for that same tool will just be ignored. But that doesn't seem to happen, as it not only gives a warning, but seems to remove the default version completely.
>>
>> $ tree -a --charset=ascii /tmp/prj
>> /tmp/prj
>> |-- A
>> |   `-- .modulerc
>> `-- B
>>     `-- .modulerc
>>
>> $ cat /tmp/prj/A/.modulerc
>> #%Module
>> if { ![info exists X_DEFAULT ] } {
>>   set X_DEFAULT X/v1
>>   module-version $X_DEFAULT default
>> }
>>
>> $ echo $MODULEPATH
>> /tmp/prj/A:/tmp/tool
>>
>> $ module --version
>> Modules Release 5.0.0 (2021-09-12)
>>
>> $ module avail -t --color=never
>> /tmp/tool:
>> X/v1(default)
>> X/v2
>> X/v3
>> Y/v1(default)
>> Y/v2
>>
>> $ module load X
>> WARNING: Symbolic version 'X/default' already defined
>>
>> $ module list
>> Currently Loaded Modulefiles:
>>  1) X/v3
>>
>> I expected X/v1 to be loaded here....
>> Now it works if I load it like this (but it still complains about that WARNING: Symbolic version already defined)
>>
>> $ module unload X
>> $ module load X/default
>> WARNING: Symbolic version 'X/default' already defined
>> $ module list
>> Currently Loaded Modulefiles:
>>  1) X/v1
>>
>> Looking at the debug output there's a few things that jump out to me:
>>
>> It seems to remove that X_DEFAULT variable I set? And then it still seems to want to eval /tmp/tool/.modulerc, it calls module-version, which according to the tcl source code should not call setModuleResolution when it finds that duplicate, but that still seems to happen, and it defaults back to the latest version
>>
>> Evaluate modulerc: '/tmp/prj/A/.modulerc'
>> DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
>> DEBUG [#1:.modulerc] module-version: X/v1 default
>> DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
>> Evaluate modulerc: '/tmp/tool/.modulerc'
>> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
>> DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
>> DEBUG [#1:.modulerc] module-version: X/v2 default
>> WARNING: Symbolic version 'X/default' already defined
>> DEBUG setModuleResolution: X/default resolved to X/v3
>> DEBUG setModuleResolution: define auto symbolic version 'X/default' targeting X/v3
>>
>>
>> Here is the complete output of "module -D load X"
>>
>> DEBUG setState: cmdline set to '/opt/modules/5.0.0/libexec/modulecmd.tcl tcsh -D load X'
>> DEBUG setConf: tcl_ext_lib set to '/opt/modules/5.0.0/lib/libtclenvmodules.so'
>> DEBUG Load Tcl extension library (/opt/modules/5.0.0/lib/libtclenvmodules.so)
>> DEBUG setState: tcl_ext_lib_loaded set to '1'
>> DEBUG setConf: siteconfig set to '/opt/modules/5.0.0/etc/siteconfig.tcl'
>> DEBUG sourceSiteConfig: Source site configuration (/opt/modules/5.0.0/etc/siteconfig.tcl)
>> DEBUG setState: siteconfig_loaded set to '1'
>> DEBUG setConf: locked_configs set to ''
>> DEBUG setState: shell set to 'tcsh'
>> DEBUG setState: shelltype set to 'csh'
>> DEBUG setState: subcmd set to 'load'
>> DEBUG setState: subcmd_args set to 'X'
>> DEBUG setState: init_error_report set to '1'
>> DEBUG setConf: verbosity set to 'debug'
>> DEBUG setState: is_stderr_tty set to '1'
>> DEBUG setConf: term_background set to 'dark'
>> DEBUG setConf: colors set to 'hi=1:db=2:tr=2:se=2:er=91:wa=93:me=95:in=94:mp=1;94:di=94:al=96:va=93:sy=95:de=4:cm=92:aL=100:L=90;47:H=2:F=41:nF=43:S=46:sS=44'
>> DEBUG setConf: color set to '1'
>> DEBUG setConf: pager set to '/usr/bin/less -eFKRX'
>> DEBUG setState: paginate set to '1'
>> DEBUG setState: report_format set to 'regular'
>> DEBUG setState: reportfd set to 'file6'
>> DEBUG lappendState: modulefile appended with '{}'
>> DEBUG module: cmd='load', args='X'
>> DEBUG setConf: avail_indepth set to '1'
>> DEBUG setConf: search_match set to 'starts_with'
>> DEBUG parseModuleCommandArgs: (show_oneperline=0, show_mtime=0, show_filter=, search_filter=, search_match=starts_with, dump_state=0, addpath_pos=prepend, not_req=0, otherargs=X)
>> DEBUG setConf: advanced_version_spec set to '1'
>> DEBUG setConf: variant_shortcut set to ''
>> DEBUG setModuleVersSpec: Set module 'X' (escglob 'X'),  module name '.' (re '\.'), module root 'X', version cmp 'eq', version(s) '', variant(s) '' and module name version spec 'X' for argument 'X'
>> DEBUG lappendState: always_read_full_file appended with '1'
>> DEBUG lappendState: commandname appended with 'load'
>> DEBUG lappendState: try_modulefile appended with '0'
>> DEBUG runModulerc: running...
>> DEBUG setState: rc_running set to '1'
>> DEBUG unsetState: rc_running unset
>> DEBUG cmdModuleLoad: loading X (context=load, uasked=1)
>> DEBUG lappendState: mode appended with 'load'
>> DEBUG getPathToModule: finding 'X' in '' (report_issue=1, look_loaded=no, excdir='')
>> DEBUG setState: is_win set to '0'
>> DEBUG setState: path_separator set to ':'
>> DEBUG resolvStringWithEnv: '/tmp/prj/A' resolved to '/tmp/prj/A'
>> DEBUG setState: cwd set to '/tmp/prj'
>> DEBUG resolvStringWithEnv: '/tmp/tool' resolved to '/tmp/tool'
>> DEBUG setConf: icase set to 'search'
>> DEBUG setConf: extended_default set to '1'
>> DEBUG setConf: implicit_default set to '1'
>> DEBUG getModules: get 'X' in /tmp/prj/A (fetch_mtime=0, search=rc_defs_included resolve, filter=)
>> DEBUG findModules: finding 'X' in /tmp/prj/A (depthlvl=0, fetch_mtime=0)
>> DEBUG setConf: ignored_dirs set to 'CVS RCS SCCS .svn .git .SYNC .sos'
>> DEBUG findModules: found .modulerc
>> DEBUG findModules: create cache entry '/tmp/prj/A:X:0:0'
>> DEBUG lappendState: modulefile appended with '/tmp/prj/A/.modulerc'
>> DEBUG lappendState: modulename appended with '.modulerc'
>> DEBUG lappendState: specifiedname appended with '.modulerc'
>> DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with '{[#1:.modulerc] }'
>> DEBUG [#1:.modulerc] setState: inhibit_errreport set to '0'
>> Evaluate modulerc: '/tmp/prj/A/.modulerc'
>> DEBUG [#1:.modulerc] execute-modulerc: creating interp __modrc_1
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_version
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_interactive
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var ModuleToolVersion
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var ModuleTool
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var auto_path
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_pkgPath
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_patchLevel
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_platform
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_library
>> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc proc list auto_load clock unknown tclLog auto_qualify auto_execok auto_import auto_load_index
>> DEBUG [#1:.modulerc] resetInterpState: saving for __modrc_1 command list module-version tell socket subst open eof pwd glob list module-hide pid exec auto_load_index time unknown eval module-forbid lassign lrange module-virtual is-loaded fblocked lsearch auto_import gets case lappend proc break variable llength module-info auto_execok return linsert is-used error catch clock info split readModuleContent array if fconfigure concat join lreplace source fcopy global switch auto_qualify update close cd for auto_load file append lreverse format unload read package set binary namespace scan versioncmp apply module-tag trace seek while chan flush after vwait dict system continue uplevel foreach lset rename fileevent regexp module-alias lrepeat upvar reportInternalBug formatErrStackTrace encoding expr unset load regsub interp exit setModulesVersion puts incr lindex lsort tclLog uname string
>> DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
>> DEBUG [#1:.modulerc] module-version: X/v1 default
>> DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
>> DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
>> DEBUG setState: debug_msg_prefix set to ''
>> DEBUG setState: specifiedname set to ''
>> DEBUG setState: modulename set to ''
>> DEBUG setState: modulefile set to '{}'
>> DEBUG setState: hiding_threshold set to '0'
>> Get modules: {X/default} matching 'X' in '/tmp/prj/A'
>> DEBUG getModules: create cache entry '/tmp/prj/A:X:0:rc_defs_included resolve:'
>> DEBUG getModules: get 'X' in /tmp/tool (fetch_mtime=0, search=rc_defs_included resolve, filter=)
>> DEBUG findModules: finding 'X' in /tmp/tool (depthlvl=0, fetch_mtime=0)
>> DEBUG checkValidModule: /tmp/tool/X/v1
>> DEBUG readModuleContent: /tmp/tool/X/v1
>> DEBUG setConf: mcookie_version_check set to '1'
>> DEBUG checkValidModule: /tmp/tool/X/v3
>> DEBUG readModuleContent: /tmp/tool/X/v3
>> DEBUG checkValidModule: /tmp/tool/X/.common
>> DEBUG readModuleContent: /tmp/tool/X/.common
>> DEBUG checkValidModule: /tmp/tool/X/v2
>> DEBUG readModuleContent: /tmp/tool/X/v2
>> DEBUG findModules: found X/v3 .modulerc X/.common X/v1 X/v2
>> DEBUG findModules: create cache entry '/tmp/tool:X:0:0'
>> DEBUG lappendState: modulefile appended with '/tmp/tool/.modulerc'
>> DEBUG lappendState: modulename appended with '.modulerc'
>> DEBUG lappendState: specifiedname appended with '.modulerc'
>> DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with '{[#1:.modulerc] }'
>> Evaluate modulerc: '/tmp/tool/.modulerc'
>> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var errorMsg
>> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
>> DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
>> DEBUG [#1:.modulerc] module-version: X/v2 default
>> WARNING: Symbolic version 'X/default' already defined
>> DEBUG [#1:.modulerc] module-version: Y/v1 default
>> DEBUG [#1:.modulerc] setModuleResolution: Y/default resolved to Y/v1
>> DEBUG [#1:.modulerc] setModuleResolution: Y resolved to Y/v1
>> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to Y/v1
>> DEBUG [#1:.modulerc] setModuleResolution: module-version Y/default = Y/v1
>> DEBUG setState: debug_msg_prefix set to ''
>> DEBUG setState: specifiedname set to ''
>> DEBUG setState: modulename set to ''
>> DEBUG setState: modulefile set to '{}'
>> DEBUG setModuleResolution: X/default resolved to X/v3
>> DEBUG setModuleResolution: define auto symbolic version 'X/default' targeting X/v3
