Re: Bug#1139559: ITP: python-hypothesis-jsonschema -- hypothesis JSON schema data generator

--0000000000002d6fb00654798426
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Did you try to get in touch and ask them to tag things? Sometimes it works.

Salvo Tomaselli

I difensori della morale tradizionale sono raramente persone di cuore. Si =
=C3=A8
tentati di pensare che essi si servano della morale come di legittimo sfogo
al loro desiderio di fare del male agli altri.
               -- Bertrand Russell, Perch=C3=A9 non sono cristiano. 1957

http://ltworf.github.io/

Il Sab 13 Giu 2026, 06:52 Simon Josefsson <[email protected]> ha scritto:

> Jeroen Ploemen <[email protected]> writes:
>
> > On Tue, 09 Jun 2026 19:02:53 +0200
> > Simon Josefsson <[email protected]> wrote:
> >
> >> Review of this packaging would be appreciated.
> >
> > Most of the upstream testsuite makes use of gen_schemas, which does
> > not appear to have been packaged in Debian. As a result, no tests are
> > run on build and 'Testsuite: autopkgtest-pkg-pybuild' should not be
> > used: all that does under these circumstances is pretend the package
> > comes with an autopkgtest. Try autopkgtest-pkg--python instead.
>
> Thank you for looking at the package!
>
> I didn't know about the various python testsuite options.  I changed to
> use autopkgtest-pkg-python now.  For future reference, I found
> explanation of autopkgtest-pkg-pybuild vs autopkgtest-pkg-python here:
>
>
> https://manpages.debian.org/testing/autodep8/autodep8.1.en.html#EXAMPLES_=
OF_PRODUCED_TEST_CONTROL_FILES
>
> I have uploaded this to NEW queue.
>
> >> I considered using github instead of pypi tarballs, but it seems
> >> upstream doesn't push tags to github so figuring which commits to
> >> pin to and writing the watch file seems complicated...  reported as
> >> https://github.com/python-jsonschema/hypothesis-jsonschema/issues/113
> >
> > You are free to use whichever best suits your needs, there's no team
> > policy that requires or even prefers one over the other. Usually, it
> > boils down to the one that includes tests and documentation.
>
> Okay.  My own preference is now to to use git sources directly.  But for
> this package, understanding which git commit they used for each release
> seems complicated since they don't publish git tags.  And writing a
> debian/watch for this situation seems impossible.  I suspect this isn't
> intentional by upstream, just some artifact of their workflow (the
> homepage URL on pypi.org also seems odd... [1]), so this could all be
> clarified whenever upstream responds.
>
> /Simon
>
> [1] https://github.com/python-jsonschema/hypothesis-jsonschema/issues/114
>

--0000000000002d6fb00654798426
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div>Did you try to get in touch and ask them to tag thin=
gs? Sometimes it works.</div><div><br></div><div data-smartmail=3D"gmail_si=
gnature"><div dir=3D"ltr">Salvo Tomaselli<br><br>






<span style=3D"font-family:monospace"><span style=3D"color:#000000;backgrou=
nd-color:#ffffff">I difensori della morale tradizionale sono raramente pers=
one di cuore. Si =C3=A8</span><span style=3D"color:#000000;background-color=
:#ffffff">
</span><br><span style=3D"color:#000000;background-color:#ffffff">tentati d=
i pensare che essi si servano della morale come di legittimo sfogo</span><s=
pan style=3D"color:#000000;background-color:#ffffff">
</span><br><span style=3D"color:#000000;background-color:#ffffff">al loro d=
esiderio di fare del male agli altri.</span><span style=3D"color:#000000;ba=
ckground-color:#ffffff">
</span><br> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0<span style=3D"color:#000000;background-color:#f=
fffff">-- Bertrand Russell, Perch=C3=A9 non sono cristiano. 1957</span><br>=
<span style=3D"color:#000000;background-color:#ffffff">
</span><br></span>


