IDEA: missing-tests-pypi-error? condition

IDEA: missing-tests-pypi-error? condition

[jgart]

Hi Guixers,

I was recently reviewing a patch by a contributor and was discussing the notion of preferring upstream to PyPI when tests are missing on PyPI.

WDYT if we also signalled a special condition when there are missing tests in PyPI?

I'm thinking of something similar to here:

https://git.savannah.gnu.org/cgit/guix.git/tree/guix/import/pypi.scm?h=master#n562

Maybe it can be a missing-tests-pypi-error?

This email is gluten-free, vegan, and 100% organic,

jgart

Re: IDEA: missing-tests-pypi-error? condition

[Munyoki Kilyungi]

[-- Attachment #1: Type: text/plain, Size: 991 bytes --]

"jgart" <jgart@dismail.de> aliandika:

> Hi Guixers,
>
> I was recently reviewing a patch by a contributor and was discussing the notion of preferring upstream to PyPI when tests are missing on PyPI.
>
What's the policy on this.  Shouldn't we prefer
upstream over PyPI?  Sometimes, PyPI versions lag
behind upstream versions.

> WDYT if we also signalled a special condition when there are missing tests in PyPI?
>
I posit that this should be the responsibility of
the contributor.  Adding extra glue can add
unnecessary complexity.

> I'm thinking of something similar to here:
>
> https://git.savannah.gnu.org/cgit/guix.git/tree/guix/import/pypi.scm?h=master#n562
>
> Maybe it can be a missing-tests-pypi-error?
>
> This email is gluten-free, vegan, and 100% organic,
>
> jgart
>

-- 
(Life is like a pencil that will surely run out,
    but will leave the beautiful writing of life.)
(D4F09EB110177E03C28E2FE1F5BBAE1E0392253F
    (hkp://keys.openpgp.org))

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 865 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[Maxim Cournoyer]

Hi,

Munyoki Kilyungi <me@bonfacemunyoki.com> writes:

> "jgart" <jgart@dismail.de> aliandika:
>
>> Hi Guixers,
>>
>> I was recently reviewing a patch by a contributor and was discussing the notion of preferring upstream to PyPI when tests are missing on PyPI.
>>
> What's the policy on this.  Shouldn't we prefer
> upstream over PyPI?  Sometimes, PyPI versions lag
> behind upstream versions.

There's no written policy about it at this time, but I think it's common
knowledge in Guix that we try hard to:

1. Unbundle dependencies (that one is documented)
2. Run test suites (that one isn't)

We could hint at this globally in the contributing section, and perhaps
expound with more specific details in the Python Modules contributing
section, mentioning that when the PyPI source archive lacks tests,
fetching from the git repository should be preferred.

Another thing we could do is default to fetch from git for the Python
importer.

>> WDYT if we also signalled a special condition when there are missing tests in PyPI?
>>
> I posit that this should be the responsibility of
> the contributor.  Adding extra glue can add
> unnecessary complexity.

I think an error would help catching this situation; the
gnu-build-system would fail running 'make check' when there is no check
target for example, but in the Python world, running 'python setup.py
test' often succeeds even 0 tests were run, which is easy to overlook.
I'm not sure how easy it'd be to check for missing tests though.

-- 
Thanks,
Maxim

Re: IDEA: missing-tests-pypi-error? condition

[Munyoki Kilyungi]

[-- Attachment #1: Type: text/plain, Size: 2129 bytes --]

Hi Maxim!

Maxim Cournoyer <maxim.cournoyer@gmail.com>
aliandika:

> Hi,
>
> Munyoki Kilyungi <me@bonfacemunyoki.com> writes:
>
>> "jgart" <jgart@dismail.de> aliandika:
>>
>>> Hi Guixers,
>>>
>>> I was recently reviewing a patch by a contributor and was discussing the notion of preferring upstream to PyPI when tests are missing on PyPI.
>>>
>> What's the policy on this.  Shouldn't we prefer
>> upstream over PyPI?  Sometimes, PyPI versions lag
>> behind upstream versions.
>
> There's no written policy about it at this time, but I think it's common
> knowledge in Guix that we try hard to:
>
> 1. Unbundle dependencies (that one is documented)
> 2. Run test suites (that one isn't)
>
> We could hint at this globally in the contributing section, and perhaps
> expound with more specific details in the Python Modules contributing
> section, mentioning that when the PyPI source archive lacks tests,
> fetching from the git repository should be preferred.
>
> Another thing we could do is default to fetch from git for the Python
> importer.
>
Thanks for this info.  I'm dedicating an hour of
my day from work to try to contrib to the
eco-system with Jgarte's help.  What's, if I may
naively ask, the process for documenting this
knowledge?  I can take a stub at trying to
document this and asking for reviews.