>> DEBUG setModuleResolution: X/latest resolved to X/v3
>> DEBUG setModuleResolution: define auto symbolic version 'X/latest' targeting X/v3
>> Get modules: {X X/default X/v3 X/latest X/v1 X/v2} matching 'X' in '/tmp/tool'
>> DEBUG getModules: create cache entry '/tmp/tool:X:0:rc_defs_included resolve:'
>> DEBUG getEqArrayKey: key 'X' in array 'mod_list' matches 'X'
>> DEBUG getEqArrayKey: key 'X/v3' in array 'mod_list' matches 'X/v3'
>> Select module: 'X/v3' (/tmp/tool/X/v3) matching 'X/v3'
>> DEBUG lappendState: evalid appended with 'load-X/v3-0'
>> DEBUG lappendState: msgrecordid appended with 'load-X/v3-0'
>> DEBUG setConf: unload_match_order set to 'returnlast'
>> DEBUG setState: lm_info_cached set to '1'
>> DEBUG setState: sub1_separator set to '&'
>> DEBUG setState: sub2_separator set to '|'
>> DEBUG cacheCurrentModules: 0 loaded
>> DEBUG getLoadedMatchingName: '' matches 'X/v3'
>> DEBUG pushSettings: settings saved (#1)
>> DEBUG setConf: auto_handling set to '1'
>> DEBUG getUnmetDependentLoadedModuleList: get dependent of upcoming loaded 'X/v3'
>> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
>> DEBUG getDependentLoadedModuleList: get loaded mod dependent of '' (strong=0, direct=0, nporeq=0, loading=0, sat_constraint=1, being_unload=0)
>> DEBUG getDependentLoadedModuleList: got ''
>> DEBUG getUnmetDependentLoadedModuleList: got ''
>> DEBUG cmdModuleLoad: depre mod list is ''
>> DEBUG lappendState: modulefile appended with '/tmp/tool/X/v3'
>> DEBUG lappendState: modulename appended with 'X/v3'
>> DEBUG lappendState: modulenamevr appended with 'X/v3'
>> DEBUG lappendState: specifiedname appended with 'X'
>> DEBUG [#1:load:X/v3] lappendState: debug_msg_prefix appended with '{[#1:load:X/v3] }'
>> DEBUG [#1:load:X/v3] setState: inhibit_interp set to '0'
>> DEBUG [#1:load:X/v3] execute-modulefile: creating interp __modfile_load_1
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_version
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_interactive
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var ModuleToolVersion
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var ModuleTool
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var auto_path
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_pkgPath
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_patchLevel
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_platform
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_library
>> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile proc list auto_load clock unknown tclLog auto_qualify auto_execok auto_import auto_load_index
>> DEBUG [#1:load:X/v3] resetInterpState: saving for __modfile_load_1 command list module-version tell socket subst conflict open eof pwd glob list module-hide pid exec auto_load_index time unknown eval module-forbid lassign lrange module-virtual module-user append-path reportWarning is-loaded fblocked lsearch auto_import gets report case lappend proc setenv break variable llength set-function module-info prereq auto_execok return linsert is-used error catch clock info split variant readModuleContent array if fconfigure concat join lreplace source fcopy global switch isVerbosityLevel auto_qualify update module-log unset-alias reportError close reportCmdTrace cd for auto_load file append lreverse source-sh getenv format unload read package set binary namespace scan versioncmp apply module-tag raiseErrorCount trace unsetenv seek while isWin chan flush after vwait is-saved dict system getvariant continue uplevel is-avail remove-path foreach lset rename chdir fileevent module-whatis regexp module-alias x-resource lrepeat upvar formatErrStackTrace reportInternalBug encoding expr unset prepend-path load regsub module-verbosity module-trace unset-function exit interp set-alias _puts puts incr lindex lsort tclLog module uname string
>> DEBUG [#1:load:X/v3] readModuleContent: /tmp/tool/X/v3
>> DEBUG [#1:load:X/v3] module-info: name
>> DEBUG [#1:load:X/v3] conflict: X
>> DEBUG [#1:load:X/v3] is-loaded: X
>> DEBUG [#1:load:X/v3] getLoadedMatchingName: '' matches 'X'
>> DEBUG [#1:load:X/v3] append-path: PATH /tool/X/v3
>> DEBUG [#1:load:X/v3] add-path: (PATH /tool/X/v3) pos=append
>> DEBUG [#1:load:X/v3] setModuleVersSpec: Set module 'X/v3' (escglob 'X/v3'),  module name 'X' (re ''), module root 'X', version cmp 'eq', version(s) '', variant(s) '' and module name version spec 'X/v3' for argument 'X/v3'
>> DEBUG [#1:load:X/v3] execute-modulefile: exiting /tmp/tool/X/v3
>> DEBUG setState: debug_msg_prefix set to ''
>> DEBUG setState: specifiedname set to ''
>> DEBUG setState: modulename set to ''
>> DEBUG setState: modulenamevr set to ''
>> DEBUG setState: modulefile set to '{}'
>> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
>> DEBUG setState: force set to '0'
>> DEBUG add-path: (LOADEDMODULES X/v3) pos=append
>> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0, ign_refcount=0, var=LOADEDMODULES, val=X/v3, nbval=1)
>> DEBUG set-env: LOADEDMODULES=X/v3
>> DEBUG interp-sync-env: set var='::env(LOADEDMODULES)', val='X/v3' on interp(s) __modfile_load_1 __modrc_1
>> DEBUG add-path: (--duplicates _LMFILES_ /tmp/tool/X/v3) pos=append
>> DEBUG parsePathCommandArgs: (delim=:, allow_dup=1, idx_val=0, ign_refcount=0, var=_LMFILES_, val=/tmp/tool/X/v3, nbval=1)
>> DEBUG set-env: _LMFILES_=/tmp/tool/X/v3
>> DEBUG interp-sync-env: set var='::env(_LMFILES_)', val='/tmp/tool/X/v3' on interp(s) __modfile_load_1 __modrc_1
>> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
>> DEBUG setModuleDependency: set requirements of 'X/v3' to ''
>> DEBUG setModuleDependency: set NPO requirements of 'X/v3' to ''
>> DEBUG add-path: (__MODULES_LMCONFLICT X/v3&X) pos=append
>> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0, ign_refcount=0, var=__MODULES_LMCONFLICT, val=X/v3&X, nbval=1)
>> DEBUG set-env: __MODULES_LMCONFLICT=X/v3&X
>> DEBUG interp-sync-env: set var='::env(__MODULES_LMCONFLICT)', val='X/v3&X' on interp(s) __modfile_load_1 __modrc_1
>> DEBUG add-path: (__MODULES_LMALTNAME X/v3&as|X/default&as|X/latest) pos=append
>> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0, ign_refcount=0, var=__MODULES_LMALTNAME, val=X/v3&as|X/default&as|X/latest, nbval=1)
>> DEBUG set-env: __MODULES_LMALTNAME=X/v3&as|X/default&as|X/latest
>> DEBUG interp-sync-env: set var='::env(__MODULES_LMALTNAME)', val='X/v3&as|X/default&as|X/latest' on interp(s) __modfile_load_1 __modrc_1
>> DEBUG popSettings: previously saved settings flushed (#1)
>> DEBUG setConf: term_width set to '0'
>> DEBUG setState: term_columns set to '138'
>> DEBUG unsetState: report_sep_next unset
>> DEBUG setState: report_sep_next set to '1'
>> Loading X/v3
>>   Evaluate modulefile: '/tmp/tool/X/v3' as 'X/v3'
>> DEBUG unsetState: report_sep_next unset
>> DEBUG setState: report_sep_next set to '1'
>> DEBUG setState: evalid set to ''
>> DEBUG setState: msgrecordid set to ''
>> DEBUG setState: mode set to ''
>> DEBUG setState: try_modulefile set to ''
>> DEBUG setState: commandname set to ''
>> DEBUG setState: always_read_full_file set to ''
>> DEBUG renderSettings: called.
>> DEBUG setState: autoinit set to '0'
>> DEBUG setState: error_count set to '0'
>> DEBUG setState: return_false set to '0'
>> DEBUG renderTrue: called.
>>
>>
>>
>> On Thu, Oct 7, 2021 at 12:10 PM Thomas M. Payerle <pa...@um...> wrote:
>>>
>>> I haven't use the hack recently, and only used it on older (3.x) versions of modules, but in my cases just using set and a Tcl var worked.  But also my situation was only dealing with a single modulerc file as well.
>>>
>>> I would suggest trying full-fledged env variables and see if that makes a difference.
>>>
>>> Otherwise, you might want to see if you can create and post a minimalist example of the issue --- I am only guessing that your issue is the same as the one I am referring to (I do not even recall the exact error message for the issue I am referring to, but it was similar to yours), but your issue might be something completely different.  It is hard to tell from just an error message.
>>>
>>> On Thu, Oct 7, 2021 at 11:53 AM Robbin Bonthond <ro...@bo...> wrote:
>>>>
>>>> hi,
>>>>
>>>> Thank you for the hint, do I need to use environment variables? it seems that a "set x y" isn't global between various .modulerc files.
>>>>
>>>> robbin
>>>>
>>>> On Wed, Oct 6, 2021 at 6:59 PM Thomas M. Payerle <pa...@um...> wrote:
>>>>>
>>>>> You might want to look at https://kb.hlrs.de/platforms/index.php/Module_command
>>>>> Toward the end, Global Default section.
>>>>>
>>>>> Basically, for each module version invocation, create an unique env variable and test for the env variable before doing the module version call.
>>>>>
>>>>> It's an ugly hack, but it gets the job done
>>>>>
>>>>>
>>>>> On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...> wrote:
>>>>>>
>>>>>> Hello,
>>>>>>
>>>>>> Is there a way to turn off the warning "WARNING: Symbolic version 'tool/default' already defined"?
>>>>>>
>>>>>> Robbin
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Modules-interest mailing list
>>>>>> Mod...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Tom Payerle
>>>>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>>>>> 5825 University Research Park               (301) 405-6135
>>>>> University of Maryland
>>>>> College Park, MD 20740-3831
>>>>> _______________________________________________
>>>>> Modules-interest mailing list
>>>>> Mod...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>>
>>>> _______________________________________________
>>>> Modules-interest mailing list
>>>> Mod...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>
>>>
>>>
>>> --
>>> Tom Payerle
>>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>>> 5825 University Research Park               (301) 405-6135
>>> University of Maryland
>>> College Park, MD 20740-3831
>>> _______________________________________________
>>> Modules-interest mailing list
>>> Mod...@li...
>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest

Re: [Modules] nested .modulerc

[Robbin B. <ro...@bo...>]

PS:

$ module config implicit_default 0
$ module avail -t --color=never
/tmp/tool:
X/v1(default)
X/v2
X/v3
Y/v1(default)
Y/v2
$ module load X
WARNING: Symbolic version 'X/default' already defined
ERROR: No default version defined for 'X'


On Thu, Oct 7, 2021 at 5:52 PM Robbin Bonthond <ro...@bo...> wrote:

> so we have various tool versions installed in a central place, and a
> .modulerc in the root there contains all defaults.
>
> something like
>
> $ tree -a --charset=ascii /tmp/tool
> /tmp/tool
> |-- .modulerc
> |-- X
> |   |-- .common
> |   |-- v1 -> .common
> |   |-- v2 -> .common
> |   `-- v3 -> .common
> `-- Y
>     |-- .common
>     |-- v1 -> .common
>     `-- v2 -> .common
>
> That /tmp/tool/.modulerc contains something like this:
>
> $ cat /tmp/tool/.modulerc
> #%Module
> if { ![info exists X_DEFAULT ] } {
>   set X_DEFAULT X/v2
>   module-version $X_DEFAULT default
> }
> if { ![info exists Y_DEFAULT ] } {
>   set Y_DEFAULT Y/v1
>   module-version $Y_DEFAULT default
> }
>
> Then we have various projects that can take a while to complete, and each
> can have it's own set of defaults.
>
> We thought about doing that by having each project have it's own .modulerc
> with a set of defaults, and adding that to the MODULEPATH
> Now ideally we would just define a module-version command in a .modulerc,
> and once that's set, all subsequent module-versions for that same tool will
> just be ignored. But that doesn't seem to happen, as it not only gives a
> warning, but seems to remove the default version completely.
>
> $ tree -a --charset=ascii /tmp/prj
> /tmp/prj
> |-- A
> |   `-- .modulerc
> `-- B
>     `-- .modulerc
>
> $ cat /tmp/prj/A/.modulerc
> #%Module
> if { ![info exists X_DEFAULT ] } {
>   set X_DEFAULT X/v1
>   module-version $X_DEFAULT default
> }
>
> $ echo $MODULEPATH
> /tmp/prj/A:/tmp/tool
>
> $ module --version
> Modules Release 5.0.0 (2021-09-12)
>
> $ module avail -t --color=never
> /tmp/tool:
> X/v1(default)
> X/v2
> X/v3
> Y/v1(default)
> Y/v2
>
> $ module load X
> WARNING: Symbolic version 'X/default' already defined
>
> $ module list
> Currently Loaded Modulefiles:
>  1) X/v3
>
> I expected X/v1 to be loaded here....
> Now it works if I load it like this (but it still complains about that
> WARNING: Symbolic version already defined)
>
> $ module unload X
> $ module load X/default
> WARNING: Symbolic version 'X/default' already defined
> $ module list
> Currently Loaded Modulefiles:
>  1) X/v1
>
> Looking at the debug output there's a few things that jump out to me:
>
> It seems to remove that X_DEFAULT variable I set? And then it still seems
> to want to eval /tmp/tool/.modulerc, it calls module-version, which
> according to the tcl source code should not call setModuleResolution when
> it finds that duplicate, but that still seems to happen, and it defaults
> back to the latest version
>
> Evaluate modulerc: '/tmp/prj/A/.modulerc'
> DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
> DEBUG [#1:.modulerc] module-version: X/v1 default
> DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
> Evaluate modulerc: '/tmp/tool/.modulerc'
> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
> DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
> DEBUG [#1:.modulerc] module-version: X/v2 default
> WARNING: Symbolic version 'X/default' already defined
>
> DEBUG setModuleResolution: X/default resolved to X/v3
> DEBUG setModuleResolution: define auto symbolic version 'X/default'
> targeting X/v3
>
>
> Here is the complete output of "module -D load X"
>
> DEBUG setState: cmdline set to '/opt/modules/5.0.0/libexec/modulecmd.tcl
> tcsh -D load X'
> DEBUG setConf: tcl_ext_lib set to
> '/opt/modules/5.0.0/lib/libtclenvmodules.so'
> DEBUG Load Tcl extension library
> (/opt/modules/5.0.0/lib/libtclenvmodules.so)
> DEBUG setState: tcl_ext_lib_loaded set to '1'
> DEBUG setConf: siteconfig set to '/opt/modules/5.0.0/etc/siteconfig.tcl'
> DEBUG sourceSiteConfig: Source site configuration
> (/opt/modules/5.0.0/etc/siteconfig.tcl)
> DEBUG setState: siteconfig_loaded set to '1'
> DEBUG setConf: locked_configs set to ''
> DEBUG setState: shell set to 'tcsh'
> DEBUG setState: shelltype set to 'csh'
> DEBUG setState: subcmd set to 'load'
> DEBUG setState: subcmd_args set to 'X'
> DEBUG setState: init_error_report set to '1'
> DEBUG setConf: verbosity set to 'debug'
> DEBUG setState: is_stderr_tty set to '1'
> DEBUG setConf: term_background set to 'dark'
> DEBUG setConf: colors set to
> 'hi=1:db=2:tr=2:se=2:er=91:wa=93:me=95:in=94:mp=1;94:di=94:al=96:va=93:sy=95:de=4:cm=92:aL=100:L=90;47:H=2:F=41:nF=43:S=46:sS=44'
> DEBUG setConf: color set to '1'
> DEBUG setConf: pager set to '/usr/bin/less -eFKRX'
> DEBUG setState: paginate set to '1'
> DEBUG setState: report_format set to 'regular'
> DEBUG setState: reportfd set to 'file6'
> DEBUG lappendState: modulefile appended with '{}'
> DEBUG module: cmd='load', args='X'
> DEBUG setConf: avail_indepth set to '1'
> DEBUG setConf: search_match set to 'starts_with'
> DEBUG parseModuleCommandArgs: (show_oneperline=0, show_mtime=0,
> show_filter=, search_filter=, search_match=starts_with, dump_state=0,
> addpath_pos=prepend, not_req=0, otherargs=X)
> DEBUG setConf: advanced_version_spec set to '1'
> DEBUG setConf: variant_shortcut set to ''
> DEBUG setModuleVersSpec: Set module 'X' (escglob 'X'),  module name '.'
> (re '\.'), module root 'X', version cmp 'eq', version(s) '', variant(s) ''
> and module name version spec 'X' for argument 'X'
> DEBUG lappendState: always_read_full_file appended with '1'
> DEBUG lappendState: commandname appended with 'load'
> DEBUG lappendState: try_modulefile appended with '0'
> DEBUG runModulerc: running...
> DEBUG setState: rc_running set to '1'
> DEBUG unsetState: rc_running unset
> DEBUG cmdModuleLoad: loading X (context=load, uasked=1)
> DEBUG lappendState: mode appended with 'load'
> DEBUG getPathToModule: finding 'X' in '' (report_issue=1, look_loaded=no,
> excdir='')
> DEBUG setState: is_win set to '0'
> DEBUG setState: path_separator set to ':'
> DEBUG resolvStringWithEnv: '/tmp/prj/A' resolved to '/tmp/prj/A'
> DEBUG setState: cwd set to '/tmp/prj'
> DEBUG resolvStringWithEnv: '/tmp/tool' resolved to '/tmp/tool'
> DEBUG setConf: icase set to 'search'
> DEBUG setConf: extended_default set to '1'
> DEBUG setConf: implicit_default set to '1'
> DEBUG getModules: get 'X' in /tmp/prj/A (fetch_mtime=0,
> search=rc_defs_included resolve, filter=)
> DEBUG findModules: finding 'X' in /tmp/prj/A (depthlvl=0, fetch_mtime=0)
> DEBUG setConf: ignored_dirs set to 'CVS RCS SCCS .svn .git .SYNC .sos'
> DEBUG findModules: found .modulerc
> DEBUG findModules: create cache entry '/tmp/prj/A:X:0:0'
> DEBUG lappendState: modulefile appended with '/tmp/prj/A/.modulerc'
> DEBUG lappendState: modulename appended with '.modulerc'
> DEBUG lappendState: specifiedname appended with '.modulerc'
> DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with
> '{[#1:.modulerc] }'
> DEBUG [#1:.modulerc] setState: inhibit_errreport set to '0'
> Evaluate modulerc: '/tmp/prj/A/.modulerc'
> DEBUG [#1:.modulerc] execute-modulerc: creating interp __modrc_1
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_version
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var
> tcl_interactive
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var
> ModuleToolVersion
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var ModuleTool
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var auto_path
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_pkgPath
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_patchLevel
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_platform
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_library
> DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc proc list
> auto_load clock unknown tclLog auto_qualify auto_execok auto_import
> auto_load_index
> DEBUG [#1:.modulerc] resetInterpState: saving for __modrc_1 command list
> module-version tell socket subst open eof pwd glob list module-hide pid
> exec auto_load_index time unknown eval module-forbid lassign lrange
> module-virtual is-loaded fblocked lsearch auto_import gets case lappend
> proc break variable llength module-info auto_execok return linsert is-used
> error catch clock info split readModuleContent array if fconfigure concat
> join lreplace source fcopy global switch auto_qualify update close cd for
> auto_load file append lreverse format unload read package set binary
> namespace scan versioncmp apply module-tag trace seek while chan flush
> after vwait dict system continue uplevel foreach lset rename fileevent
> regexp module-alias lrepeat upvar reportInternalBug formatErrStackTrace
> encoding expr unset load regsub interp exit setModulesVersion puts incr
> lindex lsort tclLog uname string
> DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
> DEBUG [#1:.modulerc] module-version: X/v1 default
> DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
> DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
> DEBUG setState: debug_msg_prefix set to ''
> DEBUG setState: specifiedname set to ''
> DEBUG setState: modulename set to ''
> DEBUG setState: modulefile set to '{}'
> DEBUG setState: hiding_threshold set to '0'
> Get modules: {X/default} matching 'X' in '/tmp/prj/A'
> DEBUG getModules: create cache entry '/tmp/prj/A:X:0:rc_defs_included
> resolve:'
> DEBUG getModules: get 'X' in /tmp/tool (fetch_mtime=0,
> search=rc_defs_included resolve, filter=)
> DEBUG findModules: finding 'X' in /tmp/tool (depthlvl=0, fetch_mtime=0)
> DEBUG checkValidModule: /tmp/tool/X/v1
> DEBUG readModuleContent: /tmp/tool/X/v1
> DEBUG setConf: mcookie_version_check set to '1'
> DEBUG checkValidModule: /tmp/tool/X/v3
> DEBUG readModuleContent: /tmp/tool/X/v3
> DEBUG checkValidModule: /tmp/tool/X/.common
> DEBUG readModuleContent: /tmp/tool/X/.common
> DEBUG checkValidModule: /tmp/tool/X/v2
> DEBUG readModuleContent: /tmp/tool/X/v2
> DEBUG findModules: found X/v3 .modulerc X/.common X/v1 X/v2
> DEBUG findModules: create cache entry '/tmp/tool:X:0:0'
> DEBUG lappendState: modulefile appended with '/tmp/tool/.modulerc'
> DEBUG lappendState: modulename appended with '.modulerc'
> DEBUG lappendState: specifiedname appended with '.modulerc'
> DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with
> '{[#1:.modulerc] }'
> Evaluate modulerc: '/tmp/tool/.modulerc'
> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var errorMsg
> DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
> DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