      <a rel=3D"nofollow me" href=3D"http://ltworf.github.io/" target=3D"_b=
lank">http://ltworf.github.io/</a></div></div></div><br><div class=3D"gmail=
_quote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">Il Sab =
13 Giu 2026, 06:52 Simon Josefsson &lt;<a href=3D"mailto:[email protected]=
g">[email protected]</a>&gt; ha scritto:<br></div><blockquote class=3D"gm=
ail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,=
204,204);padding-left:1ex">Jeroen Ploemen &lt;<a href=3D"mailto:jcfp@debian=
.org" target=3D"_blank" rel=3D"noreferrer">[email protected]</a>&gt; writes:<=
br>
<br>
&gt; On Tue, 09 Jun 2026 19:02:53 +0200<br>
&gt; Simon Josefsson &lt;<a href=3D"mailto:[email protected]" target=3D"_=
blank" rel=3D"noreferrer">[email protected]</a>&gt; wrote:<br>
&gt;<br>
&gt;&gt; Review of this packaging would be appreciated.<br>
&gt;<br>
&gt; Most of the upstream testsuite makes use of gen_schemas, which does<br=
>
&gt; not appear to have been packaged in Debian. As a result, no tests are<=
br>
&gt; run on build and &#39;Testsuite: autopkgtest-pkg-pybuild&#39; should n=
ot be<br>
&gt; used: all that does under these circumstances is pretend the package<b=
r>
&gt; comes with an autopkgtest. Try autopkgtest-pkg--python instead.<br>
<br>
Thank you for looking at the package!<br>
<br>
I didn&#39;t know about the various python testsuite options.=C2=A0 I chang=
ed to<br>
use autopkgtest-pkg-python now.=C2=A0 For future reference, I found<br>
explanation of autopkgtest-pkg-pybuild vs autopkgtest-pkg-python here:<br>
<br>
<a href=3D"https://manpages.debian.org/testing/autodep8/autodep8.1.en.html#=
EXAMPLES_OF_PRODUCED_TEST_CONTROL_FILES" rel=3D"noreferrer noreferrer" targ=
et=3D"_blank">https://manpages.debian.org/testing/autodep8/autodep8.1.en.ht=
ml#EXAMPLES_OF_PRODUCED_TEST_CONTROL_FILES</a><br>
<br>
I have uploaded this to NEW queue.<br>
<br>
&gt;&gt; I considered using github instead of pypi tarballs, but it seems<b=
r>
&gt;&gt; upstream doesn&#39;t push tags to github so figuring which commits=
 to<br>
&gt;&gt; pin to and writing the watch file seems complicated...=C2=A0 repor=
ted as<br>
&gt;&gt; <a href=3D"https://github.com/python-jsonschema/hypothesis-jsonsch=
ema/issues/113" rel=3D"noreferrer noreferrer" target=3D"_blank">https://git=
hub.com/python-jsonschema/hypothesis-jsonschema/issues/113</a><br>
&gt;<br>
&gt; You are free to use whichever best suits your needs, there&#39;s no te=
am<br>
&gt; policy that requires or even prefers one over the other. Usually, it<b=
r>
&gt; boils down to the one that includes tests and documentation.<br>
<br>
Okay.=C2=A0 My own preference is now to to use git sources directly.=C2=A0 =
But for<br>
this package, understanding which git commit they used for each release<br>
seems complicated since they don&#39;t publish git tags.=C2=A0 And writing =
a<br>
debian/watch for this situation seems impossible.=C2=A0 I suspect this isn&=
#39;t<br>
intentional by upstream, just some artifact of their workflow (the<br>
homepage URL on <a href=3D"http://pypi.org" rel=3D"noreferrer noreferrer" t=
arget=3D"_blank">pypi.org</a> also seems odd... [1]), so this could all be<=
br>
clarified whenever upstream responds.<br>
<br>
/Simon<br>
<br>
[1] <a href=3D"https://github.com/python-jsonschema/hypothesis-jsonschema/i=
ssues/114" rel=3D"noreferrer noreferrer" target=3D"_blank">https://github.c=
om/python-jsonschema/hypothesis-jsonschema/issues/114</a><br>
</blockquote></div>

--0000000000002d6fb00654798426--