[bug#72925] [PATCH v10] gnu: Add jpm.

[Omar Bassam]

Suhail Singh <suhailsingh247@gmail.com> writes:

> Omar Bassam <omar.bassam88@gmail.com> writes:
>
>>>>>>>>> What we need is _some_ mechanism to ensure that when jpm invokes gcc 
>>>>>>>>> (or
>>>>>>>>> g++), the compiler is able to locate the appropriate header files.
>>>>>>>>>
>>>>>>>>> This should be doable without propagating any other inputs.  For 
>>>>>>>>> example
>>>>>>>>> by ensuring that jpm sets appropriate environment variables (such as
>>>>>>>>> $CPATH , $C_INCLUDE_PATH , $CPLUS_INCLUDE_PATH etc.) or flags when
>>>>>>>>> invoking the compiler.  If so, that would be the preferred approach.  
>>>>>>>>> We
>>>>>>>>> only want to propagate those inputs that are strictly necessary.
>>>>>>>>>
>>>>>>>>> ...
>>>>>>>>>
>>>>>>>>> I look forward to seeing what you come up with in v11 :)
>>>
>>> I.e., it's not clear to me that propagating gcc and g++ is necessary.
>>> And if the same can be achieved by passing appropriate environment
>>> variables, why not?  Could you please answer?
>>>
>>> Regardless, we are in agreement that the propagation of gcc-toolchain is
>>> not necessary and should be removed.
>>>
>>
>> I've now removed gcc from the propagated-inputs I've tested passing the
>> gcc to jpm using the --cc flag as follows "jpm build --cc=/path/to/gcc".
>
> Am I understanding correctly that, with v11 in order for the user to be
> able to install a Janet package such as "sh", they have to find the
> location of gcc being used by jpm and pass that to it?
>
> If so, why?  This is not what I tried to explain above.  The jpm package
> "knows" the specific version of gcc (and g++) that have been patched in,
> so why does it require the user to take additional action?  I.e., the
> goal, IMHO, should be that in a pure container that contains nss-certs,
> one is simply able to run "jpm install -l sh" and it works.
>

No, the user doesn't need to specify the flag if the gcc package is
available in their shell (or pure container shell). I was just testing
that you can pass the C flags that you were concerned about from before.
I am still not sure if I fully understood your C paths cocern correctly.

>> I hope I understood your concern correctly this time.
>
> I don't believe so.
>
>> If not, please feel free to add to the patch whatever you think is
>> needed as I'm not a C compiling expert.
>
> Sorry, but I am no C compiling expert either.
>
> In addition, for the past few exchanges, it's not clear that I am able
> to bring value to the discussion.  As evidenced by the recent exchanges,
> I seem to be struggling with expressing myself effectively.  As such, I
> will step-away from actively reviewing this issue.
>

On the contrary, I've learned so much from your comments and I really
appreciate your patience. I believe it's my lack of understanding of
guix packagin philosophy as this is my first time diving into the source
code of Guix.

> I hope you are able to get jpm upstreamed with help from jgart and
> others more experienced than myself.

Thank you, if it ever get upstreamed it will be mostly becuase of your
comments, knowledge sharing and patience that I really appreciate.

Many Thanks,
Omar