> DEBUG [#1:.modulerc] module-version: X/v2 default
> WARNING: Symbolic version 'X/default' already defined
> DEBUG [#1:.modulerc] module-version: Y/v1 default
> DEBUG [#1:.modulerc] setModuleResolution: Y/default resolved to Y/v1
> DEBUG [#1:.modulerc] setModuleResolution: Y resolved to Y/v1
> DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to Y/v1
> DEBUG [#1:.modulerc] setModuleResolution: module-version Y/default = Y/v1
> DEBUG setState: debug_msg_prefix set to ''
> DEBUG setState: specifiedname set to ''
> DEBUG setState: modulename set to ''
> DEBUG setState: modulefile set to '{}'
> DEBUG setModuleResolution: X/default resolved to X/v3
> DEBUG setModuleResolution: define auto symbolic version 'X/default'
> targeting X/v3
> DEBUG setModuleResolution: X/latest resolved to X/v3
> DEBUG setModuleResolution: define auto symbolic version 'X/latest'
> targeting X/v3
> Get modules: {X X/default X/v3 X/latest X/v1 X/v2} matching 'X' in
> '/tmp/tool'
> DEBUG getModules: create cache entry '/tmp/tool:X:0:rc_defs_included
> resolve:'
> DEBUG getEqArrayKey: key 'X' in array 'mod_list' matches 'X'
> DEBUG getEqArrayKey: key 'X/v3' in array 'mod_list' matches 'X/v3'
> Select module: 'X/v3' (/tmp/tool/X/v3) matching 'X/v3'
> DEBUG lappendState: evalid appended with 'load-X/v3-0'
> DEBUG lappendState: msgrecordid appended with 'load-X/v3-0'
> DEBUG setConf: unload_match_order set to 'returnlast'
> DEBUG setState: lm_info_cached set to '1'
> DEBUG setState: sub1_separator set to '&'
> DEBUG setState: sub2_separator set to '|'
> DEBUG cacheCurrentModules: 0 loaded
> DEBUG getLoadedMatchingName: '' matches 'X/v3'
> DEBUG pushSettings: settings saved (#1)
> DEBUG setConf: auto_handling set to '1'
> DEBUG getUnmetDependentLoadedModuleList: get dependent of upcoming loaded
> 'X/v3'
> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
> DEBUG getDependentLoadedModuleList: get loaded mod dependent of ''
> (strong=0, direct=0, nporeq=0, loading=0, sat_constraint=1, being_unload=0)
> DEBUG getDependentLoadedModuleList: got ''
> DEBUG getUnmetDependentLoadedModuleList: got ''
> DEBUG cmdModuleLoad: depre mod list is ''
> DEBUG lappendState: modulefile appended with '/tmp/tool/X/v3'
> DEBUG lappendState: modulename appended with 'X/v3'
> DEBUG lappendState: modulenamevr appended with 'X/v3'
> DEBUG lappendState: specifiedname appended with 'X'
> DEBUG [#1:load:X/v3] lappendState: debug_msg_prefix appended with
> '{[#1:load:X/v3] }'
> DEBUG [#1:load:X/v3] setState: inhibit_interp set to '0'
> DEBUG [#1:load:X/v3] execute-modulefile: creating interp __modfile_load_1
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_version
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
> tcl_interactive
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
> ModuleToolVersion
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var ModuleTool
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var auto_path
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_pkgPath
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
> tcl_patchLevel
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_platform
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_library
> DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile proc list
> auto_load clock unknown tclLog auto_qualify auto_execok auto_import
> auto_load_index
> DEBUG [#1:load:X/v3] resetInterpState: saving for __modfile_load_1 command
> list module-version tell socket subst conflict open eof pwd glob list
> module-hide pid exec auto_load_index time unknown eval module-forbid
> lassign lrange module-virtual module-user append-path reportWarning
> is-loaded fblocked lsearch auto_import gets report case lappend proc setenv
> break variable llength set-function module-info prereq auto_execok return
> linsert is-used error catch clock info split variant readModuleContent
> array if fconfigure concat join lreplace source fcopy global switch
> isVerbosityLevel auto_qualify update module-log unset-alias reportError
> close reportCmdTrace cd for auto_load file append lreverse source-sh getenv
> format unload read package set binary namespace scan versioncmp apply
> module-tag raiseErrorCount trace unsetenv seek while isWin chan flush after
> vwait is-saved dict system getvariant continue uplevel is-avail remove-path
> foreach lset rename chdir fileevent module-whatis regexp module-alias
> x-resource lrepeat upvar formatErrStackTrace reportInternalBug encoding
> expr unset prepend-path load regsub module-verbosity module-trace
> unset-function exit interp set-alias _puts puts incr lindex lsort tclLog
> module uname string
> DEBUG [#1:load:X/v3] readModuleContent: /tmp/tool/X/v3
> DEBUG [#1:load:X/v3] module-info: name
> DEBUG [#1:load:X/v3] conflict: X
> DEBUG [#1:load:X/v3] is-loaded: X
> DEBUG [#1:load:X/v3] getLoadedMatchingName: '' matches 'X'
> DEBUG [#1:load:X/v3] append-path: PATH /tool/X/v3
> DEBUG [#1:load:X/v3] add-path: (PATH /tool/X/v3) pos=append
> DEBUG [#1:load:X/v3] setModuleVersSpec: Set module 'X/v3' (escglob
> 'X/v3'),  module name 'X' (re ''), module root 'X', version cmp 'eq',
> version(s) '', variant(s) '' and module name version spec 'X/v3' for
> argument 'X/v3'
> DEBUG [#1:load:X/v3] execute-modulefile: exiting /tmp/tool/X/v3
> DEBUG setState: debug_msg_prefix set to ''
> DEBUG setState: specifiedname set to ''
> DEBUG setState: modulename set to ''
> DEBUG setState: modulenamevr set to ''
> DEBUG setState: modulefile set to '{}'
> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
> DEBUG setState: force set to '0'
> DEBUG add-path: (LOADEDMODULES X/v3) pos=append
> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
> ign_refcount=0, var=LOADEDMODULES, val=X/v3, nbval=1)
> DEBUG set-env: LOADEDMODULES=X/v3
> DEBUG interp-sync-env: set var='::env(LOADEDMODULES)', val='X/v3' on
> interp(s) __modfile_load_1 __modrc_1
> DEBUG add-path: (--duplicates _LMFILES_ /tmp/tool/X/v3) pos=append
> DEBUG parsePathCommandArgs: (delim=:, allow_dup=1, idx_val=0,
> ign_refcount=0, var=_LMFILES_, val=/tmp/tool/X/v3, nbval=1)
> DEBUG set-env: _LMFILES_=/tmp/tool/X/v3
> DEBUG interp-sync-env: set var='::env(_LMFILES_)', val='/tmp/tool/X/v3' on
> interp(s) __modfile_load_1 __modrc_1
> DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
> DEBUG setModuleDependency: set requirements of 'X/v3' to ''
> DEBUG setModuleDependency: set NPO requirements of 'X/v3' to ''
> DEBUG add-path: (__MODULES_LMCONFLICT X/v3&X) pos=append
> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
> ign_refcount=0, var=__MODULES_LMCONFLICT, val=X/v3&X, nbval=1)
> DEBUG set-env: __MODULES_LMCONFLICT=X/v3&X
> DEBUG interp-sync-env: set var='::env(__MODULES_LMCONFLICT)', val='X/v3&X'
> on interp(s) __modfile_load_1 __modrc_1
> DEBUG add-path: (__MODULES_LMALTNAME X/v3&as|X/default&as|X/latest)
> pos=append
> DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
> ign_refcount=0, var=__MODULES_LMALTNAME, val=X/v3&as|X/default&as|X/latest,
> nbval=1)
> DEBUG set-env: __MODULES_LMALTNAME=X/v3&as|X/default&as|X/latest
> DEBUG interp-sync-env: set var='::env(__MODULES_LMALTNAME)',
> val='X/v3&as|X/default&as|X/latest' on interp(s) __modfile_load_1 __modrc_1
> DEBUG popSettings: previously saved settings flushed (#1)
> DEBUG setConf: term_width set to '0'
> DEBUG setState: term_columns set to '138'
> DEBUG unsetState: report_sep_next unset
> DEBUG setState: report_sep_next set to '1'
> Loading X/v3
>   Evaluate modulefile: '/tmp/tool/X/v3' as 'X/v3'
> DEBUG unsetState: report_sep_next unset
> DEBUG setState: report_sep_next set to '1'
> DEBUG setState: evalid set to ''
> DEBUG setState: msgrecordid set to ''
> DEBUG setState: mode set to ''
> DEBUG setState: try_modulefile set to ''
> DEBUG setState: commandname set to ''
> DEBUG setState: always_read_full_file set to ''
> DEBUG renderSettings: called.
> DEBUG setState: autoinit set to '0'
> DEBUG setState: error_count set to '0'
> DEBUG setState: return_false set to '0'
> DEBUG renderTrue: called.
>
>
>
> On Thu, Oct 7, 2021 at 12:10 PM Thomas M. Payerle <pa...@um...> wrote:
>
>> I haven't use the hack recently, and only used it on older (3.x) versions
>> of modules, but in my cases just using set and a Tcl var worked.  But also
>> my situation was only dealing with a single modulerc file as well.
>>
>> I would suggest trying full-fledged env variables and see if that makes a
>> difference.
>>
>> Otherwise, you might want to see if you can create and post a minimalist
>> example of the issue --- I am only guessing that your issue is the same as
>> the one I am referring to (I do not even recall the exact error message for
>> the issue I am referring to, but it was similar to yours), but your issue
>> might be something completely different.  It is hard to tell from just an
>> error message.
>>
>> On Thu, Oct 7, 2021 at 11:53 AM Robbin Bonthond <ro...@bo...>
>> wrote:
>>
>>> hi,
>>>
>>> Thank you for the hint, do I need to use environment variables? it seems
>>> that a "set x y" isn't global between various .modulerc files.
>>>
>>> robbin
>>>
>>> On Wed, Oct 6, 2021 at 6:59 PM Thomas M. Payerle <pa...@um...>
>>> wrote:
>>>
>>>> You might want to look at
>>>> https://kb.hlrs.de/platforms/index.php/Module_command
>>>> Toward the end, Global Default section.
>>>>
>>>> Basically, for each module version invocation, create an unique env
>>>> variable and test for the env variable before doing the module version call.
>>>>
>>>> It's an ugly hack, but it gets the job done
>>>>
>>>>
>>>> On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...>
>>>> wrote:
>>>>
>>>>> Hello,
>>>>>
>>>>> Is there a way to turn off the warning "WARNING: Symbolic version
>>>>> 'tool/default' already defined"?
>>>>>
>>>>> Robbin
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Modules-interest mailing list
>>>>> Mod...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>>>
>>>>
>>>>
>>>> --
>>>> Tom Payerle
>>>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>>>> 5825 University Research Park               (301) 405-6135
>>>> University of Maryland
>>>> College Park, MD 20740-3831
>>>> _______________________________________________
>>>> Modules-interest mailing list
>>>> Mod...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>>
>>> _______________________________________________
>>> Modules-interest mailing list
>>> Mod...@li...
>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>
>>
>>
>> --
>> Tom Payerle
>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>> 5825 University Research Park               (301) 405-6135
>> University of Maryland
>> College Park, MD 20740-3831
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>
>