>>> WDYT if we also signalled a special condition when there are missing tests in PyPI?
>>>
>> I posit that this should be the responsibility of
>> the contributor.  Adding extra glue can add
>> unnecessary complexity.
>
> I think an error would help catching this situation; the
> gnu-build-system would fail running 'make check' when there is no check
> target for example, but in the Python world, running 'python setup.py
> test' often succeeds even 0 tests were run, which is easy to overlook.
> I'm not sure how easy it'd be to check for missing tests though.
>

TIL'ed.
-- 
(Life is like a pencil that will surely run out,
    but will leave the beautiful writing of life.)
(D4F09EB110177E03C28E2FE1F5BBAE1E0392253F
    (hkp://keys.openpgp.org))

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 865 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[Maxim Cournoyer]

Hi,

Munyoki Kilyungi <me@bonfacemunyoki.com> writes:

> Hi Maxim!
>
> Maxim Cournoyer <maxim.cournoyer@gmail.com>
> aliandika:
>
>> Hi,
>>
>> Munyoki Kilyungi <me@bonfacemunyoki.com> writes:
>>
>>> "jgart" <jgart@dismail.de> aliandika:
>>>
>>>> Hi Guixers,
>>>>
>>>> I was recently reviewing a patch by a contributor and was discussing the notion of preferring upstream to PyPI when tests are missing on PyPI.
>>>>
>>> What's the policy on this.  Shouldn't we prefer
>>> upstream over PyPI?  Sometimes, PyPI versions lag
>>> behind upstream versions.
>>
>> There's no written policy about it at this time, but I think it's common
>> knowledge in Guix that we try hard to:
>>
>> 1. Unbundle dependencies (that one is documented)
>> 2. Run test suites (that one isn't)
>>
>> We could hint at this globally in the contributing section, and perhaps
>> expound with more specific details in the Python Modules contributing
>> section, mentioning that when the PyPI source archive lacks tests,
>> fetching from the git repository should be preferred.
>>
>> Another thing we could do is default to fetch from git for the Python
>> importer.
>>
> Thanks for this info.  I'm dedicating an hour of
> my day from work to try to contrib to the
> eco-system with Jgarte's help.  What's, if I may
> naively ask, the process for documenting this
> knowledge?  I can take a stub at trying to
> document this and asking for reviews.

That's a pretty cool workplace you've got :-).

The process would be the same as for authoring any other changes to
Guix.  You'll want to 'make doc/guix.info' to test your changes (or just
'make'), and you can review the new work with 'info doc/guix.info'.

'info Texinfo' can be used as a handy reference for the Texinfo
specifics.

-- 
Thanks,
Maxim

Re: IDEA: missing-tests-pypi-error? condition

[jgart]

> 'make'), and you can review the new work with 'info doc/guix.info'.

Hi,

Is there a way to open the guix manual directly?

Not sure why info opens a "main page" with all my texinfo manuals when I give it doc/guix.info as an argument... 🦆

https://up.nixnet.services/89q7q1f9.png

https://up.nixnet.services/k55f0n8d.png

Checking the manual with man texinfo doesn't seem very helpful:

Doing info texinfo is TMI:

https://up.nixnet.services/7tzldvyw.png

Doing man texinfo is TLI:

https://up.nixnet.services/nvk70ry0.png

off to learn something new (or relearn something old),

jgart

Re: IDEA: missing-tests-pypi-error? condition

[Luis Felipe]

[-- Attachment #1.1.1: Type: text/plain, Size: 811 bytes --]

Hi,

El 8/10/23 a las 13:11, jgart escribió:
> Doing info texinfo is TMI:
>
> https://up.nixnet.services/7tzldvyw.png

According to the Guix manual 
(https://guix.gnu.org/es/manual/devel/en/html_node/Writing-Documentation.html), 
people is encouraged to contribute documentation in any format.

So how about filing an issue (with CC to guix-devel) for this particular 
case and propose changes or additions in plain prose? Maybe someone else 
can help format the result and provide a patch.

Also, for what it's worth, I have a project whose documentation is kinda 
like a summary of Texinfo elements (commands). Maybe it is between TMI 
and TLI?

   Source: https://codeberg.org/luis-felipe/texinfo-css
   HTML: https://luis-felipe.gitlab.io/texinfo-css/

See the Elements section.


[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 2881 bytes --]

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[Maxim Cournoyer]

Hi,

"jgart" <jgart@dismail.de> writes:

>> 'make'), and you can review the new work with 'info doc/guix.info'.
>
> Hi,
>
> Is there a way to open the guix manual directly?

Did you try it?  It works for me.  Or just 'info guix' to open the Guix
manual from found from INFOPATH.

-- 
Thanks,
Maxim

Re: IDEA: missing-tests-pypi-error? condition

[Josselin Poiret]

[-- Attachment #1: Type: text/plain, Size: 260 bytes --]

Hi jgart, 

> Is there a way to open the guix manual directly?
>
> Not sure why info opens a "main page" with all my texinfo manuals when
> I give it doc/guix.info as an argument... 🦆

You want `info -f doc/guix.info`.