Re: [Modules] nested .modulerc

[Robbin B. <ro...@bo...>]

so we have various tool versions installed in a central place, and a
.modulerc in the root there contains all defaults.

something like

$ tree -a --charset=ascii /tmp/tool
/tmp/tool
|-- .modulerc
|-- X
|   |-- .common
|   |-- v1 -> .common
|   |-- v2 -> .common
|   `-- v3 -> .common
`-- Y
    |-- .common
    |-- v1 -> .common
    `-- v2 -> .common

That /tmp/tool/.modulerc contains something like this:

$ cat /tmp/tool/.modulerc
#%Module
if { ![info exists X_DEFAULT ] } {
  set X_DEFAULT X/v2
  module-version $X_DEFAULT default
}
if { ![info exists Y_DEFAULT ] } {
  set Y_DEFAULT Y/v1
  module-version $Y_DEFAULT default
}

Then we have various projects that can take a while to complete, and each
can have it's own set of defaults.

We thought about doing that by having each project have it's own .modulerc
with a set of defaults, and adding that to the MODULEPATH
Now ideally we would just define a module-version command in a .modulerc,
and once that's set, all subsequent module-versions for that same tool will
just be ignored. But that doesn't seem to happen, as it not only gives a
warning, but seems to remove the default version completely.

$ tree -a --charset=ascii /tmp/prj
/tmp/prj
|-- A
|   `-- .modulerc
`-- B
    `-- .modulerc

$ cat /tmp/prj/A/.modulerc
#%Module
if { ![info exists X_DEFAULT ] } {
  set X_DEFAULT X/v1
  module-version $X_DEFAULT default
}

$ echo $MODULEPATH
/tmp/prj/A:/tmp/tool

$ module --version
Modules Release 5.0.0 (2021-09-12)

$ module avail -t --color=never
/tmp/tool:
X/v1(default)
X/v2
X/v3
Y/v1(default)
Y/v2

$ module load X
WARNING: Symbolic version 'X/default' already defined

$ module list
Currently Loaded Modulefiles:
 1) X/v3

I expected X/v1 to be loaded here....
Now it works if I load it like this (but it still complains about that
WARNING: Symbolic version already defined)

$ module unload X
$ module load X/default
WARNING: Symbolic version 'X/default' already defined
$ module list
Currently Loaded Modulefiles:
 1) X/v1

Looking at the debug output there's a few things that jump out to me:

It seems to remove that X_DEFAULT variable I set? And then it still seems
to want to eval /tmp/tool/.modulerc, it calls module-version, which
according to the tcl source code should not call setModuleResolution when
it finds that duplicate, but that still seems to happen, and it defaults
back to the latest version

Evaluate modulerc: '/tmp/prj/A/.modulerc'
DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
DEBUG [#1:.modulerc] module-version: X/v1 default
DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
Evaluate modulerc: '/tmp/tool/.modulerc'
DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
DEBUG [#1:.modulerc] module-version: X/v2 default
WARNING: Symbolic version 'X/default' already defined

DEBUG setModuleResolution: X/default resolved to X/v3
DEBUG setModuleResolution: define auto symbolic version 'X/default'
targeting X/v3


Here is the complete output of "module -D load X"

DEBUG setState: cmdline set to '/opt/modules/5.0.0/libexec/modulecmd.tcl
tcsh -D load X'
DEBUG setConf: tcl_ext_lib set to
'/opt/modules/5.0.0/lib/libtclenvmodules.so'
DEBUG Load Tcl extension library
(/opt/modules/5.0.0/lib/libtclenvmodules.so)
DEBUG setState: tcl_ext_lib_loaded set to '1'
DEBUG setConf: siteconfig set to '/opt/modules/5.0.0/etc/siteconfig.tcl'
DEBUG sourceSiteConfig: Source site configuration
(/opt/modules/5.0.0/etc/siteconfig.tcl)
DEBUG setState: siteconfig_loaded set to '1'
DEBUG setConf: locked_configs set to ''
DEBUG setState: shell set to 'tcsh'
DEBUG setState: shelltype set to 'csh'
DEBUG setState: subcmd set to 'load'
DEBUG setState: subcmd_args set to 'X'
DEBUG setState: init_error_report set to '1'
DEBUG setConf: verbosity set to 'debug'
DEBUG setState: is_stderr_tty set to '1'
DEBUG setConf: term_background set to 'dark'
DEBUG setConf: colors set to
'hi=1:db=2:tr=2:se=2:er=91:wa=93:me=95:in=94:mp=1;94:di=94:al=96:va=93:sy=95:de=4:cm=92:aL=100:L=90;47:H=2:F=41:nF=43:S=46:sS=44'
DEBUG setConf: color set to '1'
DEBUG setConf: pager set to '/usr/bin/less -eFKRX'
DEBUG setState: paginate set to '1'
DEBUG setState: report_format set to 'regular'
DEBUG setState: reportfd set to 'file6'
DEBUG lappendState: modulefile appended with '{}'
DEBUG module: cmd='load', args='X'
DEBUG setConf: avail_indepth set to '1'
DEBUG setConf: search_match set to 'starts_with'
DEBUG parseModuleCommandArgs: (show_oneperline=0, show_mtime=0,
show_filter=, search_filter=, search_match=starts_with, dump_state=0,
addpath_pos=prepend, not_req=0, otherargs=X)
DEBUG setConf: advanced_version_spec set to '1'
DEBUG setConf: variant_shortcut set to ''
DEBUG setModuleVersSpec: Set module 'X' (escglob 'X'),  module name '.' (re
'\.'), module root 'X', version cmp 'eq', version(s) '', variant(s) '' and
module name version spec 'X' for argument 'X'
DEBUG lappendState: always_read_full_file appended with '1'
DEBUG lappendState: commandname appended with 'load'
DEBUG lappendState: try_modulefile appended with '0'
DEBUG runModulerc: running...
DEBUG setState: rc_running set to '1'
DEBUG unsetState: rc_running unset
DEBUG cmdModuleLoad: loading X (context=load, uasked=1)
DEBUG lappendState: mode appended with 'load'
DEBUG getPathToModule: finding 'X' in '' (report_issue=1, look_loaded=no,
excdir='')
DEBUG setState: is_win set to '0'
DEBUG setState: path_separator set to ':'
DEBUG resolvStringWithEnv: '/tmp/prj/A' resolved to '/tmp/prj/A'
DEBUG setState: cwd set to '/tmp/prj'
DEBUG resolvStringWithEnv: '/tmp/tool' resolved to '/tmp/tool'
DEBUG setConf: icase set to 'search'
DEBUG setConf: extended_default set to '1'
DEBUG setConf: implicit_default set to '1'
DEBUG getModules: get 'X' in /tmp/prj/A (fetch_mtime=0,
search=rc_defs_included resolve, filter=)
DEBUG findModules: finding 'X' in /tmp/prj/A (depthlvl=0, fetch_mtime=0)
DEBUG setConf: ignored_dirs set to 'CVS RCS SCCS .svn .git .SYNC .sos'
DEBUG findModules: found .modulerc
DEBUG findModules: create cache entry '/tmp/prj/A:X:0:0'
DEBUG lappendState: modulefile appended with '/tmp/prj/A/.modulerc'
DEBUG lappendState: modulename appended with '.modulerc'
DEBUG lappendState: specifiedname appended with '.modulerc'
DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with
'{[#1:.modulerc] }'
DEBUG [#1:.modulerc] setState: inhibit_errreport set to '0'
Evaluate modulerc: '/tmp/prj/A/.modulerc'
DEBUG [#1:.modulerc] execute-modulerc: creating interp __modrc_1
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_version
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_interactive
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var
ModuleToolVersion
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var ModuleTool
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var auto_path
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_pkgPath
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_patchLevel
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_platform
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc var tcl_library
DEBUG [#1:.modulerc] dumpInterpState: saving for __modrc proc list
auto_load clock unknown tclLog auto_qualify auto_execok auto_import
auto_load_index
DEBUG [#1:.modulerc] resetInterpState: saving for __modrc_1 command list
module-version tell socket subst open eof pwd glob list module-hide pid
exec auto_load_index time unknown eval module-forbid lassign lrange
module-virtual is-loaded fblocked lsearch auto_import gets case lappend
proc break variable llength module-info auto_execok return linsert is-used
error catch clock info split readModuleContent array if fconfigure concat
join lreplace source fcopy global switch auto_qualify update close cd for
auto_load file append lreverse format unload read package set binary
namespace scan versioncmp apply module-tag trace seek while chan flush
after vwait dict system continue uplevel foreach lset rename fileevent
regexp module-alias lrepeat upvar reportInternalBug formatErrStackTrace
encoding expr unset load regsub interp exit setModulesVersion puts incr
lindex lsort tclLog uname string
DEBUG [#1:.modulerc] readModuleContent: /tmp/prj/A/.modulerc
DEBUG [#1:.modulerc] module-version: X/v1 default
DEBUG [#1:.modulerc] setModuleResolution: X/default resolved to X/v1
DEBUG [#1:.modulerc] setModuleResolution: X resolved to X/v1
DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to X/v1
DEBUG [#1:.modulerc] setModuleResolution: module-version X/default = X/v1
DEBUG setState: debug_msg_prefix set to ''
DEBUG setState: specifiedname set to ''
DEBUG setState: modulename set to ''
DEBUG setState: modulefile set to '{}'
DEBUG setState: hiding_threshold set to '0'
Get modules: {X/default} matching 'X' in '/tmp/prj/A'
DEBUG getModules: create cache entry '/tmp/prj/A:X:0:rc_defs_included
resolve:'
DEBUG getModules: get 'X' in /tmp/tool (fetch_mtime=0,
search=rc_defs_included resolve, filter=)
DEBUG findModules: finding 'X' in /tmp/tool (depthlvl=0, fetch_mtime=0)
DEBUG checkValidModule: /tmp/tool/X/v1
DEBUG readModuleContent: /tmp/tool/X/v1
DEBUG setConf: mcookie_version_check set to '1'
DEBUG checkValidModule: /tmp/tool/X/v3
DEBUG readModuleContent: /tmp/tool/X/v3
DEBUG checkValidModule: /tmp/tool/X/.common
DEBUG readModuleContent: /tmp/tool/X/.common
DEBUG checkValidModule: /tmp/tool/X/v2
DEBUG readModuleContent: /tmp/tool/X/v2
DEBUG findModules: found X/v3 .modulerc X/.common X/v1 X/v2
DEBUG findModules: create cache entry '/tmp/tool:X:0:0'
DEBUG lappendState: modulefile appended with '/tmp/tool/.modulerc'
DEBUG lappendState: modulename appended with '.modulerc'
DEBUG lappendState: specifiedname appended with '.modulerc'
DEBUG [#1:.modulerc] lappendState: debug_msg_prefix appended with
'{[#1:.modulerc] }'
Evaluate modulerc: '/tmp/tool/.modulerc'
DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var errorMsg
DEBUG [#1:.modulerc] resetInterpState: removing on __modrc_1 var X_DEFAULT
DEBUG [#1:.modulerc] readModuleContent: /tmp/tool/.modulerc
DEBUG [#1:.modulerc] module-version: X/v2 default
WARNING: Symbolic version 'X/default' already defined
DEBUG [#1:.modulerc] module-version: Y/v1 default
DEBUG [#1:.modulerc] setModuleResolution: Y/default resolved to Y/v1
DEBUG [#1:.modulerc] setModuleResolution: Y resolved to Y/v1
DEBUG [#1:.modulerc] setModuleResolution: set symbols 'default' to Y/v1
DEBUG [#1:.modulerc] setModuleResolution: module-version Y/default = Y/v1
DEBUG setState: debug_msg_prefix set to ''
DEBUG setState: specifiedname set to ''
DEBUG setState: modulename set to ''
DEBUG setState: modulefile set to '{}'
DEBUG setModuleResolution: X/default resolved to X/v3
DEBUG setModuleResolution: define auto symbolic version 'X/default'
targeting X/v3
DEBUG setModuleResolution: X/latest resolved to X/v3
DEBUG setModuleResolution: define auto symbolic version 'X/latest'
targeting X/v3
Get modules: {X X/default X/v3 X/latest X/v1 X/v2} matching 'X' in
'/tmp/tool'
DEBUG getModules: create cache entry '/tmp/tool:X:0:rc_defs_included
resolve:'
DEBUG getEqArrayKey: key 'X' in array 'mod_list' matches 'X'
DEBUG getEqArrayKey: key 'X/v3' in array 'mod_list' matches 'X/v3'
Select module: 'X/v3' (/tmp/tool/X/v3) matching 'X/v3'
DEBUG lappendState: evalid appended with 'load-X/v3-0'
DEBUG lappendState: msgrecordid appended with 'load-X/v3-0'
DEBUG setConf: unload_match_order set to 'returnlast'
DEBUG setState: lm_info_cached set to '1'
DEBUG setState: sub1_separator set to '&'
DEBUG setState: sub2_separator set to '|'
DEBUG cacheCurrentModules: 0 loaded
DEBUG getLoadedMatchingName: '' matches 'X/v3'
DEBUG pushSettings: settings saved (#1)
DEBUG setConf: auto_handling set to '1'
DEBUG getUnmetDependentLoadedModuleList: get dependent of upcoming loaded
'X/v3'
DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
DEBUG getDependentLoadedModuleList: get loaded mod dependent of ''
(strong=0, direct=0, nporeq=0, loading=0, sat_constraint=1, being_unload=0)
DEBUG getDependentLoadedModuleList: got ''
DEBUG getUnmetDependentLoadedModuleList: got ''
DEBUG cmdModuleLoad: depre mod list is ''
DEBUG lappendState: modulefile appended with '/tmp/tool/X/v3'
DEBUG lappendState: modulename appended with 'X/v3'
DEBUG lappendState: modulenamevr appended with 'X/v3'
DEBUG lappendState: specifiedname appended with 'X'
DEBUG [#1:load:X/v3] lappendState: debug_msg_prefix appended with
'{[#1:load:X/v3] }'
DEBUG [#1:load:X/v3] setState: inhibit_interp set to '0'
DEBUG [#1:load:X/v3] execute-modulefile: creating interp __modfile_load_1
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_version
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
tcl_interactive
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
ModuleToolVersion
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var ModuleTool
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var auto_path
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_pkgPath
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var
tcl_patchLevel
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_platform
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile var tcl_library
DEBUG [#1:load:X/v3] dumpInterpState: saving for __modfile proc list
auto_load clock unknown tclLog auto_qualify auto_execok auto_import
auto_load_index
DEBUG [#1:load:X/v3] resetInterpState: saving for __modfile_load_1 command
list module-version tell socket subst conflict open eof pwd glob list
module-hide pid exec auto_load_index time unknown eval module-forbid
lassign lrange module-virtual module-user append-path reportWarning
is-loaded fblocked lsearch auto_import gets report case lappend proc setenv
break variable llength set-function module-info prereq auto_execok return
linsert is-used error catch clock info split variant readModuleContent
array if fconfigure concat join lreplace source fcopy global switch
isVerbosityLevel auto_qualify update module-log unset-alias reportError
close reportCmdTrace cd for auto_load file append lreverse source-sh getenv
format unload read package set binary namespace scan versioncmp apply
module-tag raiseErrorCount trace unsetenv seek while isWin chan flush after
vwait is-saved dict system getvariant continue uplevel is-avail remove-path
foreach lset rename chdir fileevent module-whatis regexp module-alias
x-resource lrepeat upvar formatErrStackTrace reportInternalBug encoding
expr unset prepend-path load regsub module-verbosity module-trace
unset-function exit interp set-alias _puts puts incr lindex lsort tclLog
module uname string
DEBUG [#1:load:X/v3] readModuleContent: /tmp/tool/X/v3
DEBUG [#1:load:X/v3] module-info: name
DEBUG [#1:load:X/v3] conflict: X
DEBUG [#1:load:X/v3] is-loaded: X
DEBUG [#1:load:X/v3] getLoadedMatchingName: '' matches 'X'
DEBUG [#1:load:X/v3] append-path: PATH /tool/X/v3
DEBUG [#1:load:X/v3] add-path: (PATH /tool/X/v3) pos=append
DEBUG [#1:load:X/v3] setModuleVersSpec: Set module 'X/v3' (escglob
'X/v3'),  module name 'X' (re ''), module root 'X', version cmp 'eq',
version(s) '', variant(s) '' and module name version spec 'X/v3' for
argument 'X/v3'
DEBUG [#1:load:X/v3] execute-modulefile: exiting /tmp/tool/X/v3
DEBUG setState: debug_msg_prefix set to ''
DEBUG setState: specifiedname set to ''
DEBUG setState: modulename set to ''
DEBUG setState: modulenamevr set to ''
DEBUG setState: modulefile set to '{}'
DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
DEBUG setState: force set to '0'
DEBUG add-path: (LOADEDMODULES X/v3) pos=append
DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
ign_refcount=0, var=LOADEDMODULES, val=X/v3, nbval=1)
DEBUG set-env: LOADEDMODULES=X/v3
DEBUG interp-sync-env: set var='::env(LOADEDMODULES)', val='X/v3' on
interp(s) __modfile_load_1 __modrc_1
DEBUG add-path: (--duplicates _LMFILES_ /tmp/tool/X/v3) pos=append
DEBUG parsePathCommandArgs: (delim=:, allow_dup=1, idx_val=0,
ign_refcount=0, var=_LMFILES_, val=/tmp/tool/X/v3, nbval=1)
DEBUG set-env: _LMFILES_=/tmp/tool/X/v3
DEBUG interp-sync-env: set var='::env(_LMFILES_)', val='/tmp/tool/X/v3' on
interp(s) __modfile_load_1 __modrc_1
DEBUG doesModuleConflict: 'X/v3' conflicts with '' (declared as '')
DEBUG setModuleDependency: set requirements of 'X/v3' to ''
DEBUG setModuleDependency: set NPO requirements of 'X/v3' to ''
DEBUG add-path: (__MODULES_LMCONFLICT X/v3&X) pos=append
DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
ign_refcount=0, var=__MODULES_LMCONFLICT, val=X/v3&X, nbval=1)
DEBUG set-env: __MODULES_LMCONFLICT=X/v3&X
DEBUG interp-sync-env: set var='::env(__MODULES_LMCONFLICT)', val='X/v3&X'
on interp(s) __modfile_load_1 __modrc_1
DEBUG add-path: (__MODULES_LMALTNAME X/v3&as|X/default&as|X/latest)
pos=append
DEBUG parsePathCommandArgs: (delim=:, allow_dup=0, idx_val=0,
ign_refcount=0, var=__MODULES_LMALTNAME, val=X/v3&as|X/default&as|X/latest,
nbval=1)
DEBUG set-env: __MODULES_LMALTNAME=X/v3&as|X/default&as|X/latest
DEBUG interp-sync-env: set var='::env(__MODULES_LMALTNAME)',
val='X/v3&as|X/default&as|X/latest' on interp(s) __modfile_load_1 __modrc_1
DEBUG popSettings: previously saved settings flushed (#1)
DEBUG setConf: term_width set to '0'
DEBUG setState: term_columns set to '138'
DEBUG unsetState: report_sep_next unset
DEBUG setState: report_sep_next set to '1'
Loading X/v3
  Evaluate modulefile: '/tmp/tool/X/v3' as 'X/v3'
DEBUG unsetState: report_sep_next unset
DEBUG setState: report_sep_next set to '1'
DEBUG setState: evalid set to ''
DEBUG setState: msgrecordid set to ''
DEBUG setState: mode set to ''
DEBUG setState: try_modulefile set to ''
DEBUG setState: commandname set to ''
DEBUG setState: always_read_full_file set to ''
DEBUG renderSettings: called.
DEBUG setState: autoinit set to '0'
DEBUG setState: error_count set to '0'
DEBUG setState: return_false set to '0'
DEBUG renderTrue: called.