HTH,
-- 
Josselin Poiret

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 682 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[jgart]

> You want `info -f doc/guix.info`.

Ah yes, I actually used that flag once before and had forgotten it...

Thanks!

Re: IDEA: missing-tests-pypi-error? condition

[Maxim Cournoyer]

Hi,

"jgart" <jgart@dismail.de> writes:

>> You want `info -f doc/guix.info`.
>
> Ah yes, I actually used that flag once before and had forgotten it...

FWIW, it works for me even without '-f'.

-- 
Thanks,
Maxim

Re: IDEA: missing-tests-pypi-error? condition

[Efraim Flashner]

[-- Attachment #1: Type: text/plain, Size: 617 bytes --]

On Mon, Oct 09, 2023 at 01:17:10PM -0400, Maxim Cournoyer wrote:
> Hi,
> 
> "jgart" <jgart@dismail.de> writes:
> 
> >> You want `info -f doc/guix.info`.
> >
> > Ah yes, I actually used that flag once before and had forgotten it...
> 
> FWIW, it works for me even without '-f'.

To add in some more similar and useful knowledge, you can do the same
thing with man, `man /path/to/man/page.gz`

-- 
Efraim Flashner   <efraim@flashner.co.il>   רנשלפ םירפא
GPG key = A28B F40C 3E55 1372 662D  14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[jgart]

> Sometimes, PyPI versions lag behind upstream versions.

Hi Bonz,

Yes, in my experience that has sometimes been the case because maintainers don't make timely releases to PyPI.

> What's the policy on this. Shouldn't we prefer upstream over PyPI? 

Our importer works with the JSON API that PyPI provides at https://pypi.org/pypi/baz/json

See here:

https://git.savannah.gnu.org/cgit/guix.git/tree/guix/import/pypi.scm?h=master#n141

That API provides a lot of package data that GitHub does not track in a unified and structured way.

> Adding extra glue can add unnecessary complexity.

You're right. I thought about that after I sent the email with the idea ;()

Doing my best to make inboxes a happier place,

jgart

Re: IDEA: missing-tests-pypi-error? condition

[Munyoki Kilyungi]

[-- Attachment #1: Type: text/plain, Size: 1025 bytes --]

"jgart" <jgart@dismail.de> aliandika:

>> Sometimes, PyPI versions lag behind upstream versions.
>
> Hi Bonz,
>
> Yes, in my experience that has sometimes been the case because maintainers don't make timely releases to PyPI.
>
>> What's the policy on this. Shouldn't we prefer upstream over PyPI? 
>
> Our importer works with the JSON API that PyPI provides at https://pypi.org/pypi/baz/json
>
> See here:
>
> https://git.savannah.gnu.org/cgit/guix.git/tree/guix/import/pypi.scm?h=master#n141
>
> That API provides a lot of package data that GitHub does not track in a unified and structured way.
>
Then given the above ...
>> Adding extra glue can add unnecessary complexity.
>
> You're right. I thought about that after I sent the email with the idea ;()
>

PyPi unless we need to use upstream is the way to
go ;)

-- 
(Life is like a pencil that will surely run out,
    but will leave the beautiful writing of life.)
(D4F09EB110177E03C28E2FE1F5BBAE1E0392253F
    (hkp://keys.openpgp.org))

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 865 bytes --]

Re: IDEA: missing-tests-pypi-error? condition

[Maxim Cournoyer]

Hi,

Munyoki Kilyungi <me@bonfacemunyoki.com> writes:

> "jgart" <jgart@dismail.de> aliandika:
>
>>> Sometimes, PyPI versions lag behind upstream versions.
>>
>> Hi Bonz,
>>
>> Yes, in my experience that has sometimes been the case because maintainers don't make timely releases to PyPI.
>>
>>> What's the policy on this. Shouldn't we prefer upstream over PyPI? 
>>
>> Our importer works with the JSON API that PyPI provides at https://pypi.org/pypi/baz/json
>>
>> See here:
>>
>> https://git.savannah.gnu.org/cgit/guix.git/tree/guix/import/pypi.scm?h=master#n141
>>
>> That API provides a lot of package data that GitHub does not track in a unified and structured way.
>>
> Then given the above ...
>>> Adding extra glue can add unnecessary complexity.
>>
>> You're right. I thought about that after I sent the email with the idea ;()
>>
>
> PyPi unless we need to use upstream is the way to
> go ;)

Yes, I was thinking continuing to use PyPI, but attempt to find the
repository URL from PyPI's metadata and used that when generating a
package definition with the pypi importer.

I don't know how feasible that would be.

-- 
Thanks,
Maxim