On Thu, Oct 7, 2021 at 12:10 PM Thomas M. Payerle <pa...@um...> wrote:

> I haven't use the hack recently, and only used it on older (3.x) versions
> of modules, but in my cases just using set and a Tcl var worked.  But also
> my situation was only dealing with a single modulerc file as well.
>
> I would suggest trying full-fledged env variables and see if that makes a
> difference.
>
> Otherwise, you might want to see if you can create and post a minimalist
> example of the issue --- I am only guessing that your issue is the same as
> the one I am referring to (I do not even recall the exact error message for
> the issue I am referring to, but it was similar to yours), but your issue
> might be something completely different.  It is hard to tell from just an
> error message.
>
> On Thu, Oct 7, 2021 at 11:53 AM Robbin Bonthond <ro...@bo...>
> wrote:
>
>> hi,
>>
>> Thank you for the hint, do I need to use environment variables? it seems
>> that a "set x y" isn't global between various .modulerc files.
>>
>> robbin
>>
>> On Wed, Oct 6, 2021 at 6:59 PM Thomas M. Payerle <pa...@um...> wrote:
>>
>>> You might want to look at
>>> https://kb.hlrs.de/platforms/index.php/Module_command
>>> Toward the end, Global Default section.
>>>
>>> Basically, for each module version invocation, create an unique env
>>> variable and test for the env variable before doing the module version call.
>>>
>>> It's an ugly hack, but it gets the job done
>>>
>>>
>>> On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...>
>>> wrote:
>>>
>>>> Hello,
>>>>
>>>> Is there a way to turn off the warning "WARNING: Symbolic version
>>>> 'tool/default' already defined"?
>>>>
>>>> Robbin
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Modules-interest mailing list
>>>> Mod...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>>
>>>
>>>
>>> --
>>> Tom Payerle
>>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>>> 5825 University Research Park               (301) 405-6135
>>> University of Maryland
>>> College Park, MD 20740-3831
>>> _______________________________________________
>>> Modules-interest mailing list
>>> Mod...@li...
>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>
>
>
> --
> Tom Payerle
> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
> 5825 University Research Park               (301) 405-6135
> University of Maryland
> College Park, MD 20740-3831
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
>

Re: [Modules] module source - changed as of 5.0.0 ?

[Xavier D. <xav...@gm...>]

Hi Paul,

as you found, "source" is the way to go within a modulefile instead of
"module source". It avoids extra modulecmd steps as code is directly
evaluated rather evaluated through another modulefile evaluation.

Over certain evaluation modes it is also more efficient, for instance
sourced content will be reported on a "module display".

I need to clarify the documentation on the module sub-commands that
are allowed when used within a modulefile.

Regards,
Xavier

Le jeu. 7 oct. 2021 à 19:43, Paul Markfort <pau...@gm...> a écrit :
>
> Finally found it in the changes.
>
> https://modules.readthedocs.io/en/latest/NEWS.html
>
> I will test using the tcl "source" command, instead of "module source"
> However, I suspect there are subtle differences between the two in certain module modes.
>
> You should update the module man page to clarify that you should use simply "source" in module files.  Perhaps you should update the error message when you use module source in a module file.
>
>
>
> On 10/7/2021 11:39 AM, Paul Markfort wrote:
> >
> > Is it by design, or by error, that the "module source" command can no longer be used in module files (it still works directly).
> >
> > When I upgraded to 5.0.0, I had to change this in my initial module file:
> > (old line is commented out - it worked in  4.8.0).
> >
> >      #       module source  "${_modules_home}/common/base/basealias"
> >      module switch  base/basealias base/basealias
> >
> >
> > In the specific situation (the affected module just defines aliases) - using module switch is better than source (because I don't have to include the full path), but I can think of situations where you don't want a command that unloads the module before loading it again.
> >
> >
> >
> >
> >
> >
>
> --
> --------------------------------------------------------
> The views and opinions expressed above are strictly
> those of the author(s).  The content of this message has
> not been reviewed nor approved by any entity whatsoever.
> --------------------------------------------------------
> Paul FM         Info: http://paulfm.com/~paulfm/
> --------------------------------------------------------
>
>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest

Re: [Modules] module source - changed as of 5.0.0 ?

[Paul M. <pau...@gm...>]

Finally found it in the changes.

https://modules.readthedocs.io/en/latest/NEWS.html

I will test using the tcl "source" command, instead of "module source"
However, I suspect there are subtle differences between the two in certain module modes.

You should update the module man page to clarify that you should use simply "source" in module files.  Perhaps you should update the error message when you use module source in a module file.



On 10/7/2021 11:39 AM, Paul Markfort wrote:
> 
> Is it by design, or by error, that the "module source" command can no longer be used in module files (it still works directly).
> 
> When I upgraded to 5.0.0, I had to change this in my initial module file:
> (old line is commented out - it worked in  4.8.0).
> 
>      #       module source  "${_modules_home}/common/base/basealias"
>      module switch  base/basealias base/basealias
> 
> 
> In the specific situation (the affected module just defines aliases) - using module switch is better than source (because I don't have to include the full path), but I can think of situations where you don't want a command that unloads the module before loading it again.
> 
> 
> 
> 
> 
> 

-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] nested .modulerc

[Thomas M. P. <pa...@um...>]

I haven't use the hack recently, and only used it on older (3.x) versions
of modules, but in my cases just using set and a Tcl var worked.  But also
my situation was only dealing with a single modulerc file as well.

I would suggest trying full-fledged env variables and see if that makes a
difference.

Otherwise, you might want to see if you can create and post a minimalist
example of the issue --- I am only guessing that your issue is the same as
the one I am referring to (I do not even recall the exact error message for
the issue I am referring to, but it was similar to yours), but your issue
might be something completely different.  It is hard to tell from just an
error message.

On Thu, Oct 7, 2021 at 11:53 AM Robbin Bonthond <ro...@bo...> wrote:

> hi,
>
> Thank you for the hint, do I need to use environment variables? it seems
> that a "set x y" isn't global between various .modulerc files.
>
> robbin
>
> On Wed, Oct 6, 2021 at 6:59 PM Thomas M. Payerle <pa...@um...> wrote:
>
>> You might want to look at
>> https://kb.hlrs.de/platforms/index.php/Module_command
>> Toward the end, Global Default section.
>>
>> Basically, for each module version invocation, create an unique env
>> variable and test for the env variable before doing the module version call.
>>
>> It's an ugly hack, but it gets the job done
>>
>>
>> On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...>
>> wrote:
>>
>>> Hello,
>>>
>>> Is there a way to turn off the warning "WARNING: Symbolic version
>>> 'tool/default' already defined"?
>>>
>>> Robbin
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Modules-interest mailing list
>>> Mod...@li...
>>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>>
>>
>>
>> --
>> Tom Payerle
>> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
>> 5825 University Research Park               (301) 405-6135
>> University of Maryland
>> College Park, MD 20740-3831
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
>


-- 
Tom Payerle
DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
5825 University Research Park               (301) 405-6135
University of Maryland
College Park, MD 20740-3831

[Modules] module source - changed as of 5.0.0 ?

[Paul M. <pau...@gm...>]

Is it by design, or by error, that the "module source" command can no longer be used in module files (it still works directly).

When I upgraded to 5.0.0, I had to change this in my initial module file:
(old line is commented out - it worked in  4.8.0).

	#       module source  "${_modules_home}/common/base/basealias"
	module switch  base/basealias base/basealias


In the specific situation (the affected module just defines aliases) - using module switch is better than source (because I don't have to include the full path), but I can think of situations where you don't want a command that unloads the module before loading it again.






-- 
--------------------------------------------------------
The views and opinions expressed above are strictly
those of the author(s).  The content of this message has
not been reviewed nor approved by any entity whatsoever.
--------------------------------------------------------
Paul FM         Info: http://paulfm.com/~paulfm/
--------------------------------------------------------

Re: [Modules] nested .modulerc

[Robbin B. <ro...@bo...>]

hi,

Thank you for the hint, do I need to use environment variables? it seems
that a "set x y" isn't global between various .modulerc files.

robbin

On Wed, Oct 6, 2021 at 6:59 PM Thomas M. Payerle <pa...@um...> wrote:

> You might want to look at
> https://kb.hlrs.de/platforms/index.php/Module_command
> Toward the end, Global Default section.
>
> Basically, for each module version invocation, create an unique env
> variable and test for the env variable before doing the module version call.
>
> It's an ugly hack, but it gets the job done
>
>
> On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...>
> wrote:
>
>> Hello,
>>
>> Is there a way to turn off the warning "WARNING: Symbolic version
>> 'tool/default' already defined"?
>>
>> Robbin
>>
>>
>>
>>
>> _______________________________________________
>> Modules-interest mailing list
>> Mod...@li...
>> https://lists.sourceforge.net/lists/listinfo/modules-interest
>>
>
>
> --
> Tom Payerle
> DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
> 5825 University Research Park               (301) 405-6135
> University of Maryland
> College Park, MD 20740-3831
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
>

Re: [Modules] nested .modulerc

[Thomas M. P. <pa...@um...>]

You might want to look at
https://kb.hlrs.de/platforms/index.php/Module_command
Toward the end, Global Default section.

Basically, for each module version invocation, create an unique env
variable and test for the env variable before doing the module version call.

It's an ugly hack, but it gets the job done


On Wed, Oct 6, 2021 at 4:24 PM Robbin Bonthond <ro...@bo...> wrote:

> Hello,
>
> Is there a way to turn off the warning "WARNING: Symbolic version
> 'tool/default' already defined"?
>
> Robbin
>
>
>
>
> _______________________________________________
> Modules-interest mailing list
> Mod...@li...
> https://lists.sourceforge.net/lists/listinfo/modules-interest
>


-- 
Tom Payerle
DIT-ACIGS/Mid-Atlantic Crossroads        pa...@um...
5825 University Research Park               (301) 405-6135
University of Maryland
College Park, MD 20740-3831

[Modules] nested .modulerc

[Robbin B. <ro...@bo...>]

Hello,

Is there a way to turn off the warning "WARNING: Symbolic version
'tool/default' already defined"?

Robbin

[Modules] Hacktoberfest 2021

[Xavier D. <xav...@gm...>]

Dear All,

Modules project is getting ready for Hacktoberfest 2021:

https://hacktoberfest.digitalocean.com/

For this occasion, some *good first issues* have been created to let
people know how they could help:

https://github.com/cea-hpc/modules/issues?q=is%3Aissue+is%3Aopen+label%3Ahacktoberfest

Regards,
Xavier

[Modules] Log module commands

[Xavier D. <xav...@gm...>]

Hello,

The *Log module commands* cookbook recipe has just been updated to
describe how to log every modulefile evaluations and module/ml command
calls:

https://modules.readthedocs.io/en/latest/cookbook/log-module-commands.html

Instead of superseding the module procedure as done in the previous
version of this recipe, the `trace` Tcl command is now used to monitor
each procedure call.

Regards,
Xavier

[Modules] Modules 5.0.0 released

[Xavier D. <xav...@gm...>]

Hello All,

Modules 5.0 is here! This new major release improves the 'module'
experience by enabling by default all the advanced behaviors developed
over the Modules 4 versions. Changes are also introduced to improve
the consistency of the whole solution.

Details on the major changes included in this new release are provided in
the MIGRATING document. The list of all changes made in this new
version are described in the NEWS document.

 http://modules.readthedocs.io/en/stable/MIGRATING.html
 http://modules.readthedocs.io/en/stable/NEWS.html

The Changes document gives an in-depth view of the modified behaviors
and new features between Modules 4.8 and 5.0 versions.

 http://modules.readthedocs.io/en/stable/changes.html

The tarball of this new version can be downloaded at:

 http://downloads.sourceforge.net/modules/modules-5.0.0.tar.gz

The zipball to install this new version on Windows platform can be
downloaded at:

 http://downloads.sourceforge.net/modules/modules-5.0.0-win.zip

If you encounter any issue, please let us know by creating a ticket on
the project bug tracker at:

 https://github.com/cea-hpc/modules/issues

I would like to thanks all of you who improve Modules by reporting
bugs or sharing new ideas.

Best regards,
Xavier

[Modules] Modules 5.0 is almost there

[Xavier D. <xav...@gm...>]

Hi All,

Modules 5.0 should be out in a few days. The 'changes' doc that
describes the updated behaviors and new features compared to Modules 4
has just been finished:

https://modules.readthedocs.io/en/latest/changes.html#modules-5

All the changes made for this new major version are also listed in the
NEWS document:

https://modules.readthedocs.io/en/latest/NEWS.html#modules-5-0-0-not-yet-released

Now I need to update the MIGRATING (aka "New features") document and
after that everything will be ready for launch.

Regards,
Xavier